Temperature Measurement and the Fluke 80TK

Ask any questions relating to the PicoScope hardware or kit contents here.
If you have any questions prior to purchasing the kit post them here.
Post Reply
User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 386
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK

Temperature Measurement and the Fluke 80TK

Post by FioranoCars »

Hi
Has anyone used the Fluke 80TK to measure temperatures? How did you find it, is it any good or are there better alternatives?

We would like to be able to measure temperature in several scenarios, so overall range is important, from exhaust catalysts to interior ambient and air con output, to motors and pumps or condensors and radiators.

The goal is to be able to verify the reasonableness of components switch on / off etc, especially for intermittent faults, where having a graph over time of several cycles or operations would be a major benefit.

Has anyone built their own using standard K type probes, is that easy to do?

Thanks in advance

Richard
Richard Lukins
FioranoCars

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: Temperature Measurement and the Fluke 80TK

Post by Fat Freddy »

Check your email. Probe sent.

Best do some calibration tests first. The zero compensation may need adjusting but should be accurate to a couple of degrees. Keep sample rates low and be prepared to filter the results.
From the few runs I have done using that probe (EGT's 700*C +) the results are good.

Let me know if you have any questions.


HTH
FF

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 386
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK

Re: Temperature Measurement and the Fluke 80TK

Post by FioranoCars »

Hi FF

Thanks, though not yet bought the Fluke unit, just looking to find out if anyone has used this (clearly your have!) and any issues (I assume none, as you did not mention any issues except calibration)...

Has anyone else done any temperature work and what have you used? Anyone using a direct K type thermocouple or using it with a pre-built or home-made cold joint etc?

Thanks
Richard
Richard Lukins
FioranoCars

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: Temperature Measurement and the Fluke 80TK

Post by Fat Freddy »

That probe is for a direct connection to a K type thermocouple.

I had assummed (possibly incorrectly) the connector made up some sort of referance/cold junction. As I'm able to measure a voltage at room tempreture. I'm not sure what causes that junction although it can be worked out and that is what I was referring to in calibrating.

Knowing the output from a K type is fairly linear, I work out at what tempreture I have 0mV generated and that is what is added to the sum (26*C in the case of that probe).

Sounds complicated but it's very easy to use and appears to be very accurate when compared to using a multi meter and K type to measure temp.
It's worth spending a few quid on a thermocouple just to suss it out

HTH
FF

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 386
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK

Re: Temperature Measurement and the Fluke 80TK

Post by FioranoCars »

I think we may be talking at slightly crossed purposes, I was hoping to avoid using the Fluke 80TK and just plug a K type directly or via a cold joint into the Pico ... but I think I can answer that now for myself!

The Pico 4x23 scopes have a resolution of 10mv, where a k type probe has a Thermoelectric Voltage in Millivolts (ie the output) of -2.2mv at -50degC to 0mv at 0degC to 4.1mv at 100degC to 10mv at about 250degC and about 20mv at 500degC. So based on that, I think the Pico scope is not going to be able to give any real readings for 0-750degrees as it would be about 4 steps from 0 to 750, so no accuracy or reliability?

I guess I should try tapping into a sensor on a car's exhaust to prove it... that'll be this afternoon then!

So unless someone can correct me, it looks like I'll need the Fluke (or similar amplifier - anyone using anything else?) to amplify the signal into a range the Pico can reliably plot to a usable scale.

I'll post a waveform data file a bit later showing the readings I get from the thermo sensor.

thanks
Richard
Richard Lukins
FioranoCars

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: Temperature Measurement and the Fluke 80TK

Post by Fat Freddy »

... but I think I can answer that now for myself!

The Pico 4x23 scopes have a resolution of 10mv, where a k type probe has a Thermoelectric Voltage in Millivolts (ie the output) of -2.2mv at -50degC to 0mv at 0degC to 4.1mv at 100degC to 10mv at about 250degC and about 20mv at 500degC. So based on that, I think the Pico scope is not going to be able to give any real readings for 0-750degrees as it would be about 4 steps from 0 to 750, so no accuracy or reliability?
Don't forget its 10mV a div, 12 bit resolution, and an accuracy of +/- 1%.

If I have done my maths correctly (a fairly complicated formula) it should be accurate to 0.6*C +/- 1%. I could be wrong but the accuracy of the Pico is alot better than 10mV, the lowest set up is +/- 50mV.


HTH
FF

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: Temperature Measurement and the Fluke 80TK

Post by Fat Freddy »

FioranoCars wrote:I guess I should try tapping into a sensor on a car's exhaust to prove it... that'll be this afternoon then!

FioranoCars wrote:I'll post a waveform data file a bit later showing the readings I get from the thermo sensor.
Any update/images on this yet?

FF

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 386
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK

Re: Temperature Measurement and the Fluke 80TK

Post by FioranoCars »

Hi
Sadly off work due to an illness, but hope to be back to form next week :-(

Will post some results then. In the meantime I have downloaded the data I need to create a lookup table for the K type probes. On further reflection the accuracy is going to be pretty poor, with an offset of ambient temperature and then an accuracy of about +/- 5% (maybe more). To build in Ambient Temperature a maths channel (or edit the custom probe each day) with a manual adjustment could be performed.

The overall accuracy is down to the cold joint materials, and in automotive applications the wires already will be crimped to some sort of "metal" connector before the official cold joint, causing an error level. I'm not sure how much this might be ...

But for my purposes, to check the operation of the probe, verse the failure of the cold joint, and to see that the probe functions in an "reasonable" output during warmup and once operating temperatures are reached, is really what I'm after. Our cars suffer cold joint failures just as often as probes, and we wanted to be able to prove which failed, quickly (ie a template with a good waveform to show what should be happening).

I guess it should work (using a direct K type probe without extra hardware) for rough measurements, but not to the point of proving the boiling point of water! ... I will post results of exhaust systems test shortly.

Thanks
Richard
Richard Lukins
FioranoCars

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: Temperature Measurement and the Fluke 80TK

Post by Fat Freddy »

I think the look-up table would give, in theory, a far more accurate result. I had considered making one up myself but found using off the shelf thermocouples, the offset required daily adjustment depending on ambient as you have pointed out. So opted to take advantage of the K type linearity output instead.

The accuracy is still reasonable. The last check was <3% up to 1300*C, using a DMM as comparison. Although that in itself will have a degree of error along with the Pico's +/- 1%.
Not easy to do. Something that will be proven over time and the required adjustments done or new probes made depending on the application.
Noise will always be a big problem.

Looking forward to seeing the results.

Regards
FF

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 386
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK

Re: Temperature Measurement and the Fluke 80TK

Post by FioranoCars »

Started testing the Custom Probe settings and the first cars Thermocouple (a Lambo) seems to be completely out of scale.

The raw pre cold joint readings are over 50mv, which would be circa 1300c instead of 300-450 of the exhaust, so just going to do some custom conversion tables for this car and try a Ferrari next week, fingers crossed the Ferrari uses standard thermocouple K type ranges?!

Also got a Fluke 80TK, so will have a play with that too.
Richard Lukins
FioranoCars

Post Reply