Voting resources, early voting, and poll worker information - VOTE. ... Adafruit is open and shipping.
0

ad8495 Breakout board schematic
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

ad8495 Breakout board schematic

by sadavis on Mon Aug 10, 2020 3:01 pm

Strangely, your documentation does NOT include a schematic of the Breakout board other than the link to GitHub.. and the file I downloaded from there won't load into my version (5.11) of Eagle. I'm not sure if it's a version problem or the fact that I didn't download the file correctly from Github.

I seem to have a pretty wide reported temperature (voltage) variance from the straight line curves for TypeK thermocouples .. like over a hundred degrees high at ~400 degrees. I've read through the AD Datasheet on the 8495 and the only thing I can reasonably question is the cold temperature sensing... In my case, the board AMBIENT temperature changes quite a bit during the measurement process.. and between measurements. This unit is NOT mounted in a 'controlled temperature location', but I still need the temp output voltage to be valid and reasonably close (+/- 5 degrees would be great).

I've tried several different sensors from several different vendors and they ALL read way high in my range of interest, but read reasonably close at room temperature. Can you provide a schematic - it should be pretty simple, but I'm not sure if I'm dealing with a multilayer board or not.

I'm running the board at a regulated +5 volts and all the charts I can find for 0 - 5 volts output say I should be able to cover my range pretty accurately, but the voltage coming off the board doesn't track the charts.

Suggestions?
Steve

sadavis
 
Posts: 2
Joined: Mon Aug 10, 2020 2:43 pm

Re: ad8495 Breakout board schematic

by dastels on Mon Aug 10, 2020 3:19 pm

I don't have experience with that board, but the schematic is at https://learn.adafruit.com/ad8495-thermocouple-amplifier/downloads

dastels
 
Posts: 3519
Joined: Tue Oct 20, 2015 3:22 pm

Re: ad8495 Breakout board schematic

by sadavis on Mon Aug 10, 2020 4:37 pm

Thanks for the quick reply. It is basically what I expected, but I needed to be sure. After looking it over, I'm pretty sure I understand how it SHOULD work but I'm still stuck with my readings that are much higher than I think they really are. Obviously, more investigation is needed on my end.
Steve

sadavis
 
Posts: 2
Joined: Mon Aug 10, 2020 2:43 pm

Re: ad8495 Breakout board schematic

by adafruit_support_mike on Tue Aug 11, 2020 9:13 pm

The AD8495 does what's called 'cold-junction compensation'.

Officially, all thermocouple voltage tables are calculated on the assumption that the readout end of the negative wire is at 0C. Early thermocouple systems had to keep that end in an ice bath.

Obviously that's inconvenient, so once ICs became good enough, manufacturers started to build a thermal sensor into the chip at the point where the negative lead connects. Once you know that temperature, you can add an offset voltage to the thermocouple voltage and get accurate values from the 0C tables.

The AD8495 datasheet specs 0.5C ambient temperature rejection across the 0C to 50C range, meaning the output from a thermocouple whose hot end is connected to a fixed temperature will only change by 0.5C as the board temperature goes from 0C to 50C.

The datasheet doesn't specify a slew rate for cold junction compensation, so sudden changes in board temperature could create small jumps that smooth out over time.

sadavis wrote:I'm pretty sure I understand how it SHOULD work but I'm still stuck with my readings that are much higher than I think they really are.

The first thing to do is get another source of calibrated measurements for the temperature of interest.

The hardest problem in measurement theory is dealing with uncertainty. There are unknowns in the things we want to measure, the instruments, and the references the instruments use. No amount of improvement in one area can fix problems caused by unknowns in another area, and there's no way to know what area is causing problems (or how much problem each area causes) until you calibrate them individually.

Making sure you have a reliable measurement for the temperature of interest will help the debugging process tremendously.

adafruit_support_mike
 
Posts: 61422
Joined: Thu Feb 11, 2010 2:51 pm

Please be positive and constructive with your questions and comments.