mind telling us why you think thermocouples are better? most believe thermisters, while a bit slower to stabilize are more accurate and consistant

 

Ya not better for a digital gauge x2
Only a few left that use thermocouples now.

You can view iManifold temps out to .001°F

That would drive me bonkers on a SMAN with that tiny screwdriver lol

I think Thermocouples get used for higher temps since they won't melt

 

Thermistors are better than thermocouples.

 

Well every day checks are out of the range of the thermistors on these new tools.

For example a fujitsu A/C will slow the frequency down on a compressor if the discharge pipe temp gets to 120C, this cannot be checked with a thermistor clamp thats only rated to 95 or 100C.

A screw compressor oil temp can get to 150C even higher in a fault condition, which causes many issues for example some oil flow switches on screw racks are only rated to 100C, so i need to know exactly what temps are being reached in a fault condition.

Auto TXV also are very fast as the power elements are in the valve unlike a regular TXV.

 

Fujitsu use thermistors.

What do you guys use to measure high temps, and what specifically are you measuring that requires such high accuracy?

Did I misunderstand something?

 

So how does the Fujitsu measure discharge line temp?
A Cooper 4011 is as fast as any valve or heat transfer through a refrigeration pipe Imo

 

Fujitsu use thermistors.

What do you guys use to measure high temps, and what specifically are you measuring that requires such high accuracy?

 

It's not so much an accuracy thing. You can buy a thermocouple with the same accuracy as a thermistor like 1000# of bricks or 1000# of feathers.

 

Sorry for the incomplete thought on accuracy redback.
The issues are sources of error for thermocouples in the field expecilaly when tied to a set of digital gauges.

 

Can you list some of these sources

 

TM, TC or the RTD ?

View attachment thermistors vs RTDs vs TCs.pdf

This is a comparison btw thermistors, RTDs and TCs. For comparison, an industrial 3 wire 100 Ohm RTD with a 316ss thermowell for measuring a process temp. and transmit a 4-20mA output would run about twice the price of a decent DMM. You still have to do something with that analog signal to make sense of it. Some smart RTDs do have a digital display on the head or you can buy a separate digital display for a little more than the price of a good DMM.

Thermistors are cheap, low temp. down and dirty when 'good enough' is good enough. A TC, based upon matching the type of dissimilar metals to the temp. is reliable enough even for UL listing tests. For instance, a type 'J' iron/ Constantan TC is spec'd for measuring clearances to combustibles around heat producing appliances and venting looking for temp. rises above ambient at either 90°F or 117°F depending whether the surface is inside and enclosed space such as a wall cavity vs. an external surface such as a floor, wall, mantel, or ceiling. However, typ 'K' thermocouples are used to test flue gas temps. to 2,200°F for factory chimneys sold in Canada/ 2,100°F in the US (UL103HT). You can take a copper and iron wire, clean ends, twist together, stick into a flame, attach a DMM measuring millivolts dc and get a reading up to about 25-30 mv max but usually much lower. If you were to take a zillion pairs of twisted dissimilar wires and daisy chain them in series, place them against a flat black background and set them in direct sunlight, you will have your own solar panel. :bsing:

 

Thanks iceman

With my fluke 902 when the meter is stable on a bench giving a temp reading of 22C and i pick the meter up the temp drops by a whopping 5.5C, meaning it can't meet rated spec accuracy when being handled due to cold junction fault, and it was a $436 meter, does that mean i should throw my fieldpiece ST4 in the trash just because the fluke 902 thermocouple thermometer has major cold junction issues?

Im keen for more info on the static electricity and magnetic field issues, i do have clamp meters that jump around cause of magnetic field from contactors but i just adjust and use them differently. Can you give actual examples that i face in the field from magnetic field and static electricity, and how much they effect accuracy of a thermocouple thermometer.

 

Fluorescent lights
Cell phones
Microwaves
Electric motors
Electric wiring
WiFi
Bluetooth
Two way radios
Charging refrigerant through a manifold
Static electricity on yourself
Straight up magnets

http://vid268.photobucket.com/albums/jj31/itsiceman/Thermocouple%20Error/4080ab96.mp4


https://m.youtube.com/watch?v=XG0otLjj4h4


https://m.youtube.com/watch?v=3tMDMZESC5M

There is another video floating around with error around a electric motor.
I pretty much stopped using them and posts about them being better stopped popping up so no more videos lately

 

Thermistors have a steeper curve, so it's easier for the instrument to read a small change. The slope is curvy so its more difficult to convert. Thermocouples produce a weaker signal, but there are only a few types made. Type K probe works with anything with a type K scale. The sensor itself is metal, so the response is faster. Many China made junk use thermistors and get very good accuracy if the actual thermistor is mapped out and programmed. Good for OEMs who can use production line equipment to vary the environment and program the table.

