New 4.3 Thermocouple Build Reading 218.5 in Boiling Water

[Mike Barrett]

My new 4.3 thermocouple build is reading 218.5 in boiling water with the TC. Is that within spec for a successful build or are there things I should check? I got the TC reading around 211.7 by changing mv/C to 5.17 (the calculated 5.15 still read a bit high). The thermistor probes are working as expected.

I saw another thread where someone mentioned that their new 4.3 TC build was reading 217 so they swapped in the ATmega from a 4.2 build and that solved the issue. I gave it a shot with the ATmega from my 4.2 build but the TC read about the same.

I'm just wondering if I'm OK to calibrate it out or if I should be hunting for a problem.

Thanks,
Mike

 

[John Bostwick]

That was my 4.3,

Check for shorts with the TC amp. You should see 211° or 212° I think Tc'a have +-2°.

 

[Mike Barrett]

Thanks John. I checked the TC amp and I don't see any shorts. I've attached an image of the TC area of the board for reference.

The surface mount resistors also check out OK. Anything else I should look in to?

 

[WBegg]

What type of TC are you using? Make sure it's K-type? Did you buy a cheap chinese TC?

 

[Mike Barrett]

I have two thermocouples. An Auber TC-K6A and an Auber TC-K3MM-C. Both are K-type and both are reading around 218.5.

 

[John Bostwick]

Your board looks good. A thermocouple tends to be accurate as long as the weld is not broken ,it should work. I would install the noise software and see if you are getting any noise. You find it on one of the setup pages

 

[Mike Barrett]

I tried turning on the AC line filter in the setup and that didn't help. I installed the hm-noise hex and while the K3MM probe was fine (bounced between 476 and 478), the K6A bounced between 468 and 478. I decided to remove AC from the equation entirely and switched to 12v battery power. That settled the K6A down so that it bounced between 480 and 482. Both probes are still reading high, though a bit better than before. They're around 217 now. Other suggestions?

Thanks for your help with this!

 

[JasonB]

I tried turning on the AC line filter in the setup and that didn't help. I installed the hm-noise hex and while the K3MM probe was fine (bounced between 476 and 478), the K6A bounced between 468 and 478. I decided to remove AC from the equation entirely and switched to 12v battery power. That settled the K6A down so that it bounced between 480 and 482. Both probes are still reading high, though a bit better than before. They're around 217 now. Other suggestions?

Thanks for your help with this!

Did you find a solution here? I have a freshly built heatermeter where all my thermoworks pro probes are reading between 210.9 and 211.3 in a boiling water test (Thermapen is 211.2 so I will call the termistors good!) but the thermocouple port is reading 6ish degrees higher.

I was going to follow the Theromocouple Calibration in the wiki HERE but then I read in searches that you should not have to do that, that the thermocouple should just work.

I also learned that my previously owned iGrill probes work; I was initially a little bummed I bought the thermoworks until I saw how fast they worked and how well they are built. I will be perfectly happy running just the thernoworks pro probes if i can't get the thermocouple working quite right.

 

[Bryan Mayland]

but the thermocouple port is reading 6ish degrees higher.

Can I ask what thermocouple you're using? I bought their Crocodile Clip Oven Probe for testing and I was surprised to see that it was reading almost 20F higher than my other thermocouple I had attached to the grill. I assumed there had to be a HeaterMeter problem but I pulled the plugs out of HeaterMeter and plugged them into their NIST-certified thermocouple reader and it read within 1F of what HeaterMeter was saying on both probes.

I had assumed that my older probe had just gone bad so I left the thermoworks in charge at 225F on a 6lb boston butt. Started at 10pm and the next day at 6pm I had passed the stall but the meat temperature had started curving off around 180F and looked like it wasn't even going to ever reach 190F. I tossed a pro-series termistor on the grill grate and let it warm up and it read ~200F. This matched what the older thermocouple read, as well as the dial thermometer in the lid. Plugging the new thermoworks thermocouple back into the reader, it read 224F.

For some reason the thermoworks thermocouple was just plain wrong, not according to HeaterMeter, but according to their own thermocouple reading device. After I pulled the meat and put it in the oven, I turned up the temp on the grill and found that the thermoworks thermocouple read high by 10-20F anywhere I looked all the way up to 500F. I need to investigate further, but it isn't a heatermeter problem, it is a problem with thermocouples in general that they can be wrong?

