Scientific-grade temperature measurement

[kevm14]

I wanted this to measure the temp of my furnace and hot water heater flue gas but measuring other things seems useful.

There are many generic ways to do this. However, I elected to put a kit together that is all Fluke. I have an old Fluke 8060A that my uncle gave me a long time ago. It is my official multimeter though I do have that nice one I got from HF with clamp on current meter.

Info: https://www.tequipment.net/Fluke8060A.html

It does more than you'd guess just looking at the different settings.

I decided to look into whether it would even be possible to outfit this with temperature measuring capability. Turns out, yes!

In fact I would simply need two things:
1) Any K-type thermocouple (there are actually many different styles optimized for different things)
2) A converter box that takes the output from the thermocouple and scales it to a calibrated voltage reading that the DVM displays, and you can select Communist units or Freedom units.

For the converter, the unit is the Fluke 80TK. Here it is brand new on Amazon.
https://www.amazon.com/Fluke-80TK-Therm ... 4283&psc=1

Yikes, $133! OK. But how about used on eBay?

Yeah, used is way better! Only $27 shipped. Here: https://www.ebay.com/itm/80TK-FLUKE-K-T ... 2749.l2649

But I still need a probe. Cheap probes on Amazon are $10, such as this: https://www.amazon.com/T-PRO-Thermocoup ... RydWU&th=1

That's nice but some complained about calibration (shocking). Seems like I might look for an all-Fluke solution. So what are the options there?

There is the 80PK-22 immersion probe. Immersion seems like a good general purpose option because it is good for liquids and gels but also works in air (albeit with a delayed response compared to ones made for air operation).

Temp range: -40 to 1090 Degree Celsius. That's as wide as I could possibly need I think.

Here's a new 80PK-22 on Amazon: https://www.amazon.com/Fluke-80PK-22-Su ... r=1-3&th=1

Geez, $92. How about used? Yep.

https://www.ebay.com/itm/Fluke-80PK-22- ... 2749.l2649

$30 and shows working in the ad.

So how did I do? Well I added (uncalibrated but probably more than acceptable for DIY/hobby use) temp monitoring to my Fluke, with Fluke accessories for $57.

Yes for $20 you can get a dedicated temperature probe system, such as this, which even comes with 2 probes: https://www.amazon.com/Signstek-Thermoc ... hi&sr=1-21

But I'm pretty sure I'm better off with what I put together than this.

I finally received the probe from Fargo, ND (hilarious) so I immediately plugged everything together to test. I tried air and water. Results are good:
Furnace at the end of its heating cycle, cooling down the heat exchanger:

Cold water from kitchen sink:

Hot water from kitchen sink:

It works!

[Fast_Ed]

I was reading this thinking you could have bought a cheap thermometer and stuck it in there. A degree or two of innacuracy would probably be fine in this case.

But then I realized that I've never actually seen someone go through the mental effort to use a k type thermocouple and the temp probe piece of a multimeter. That's pretty cool.

So 95° air through the vents is what you would expect?

[kevm14]

Yeah I freely admit there was a cheaper way to get this done, and linked to it. But I think this is cool.

I don't know how accurate it is but from the eBay listing I know some things:

You are buying a Fluke 80PK-22 SureGrip Immersion Thermocouple Probe. Unit is a k-type thermocouple probe. Temperature probe for use with liquids and gels. Unit has mini connector and a temperature range of -40 *C to 1090 *C (-40 *F to 1994 *F). Unit accuracy of + 2.2*C (+ 4.0 *F) to 293*C (559.4*F) and +/-0.75% of reading after that. Only the 80PK-22 is included on this listing, no other extension wires or readouts are included. I am willing to ship anywhere.

And he shows some images next to another device and they are damn close. Plus, I want to buy used equipment from someone who has stuff like that. For those times when you need to measure temp to 3 decimal places. FWIW, this appears to match his other Fluke meter to the nearest 0.1 degree. That's....pretty tight.

Now that's just for the probe. The 80TK converter thing stacks up its own error. And aside from calibration drift, the rating is the following:

Plus this info sheet is dated 1985. We built a time machine in 1985 so obviously this is quality.

But yeah then the DVM itself has its own accuracy and possible calibration. I guess I'm just assuming it's tighter than some $20 thing on Amazon. Based on some reviews, it probably is. Would be interesting to test.

As to your last question....I guess 95 is what I would expect at the tail end of the heating cycle after the oil burner had already shut off. I felt the air before I measured it and was like "hmm, 90s?" Conclusion: didn't need this. Except I did because I'm not using that method for my flue gas......

[kevm14]

Checked furnace vent outlet temps again while it was running. This time it peaked at almost 123F before the furnace shut off. Not sure how much hotter it would get. Obviously the ducts warm up. This is probably pretty close to max at least under these conditions.

OK I need something hotter to test.

Hot water heater exhaust gas. A toasty 458F. That's fairly normal, maybe a tad warmer than I expected. It hit overload in the 200mV range (at 200F). So I switched to 2V which then scales the output to x1000 degrees.

Then I got greedy. Fire is less than 2000F right? Let's test. Looks like 1459F.

The end of the probe was glowing red when I pulled it out. Don't believe I damaged it and I guess I didn't expect to if it is really rated to 2000F.