The mapping table is embedded somewhere in the embedded system :gah: that you can't rewrite without the software used in production.

 

The proprietary embedded systems strikes again...

 

Here is some good material on the subject. Biggest reason k-type is no good for digital manifolds is noise generated by static electricity generated by flowing refrigerant

https://www.picotech.com/library/ap...pplication-note/improving-the-accuracy-of-temperature-measurements#introduction

 

Here is some good material on the subject. Biggest reason k-type is no good for digital manifolds is noise generated by static electricity generated by flowing refrigerant

The linesets are metal and connected to the compressor which is grounded. I have to disagree on your reasoning. Some instruments poll the probe differently. Some cheap Chinese meter I have is way sensitive to any kind of noise when you're using the "K type thermometer" port, but my Fluke 87 is much more stable under the same conditions using the same exact probe. Thermocouples are used in all kinds of industrial environment many of which are ISO 9001 certified facilities.

The analog front-end amplifier for thermistor is much easier to build than a noise resistant K-type design and its just a matter of plugging in the look-up table into the proprietary embedded system which is done with ATE or by hand in many China made stuff. The manifold is just an ohmmeter with a calculator that translates ohms to degrees using a look-up table or a formula stored in the proprietary embedded controller.

The proprietary embedded systems strikes again...

Thanks to them, thermistors are not interchangeable unless the piece of tool/instrument happens to use one of the industry standard spec thermistors.

 

Fixed link and found the other

http://vid268.photobucket.com/albums/jj31/itsiceman/Thermocouple Error/4080ab96.mp4

http://vid2.photobucket.com/albums/y41/elkman/K-TypeProblem.flv

 

I took another video with a UEi two input thermometer that I'll look for. I was tring to take temperatures in a glass door wine box. I just got close to the door from the outside and sent both readings bonkers. I believe it was the lights on that one.

 

My cheap DMM is like that, but that 87 that we both have don't act up.

Never had the need to measure a smoking hot discharge line to diagnose a problem.
I'll test some of mine in a oven if you want but if Fujitsu is using a simple thermistor and probally a stainless steel clamp I don't know why I need to prove anything further hear.

When you shove a meat thermometer in meat, how do you know if it's the probe catching up to meat or if the meat temperature is rising too? Thermocouples are usually faster.
http://www.deltat.com/thermistor.html The last couple of degrees take a loooong time to catch up. Fever thermometers don't wait for the thermistor to completely equalize. If it did, it takes about five minutes. It monitors the rise curve and guesses the target. This is fine if you're chasing a steady target.

 

I remember one time in a warehouse having my gauges bounce around when forklifts were running by. Another time on a rubber roof.
Besides you using the wrong thermistor for taking high temps I don't think you've brought any good points why to use thermocouples with a digital gauges while you have been given plenty of reasons why not to use thermocouples. Are you arguing just to argue now?

 

in post #5 i questioned what do you guys use for high temps?

If you use a thermistor that can measure discharge pipe temp when compressors have leaking valve plates reversing valves and and other faults that put out heat hot enough to cook oil, motor windings and paint, then what is it?

 

Never had the need to measure a smoking hot discharge line to diagnose a problem.
I'll test some of mine in a oven if you want but if Fujitsu is using a simple thermistor and probally a stainless steel clamp I don't know why I need to prove anything further hear.

 

Even when you stop charging, there is still residual static. Until it is fully discharged it will affect the readings. Static electricity can be generated at such a high voltages it will also destroy the electronics.

 

I did pull a digital manifold apart a while ago, pretty sure their was a varistor soldered to a washer under one of the pressure transducers, never checked much further, as i was checking a gas leak.

So how does the i manifold prevent static?

 

High voltage diodes

 

yes you can find slow thermistors but there are slow thermocouples as well especially the ones with shielding from external error. just slapping a bare Ktype on another piece of metal is room for more error

 

Was looking at Cooper Atkins and most the regular sensors are rated to 300°F
They also have one for cooking with a silicon wiring good for about 400°F