Temp differences


 

J. Winn

TVWBB Fan
I was getting what appeared to be some weird temps from the pit probe. So I took my other thermoworks needle probe and stuck it through some foil to secure through one of the top vent holes on my mini wsm. The temperatures between the two probes, Ktype and thermoworks needle, are quite different as you can see(probe 3 and pit). Not sure what to believe. I calibrated all ports using boiling water to 212F.
image.jpg
 
Last edited:
It would help to see your config, but I see yellow noise levels on the pit probe which might indicate that the power supply grounding is giving you false readings? You might want to flash the hm-noise firmware and see if the values are just all over the place.
 
Ok I'll set it up again and get a picture. I will also try and get the noise graphs. I will report back.
 
I flashed the noise firmware. What am I looking for exactly? Also, I flashed the latest snapshot so I could give the AC filter a try but it's not showing on the config page.
 
Last edited:
Here are some pictures of how I had it set up.
Probe3 was like this:
image.jpg

The pit probe was like this at grate level with the wire through the top vent:
image.jpg
 
Last edited:
Put them together. There can be temperature differences when you have the probes at different levels and in different areas of the bbq.

Also, check them in boiling water to get a better idea if there are temp differences
 
If you stuck 10 probes in different places in your grill they would all read different. The top vent and grill grate are usually a bit different, a 20F swing isn't out of the ballpark in certain scenarios.

If you want to compare the readings from the probes it has to be apples to apples, meaning, same location. Even still, a thin probe will react to temp changes faster than thick ones, so even tied together they might not read exactly the same if the temp is fluctuating. You could wrap the probes together in some foil or better yet in a pot of water, then they should track pretty tight. If you want to verify that your probes are tracking temp the same/properly put the probes (not the cables) in a pot of water on the grill, heat it up and watch the HM graph.....

If you need a reliability check on the pit probe during a cook, next time put the food probe right next to the pit probe...
 
I figured there would be some difference between the two locations but not quite that much I guess. There was roughly probably only about 4" between the two probes if I had to guess. What is odd to me is I calibrated all probes/jacks using boiling water to 212. Yesterday I plugged in the HM and the two needle probes and thermocouple and just let it set on the table for awhile. The two needle probes were both +/-1 degree or less but the thermocouple was reading at least 10 degrees warmer, I left if plugged in for awhile and periodically checked it to see if it changed at all.
Bryan mentioned noise. I have installed the noise firmware but I am not exactly sure what I am looking at or how to check the noise on the individual probes. I also saw that there is a AC filter in the latest snapshot that I was going to try but when I flashed the latest snapshot the option still was not available. I also tried powering the whole thing through the Pi rather than the 12V supply to see if the results would differ, but the results where the same.
 
Last edited:
Oh yeah they will definitely be a LOT different if they are in different locations. On my egg the temperature can be close, or one can be 15 degrees hotter. Which probe is hotter depends on how it feels that hour, as they can swap which one is hotter during a cook. To get the AC filter you need the linkmeter snapshot too (you're missing the configuration option if you just do the AVR snapshot).

The noise graph on the home page, you're looking to see what the range of the error is and if it is consistent. It should theoretically be a straight line or have at most 2 numbers it is going back and forth between. If it is all over the place drifting about you have a problem. If it is a sine wave then with a period equal to your AC mains frequency, it's a AC problem.

As far as "calibration", HeaterMeter can't really be calibrated because you're only doing a 1 point calibration. Thermistors are non-linear so calibrating will only get you accurate temperatures at the calibration point. The thermocouple is a little better but it should be 5.000V +/-2.5%.
 
Oh yeah they will definitely be a LOT different if they are in different locations. On my egg the temperature can be close, or one can be 15 degrees hotter. Which probe is hotter depends on how it feels that hour, as they can swap which one is hotter during a cook. To get the AC filter you need the linkmeter snapshot too (you're missing the configuration option if you just do the AVR snapshot).

The noise graph on the home page, you're looking to see what the range of the error is and if it is consistent. It should theoretically be a straight line or have at most 2 numbers it is going back and forth between. If it is all over the place drifting about you have a problem. If it is a sine wave then with a period equal to your AC mains frequency, it's a AC problem.

As far as "calibration", HeaterMeter can't really be calibrated because you're only doing a 1 point calibration. Thermistors are non-linear so calibrating will only get you accurate temperatures at the calibration point. The thermocouple is a little better but it should be 5.000V +/-2.5%.

Gotcha. So when looking at the homepage next to the graph I see numbers 0-5 and they are clickable; what does this do. Also, am I suppose to be looking at the yellow line, blue line, or red line? Thanks again!
 
The clickable numbers change what ADC input you're looking at (with a 10 second delay). 5 is the Pit, 4 is the first food and so on. The legend tells you that the red and blue lines are the average and the mean, so the yellow is the data.
 
The clickable numbers change what ADC input you're looking at (with a 10 second delay). 5 is the Pit, 4 is the first food and so on. The legend tells you that the red and blue lines are the average and the mean, so the yellow is the data.

Gotcha. Here is what I'm looking at. All 3 probes were sitting on the table next to each other for this.
Here are a couple screenshots with 5 selected:



Here is 4:



Here is 3:



Here is 2:

 
Last edited:
Anyone? To me it looks like I may have an issue with probe 0 and 1. I'm just looking for confirmation and what I can do to fix it.
 
Last edited:
I still haven't seen your configuration, but your noise levels look fine with no blower running / in the electrical outlet you used to produce those graphs. The configuration is most important because I want to see how you've calibrated the probes. The noise can only really accurately be measured when the HeaterMeter is in operation, because with it sitting idle in an outlet in your home may be very different that the environment when it is in use.
 
I guess I'm not exactly sure what you mean by configuration. The probes were like I posted in the pictures with an additional probe in the pork butt and I had no blower running. For calibrating the probes I just boiled some water and had to offset the pit probe -4 degrees, the three food probes needed no offset. The only difference was the outlet I used, I can use that one and see what I am getting.
 
How do I flash the latest Linkmeter snapshot so I can try the filter option? Is the .gz hosted somewhere? Thanks.

Edit: nevermind I found it
 
Last edited:

 

Back
Top