 

[Andrew F]

I hate to ask, but are you suspending the probes in boiling water and not letting them touch the sides and bottom of the pan? And are you holding them in boiling water, and not clipping them to the side of the pan while it comes up to a boil?

 

[JasonB]

Can I ask what thermocouple you're using? I bought their Crocodile Clip Oven Probe for testing and I was surprised to see that it was reading almost 20F higher than my other thermocouple I had attached to the grill. I assumed there had to be a HeaterMeter problem but I pulled the plugs out of HeaterMeter and plugged them into their NIST-certified thermocouple reader and it read within 1F of what HeaterMeter was saying on both probes.

I had assumed that my older probe had just gone bad so I left the thermoworks in charge at 225F on a 6lb boston butt. Started at 10pm and the next day at 6pm I had passed the stall but the meat temperature had started curving off around 180F and looked like it wasn't even going to ever reach 190F. I tossed a pro-series termistor on the grill grate and let it warm up and it read ~200F. This matched what the older thermocouple read, as well as the dial thermometer in the lid. Plugging the new thermoworks thermocouple back into the reader, it read 224F.

For some reason the thermoworks thermocouple was just plain wrong, not according to HeaterMeter, but according to their own thermocouple reading device. After I pulled the meat and put it in the oven, I turned up the temp on the grill and found that the thermoworks thermocouple read high by 10-20F anywhere I looked all the way up to 500F. I need to investigate further, but it isn't a heatermeter problem, it is a problem with thermocouples in general that they can be wrong?

For me the thermocouple is an Auber alligator clip model. I only ordered one so I cant test against another, I may order a cheapie from Amazon just to test.

Would it be wrong to tune the variance out by adjusting the mv/C against a known temp (like the wiki states) regardless of the cause of the variance? I guess it the probe is straight up flaky I cant expect the flakyness to be consistent enough to tune it out..... I did not try either the termisistor or thermocouple + heatermeter with an ice test to get another reference point, should I even expect that to work at all?

 

[JasonB]

I hate to ask, but are you suspending the probes in boiling water and not letting them touch the sides and bottom of the pan? And are you holding them in boiling water, and not clipping them to the side of the pan while it comes up to a boil?

Completely valid question, yes I donned my silicone glove (I am a wuss ) and dangled the probe in the water not touching sides and bottom

 

[BrunoBronosky]

Can I ask what thermocouple you're using? I bought their Crocodile Clip Oven Probe for testing and I was surprised to see that it was reading almost 20F higher than my other thermocouple I had attached to the grill. I assumed there had to be a HeaterMeter problem but I pulled the plugs out of HeaterMeter and plugged them into their NIST-certified thermocouple reader and it read within 1F of what HeaterMeter was saying on both probes.

I had assumed that my older probe had just gone bad so I left the thermoworks in charge at 225F on a 6lb boston butt. Started at 10pm and the next day at 6pm I had passed the stall but the meat temperature had started curving off around 180F and looked like it wasn't even going to ever reach 190F. I tossed a pro-series thermistor on the grill grate and let it warm up and it read ~200F. This matched what the older thermocouple read, as well as the dial thermometer in the lid. Plugging the new thermoworks thermocouple back into the reader, it read 224F.

For some reason the thermoworks thermocouple was just plain wrong, not according to HeaterMeter, but according to their own thermocouple reading device. After I pulled the meat and put it in the oven, I turned up the temp on the grill and found that the thermoworks thermocouple read high by 10-20F anywhere I looked all the way up to 500F. I need to investigate further, but it isn't a heatermeter problem, it is a problem with thermocouples in general that they can be wrong?

I assume this is what resulted in this statement on the HeaterMeter-Probes page:

Crocodile Clip Oven Probe NOT RECOMMENDED the exposed tip is too sensitive to passing heated air currents

Did you ever try adding something to the thermocouple to increase the thermal inertia? Maybe a metal or silicone sleeve?

