BRADLEY SMOKER | "Taste the Great Outdoors"

Miscellaneous Topics => General Discussions => Topic started by: iceman on February 04, 2008, 09:36:53 AM

Title: Need everyones help on a PID problem
Post by: iceman on February 04, 2008, 09:36:53 AM
Wildcat just got the PID I built him and needs help on how to run the auto tune. I'm out of town for a few weeks and don't have the manual so I can't give him directions. Can you guys help him out for me so he doesn't have to wait until I get home?
Thanks folks  ;) :)
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 09:44:32 AM
Not a problem Ice!

I'm sure between everybody we should be able to help him out. One question though is it an Auber PID?

Mike
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 04, 2008, 09:55:26 AM
Quote from: Mr Walleye on February 04, 2008, 09:44:32 AM
Not a problem Ice!

I'm sure between everybody we should be able to help him out. One question though is it an Auber PID?

Mike

Yep. It's the 1/16 DIN model SYL-2352 with a ThermoWorks oven probe THS-113-173
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 10:28:29 AM
Hope this makes sense.  Press and hold the "Set" button for about 2 seconds or until the display changes.  Press a few more times until the "At" field shows up in the upper display, and it should have a "3" in the lower field...change the "3" to a "2" using the down arrow key.  Should start autotuning when it goes back into operation in about 8 seconds.  When autotune is complete, it will automatically go back to the "3" setting, and remember your P, I, and D settings.  You can then check them by going back into the setting mode and going down to the respective readings, although it will have the "I" first.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 11:59:38 AM
I really appreciate any help I can get.

Do I need to be smoking/cooking something when I do this?

How long will the process take?

How do I find out the PID settings?
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 12:15:01 PM
More questions - should vent be open?
I always cook with the generator on.  Should this be on during this auto tune?
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 12:18:07 PM
It's best to have a load in there when you do it...to find out the settings, hold the set button again until you get into the menu, then scroll through using the down arrow...the I setting will be the first one that shows, then the P, then the D...they'll be in line right after the Autotune part.

I think it takes about an hour to get done, I didn't watch mine, but when I checked on it after an hour or so, it was done.

And, since you posted again before I got this posted...have it set up the way you do for most of your smokes.  If you keep the smoke gen on, leave it on.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 12:21:48 PM
Thanks TT.  I will see if I can scare up some meat and will get back after I give it a try.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 12:28:37 PM
WildCat

I don't have my manual here at work but Tiny appears to have the procedure down. Don't be scared to familiarize yourself with the buttons and the menu structure on the unit. It really is easy once you go through it once. There is one more setting I changed but I will have to get back to you on it because I don't have that information here at work.

It can take up to an hour for the Autotune to run. You will know when it's complete because it will return to its normal view (no longer flashing). I would recommend having some kind of a load in it to simulate the typical load you would smoke. You should bring the unit up to within 10 degrees of the set temp, then engage the AutoTune and let her rip. Also, if you usually run the smoke generator all the time make sure you keep it on as this will have an effect on it.

Like I said in another post I ran Autotune several times. The first few times without much of a load in it. The final time I ran the Autotune I had a butt in. I engaged the Autotune after it had been in for a couple of hours (still during smoke so the generator was on). You can run the Autotune as many times as you want so don't be concerned you are going to hurt anything. The only thing I would do if you are going to run multiple Autotunes is write down the current numbers along with the setup info during the Autotune.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 12:43:10 PM
Actually, Mike, I cheated...I still have the pdf of the manual you e-mailed me before I even had mine in hand.  Just got done printing another copy to have on my desk, so the questions are probably done. lol
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 12:50:10 PM
Thanks Mike.

TT I went and picked up some more chicken.  The BS is heating as I type this.  I followed your instructions but could not find anything that said "At".  I kept pushing the buttons until the normal display came up.  I tried this twice and could not find the "At".
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 01:02:34 PM
Shoot...maybe I'm not as smart as I thought I was...and the manual ain't as clear as it could be.  Maybe get into the menu by holding the set button, then use the down arrow.........on second thought, I'm gonna check with Dad to see if I can run home quick (there's something else I want to do too), so I'll be back to you yet this afternoon...hopefully within the next half hour or so.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 01:07:29 PM
The PID probe is currently set at 220.  The PID indicater is bouncing around between 217 and 227.  By bouncing around, I mean it is showing multiple numbers within seconds of each other.  I assume it is doing this because it needs to be auto tuned.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 01:31:12 PM
  Wildcat, maybe one of the others can shed more light on the numbers bouncing since I have never used an Auber temp controller. But I do work with industrial temp controllers just about every day and when I see numbers bouncing faster than the furnace could possibly be reacting to, it is normally a loose connection somewhere between the controller and the thermocouple unless the T/C is failing but your is brand new so I would rule that out.
  Go ahead and do your autotune to see if it helps but if not I would begin by checking that the T/C wires are tight. Hopefully someone here that uses an Auber will have more info.
  When running an autotune you do not need to actually cook something, you can simulate a load with bricks or something like that. You do want the smoker to be as close to a normal operation as possible when running the autotune.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 01:34:10 PM
WildCat

Do you have the manual there with you?

If you do find where it lists the menu structure and locate the autotune sybol. It really doesn't resemble an "At" because of the limitations of the numeric screen. I know the first couple of times I scanned through the menu I missed it too.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 01:38:31 PM
Thanks pensrock.  At the moment I am trying to figure out how to do the auto tune.

Mike - Pat did send me some kind of literature on this thing.  I will get it out an look again.  When I looked at it when it arrived it was all Greek to me.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 01:45:32 PM
WildCat

On one of the pages it does show a chart of some sort on what you will see as you step through the menu. Have a look at it and you should find where you will find the At function. I just had a look at the Auber site but they don't have the manual in .pdf format. A year or 2 ago I did get one from them (that's the one I sent Tiny) but for some reason I don't have it here at the office.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 01:49:45 PM
I'm back.  All I had to do was press and hold the set button, then the display changed.  To progress through the menu, I did a "press and release" of the "Set" button..first up was "ALM 1", then "ALM 2", then "Hy 1", then "Hy 2", then "Hy", and finally "At" (the "T" does look a bit funny, kinda like an inverted "F").

If this don't work, call me on my cell tonight after 5:45 Central time, PMing my cell # to you as soon as this posts.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 01:52:42 PM
I think Tiny has it.

The only thing I would add is not to start the autotune until the temp (with load) has reached within 10 degrees of your target temp.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 01:57:35 PM
Thanks TT and Mike.  After reading Mike's post, I looked in the manual and saw what the "At" looked like.  I then followed TT's instructions and it is now auto tuning (I think).  The lower (green) numbers are flashing between my set temp and the "At" symbol.  I only hope this is all it needs.  I still have a little concern about displaying multiple temp readings within seconds that vary by about 10 degrees.  Hopefully this auto tune will resolve the problem.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 02:04:01 PM
Yup, I would let it run through the autotune, then see how it works.

Keep us posted.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 02:06:27 PM
Looks like ya got it under control.  Give me a call if you need something later.
Title: Re: Need everyones help on a PID problem
Post by: MRH on February 04, 2008, 02:11:19 PM
Wildcat,

Here is the directions right from the manual hope they help:

6.2.3  Fuzzy Logic enhanced PID control and autotuning

For the first time user, it is a good idea to use the auto tuning feature to let the controller automatically determine the PID constants. To start auto tuning, you need to set At=2 (see 6.5.3 for details).  During auto tuning, the instrument executes on-off control.  After 2-3 times on-off action, the microprocessor in the instrument will anelyze the period, amplitude, waveform of the oscillation generated by the on-off control, and calculate the optimal control parameter value.  The instrument begins to perform accurate artificial intelligence control after auto-tuning is finished.  If you want to exit from auto-tuning mode, press and hold the (A/M) key for about 2 seconds until the blinking of "At" symbol is stopped in the lower display window.  Generally, it will meet your need to perform auto tuning one time only.  After the auto tuning is finished, the instrument will set parameter "At" to 3, which will disable triggering of auto-tuning with the (A/M) key.  This will prevent an accidental repeat of the auto-tuning process.

6.5.3 Control mode parameter "At"

At=0 ON OFF control, suitable for the applications which don't need high precision
AT=1 Get  the controller ready to start the Auto tunning by pressing A/M key.
At=2  Start auto tuning. The function is the same as starting auto tuning from front panel.
At=3 This configuration is automatically set after auto tuning is done.  Auto tuning from the fron panel is inhibited to prevent accidental re-starting of the auto tuning process. to start auto tuning again, set At=1 or At=2.


Thanks to my wife for typing, you would be done before I got it typed!!
Mark


Looks like it took me too long.... Oh well here it is anyway ;D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 02:26:59 PM
Thanks MRH.

OK guys - here are the results.  I ran auto tune 3 times.  First set P=12 I=126 D=4.  Forgot to write down second run results.  Third run P=196 I=9500 D=107.

PID temp probe reading 214 to 222 (bouncing around these numbers within seconds) and within a few minutes 218 to 226.  Many other sets every few minutes.  I have the PID set for 220.

Chicken (skinless) currently at 130 and Maverick reading a box temp of 198.  Door gauge on BS reading 198 to 200.

What next?
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 02:33:36 PM
WildCat

Where abouts in the cabinet do you have the TC mounted?

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 02:39:43 PM
Mike, you may be onto something there. Some T/C's cannot touch ground or it will cause interferance. I would move the T/C or rest it on something that is not metal. Or even just hold it in your hand to see if it settles down.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 02:40:37 PM
WildCat

When you say bouncing between 214 & 222 do you mean you would see 214, then in a few seconds it changes to 222?

