uni-t ut71c temp sensor upgrade

[spikey1973]

hey guys,

well yeah i own a ut71c, and i really don't wanna go in to the discussion if the was or wasn't a wise choise.
but do have a question on something else.
the ut71 has a temp sensor setting, which when you use it it says there is a max input of 1000!? (not saying temp there)
on the otherhand the manual says that the "included contact point temp prode" can only be used till 230C

this looks to me like the situation that the dmm itself can handle higher temps then the contact point.
as i would like to use this in regards to my 3d printer i would like the max temp bar to be slightly higher then 230 as 230 is about the temp for ABS.
300- 400C would be nice, although 400 maybe also a bit much.

anyway so a solution to this could be to just change the contact point, actually have some spares lying around from my printer, new hot-end came included with new.
but then there is the problem of not knowing jack s#/t about these sensors but i expect ech sensor / contact point has it's own measurement curve.
adding to that, is that i have no clue what kind of probe is included.

meaning that if i just change the prope i could end up with completely wrong temps and with no means to re callibrate (properly) / reprogram.

so basically what i'm trying to ask is this..
1) is there a way to find out what probe is used? or is this even needed?
2) if there are probes with the same / similar outputs (same kind) which can measure higher temps, or can i simply use the probes from my printer.
that is without wrecking the dmm ofcourse.

thanks in advance!

[Paul Moir]

The manual says it's a K type thermocouple, so you can use any K type thermocouple you want.  You can even make your own out of K type thermocouple wire.  Note that the maximum temperature is restricted by the insulation melting more than anything else.

The K type refers to the particular two dissimilar metals that the wire is made out of.  Fortunately K is the most popular by far.
 
EDIT:  Chances are the probes that came with your printer are also type K.  Just check them against your current probe.  A degree or so disagreement is pretty normal, but they'll likely disagree by more than 10 degrees if they're different types.

[spikey1973]

thank you so very much..

this sounds really much easier then expected.. i will go and check it out.
But really.. one more question, i have here the manual in front of me and i really wonder where you found the info in the k-type
(this would've given me a start to go and look for myself) never the less. i will give it another proper read through.

Many thanks

[Macbeth]

It is a K-type. Pretty much every DMM with a temperature probe setting is K-type, but more expensive DMMs will add all the others.

Also, the 230 C limit - it only means they have rated the meters guaranteed accuracy within that window, and the cheap K-type probe too.

The good thing is it appears to actually have a cold junction compensation, a lot of DMM's will just assume the ambient temp is 25C and work out the compensation from that. This one has a temperature probe at the inputs.

I'm pretty sure you can use your UNI-T K probe at well over 400 C. The insulation may char up a bit, and the accuracy is not going to be so good (but what's +/- 10C at these temps anyway?)

I think the reason for the 230 C limit is that the thermocouple wiring could be inside an oven and the insulation is only rated to 230 C. Your application appears to be point contact in ambient so it's going to be fine. Note that the thermocouple wires are welded, not soldered, at the business end.

[Paul Moir]

Maybe I looked at a newer manual:
http://www.batronix.com/downloads/UNI-T/UT71CDE_Manual.pdf
Page 6 where the unpacking instructions mentions it's a type K.  It also says it itself is only good to 230.  The meter's accuracy is stated up to 1000C (a -40-40 step, a 40-400 step and a 400-1000).
Since the thermocouple itself generates a (very small) voltage, if you get your wires backwards you'll read temperature going down as it actually goes up.  There are colour codes for the wires but it's like the old joke about standards:  there are so many you can choose whichever one you like.