The reason I am asking is that I am planning to use the Hi-Temp Flexible Ceramic Fiber-Insulated Probe as the thermocouple in my first HeaterMeter build (which you shipped to Chicago this week). I like the idea of it being able to survive a 1100ºF Big Green Egg carbonization burn. As I said in my planning thread: If I experience a problem with it being too (thermo-mechanically) sensitive, I (expect that I) can put a brass, aluminum, or silicone sleeve over it to dampen the swings. If the problem is electrical, I am less equipped to be able to solve that with circuitry modifications. (Though if someone versed in EE were able to suggest a circuit, I could build it.)

 

[Bryan Mayland]

Did you ever try adding something to the thermocouple to increase the thermal inertia? Maybe a metal or silicone sleeve?

The reason I am asking is that I am planning to use the Hi-Temp Flexible Ceramic Fiber-Insulated Probe as the thermocouple in my first HeaterMeter build (which you shipped to Chicago this week). I like the idea of it being able to survive a 1100ºF Big Green Egg carbonization burn. As I said in my planning thread: If I experience a problem with it being too (thermo-mechanically) sensitive, I (expect that I) can put a brass, aluminum, or silicone sleeve over it to dampen the swings. If the problem is electrical, I am less equipped to be able to solve that with circuitry modifications. (Though if someone versed in EE were able to suggest a circuit, I could build it.)

Actually, yes and it works really well! I wrapped the exposed thermocouple bit in kapton tape just to electrically insulate it, then cut a 1"/25mm piece of thin aluminum tubing and slid it over. Crimping both ends with a pair of pliers made a pretty good seal and it completely eliminates the noise that comes from it being insanely sensitive with no thermal mass.

The issue wasn't electrical at all, it was just that the temperature would vary +/-0.5F as very small currents of hot air passing the tip would be instantly read by the thermocouple. HeaterMeter had a real hard time trying to maintain temperature as it expects that amount of variation to take place over several minutes, not several seconds. The PID gains could be dropped but it really just needed a longer low pass filter to average the readings over a longer period instead of trying to react to each second's update. That would break normal users because the response would be too dampened I think. I ended up just not recommending them because you'd have to modify the probe and I didn't want to add to the confusion that is all the documentation.

 

[BrunoBronosky]

Actually, yes and it works really well! I wrapped the exposed thermocouple bit in kapton tape just to electrically insulate it, then cut a 1"/25mm piece of thin aluminum tubing and slid it over. Crimping both ends with a pair of pliers made a pretty good seal and it completely eliminates the noise that comes from it being insanely sensitive with no thermal mass.

Woot! Thanks for the followup.

it really just needed a longer low pass filter to average the readings over a longer period instead of trying to react to each second's update

Oh I like that use of "low pass". It makes me think of the concept my Dad taught me about resistors inhibit the flow of current, capacitors inhibit changes in voltage, inductors inhibit changes in current. https://electronics.stackexchange.com/a/12712/26064

I ended up just not recommending them because you'd have to modify the probe and I didn't want to add to the confusion that is all the documentation.

Good choice. I think your terse explanation is enough to steer the knuckleheads like myself in the right direction if they want to go against the simpler solution. I'll let you know what kind of results I get from the -58 to 2200°F ceramic thermocouple.

 

[Bryan Mayland]

Oh I like that use of "low pass". It makes me think of the concept my Dad taught me about resistors inhibit the flow of current, capacitors inhibit changes in voltage, inductors inhibit changes in current. https://electronics.stackexchange.com/a/12712/26064

Yeah, exactly that! There's actually a hardware low pass filter in place of ~16Hz cutoff which cuts out a lot of electrical noise, but it can't do much about the small temperature changes the thermocouple is reporting over a few seconds. A 47uF capacitor can be used instead of the 0.1uF capacitor on the Pit probe, to give it a 0.3Hz cutoff but then it would take over 30 seconds after inserting a probe before the temperature would stabilize. Same thing when opening and closing the lid. It is better to just dampen it at the source, like you've suggested with some sort of sleeve on the bare thermocouple to soak up the temperature variations of swirling convection currents in the smoker.

 

[WBegg]

You must watch elevation, also. I live at 4,500 feet where water boils at 204 degrees F (vs 212 f). I always adjust my cooking temps accordingly.