Or... do you see 222 and it slowly climbs to 222?

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 02:45:53 PM
It also seems odd that he was able to complete three autotunes in such a short period of time.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 02:52:43 PM
When I went to check the PID was reading 227 to 236 and the Maverick indicated 208.  It bounces around within seconds.  Did another auto tune and the numbers are: P=20 I=593 D=7.  The sets of numbers creep up and down but there is always a range of about 9 degrees that bounce around.

Today I have the Temp probe mounted with a metal holder that the Maverick set had.  The other day I was having the same problem with numbers bouncing and I had it set through a small chuck of ham to hold it off of metal.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 02:56:28 PM
Forgot to mention - by TC I assume you mean the temp probe from the PID that goes into the cabinet.  I have it on the third shelf below the chicken where the juices cannot drip on it.  I have the Maverick probe in the same general area.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 02:58:49 PM
Quote from: pensrock on February 04, 2008, 02:45:53 PM
It also seems odd that he was able to complete three autotunes in such a short period of time.

WildCat

To expand on  what Pensrock said... How long does it take to run the auto tune?

Something doesn't add up....

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 02:59:48 PM
If you pick the probe up so it is not touching anything inside the smoker does it still keep bouncing? Did you check to make sure the wires are tight on the back of the controller?
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 03:02:26 PM
It takes about 2 to 5 minutes for the autotune.  I will go change the temp probe so that it will just hang and will get back in a few.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 03:02:31 PM
Sorry WildCat... I have to head home for supper. I will be back shortly & see how you are making out.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 03:04:47 PM
Quote from: Wildcat on February 04, 2008, 03:02:26 PM
It takes about 2 to 5 minutes for the autotune.  I will go change the temp probe so that it will just hang and will get back in a few.

Wow

Non of the autotunes I have ran were that quick! I think the fastest was about 30 minutes. Did you pay attention to the display temps as it was autotuning? It should overshoot then undershoot about 3 times.

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 03:05:02 PM
I have never used an Auber but most autotunes take anywhere from 1-3 hours. I suspect the fact that the temp is jumping around in the instrument that it is getting kicked out of autotune.
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 04, 2008, 03:08:35 PM
I'm assuming your chicken is pretty close to done by now...just for kicks, when it is, take a couple of bricks or something and put them in the position you usually put the meat in...start autotune, then leave it alone for an hour or so, then see what it's doing.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 03:26:30 PM
Moved the probe and is simply hanging above the meat.  Chicken at 135.  Probe temp reading climbed steadly to 225.  No jumping.  Of course it did this before then started jumping.  This time it slowly went down to 211 and is now slowly climbing.  No jumping.  One problem fixed.  I think I should try another auto tune and see if it works this time.  What do you guys think?
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 03:28:23 PM
Yes, I think the autotune was shutting off due to the bouncing you were seeing on the controller.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 03:57:10 PM
Thanks to all of you.  Been running the auto tune for about 25 minutes now and it is still tuning.  Wish I knew what was causing the bouncing before.  It was not touching metal the other day.  Seems to be working now though.  I also moved my Maverick probe to be near the PID probe.  There is about an 11 degree difference between the two.  Maverick showing a lower temp.  May need to get a more accurate PID probe in the future if this thing calibrates itself alright.  In the mean time, I will just set the PID about 10 degrees higher than I want to cook at.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 04:07:24 PM
Talk to iceman when he get back home. I think he uses these type of probes, see how he mounts it. It's possible you do have a defective probe, maybe the wires inside the insulation are bad? Its hard to tell without actually being there.
Good luck, pensrock.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 04:10:01 PM
Also, your PID controller should be more acurrate than the Maverick. Most digital controllers are +- 5 degrees F. Maybe the probe is the correct reading and the Maverick is off a few degrees? Just a thought.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 04:17:05 PM
I checked the Maverick today in boiling water.  The cabinet probe went between 209 and 211 and the meat probe between 210 and 213.  I think the boiling point of water is 212.

The auto tune went for 25 minutes.  The temp swings are now from 216 to 222.  The Maverick is roughly 10 degrees lower.  May have to do another auto tune.  Any other recommendations for a smaller temp swing?

The chicken has been in there for 3 hours now and is only registering a temp of 135.  I seems strange that it is still at that temp.  It is skinless and boneless breasts and vent is a little over half open.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 05:25:05 PM
Ran auto tune one last time for tonight.  Took another 25 minutes.  Numbers are P=544 I=161 D=166.  Let it run at 220 for awhile and had a temp swing of 219 to 221.  Looks like goal is met.  Maverick temp reading 9 to 11 degrees less.  I just bumped the PID setting to 230 so that I can get the chicken done.  Will probably be dry.  If so, the dog has some treats to look forward to.

Thanks for all your help guys.  It appears that now all I need to do is get another temp probe for the PID.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 04, 2008, 05:55:29 PM
Your within two-three degrees now. It may get even better with another tuning but I would write down the numbers before changing anything. Some controllers will allow you to enter an offset to make the controller match what the maverick reads, I'm not familiar with the Auber so I do not know if you can offset it. I'm happy it looks like it is working now for you. The dog is probably happy too!
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 06:02:22 PM
I'm perfectly content with a +/- 1 degree swing off of set temp.  The only thing I need now is an accurate probe.  The Maverick is only off by no more than 3 degrees below actual temp IMHO and it reflects 10 to 11 degrees lower than the PID.  I will get some input from Ice later regarding that.  Thanks a lot for all of your input and to the rest of you guys that helped as well.  Sorry about you taking off of work TT.  That was beyond the call of duty.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 06:03:17 PM
WildCat

I said earlier that there was one other setting that I had changed and I'm home now so here it is.

It's called the "Cycle Time". The display shows the "t" for Cycle Time. You can set this to "0". This number represents the speed at which the output is turned on and off. Setting the Cycle Time lower increases the performance and accuracy. It does have to be hooked to an SSR and not a mechanical relay to do this (ours, meaning yours and mine have a SSR so we can increase the Cycle Time).

Mike
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 04, 2008, 06:16:30 PM
Quote from: pensrock on February 04, 2008, 05:55:29 PM
Your within two-three degrees now. It may get even better with another tuning but I would write down the numbers before changing anything. Some controllers will allow you to enter an offset to make the controller match what the maverick reads, I'm not familiar with the Auber so I do not know if you can offset it. I'm happy it looks like it is working now for you. The dog is probably happy too!

Auber does have an offset setting. It's setting in the menu is "Pb" in the menu and it will allow both positive and negative numbers. Although the directions say this is rarely adjusted. Certainly if your PID maintains a temperature but is consistently out by the same number of degrees, by all means adjust it.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 04, 2008, 06:31:42 PM
Thanks Mike.  I will give both of these a try before Pat gets back home.  I am slowly begining to understand how to run this electronic gadget.  As I have stated in the past, I am a complete idiot when it comes to electronics.  Even if I can not improve on what I already have, it is much better than the slider control.  Much less monitoring.  Will not let the wife know this though.  ;D
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 04, 2008, 06:35:59 PM
Quote from: Wildcat on February 04, 2008, 06:31:42 PM
Even if I can not improve on what I already have, it is much better than the slider control.  Much less monitoring.  Will not let the wife know this though.  ;D

This is approaching classified information. Any notion that you will not be exerting yourself (and passing by the adult beverage refrigerator) while constantly tending the beast must be tightly held.   :o
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 05, 2008, 07:32:37 AM
Quote from: Wildcat on February 04, 2008, 06:02:22 PM
I'm perfectly content with a +/- 1 degree swing off of set temp.  The only thing I need now is an accurate probe.  The Maverick is only off by no more than 3 degrees below actual temp IMHO and it reflects 10 to 11 degrees lower than the PID.  I will get some input from Ice later regarding that.  Thanks a lot for all of your input and to the rest of you guys that helped as well.  Sorry about you taking off of work TT.  That was beyond the call of duty.

No problem, bud...like I said, I had something else to take care of at home anyway.  Just happy to see that you've got it figured out now.
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 05, 2008, 02:06:42 PM
Quote from: Arcs_n_Sparks on February 04, 2008, 06:35:59 PM
Quote from: Wildcat on February 04, 2008, 06:31:42 PM
Even if I can not improve on what I already have, it is much better than the slider control.  Much less monitoring.  Will not let the wife know this though.  ;D

This is approaching classified information. Any notion that you will not be exerting yourself (and passing by the adult beverage refrigerator) while constantly tending the beast must be tightly held.   :o

Arcs, I for one will be monitoring (and drinking) from a distance to keep security tight on this subject. We must stand together and keep the ease of the PID a secret.
Ya hear tha La Quinta? No spilling the beans to Ann. LOL  :D ;)
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on February 05, 2008, 03:36:48 PM
Quote from: iceman on February 05, 2008, 02:06:42 PM
Quote from: Arcs_n_Sparks on February 04, 2008, 06:35:59 PM
Quote from: Wildcat on February 04, 2008, 06:31:42 PM
Even if I can not improve on what I already have, it is much better than the slider control.  Much less monitoring.  Will not let the wife know this though.  ;D

This is approaching classified information. Any notion that you will not be exerting yourself (and passing by the adult beverage refrigerator) while constantly tending the beast must be tightly held.   :o

Arcs, I for one will be monitoring (and drinking) from a distance to keep security tight on this subject. We must stand together and keep the ease of the PID a secret.
Ya hear tha La Quinta? No spilling the beans to Ann. LOL  :D ;)

Spilling the beans shall result in a loss of carrots.........wait a minute..that could be very costly for whomever tries to repossess the carrots......forget I said anything.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 05, 2008, 04:18:54 PM
 ;D ;D ;D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 04:19:14 PM
Quote from: Mr Walleye on February 04, 2008, 06:03:17 PM
WildCat

I said earlier that there was one other setting that I had changed and I'm home now so here it is.