 

[spikey1973]

thanks macbeth,

again very informative. to be truthfull, i didn't even consider the fact that the probe thing was a thermocouple
notice me never using the trerm in the question. in hind site, rather silly of me.
also i found the reference in the manual so. another blond thing to do (or better said, not to do)
then again, i had no idea that the principle of thermocouple was so simple.
i should have been able to figure all this out by myself.. but i really did not know where to start. 

non the less i feel like i gained a lot of knowledge tonight.
i'll go and mod the current setup into a nice rod setup so it will be easier to use for my application.
or just leave it as it is and mod the remaining thermocouples from the printer i got lying around after confirming it is indeed type k
looks like a nice and simple thing to do.

thanks again!

[spikey1973]

yeah i found it now, just looked at the wrong place.. or the wrong time.. all the same in the space time continuum isn't it

[Macbeth]

Seebeck effect in action. You can find NIST tables for the tiny voltages involved. Ideally you would have the other end of the dissimilar metal wires in an ice bath, but DMM's will just have a fixed constant of 23 or 25 degrees compensation, or an actual local cold junction thermistor located as thermally close to the end of the wires as possible and apply a calculation to get her alright. 

[spikey1973]

oke.. i was a bit quick,
so maybe i can ask for another thing on the topic.
i think i understand the way the supplied thermo couple work.. it has that nice spot welded bead at the top.. and the thermo probe at the input. 
but when i look at the thermol "sensor" wire that i ripped from the old hot end i find two long copper wires on each side. these wires are crimped to what looks, and feels like the different thermocouple wire bits. But there where the bead is expected i find something that looks a lot like a small smd zener diode. (probably isn't but looks like it). is this just the thermocouple contact point protected by a bit of glass or is that a completely different setup?

leaving me to wonder if the by macbeth mentioned cold junction compenstaion at the input will still be functional as it only contacts the copper wire and not the thermo couple wire.
(that is if i really understand what is going on) . on the other hand. It is the same for the supplied thermocouple. it's just that the bit of copper wire are a lot shorter, and the thermo couple wires a lot longer.
so maybe i already answered my own question here 

except for the "zener diode" part

ps: @ macbeth: where did you find reference to the cold juinction compensation with the ut71 i googled but did not get a relevent hit

[Paul Moir]

Yeah, well I wasn't expecting to find it in the unpacking instructions either.  CTRL-F in the PDF reader took me there...

Those glass things sound like thermistors, so they won't be compatible.  Measure their resistance.  Thermocouple junctions naturally look a lot like a dead short.

(PS - you don't have to weld the wires together.  In a pinch twisting them together works.  Or crimping in a crimp connector.  Or silver-soldering.)
 

[Macbeth]

ps: @ macbeth: where did you find reference to the cold juinction compensation with the ut71 i googled but did not get a relevent hit

I am only going on Wytnucls review of these meters way back when, "The cold junction temperature is tapped on a transistor PN junction on the middle of the PCB and with the provided K-type probe, measurements are good up to 230C. Accuracy is quoted at 1%+30 above 40C."


Plenty of interesting stuff in the review of this much maligned meter! 

[spikey1973]

@ paul: yeah guess they will be thermistors.. sound familiar indeed. so a no go on that part.

@ macbeth: hmm, that is a revoiew that i will have a proper read through thanks for that link and indeed much maligned meter. but reallu for what i need it most of those problems are of no concern..
i'm happy with my purchase (actually first bought a ut61e, but a friend needed a replacement dmm quick and wanted to buy one too, while i found still one ut71c being on sale. so i sold mine to him and purchased this one.

[Macbeth]

To be honest the reason the UT-71 range are shit meters to me is because they have the slowest continuity test of all meters ever. You won't hear that beep for a good half second or abouts. When you are testing a board you want to waft the test probe over the pins of a huge pin device and nail it quickly.

Also they have an artificial zero and you will find measuring the same exact volt ref in reverse can be many counts out - and consistently, so why do they not fix this in the firmware yet present a false 0.0000V?

Now the big problem to the Elven Safety crowd is the originals have absolute shit fuses and little protection which is important to linesmen working on electricity substations and electricians at the meter cupboard, but pretty much nobody else. The Euro version of this meter does have the required fuses. It costs a lot more too.