Cook in progress - trouble with probe accuracy - Help me Obi Wan, You're my only hope


 
Then what is the offset for then? I understand the nature of the thermocouple and understand that in theory all thermocouples of this type should be outputting the same voltages. However, its not a perfect world and there could be slight impurities in the metals perhaps. I assumed that's what the offsets are for, for making slight adjustments to account for these anomalies.
 
The thermocouple should be putting out roughly 5mv/c... Yours might be like mine and be off a little bit. The output should be somewhat linear so you don't need to change the offset, instead, boil water and put the thermocouple in it. Take what the heatermeter is reading and divide it by what boiling is at your altitude. Multiply this whole value by 5 and enter that into the thermocouple mv/c dialog box. So should be like this 5*(heatermeter reading at boiling/boiling for your altitude). My thermocouple was about 10 degrees off and this made it pretty much spot on.

Charles
 
CHeathman said:
The thermocouple should be putting out roughly 5mv/c... Yours might be like mine and be off a little bit. The output should be somewhat linear so you don't need to change the offset, instead, boil water and put the thermocouple in it. Take what the heatermeter is reading and divide it by what boiling is at your altitude. Multiply this whole value by 5 and enter that into the thermocouple mv/c dialog box. So should be like this 5*(heatermeter reading at boiling/boiling for your altitude). My thermocouple was about 10 degrees off and this made it pretty much spot on.

Charles
Click to expand...

This makes more sense, not sure what I was thinking. I guess I wasn't thinking that the thermocouple was linear and behaved more like a thermistor for some reason. I did do a bit of reading and everything Ive read indicates that k-type thermocouples have a tolerance of +/-1.5C* to +/-2.2C* depending on where you look. Im going to give this method a go. Thanks. I am curious now what the offset setting would be for.
 
Last edited:
Not sure really, but that block is there for standard rtd probes. Not sure when you would use it with a thermocouple.
 
At one point Bryan was going to remove the offset settings, as an option. They shouldn't be used, as if you have to use them them something else is wrong, or the probe is no good.

The offset on the thermocouple can be used if using one of my adapter boards, to adjust the thermocouple to that of boiling water when you want to be able to observe temperatures below the heatermeters 37°F minimum. It adjust the the thermocouple via a small trim pot, just like what is used on the LCD.
 
J. Winn said:
I am curious now what the offset setting would be for.
Click to expand...
It is there to add one degree or subtract a degree to make two probes read the same which makes people really happy!

The thermocouple amp has a 3*C accuracy with a 0.3% gain error, which means your mV/C value should probably be between 4.985 and 5.015 to remove the gain and you can use the offset to remove any of that 3*C error.
 
Bryan Mayland said:
It is there to add one degree or subtract a degree to make two probes read the same which makes people really happy!

The thermocouple amp has a 3*C accuracy with a 0.3% gain error, which means your mV/C value should probably be between 4.985 and 5.015 to remove the gain and you can use the offset to remove any of that 3*C error.
Click to expand...
I just re-read this and was curious. So, if I'm not in this range then what. It seems like, being that it is linear, using a ratio of temp read/temp should be x 5 would be all the error correction you would need. No?
 
J. Winn said:
Yeah it's reading 212* for boiling water. Before I had a offset of 4 or 5* to get 212* at boiling. I assume I will have better accuracy at any temperature now.
Click to expand...

Now test your assumptions. Take your maverick probe(s) and thermocouple probe, wrap them together with some tinfoil and stick them in your oven to see if they match at 400F
 
Steve_M said:
Now test your assumptions. Take your maverick probe(s) and thermocouple probe, wrap them together with some tinfoil and stick them in your oven to see if they match at 400F
Click to expand...

So interesting discovery. I went to try this because I was curious to see how they compared as well. I noticed the noise icon, which I thought was weird because I have the noise filter enabled and I had never seen it since then. I flashed the noise firmware and sure enough I saw this:
15yl4k.jpg


So I took the probes out of the oven and the noise went away immediately. Anyone know what the deal is? I put the probes back in and immediately noise again.
 
Last edited:
The noise filter doesn't remove noise, it just adjusts the sampling to match the mains frequency to cancel it out. It doesn't look like it is on there though, because if you have it set to 60Hz it would end around 16.6 on the graph and you'd have one complete sin wave.

In any instance it does look like A/C line noise on the probe.
 
I have the 50Hz filter on as I'm in Australia. But like I said as soon as I just break the plane of the oven opening with the probes instant noise. I don't even have to set them in there, just hold them in there and I get noise. As soon as you take them out the noise disappears. I hope I am explaining it well enough. I could take a video maybe to show you what I'm talking about. It's really odd.

Just tried again with the oven off and it still does it :-/ weird. Just touching the outside of the oven sets it off. I dunno, bad probe maybe? Any way to know for sure?
 
Last edited:
Oh I see, well then the noise filter is set properly. What you're definitely getting is ground loop noise from the metal probe sheaths (which are tied to ground) and your oven which probably has its chassis tied to ground as well.
 
That's what I figured. The probe was suppose to be ungrounded. I'm going to check it later on with a multimeter. I got it on eBay for like 6$ I think so that may be it; poor manufacturing.

Edit: Well that wasn't it. I checked resistance between each contact and probe as well as metal jacket and that checks out; ~25Mohm. The only continuity is between the two contacts like it should be.
 
Last edited:
I used different power supply and it eliminated the ground loop noise I was getting. Here are the results of my test after I adjusted my mV/C value to 4.889.
4sz0i8.jpg

I am happy with the results. The pit probe lags the food probe as it comes up to temp but then they even out. This is to be expect since it's an ungrounded probe. It's better this way too for the pit probe at least. The first arrow is where I adjust the mV/C value just to see what type of difference it would make and changed it back. The second arrow I opened up the oven, closed it, then turned it off.
 
You must log in or register to reply here.

 

Back
Top