It's called the "Cycle Time". The display shows the "t" for Cycle Time. You can set this to "0". This number represents the speed at which the output is turned on and off. Setting the Cycle Time lower increases the performance and accuracy. It does have to be hooked to an SSR and not a mechanical relay to do this (ours, meaning yours and mine have a SSR so we can increase the Cycle Time).

Mike

I just did this.  It was set at 12 and it is now on 0.  I will monitor and see if it makes any difference.

Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 04:24:41 PM


Auber does have an offset setting. It's setting in the menu is "Pb" in the menu and it will allow both positive and negative numbers. Although the directions say this is rarely adjusted. Certainly if your PID maintains a temperature but is consistently out by the same number of degrees, by all means adjust it.

Mike
[/quote]

Mike - I tried this but can't tell if it made any difference since neither the set temp nor the actual temp numbers changed.  It was set at 0.0 and then I set it at -12.  I put it back at the original setting until I got more input from you.

At the present time, in order to maintain a 205 F cabinet temp, I have to set the PID temp at 217.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 05:37:24 PM
WildCat

I have never played with the temperature offset before but from what I read in the manual I would think that is correct to use a setting of -12 based on your numbers. When you had it set that way did you see your temps change?

Is your PID holding temps without the swings you were encountering earlier?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 05:48:20 PM
Quote from: Mr Walleye on February 07, 2008, 05:37:24 PM
WildCat

I have never played with the temperature offset before but from what I read in the manual I would think that is correct to use a setting of -12 based on your numbers. When you had it set that way did you see your temps change?

Is your PID holding temps without the swings you were encountering earlier?

Mike

At -12 the read out set and probed temp remained the same.  I did not let it stay there since it did not change anything until I talked with you.

After fooling with everything, my temp swings went to +/- 2 degrees.  Currently doing another auto tune to try and get back to the +/- 1 degree.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 06:00:13 PM
Quote from: Wildcat on February 07, 2008, 04:19:14 PM
Quote from: Mr Walleye on February 04, 2008, 06:03:17 PM
WildCat

I said earlier that there was one other setting that I had changed and I'm home now so here it is.

It's called the "Cycle Time". The display shows the "t" for Cycle Time. You can set this to "0". This number represents the speed at which the output is turned on and off. Setting the Cycle Time lower increases the performance and accuracy. It does have to be hooked to an SSR and not a mechanical relay to do this (ours, meaning yours and mine have a SSR so we can increase the Cycle Time).

Mike

I just did this.  It was set at 12 and it is now on 0.  I will monitor and see if it makes any difference.



After another auto tune, my temp read 169 vice the 214 I have it set at.  Changed it back to 12 and now I fluctuate +/- 2 degrees from the 214 F.  May even do another autotune.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 07, 2008, 06:02:51 PM
Wildcat,

Let me know what your final PID numbers are, and I will post with the rest.

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 06:07:23 PM
Will do Sparky
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 06:12:37 PM
WildCat

Are you checking your temps with your ET73?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 06:20:01 PM
Only so that I know my true box temp and meat temp.  The information on what I have been referring to are from the PID only.  I am trying to maintain a box temp of 205 on my ET73, which at the moment requires a setting of 214 on the PID.

Also, my temp swings went to +/- 5 degrees from the 214 setting, so I am currently running another auto tune.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 07:00:17 PM
Auto tunes have only been taking about 25 minutes.  After this latest, temps swung -4 to +3 around the 214 set up.  After a few minutes it stablized at 214 to 215 box temp with it set at 214.  After about another 15 minutes it went down to 212 and slowly went back to the 214 to 215.  Will continue to monitor.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 07, 2008, 07:03:02 PM
Quote from: Wildcat on February 07, 2008, 07:00:17 PM
Will continue to monitor.

All this monitoring is making me thirsty...

I am starting to drink, at a distance.   :D :D :D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 07:05:18 PM
Yea.  I think I will make myself a crown and sprite and join ya.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 07:11:08 PM
WildCat

I'm just trying to get my brain around your numbers.

The PID set temp is 214. Currently at 214 to 215.

ET73 is reading 205.

I'm wondering where your thermocouplers are for both the PID and the ET73?


Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 07:28:13 PM
PID set for 215.  PID now reflecting box temp between 213 and 215.  Fluctuation is less now.  May change again later.  We shall see.  The ET-73 fluctuates along with the PID readings, but at temps of 199 to 203 now.  Both are right next to each other.  They are simply hanging above the meat without touching.  Not my favorite place to put them, but it is the only location where the loooong PID probe will not touch metal.  That is why I am cooking at 200 to 205 ET-73 box temp.  I figure the bottom of the meat is around 207 to 210.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 07:48:50 PM
Quote from: Arcs_n_Sparks on February 07, 2008, 06:02:51 PM
Wildcat,

Let me know what your final PID numbers are, and I will post with the rest.

Arcs_n_Sparks

Sparky - Since I seem to be back at the +/- 1 degree, I do not think I will auto tune anymore tonight.  My numbers are P=498 I=191 and D=172.

I do not think much of this particular PID probe (thermocouple).  Will probably get a different one and see what I have at that point.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 07:49:30 PM
I wonder if you try changing the offset again to -12 and give it 15 minutes or so and see what all your temp readings are.

Maybe before you try this you should record your current P, I & D settings just in case. That way you can always manualy change them back if it doesn't work.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 07:52:34 PM
OK - I'll give it a try now.  One question - Do I need to auto tune after I do this?
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 07:55:28 PM
No, you shouldn't have to. It's just adjusting the display numbers to my knowledge.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 08:05:34 PM
 ;D ;D ;D  I went and did the change.  I realized what I did wrong last time I tried to do this.  I had set it at -1.2 vice -12.0.  Duh!  Now they should read the same.  I will fine tune that setting after awhile.  Still do not like this PID probe.  It is so long that I have no place to put it other that hanging above the meat.  If I go below the meat, it hits the V tray.  The meat is on the second position from the top.  If the PID continues to read the same as the ET-73 then I will probably move the ET-73 to just under the meat and off to the side so that I can get a good reading.

Thanks again for your help.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 08:15:27 PM
No problem WildCat.

Let me know what everything reads in a few minutes. I'm really curious now.

I will have to take a picture of the bracket I built for my PID TC. I have it coming down through the vent, over to the top rear corner, down the corner to about midway down the tower, then I built and attached a bracket that the TC snaps into and it keeps it off the back wall. It's not the ideal place because it is usually a little warmer at the back of the cabinet but this way I don't have to ever move it.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 08:23:08 PM
Sounds interesting.  I have learned that it (the metal part) can not touch other metal.  I e-mailed Pat and he said that he just uses a wire thermocopule (no metal probe, and simply wraps it around one of the racks to hold it in place.

The butt was in for 4 hours about 15 minutes ago.  I changed the water bowl and spritzed the butt with some apple juice, so it will be a little bit getting back up with the ET-73.  Just before that I changed the setting on the Pb a couple of degrees and matched the ET-73.  We will see how it goes after a few more minutes.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 08:25:37 PM
Forgot to mention, the cycle time is back at its original setting.  When I had changed it to 0 everything went crazy.  That is why I changed it back and did the auto tune again.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 08:34:01 PM
That seems odd WildCat, The cycle time should increase the accuracy if anything. Once you get things levelled out temp wise you should try setting it to 0 again and let it run for a bit to test it. If it doesn't work you can always set it back. As long as you keep the same P, I & D numbers you shouldn't affect it.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 08:43:49 PM
Will do.  I also forgot to mention that when I adjusted the PID to match the ET-73 I also adjusted the set temp on the PID to 205.  Both PID and ET-73 are still in sinc, although the ET-73 climbs and falls at a slower rate.  Still running +/- 1 degree from the 205 set.

I did test the ET-73 with boiling water.  Am I correct that water boils at about 212 degrees F?  The ET-73 smoker probed tested at 210 and the meat probe at 211.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 08:48:24 PM
212 is more or less correct. You can adjust for altitude. I have seen a web site that had a little program that would calculate it for you but I can't remember where it was.

Sounds like your starting to get it dialed in though!  ;)

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 09:02:10 PM
I think I am getting there.  After adjusting the temp back to 0, it stays on my 205 setting longer.  It still touches on 206 and 204, but just briefly.  The ET-73 went up to 206 then to 207 then to 206.  I may have to adjust the Pb another notch, but I think I am zeroing in.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 09:06:08 PM
A little fine tuning and you'll have it nailed!  8)

Mike
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 07, 2008, 09:08:59 PM
Quote from: Mr Walleye on February 07, 2008, 09:06:08 PM
A little fine tuning and you'll have it nailed!  8)

Fine tuning is making me thirsty.   :D :D :D

Underscores that this is about the Porter Indicating Device.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 09:09:48 PM
Oops.  Spoke too soon.  Went back to look after typing the above and the PID was showing 203 and the ET-73 showing 201.  PID climbed back to 205 while I was watching.  May have to do another auto tune with these new settings.  If that does not do it, my guess is that it is the PID probe that is causing the difficulity.  Not enough swing to really worry about, just trying to get it the best I can.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 07, 2008, 09:11:18 PM
Plus or minus a few degrees is pretty good in my line of work. I suggest you declare success and enjoy.....

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 07, 2008, 09:14:30 PM
I would run it for a while and see how it works.

If you do run another autotune you may want to check all your settings again. I'm not sure if the autotune has any effect on them or not.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 09:15:07 PM
Declared.  Going for another thirst quincher.  I will still try to fine tune a little bit, but not overly concerned.  It is working much better at keeping the temperature constant than when I was just using the slider.  ;)
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 07, 2008, 09:16:19 PM
Thanks Mike.  I will do just that.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 08, 2008, 06:11:23 PM
WildCat

How did your butt turn out? Did you temps continue to hold?

Mike
Title: Re: Need everyones help on a PID problem
Post by: MRH on February 09, 2008, 07:06:30 AM
So the  thermocouple probe it shouldn't be touching any metal?  I have always just hung mine with a piece of wire under the first or second rack down from the top and it always seemed fine.

Mark
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 07:29:23 AM
MRH,
I think it depends on the controller, type of probe and how well the whole unit is grounded. I have seen computer controller furnaces go nuts if the probe touched a grounded piece of metal. Other controllers it does not effect. I know some probes are grounded at the tip and others are not, some need resistors across the probe wires at the controller. Which you need to use? I do not know? Maybe Arcs has some input on this, he is real knowledgeable with electronics. The other thing you could do is to contact the controller manufacture and explain what you are using the controller for and let them suggest what type of probe would work best for that application.
pensrock
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 09, 2008, 07:47:07 AM
MRH,

I'm with pensrock on this. Unless you really understand the grounding arrangement, a bare TC that comes in contact with metal will be a problem. You are dealing with millivolt signals, and ground loops will kill you. Bare TCs are used for the fastest temperature response, but nothing in the BS moves very quickly, and you are better off with an insulated TC.

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 07:52:58 AM
I use the Omega TC and Auber controller. I have never seem any problems with mine and I did create a mounting bracket out of stainless steel that the TC snaps into. I would assume this would ground the TC because the bracket I made is screwed to the back wall.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on February 09, 2008, 07:56:57 AM
Mr. Walleye,

Is the TC a grounded TC?

Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 08:04:43 AM
I'm not sure. It was a custom order. I'll see if I can find the numbers for it. Would the order number tell the storey?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 08:11:24 AM
Here's the number

Omega T/C Assembly W/Trans Joint Model TJ96-CASS-316U-3-SB-SMPW-M

I don't know if there is any way to tell or not.

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 08:35:24 AM
Mike, Arcs,
  From what I can tell from Mike's part number it is an ungrounded type K T/C. I'll show a link to the info from the manufacture.
http://www.omega.com/toc_asp/frameset.html?book=Temperature&file=TJ36-ICIN

At the very bottome of the info page it shows an example ant it appears the 'U' in the part number means 'ungrounded'. At least that is how I read it. But since it is a sheathed T/C I would assume the sheath touching metal would have no effect because the tip of the T/C should be isolated from the sheath itself. Does this make any sence?

Is there a reason everyone seems to be using type 'K' T/C's? I do not see why a type 'J' would not work since this is such a low temp application. We use type K for temp over 800 and up to 2100 F.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 08:37:00 AM
I just checked out the Omega site and it indicates the U (in "316U") in the model number signifies Ungrounded tip....

Ok.... So what does that mean?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 08:40:22 AM
Sorry Pensrock, I think we must have posted at the same time.

So, if I'm reading this correctly, a person would want an ungrounded tip on whatever TC you use?

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 09:20:29 AM
Mike,
I would think that would depend on the application and the thermocouple. At work our unsheathed T/C's are mostly open ungrounded, that means the welded tip of the two metals that make up the T/C are exposed but seperated along the legnth by a ceramic barrier. This type is also inside a ceramic protection tube, if the end of the ceramic tube breaks and the T/C touches metal it can cause interferance or noise on the signal to the controller, it is a millivolt signal so its does not take much to mess it up.
Our sheathed T/C's are for the most part also ungrounded, the two wires are much smaller and are placed inside the stainless steel or inconel sheath with ceramic power which is to keep the T/C wires seperated from the sheath. So in this case if the sheath touches metal, it would not matter because the ceramic powder is seperating the sheath from the T/C wires inside.
There are applications with certian controllers where they do use grounded tip T/C's but I am not that familiar with them so I cannot give you a good example.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 09:31:05 AM
Here is a very good page explaning the types of thermocouples and the grounded, ungrounded or exposed tips of the T/C's.
http://www.omega.com/thermocouples.html
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 09:49:55 AM
Thanks for the link Pensrock. From what I read, in our application in the Bradley I don't see why you would want a grounded tip. It may be quicker but I think the risk of touching metal in an all metal smoker would be greater for sure.

Mike
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 10:08:42 AM
I agree, why would we care if the ungrounded tip took another five seconds to get a reading?
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 10:15:53 AM
Quote from: Mr Walleye on February 08, 2008, 06:11:23 PM
WildCat

How did your butt turn out? Did you temps continue to hold?

Mike

Butt was wonderful.  Took a total of 20 hours.  It as a 9.2 pounder bone in.  Took it to 185 internal.  It stayed a long time in the 160 range.  The box temp started getting to a +/- 5 degree on a setting of 203.  Did another autotune and it eventually got to a +/- 2 degrees.  Next time I will not set the t to 0.  Had the best results before I did that with a +/- 1 degree.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 10:25:52 AM
Wildcat, At work when we use a relay output controller, we normally have the 'cycles' setting between 20 and 32. We can keep a furnace to +-1 degree at 2100 degrees. Since you are using a SSR output, which is basiclly the same as a triac or relay output controller execpt it is putting out a DC signal, you do not need to worry about contactors snapping in and out all the time or parts on the SSR wearing out so you should be able to keep the cycle setting lower than I do but I do not feel it would be necessary to go any lower than 8-12 cycles. If you can get good control at a higher cycle setting by all means, leave it alone. When I finally get my PID controller put together, I plan to start out with a cycle setting of 24-32, I feel very comfortable to being able to tune the instrument to control within +- 2 degrees.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 10:37:48 AM
Thanks pensrock.  When I do an autotune, it always sets the cycle back to its original configuration.  If I remember correctly it was 26.  I posted the number earlier on this thread somewhere.  I plan to leave it alone unless I just have to.  I plan to eventually get another probe (thermocouple) to get rid of the grounding issue though.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 10:38:16 AM
Wildcat

I agree with Pensrock, you gotta do what works. You really shouldn't have to run another auto tune if you wrote down your numbers. All you really have to do is adjust the cycle time and readjust the P, I & D numbers manually back to where it was working better.

It's interesting though, I run mine set to 0. I adjusted it there based on the information here on the forum. I believe it was BubbaGrump that indicated there wasn't any reason not to run it at 0 when used in conjunction with an SSR.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 10:41:07 AM
WildCat

According to the Auber manual the default/factory setting for the cycle time with an SSR is 2 and a relay is 4.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 10:43:55 AM
 ???  Relay?
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 09, 2008, 10:44:05 AM
ThermoWorks has a caculator for boiling water temps on it's main site.
Wildcat they also have the wire thermocouples I use for about 12 bucks. I'll get the part number for you when I get a few minutes later today. That will solve the grounding problem. Your turning into a PID expert!  :D ;)
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 09, 2008, 10:45:45 AM
Quote from: Wildcat on February 09, 2008, 10:43:55 AM
???  Relay?
You have an SSR in yours. Not a mech relay.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 10:48:01 AM
Quote from: iceman on February 09, 2008, 10:44:05 AM
ThermoWorks has a caculator for boiling water temps on it's main site.
Wildcat they also have the wire thermocouples I use for about 12 bucks. I'll get the part number for you when I get a few minutes later today. That will solve the grounding problem. Your turning into a PID expert!  :D ;)

:D Looooooooooong way from expert, but I am beginning to understand how to use this thing.  Even with a temp swing of +/- 2 degrees, it is much less of a swing than I get with the slider alone.  Good investment IMHO.  Use of the slider was easy, but required too much of my time on the long smokes.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 10:49:03 AM
Wildcat

Your Auber Controller is setup with an SSR (Solid State Relay). You can also set them up with a standard relay but they are subject to burning out if you cycle them too fast. The SSR has no moving parts and therefore it can be cycled at a much higher rate. This is correct to the best of my knowledge.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 10:59:55 AM
I just looked back on this thread.  The first time I went to adjust the cycle it was on 12 before I set it to 0.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 11:02:17 AM
Mike that is correct. Mechanical power control relays will eventually cause the contacts to burn due to the opening and closing. The other thing to consider is the contacts inside the controller, if it also has mechanical contacts. I would think your controller works more like a triac output than a mechanical output, in other words an electronic output. I do not think I would ever set the cycles to 4 for a mechanical contactor, that's way to many on/off cycling. I think the Barber Colman and Honeywell instruments that we use default to a 20 or something close to that.

Wildcat, I would be very happy with a +-2 degree swing in temps. Thats nothing, your controller is probably only capable of +-5 degrees calibration and the probe can be as high as +- ten degrees. Two degrees is great, why sweat trying to get one more degree? It can change some due to the outside temp, the size of the load in the smoker, if the puck burner is on or off. There are a lot of variables to consider. Just because it is +-2 this load does not guarentee that it won't be +-1 or +-3 the next time. Anything within the calibration accuracy of the controller is very acceptable. You should look for another thermocouple so you do not need to worry about the tip touching anything though.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 11:02:50 AM
Quote from: iceman on February 09, 2008, 10:44:05 AM
ThermoWorks has a caculator for boiling water temps on it's main site.
Wildcat they also have the wire thermocouples I use for about 12 bucks. I'll get the part number for you when I get a few minutes later today. That will solve the grounding problem. Your turning into a PID expert!  :D ;)

Thanks Pat
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on February 09, 2008, 11:04:10 AM
WildCat

If you look in your manual (on mine it's page 8 ) there is a chart that shows all the factory settings for all the settings on the controller. You can manually adjust any of them and if your not happy with it just set it back to what it was.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 11:10:47 AM
Thanks guys.  I am very pleased with a 5 degree swing (2 either side).  I will play with it a little and get the best I can, (you know - new toy syndrom) but where it is at is satisfactory.  I did notice during this last smoke that the closer it got to target meat temp, the greater the swing.  I did run the generator the whole time.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 11:19:59 AM
That's why you want everything to be as close to an actual run when doing an autotune, because if you make a big change, it is not tuned to handle it. So if you autotuned empty and without the puckburner then tried to run a large load with the puck burner you would most likly see a differance in the way it controls.
Also if you adjust the numbers in the P-I-D settings manually, you should allow some time for the instrument to adjust itself to the new settings before making a judgement of how well its working. I would let it go for a good 15 minutes before deciding if I wanted to change it back or adjust it more.
Isn't this fun? LOL..... Tuning instruments requires lots and lots of liquid refreshment!
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 09, 2008, 11:42:15 AM
I'm heading to the commercial icemaker in the garage with a bottle of rum and cola as we speak!!! This one day off before going back up north is making me thirsty.

Widcat I think either the 363 or the 383 would work for you.
 
Wire Probes
Very inexpensive and can be used in destructive testing or where you need a small, fast sensor that can be shut in an oven door, threaded through a small port, taped down, screwed down or glued in a hole.




The maximum temperature listed for the probes applies to the sensor itself. Plastic handles and thermocouple plugs are rated for short term exposure of 482°F (250°C) and extended exposure of 428°F (220°C).


--------------------------------------------------------------------------------


Fine Gauge Wire Probe
PTFE insulated with exposed junction. Easily taped, clamped or glued to surfaces, in holes. Wire dia: 0.01", 0.024" with insulation. Very fast air, gas or general purpose. Max. Temp 482°F
113-362 39" length $12.00     
113-363 78" length $14.00     
113-365 10' length $16.00     

--------------------------------------------------------------------------------


High Temp Wire Probe
Fibreglass insulation with exposed junction. Use for ovens and appliance temps. Max. Temp 662°F
113-382 39" length $12.00     
113-383 78" length $14.00     
113-385 10' length $16.00     

--------------------------------------------------------------------------------


Heavy Duty Wire Probes
PTFE insulated with exposed junction. Stranded wire with less than 0.1" overall diameter. General purpose. Max. Temp 482°F
113-372 39" length $18.00     
113-373 78" length $22.00     
113-375 10' length $29.00     

Title: Re: Need everyones help on a PID problem
Post by: pensrock on February 09, 2008, 12:14:24 PM
I think those would work well, just make sure the exposed tip of the sensor is not touching the metal surfaces. Since it is a wire T/C you should be able to twist it around a rack or something to hold it in place.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 09, 2008, 04:38:14 PM
Thanks guys.  While tuning, the more you drink the less you care.  ;D

Pat - I went to their web site and saw those.  Thanks for the info.
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 09, 2008, 07:47:05 PM
Update Wildcat;
Used 3 of the same probes you have as a test and had no issue with grounding problems. I even stuck them in a glass of water in the smoker and got good results. Touched the walls of the smoker and no swings in readings. I am sure you have a bum probe. Temp swings of 2 degrees average on all probes. Maverick is reading 5 degrees lower than PID. Did a claculation on bar pressure and elevation and found out the Mav was at fault. I checked with 2 of my fluke meters to verify and the PID was right on. Sorry for all the crap you had to go thru. I'll order you a wire probe Monday and mail it to you. Just don't let the end of the wire touch anything like pensrock said.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 12, 2008, 05:02:56 AM
Pat - I answered this before, but it got deleted with the forum problems.  Luckily I also sent you an e-mail with the same info.

No need to apologize for the probe.  It made me jump in there and I learned a lot from it.  It had its good side.  I thank you for your help.  No need to order the new probe.  I went on line and ordered the 383 myself.  You have put enough into this already.  Thank-you my half frozen friend.

In case you did not get my e-mail, I need another case or two of sauce (half regular and half mild) when you get a chance.

Rick
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 12, 2008, 09:25:39 AM
Got the e-mail Wildcat. Sauce will be out to you by the weekend when I get back. Might be in tonight for a day so I'll try to hit the chat room if it's back up and running. Hoping to pull a three day weekend and get a batch of Italian sausage made up while the weather is half decent for us.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 12, 2008, 01:52:41 PM
Thanks Ice.  There is no rush on the sauce.  I still have a couble of jars left.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on February 15, 2008, 03:01:08 PM
Got the new probe from ThermoWorks.  Really liked the little surprise gift they put in with it.  :)
Title: Re: Need everyones help on a PID problem
Post by: iceman on February 15, 2008, 03:27:47 PM
Quote from: Wildcat on February 15, 2008, 03:01:08 PM
Got the new probe from ThermoWorks.  Really liked the little surprise gift they put in with it.  :)

Ann keeps telling me "That's an expensive way to buy jelly beans".  ;D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 02:04:14 PM
Finally getting to try out the new probe tonight.  Reset the temp variance (Pb) for 0.  Also set the (t) to 0.  Picked up a brisket at Sam's this evening.  Put it in the smoker about 15 minutes ago.  Will see how she behaves and let you know.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 02:52:45 PM
Had to set the Pb back to -14.0 to match the ET-73.  ET-73 is one degree off based on boiling water test, so I guess my particular PID is off but at least it can be programed.  Had temp swing of 10 degrees (+5/-5), so I am doing an autotune.  Will post final outcome later.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 09, 2008, 05:09:44 PM
How are ya making out WildCat?

I'm not sure but I wonder if a person should set the offset (Pb) to "0" for the auto tune. I don't know if it makes any difference or not but it makes sense that you would get it holding a steady temp, then adjust the offset to match the temp that is hitting your product.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 05:27:05 PM
It does not seem to matter what I put the Pb setting at when doing the autotune.  After the autotune, the Pb gets automatically set where the PID wants it.  I then need to go back and reset the Pb.

What is really strange is that after the autotune, the temp spread was 4 degrees (2 degrees either side of the set temp), and the PID temp was matching the ET-73, which is fine in my book.  After a couple of hours the PID was about 5 degrees different than the ET-73 and the temp spread was back at the 10 degree spread.  No weird fast jumping around like the other probe did, so at least that is fixed.

I readjusted the Pb to again match the Maverick and did another autotune then again set the Pb to 0.  I will watch and see what happens.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 09, 2008, 05:32:06 PM
Keep us posted Wildcat. I'm always interested in better understanding these babies.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 05:49:35 PM
It just dawned on me that regarding the temp difference between the PID and the ET-73 changing is that I may need to replace the batteries in the ET-73 or the ET-73 may be going bad.  Does not explain the temp swing changing though.  The autotune just completed and I reset the t to 0.  The autotune had changed it to 120.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 09, 2008, 06:01:01 PM
I didn't realize until recently that the auto tune function also changed the cycle rate (t). I was going through the settings on my own and realized the auto tune had changed it to 9.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 07:36:25 PM
After another auto tune and Pb adjustment, the Maverick and PID are again in sinc, the temp spread is 3 below target and 2 above.  This is acceptable if it will continue.  This time I did not change the t to 0 afterwards, but rather left it at the 3 it set itself to.  I also had to adjust the Pb another few degrees.

I am begining to think that as the meat temp rises, the autotune needs to be re-done.  It has been stable for awhile now and the meat temp has stalled.  I will see what happens when the meat temp rises again.  This does not make sense to me, but it does seem to be taking place.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 09, 2008, 08:22:17 PM
Quote from: Wildcat on March 09, 2008, 07:36:25 PM
I am begining to think that as the meat temp rises, the autotune needs to be re-done.

I am beginning to think that you are at a point in the control looping tuning where a beverage needs to be re-done.   :D :D :D

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 09, 2008, 08:33:00 PM
 ;D ;D If it were not so late I would do so.  The autotune was holding within 2 degrees either side of set temp for a good while.  It is now 3 degrees either side.  If this will continue to hold then I think I will take a few naps during the night.  Got the Maverick alarms set.

If the temp swings remain stable, then my new numbers are:
P = 304
I = 88
D = 67
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 10, 2008, 04:40:39 AM
Temp swings went back to 2 degrees either side of set temp and stayed there.  Now if it will do likewise on the next smoke I should be good to go.
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on March 10, 2008, 07:33:05 AM
Just food for thought here, but when you did the first autotune on this particular project, was the meat just put in or was it in for a while to get some heat into the mass of meat?

I'm thinking that if the meat is cold (or room temp), and the PID learns to control the heat in that environment, that it's not going to be quite as accurate when the meat would get to the higher temps, and conversely, if the meat was almost done when autotuning, that it won't be quite as accurate when the meat is at the lower temps.

I could be way off on my thoughts though.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 10, 2008, 08:10:43 AM
Makes sense Tiny.  I did the first autotune after the meat reached about 100 degrees.  I did the final autotune while the meat was in the plateau (160 - 170 range).  I do not think I will have to autotune anymore.  Still do not understand why the difference in the Maverick and PID cabinet temp kept changing.  I plan to put new batteries in the Maverick and do another boiling water test on it before the next smoke.  I may have to invest in a thermo set.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 10, 2008, 08:30:07 AM
Wildcat

I agree with Tiny's thoughts on the cold meat theory. With regards to your temp variations between the Maverick and the PID I have noticed that when sitting and watching as the heat cutts in the temp display on my Maverick will climb quicker, then when the heater is off the Maverick will tend to drop quicker. It would appear as though the Maverick is more sensitive to the small changes. In my setup now my PID pretty much stays within 1 degree of it's set temp. My Maverick will vary a couple of degrees higher or lower depending on the cycle of the heater. Keep in mind I have my circulation fan blowing back to front no as well.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 16, 2008, 03:21:54 PM
My PID is holding pretty steady at +/- 1 degree.  There is one thing that has me puzzled.  My Maverick is off by 1 degree on the boiling water test.  I replaced all batteries.  Started smoking today and adjusted the Pb on the PID to match the Maverick.  After 4 hours the Maverick started showing a box temp 6 degrees higher than the PID.  This happened on my last smoke also.  If it does like last time, by the time the meat hits its stall period it will be off another 4 or 5 degrees.  After that it will remain the same.  I am not referring to a fluctuation, but rather a constant difference.  Last time I simply adjusted the Pb setting to match and will probably do the same this time.  Is this indicative of something not right with the PID  (settings, probe, or unit) or the Maverick.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 16, 2008, 05:18:59 PM
Quote from: Arcs_n_Sparks on March 09, 2008, 08:22:17 PM
Quote from: Wildcat on March 09, 2008, 07:36:25 PM
I am begining to think that as the meat temp rises, the autotune needs to be re-done.

I am beginning to think that you are at a point in the control looping tuning where a beverage needs to be re-done.   :D :D :D

Arcs_n_Sparks

I think my theory still applies.    :D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 16, 2008, 05:48:58 PM
Then I guess that I can assume there is no answer.  I will search elsewhere.  Beverage cooler here I come.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 16, 2008, 06:45:21 PM
Wildcat,

Your boiling water test is good at one point. Not clear how accurate it is over span. Are the probes right next to each other?

This reminds me of the "clock" problem; trying to get clocks to all read the same over time. Turns out a collection of clocks, averaged out, is more accurate than any single clock in the collection.

Given variables and not knowing how accurate each of the tempeature measuring systems are, getting them within a few degrees over span would be pretty good in my view.

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 16, 2008, 08:02:33 PM
Arcs,

Yes, the probes are right next to each other and not touching.

I understand what you said regarding the clocks.  Averaging from multiple sources does give one a relatively accurate picture.

The few degrees you mention in my case is simply too large of a spread.  The PID is working great at keeping the set temperature.  My problem at this point is not maintaining a constant temperature, but rather not knowing what that temperature really is.  The Maverick starts out reading a certain temp.  I set the PID to match.  By the end of the smoke, if I let the PID alone, the Maverick temp reading will be between 10 and 15 degrees higher than the PID.  This is not a fluctuating reading, but rather an averaged constant.

In other words and as an example, say I want to smoke/cook at around 200 (give or take a couple of degrees) for butts, but am unsure if I am at 185 or 215 or somewhere in between.

I am trying to figure out if my Maverick needs to be replaced or if the PID has a problem in dealing with being manually calibrated.  The PID seems to be functioning perfectly in all other aspects.  I suspect it is the Maverick.  I was hoping someone else would have some thoughts on the subject.  I apologize if folks are irritated or simply tired of this particular thread (or me for that matter).  If this is the case just say enough is enough and I will stop with no hard feelings and will simply find another source or wing it and start making some more purchases.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 16, 2008, 08:12:44 PM
Quote from: Wildcat on March 16, 2008, 08:02:33 PM
By the end of the smoke, if I let the PID alone, the Maverick temp reading will be between 10 and 15 degrees higher than the PID.

Was under the impression you were <5 degrees off between the two. 10-15 is more than one would expect. Perhaps the integration error in the PID is growing too large (not likely)? May want to experiment with the "I" term.

Regarding the Maverick, I have another wireless temperture system that I use to compare with my Maverick (yes, adding clocks!). Have not had a problem.

Arcs_n_Sparks

Title: Re: Need everyones help on a PID problem
Post by: West Coast Kansan on March 16, 2008, 08:21:34 PM
I got this DBS unit and it has fluctuations once set point is reached the first time.  This is a strategic feature included in the DBS to facilitate a controlled breakdown of connective tissue. You want an intermittant break down because it allows the fibers of the meat to recover and reabsorb as the geletin begins to move.  It will then rest some and then move again.  This has been the temperture sequence employed by championship bbq winners for decades now.  Only the DBS has the controls that will accurately mimic the temperatures used by purist smokers everywhere for as long as anyone remembers. Why else would you also incorporate a design where the temperature fluctuations decrease as the meat approachs set point. It is ONLY with the DBS that you can get the cake and eat it too.

The application of a steady heat such as with a PID defeats the intermittant break down of the tissue described above and should not be pursued.

Actually I have no clue what I am saying but it sounded good to me.

However, then again it may have just been a real dumb ass mistake to call something digital and put a profoundly stupid control scheme into the unit....

Personally I would like to think it was a purposeful design and that someone at Bradley would have resisted the group brain flushing and said like == excuse me, but that aint digital ==

::)  ::)  ::) So there is my story and I am sticking with it.  ;D  ;D  ;D

Dont be frustrated Cat, we are all making some pretty good BBQ and it sure is a lot easier than burning our fingers on those hot iron, temp fluctuating, ash makers that traditionalists use.  Give me an ez bake oven any time.  ;)  ;)  ;)

Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 16, 2008, 08:24:47 PM
I am, drinking, at a distance............   :D :D :D
Title: Re: Need everyones help on a PID problem
Post by: Gizmo on March 16, 2008, 08:33:58 PM
Wildcat,
The only thing I could add to possible help is to check out the new Thermapen device that has been recently posted on a new thread.  If you don't mind spending some more cash, the Thermapen's have a calibrated temp probe and I would trust is explicitly and adjust all other devices to what it reads.  With the wired probed that was shown, you could probe the meat (or water, or what ever) and keep it in the whole time.  At least that is what I am guessing since I don't have one of the new ones.  I have the older model that is hand held with probe attached.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 16, 2008, 08:46:52 PM
Thanks for the replies.

WCK.  What you said makes perfect sense. I think. :D  For me, a fairly constant low and slow provides excellent results for my taste buds.  I have the OBS.

Giz.  I have been looking at some of the stuff that Thermoworks sell even before anything was posted here.  Will probably invest a little after our trip and several birthdays.

Arcs.  I need to learn a little more before I can understand how to experiment with the "I".  Electronics is new to me.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 16, 2008, 09:05:39 PM
Quote from: Wildcat on March 16, 2008, 08:46:52 PM
Arcs.  I need to learn a little more before I can understand how to experiment with the "I".  Electronics is new to me.

I understand the frustration in finding "ground truth." Almost sounds like the Heisenberg problem: the act of measuring something changes what you were trying to understand. The only time I wind up three thermometers (PID included) is for sausage, since I don't want to render out.
Title: Re: Need everyones help on a PID problem
Post by: West Coast Kansan on March 16, 2008, 09:15:39 PM
Two drinks behind and catching up fast now.  Took a break for ice cream. Hard to be troubled when eating icecream. Been a long day but got the loin out and vac packed. it looks great. Another week ahead  :)
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 03:40:27 AM
I readjusted the Pb on the PID twice yesterday afternoon to match the Maverick.  If I had left it alone it would be 13 degrees difference between the two.  After the meat temp reached about 150 the box temp (PID and Maverick) continued to be the same all night.  This is about the same as it did last time.  I guess either the Maverick gets a little weird during the first part of the smoke/cook or the PID needs a little tweaking.  Don't know how to adjust the "I".  Probably will need to get another "clock" to figure out which one is acting up.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 17, 2008, 07:18:30 AM
Wildcat

I find that there is a larger difference between the ET73 and the PID when I first put a load in. Then once the load starts to reach alittle higher temp the difference seems to become less. What I have done is adjust the offset once the load has been in for a few hours (3 or 4 hours with a butt). By setting it up this way I just ignore the difference and once the meat starts to get up to temp it balances out. I'll just mention though that my probes are not in the same location in the cabinet. In my situation it seems the meat effects the ET73 a little more than the PID temp probe. This makes sense because my PID TC is in the back of the cabinet and I usually clip my ET73 close to the meat to see what the meat is experiencing.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 08:49:54 AM
Thanks Mike.  Both probes are next to each other on mine.  I think I might order the MicroTherm 2.  Here is the link: http://www.thermoworks.com/products/handheld/microtherma2.html

I think I will be able to nail down whether the problem is the Maverick or the calibration of the PID once I can determine what the true temperature is.

I liked the Thermopen accuracy that Iceman recommended but the 7 does not have a high enough range.  The 3 does but is not near as accurate.  The Microtherm 2 has more flexability and better accuracy to boot for just a few bucks more.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 17, 2008, 09:23:29 AM
Looks like a nice unit Wildcat.

I should be receiving my MTC & THS-113-181 probe shortly but I don't have it yet. I like how small the probe is on this especially for sausage and with a 3 second response time it seems pretty impresive. Here is a link to it.
http://www.thermoworks.com/products/handheld/mtc.html

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 09:44:18 AM
I looked at that as well.  Can't argue with the price.  Accuracy not as good but I guess a couple of degrees is close enough.  I am still thinking it over.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on March 17, 2008, 02:25:52 PM
Wildcat,
   You keep mentioning that you are adjusting the 'Pb' setting in your PID controller to get it to read the same as your Mav. I have never used an Auber controller but most controllers have an 'offset' setting which you can use to make the controller read the same as another device. Every controller I have ever used the 'Pb' setting is for the proportional band which is part of the tuning to control the output of the instrument. In other words when you run an autotune the controller will set the proportional–integral–derivative settings for you, not saying you may not need to tweak a little here and there. But I get the impression you are adjusting the 'Pb' in an attempt to match your Mav instead of using the 'offset' function. I may be wrong because I have not used that instrument but maybe someone who has one will be able to explain this better.
   Just a thought.... one other thing, most digital controllers can be set back to factory calibrations by changing the thermocouple type, leaving the set up mode, then changing it back, but it may change other settings in the controller so go through all the setups and write them down before doing something like this unless you really know what you are doing.
   One other thing to consider is the calibration tolerances of the equipment you are using. I think you are using a type K thermocouple it is most likely +-5 degrees F, unless you have a 'special limits' thermocouple which is +-2 degrees F. I really doubt is yours is special limits. Your controller is also most likely +-5 degrees F. The extention wire between the controller and thermocouple will also have some effect. This is only your PID set up, the Mav will have its own tolerances.
Just my two cents worth, I do not know if it will help any but its something to consider.
pensrock
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 17, 2008, 03:04:43 PM
I just checked the Auber manual and here are a couple of descriptions from it.

Measurement Accuracy - .2% Full Scale or +/- .2 degrees C Thermocouple Input with Internal Automatic Compensation.

Pb = Input Offset.


I'm not sure about the accuracy of the exact TC that Wildcat is using. I'm not even sure of the accuracy of the Omega TC that I'm using. Interesting comments though Pensrock. At a .2% accuracy does this equal .4 degrees at 200 degrees F?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 05:14:32 PM
Here is something a little different.  I was smoking 2 butts and when the first one was done, I took it out and repositioned the remaining one and I also moved the Maverick box temp probe and the PID back to the same location.  After the box temp recovered there was an 8 degree difference in the two and remained that way for the remaining 4 hours.  Maverick reading lower.  I did not change the Pb settings this time after the move.  Could this problem be as simple as a problem with the Maverick probe wire?  On the other smokes after the temp got to 140 - 150 the readings remained the same with each other, but they were never disturbed.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on March 17, 2008, 05:49:06 PM
That's a good question Wildcat. It would be interesting to have another thermometer and TC to check it against.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 17, 2008, 06:03:13 PM
Wildcat,

Are you using an insulated TC?
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 06:56:44 PM
Not sure.  It is the 383 from Thermoworks.  I have only used it 2 or 3 times now.  The Maverick is kind of old and I got it used.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 06:59:50 PM
Yep, here is a link to it:  http://www.thermoworks.com/products/probe/tc_wire.html
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 17, 2008, 07:04:51 PM
Okay, uninsulated junction. Are you keeping it from touching stuff? I assume you are, but was trying to figure out if you have a ground loop that is driving you crazy. Running out of ideas.....   :(

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 17, 2008, 07:33:02 PM
The wire is flexable where I can just lightly loop it onto one of the racks.  The exposed tip does not touch anything.  I might be wrong but I really think it is the Maverick or its probe that is the problem.  Once I get another thermometer I should be able to determine whether it is the Maverick or the PID that is changing (with the assumption that one of them will stay consistant with the 3rd thermometer).
Title: Re: Need everyones help on a PID problem
Post by: pensrock on March 18, 2008, 01:43:45 AM
Mike,
  The +-.2 degrees C is most likely from the max of the thermocouple not the operating temperature. I do not think you can use a 200 degree setpoint to measure the calibration. I may be wrong but normally it is measured at the high limit of the thermocouple. If it is a type K it could be 1800 to 2200 degrees F. Still this is not something to be too concerned about. I'm just trying to figure out why he sees the temp differance if he uses the offset and matches the control to the Mav.
  If the manual says the offset is 'Pb' then I guess it is so. Sounds very odd to me but I have not used an Auber instrument so I cannot say. Most instruments we use have a setting in the input section not the tuning section that says something like, 'off' or 'ofset' or 'ost' well you get the idea.
  Arcs has  a good idea of a possible ground loop.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 18, 2008, 04:40:51 AM
Is it difficult to explain "ground loop"?  I am new to this electronic stuff.  I know a little about electricity.  Is it essentially allowing a break in the flow of current or a decrease in the flow of current, thereby giving false readings?
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 18, 2008, 07:06:28 PM
Wildcat,

Basically, you follow the electrons, and keep them moving where they belong. The worst example of a ground loop problem is if you are holding a tool that has the hot side contacting the case, you grab it, and are connected to ground somehow (standing in water or holding something that is grounded). Electrons start flowing through you with a bad outcome. That is how GFIs protect you: making sure the electrons flowing out match those flowing back.

The potential problem here is more subtle. Very small currents and voltages in the TC wiring loop. If the end contacts ground, some electrons flow back to the instrument (since it is grounded in some fashion) in a way not anticipated, and disrupt the measurement.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 19, 2008, 04:42:20 AM
Thanks.  I think I understand that.  Unless there is a break in this new wire, I do not think that is the problem since the end was not touching anything.
Title: Re: Need everyones help on a PID problem
Post by: pensrock on March 19, 2008, 02:04:34 PM
Wildcat,
  If you get an actual break in the t/c wire it should read as high as the instrument will read, sometimes just a series of bars will be on the screen. If a t/c is shorted, both wires touching one another, it should read about ambient temperature. Arcs did not mean that one of the wires was actually broken but if the + side of the extention wire or t/c is rubbed through and touching a ground or some other odd thing, it can cause weird things to happen. You have a new t/c and if I read right it is an ungrounded t/c, so I doubt it is the problem. But it could have something else to do with how the PID, smoker or where you plug it in is grounded. The millivolt signal that is produced by a thermocouple is very small and can be influenced by any number of things.

Arcs,
  Beverage.... beverage... beverage...
   Now that I got your attention, is there a place on the Auber instrument to ground it? From looking at the wiring diagram on olds site I did not see a ground run to the instrument. All the temperature controllers we use require a separate ground run to the instrument.
   I'm also wondering if he has a good ground from his receptacle. My last bit of thinking for the day is.... he may want to run a ground wire to the negative side of the thermocouple at the instrument, I do not know if it would help but it should not hurt to try.
   This is an odd problem to try to figure out without being there to look at things for yourself. Now its time for me to partake in a beverage.
pensrock
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 19, 2008, 02:27:36 PM
Thanks for all the info pensrock.  I think the house wiring is fine.  The PID is plugged into a GFI plug.  The BS runs off of the PID.  Iceman built the PID for me.  The PID is reflecting that it is maintaining the set temp within a couple of degrees so it appears to be working fine.

I just ordered the MTC along with a 113-159 probe and 113-041 from ThermoWorks.  Once it gets here and I get the time and adequate weather to do another smoke (hopefully next week) I will do a cold and hot test on it and then I should be able to figure out if it is the Maverick that is acting up or the PID/BS.
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on March 19, 2008, 06:13:23 PM
Quote from: pensrock on March 19, 2008, 02:04:34 PM
Arcs,
  Beverage.... beverage... beverage...
   Now that I got your attention, is there a place on the Auber instrument to ground it?

Don't have the Auber, so cannot answer. I do, however, have a beverage...   :D
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on March 25, 2008, 04:38:14 PM
Finally got the MTC today.  Did a cold and hot test on it myself.  Off by 0.6F.  This jewel should work fine.  Leaving town for about a week in the RV this weekend so I may not be able to get a smoke in before we go to determine where the problem lies.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 12:33:22 PM
Finally got around to another smoke (brisket).  Started at 7:30 this morning, it is now 2:25 here locally in Central Time Zone.

It is NOT the Maverick that has been acting up.  I started with the Pb setting on the PID at -10.  After about an hour I had to reset to -18 to match the MT and the Maverick, which were within 1 degree of each other.  I had to do this about every hour or two all day.  At one point is was set at -4.  It is now at -10.

Each time I did this, I found that the PID would cause temp to fluctuate -3 degrees to at + 10 degrees above the 200 set temp.  13 degrees.  I tried doing auto tunes and they sometimes worked.  I also simply unplugged from power and re-plugged.  This would normally fluctuate 3 degrees either side of set temp (6 degrees).  It appears that the PID cannot maintain the variance setting of the Pb.

Any thoughts?
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 12:43:42 PM
 ???  ???  ???

That's got me puzzled WildCat! I'm going to have to think about this... I'll let you know if I come up with any thoughts on it.

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 12:50:41 PM
Another point to ponder.  Could un-plugging the tc from the PID to check with MT and then re-connecting be part of the cause of temp variance?
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 01:13:04 PM
Quote from: Wildcat on April 25, 2008, 12:50:41 PM
Another point to ponder.  Could un-plugging the tc from the PID to check with MT and then re-connecting be part of the cause of temp variance?

I don't really think so unless you were running auto tune at that point but you never know. Did you re-check all your settings after?

So I'm assuming you unplugged the TC off the PID, then plugged it into the MTC? If so, how did these readings compare? (assuming you didn't move the TC location inside the cabinet for this test)

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 01:23:01 PM
Did not move location.  The tc and Maverick are in the same place.  Yes, I did unplugged the tc from the PID to check with the MT.  I did not do this during auto tune, and I only did so while temp was relatively stable.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 01:30:28 PM
How did the readings compare using the same TC?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 01:31:11 PM
I just checked an the PID was reading 197 with the MT and Maverick at 191 and 192.  Changed the PID Pb to -13.  Been having to do this all day.  Usually there is more of a difference.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 01:39:35 PM
WildCat

I must be miss reading something here...

I was under the impression you had the PID reading a given temp, then unplugged the TC off of it and plugged it (the TC from the PID) into the MTC. I was wondering if you saw any temp change between the to units, using the same TC or did I miss read your post?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 01:44:22 PM
Everything mentioned with the PID and MT are with the same tc.  The Maverick and MT are about 1 degree off of each other.  I started with a separate tc on the PID and MT, but one of them went bad early on.  The one that went bad was a small wired tc that is designed to wrap around something to hold it in place (disposable tc).  The one I have left is heavier gauge and has an aligator clip.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 01:49:56 PM
You are correct.  I am useing the same tc as described.  I can get them to read the same by adjusting the Pb, but then the fluctuations from set temp will then start.  Once I get that under control (?) then eventually the MT and PID will start reading different again.  Usually after an hour or two and by about 7 to 15 degrees.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 01:53:53 PM
Quote from: Wildcat on April 25, 2008, 01:31:11 PM
I just checked an the PID was reading 197 with the MT and Maverick at 191 and 192.  Changed the PID Pb to -13.  Been having to do this all day.  Usually there is more of a difference.

Ok... still trying to get my brain around this... The PID (in this case) was reading 197 with a Pb setting of -7 (I'm guessing on the Pb setting), then you unplugged the TC off the PID and plugged it into the MTC and the MTC read 191. Next you adjusted the Pb to -13 to try and compensate for the difference.

Seem correct so far?

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 01:57:18 PM
Yeppers.  I just checked on it and so far the PID is maintining a reading of 201 to 203 (set temp is 200).  The Maverick is reading 201.  Afraid to check the MT right now since I would have to unplug (temporarily) the tc from the PID to do so.
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on April 25, 2008, 01:58:21 PM
My advice now is to just let it be.  You're in the ballpark with a lot tighter tolerance than most (if not all) brand new ovens.  As long as you're paying attention to your temp, and it don't get away from the target more than 10 degrees, you're good to go.  If ya gotta set it at 220 to see 210 at the meat, so be it.

I'm sorry if that sounds harsh, but I just feel you're making it much more complicated than it has to be.

Now let's go to the bar...first round's on me.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 02:07:52 PM
You covering air fare with the first round Tiny?

:D  :D  ;D

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 02:10:11 PM
Sorry you feel this way Tiny.  It will not stay this accurate for long.  It has fluctuated more than 10 degrees on both sides of the set temp all day.  Sometimes it stays on the mark.  Perhaps I am wrong in thinking that the PID would keep things closer to set temp than without it.

Thanks to all of you who have tried to help and I apologize for being a pain with this PID.  Perhaps it is time to do something else.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 02:18:43 PM
WildCat

WildCat

There is no question it should maintain more accurately than that, certianly there should not be that big a variation. I thought I would re-post this. It's just some thoughts on some slight temp variations on my setup and I thought it might be useful.

Mike

I find that there is a larger difference between the ET73 and the PID when I first put a load in. Then once the load starts to reach alittle higher temp the difference seems to become less. What I have done is adjust the offset once the load has been in for a few hours (3 or 4 hours with a butt). By setting it up this way I just ignore the difference and once the meat starts to get up to temp it balances out. I'll just mention though that my probes are not in the same location in the cabinet. In my situation it seems the meat effects the ET73 a little more than the PID temp probe. This makes sense because my PID TC is in the back of the cabinet and I usually clip my ET73 close to the meat to see what the meat is experiencing.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 02:30:59 PM
Thanks Mike.  I read that part of your post several times and it makes sense.  I am positive this is not the case at hand since all both probes are in the same location, and I am checking the PID with the MT utilizing the same tc that the PID is.  The PID is simply changing itself.

I thank you for your help my friend, but it is obvious that this post is beginning to irriate others and we should not go any further unless offline.  I will continue to try to monkey with this on my own or possibly hire someone locally, or I might just trash it.  Not worth losing friends over.  I can always look for something better than the PID, or simply just go back to useing the slider.
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on April 25, 2008, 02:37:59 PM
I guess I've misunderstood the whole thing so far then, and I apologize for that.  I still feel you're changing too many variables for the PID to function the way it's supposed to.

What I would do, is get the PID to be consistent before looking to match it up with readings on other devices.

There are going to be temp swings no matter what.  You have a mass that you're trying to heat up, and a heat source turning on and off to heat the air around it to heat the mass.  If you can get to a +/- 10 degree swing early in the smoke, that levels off near the end of your smoke, that's good enough.

Once you get to that ballpark, then worry about how much difference there is between the PID and your Maverick or other thermometer.  I wouldn't mess with the offset on the PID at all...just manually adjust up or down based on the Maverick reading.

As for the drink...if ya can't make it here, just tell me what you want and I'll drink it for ya too. :D
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on April 25, 2008, 02:42:46 PM
I just read you and Walleye's last post before mine.  I'm not irritated with the discussion...it has been informative.  I just felt that you were trying to make it way too complicated, and that's where my frustration came from.  Walleye said what pretty much what I spent the last about 15 minutes trying to say too...and I think he said it better.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 02:50:04 PM
Tiny, this has all taken place over a period of over 9 hours.  I have waited each time for the PID to be consistant for at least an hour or two before modifying.  But that is enough.  I have some crown and sprite in the kitchen, so if you got some - Here's to ya.
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 02:58:20 PM
Quote from: Wildcat on April 25, 2008, 02:50:04 PM
I have some crown and sprite in the kitchen, so if you got some - Here's to ya.

Hey... Can you guys at least pour/drink one for me! I'm still at work.....  ::)

Mike
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 25, 2008, 03:00:08 PM
Just finished mine (this made me thirsty) so I will drink the next one for ya.  :D
Title: Re: Need everyones help on a PID problem
Post by: Mr Walleye on April 25, 2008, 03:08:10 PM
Quote from: Wildcat on April 25, 2008, 03:00:08 PM
Just finished mine (this made me thirsty) so I will drink the next one for ya.  :D

Excellent!... I feel all better now!  :D

Only 52 minutes till the real thing!  ;)

Mike
Title: Re: Need everyones help on a PID problem
Post by: Arcs_n_Sparks on April 25, 2008, 06:46:17 PM
As usual, I am drinking at a distance.......   8) 8) 8)

Arcs_n_Sparks
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on April 26, 2008, 07:23:52 AM
Quote from: Wildcat on April 25, 2008, 02:50:04 PM
  I have waited each time for the PID to be consistant for at least an hour or two before modifying. 


What would happen if you didn't do any modifying?  Would your set point and the read temp maintain an offset of X degrees?
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 26, 2008, 08:54:47 AM
No.  That is one of the problems.  :-\  Most of the time I would be smoking about 15 degrees higher than the PID indicated.  Once I even tried just lowering the set temp, then the actual temp would swing 10 degrees lower than I wanted.  My problem has been two fold.  The offset  and then the temp swings being too great.  ???  That being said, after my last post last night, the off set stayed right on the mark, but the swing was about 12 degrees so I ran another auto tune.  After that the off set remained stable and the temp swing was 199 to 202 with the set temp at 200.  Absolutely perfect.  :o :) Next smoke I will leave everything alone and hope that everything remains constant.  If not I will simply adjust the set temp and see if it will again become stable.  If not I will try to find an expert locally or just go back to smoking without a PID.  Had less monitoring without one.  The only reason I wanted a PID to start off with was so that I could leave the house while smoking or get more sleep on the overnighters.
Title: Re: Need everyones help on a PID problem
Post by: Tiny Tim on April 26, 2008, 09:23:06 AM
Sounds like you got it now.  Hope it continues for you.

I've never worried too much about how close mine was.  I just set it at 10 degrees above what I really want, since my PID probe is on the lowest rack, and I cook on the next one up.  Have my Maverick alarms set 40 degrees above and below the desired temps.  And just let it be. 

Might worry about it a little more if I was doing sausage or bacon, but I'd also be using a lower temp than needed and slowly ramping it into the sweet zone.

Cheers.
Title: Re: Need everyones help on a PID problem
Post by: Wildcat on April 26, 2008, 09:26:00 AM
 :D  Yeah, I want to eventually do some of those other things, but I gotta get this thing going right first.
Title: Re: Need everyones help on a PID problem
Post by: Smoking Duck on April 26, 2008, 03:55:29 PM
I hope you get it where you want it Wildcat.  My PID is going right where I set it.  Usually my Maverick is 7 degrees less than the PID.  I figure that I'm probably somewhere right in the middle of the two, which is no real problem when doing pulled pork.  Might want it a little more exact, but I find the more I fool with something, the worse I make it.  I wish I had your patience in dealing with things like that.