DMM that measure temperature

[AjohnwParticipant @ajohnw51620]

Years ago I purchased a DMM from Henry's radio because it had a proper socket for thermistors. Not a brilliant meter but comparing it with accurate ones at work it was reasonable.

It's not very well any more so I looked around for a replacement. No luck at all. Most come with a wire thermocouple with a normal DMM plug on the end. The wire ones aren't much use for all sorts of things.

I noticed a Korean one that used an adapter but didn't come with it and no comments about what the socket looked like. A supplier then pointed me at Extech as they didn't stock the adapter for the other one and theirs comes with it. It does have a standard thermistor socket in it and two prongs to fit into the meter. It's under a 1/4 C out at 25C but takes a while for the reading to settle. Same on voltage. I'd guess there is some sort of filtering in it. The Korean one had an add filter button.

As I had just sold something and it came near to paying for it I bought a better quality meter. It's true rms and has a basic accuracy of 0.5%. Unlike others they don't specify it on all ranges but that sort of accuracy in the mV level probably explains the slow to settle aspect.

It looked to me that Fluke have gone down market and also sell some rather expensive ones. It looks like there are a number of much cheaper viable alternatives about now.

John

–

[AjohnwParticipant @ajohnw51620]

aka Digital Multimeters

[Clive FosterParticipant @clivefoster55965]

The Draper Expert DMM14 autoranging meter has a proper thermistor socket and thermistor in the box too. Mine seems reliable, as accurate as any lower end meter, and batteries last ages. I do have proper high end old school Fluke somewhere but have not bothered to dig it out for ages. The Draper was got as a more disposible alternative when the cheap Thandar (remember them!) I kept for out of shop jobs decided that 0 was abit too low for the ohms range to go. Still works fine so long as I remember to mentally correct resistance readings and still gets used.

Daydream official price of nearly £130 but can be got for around half (£70.00 ish) from several suppliers including Lawson HIS who I've dealt with and know to be reliable. Mine cost about half that some years back. Intended for automotive use so has an inductive pick up too along with several car biased ranges including dwell angle.

Clive

[John ShepherdParticipant @johnshepherd38883]

DIgital voltmeters usually use thermocouples not thermistors, if you google DVM and thermocouple you will probably find several to choose from. Hope that help

 

 

Edited By John Shepherd on 21/07/2016 20:45:10

[AjohnwParticipant @ajohnw51620]

The full specs are in the manuals on the Extech's. Never thought of looking. It's pretty good but the "Korean" one by TPI is probably a bit better. Turns out TPI might be American. That one is up with Flukes much better models for a much lower price. The basic accuracy of that one would be stated as +/- 0.05%. I should have said +/- 0.09% on the Extech. The various mode accuracies look better on the TPI one.

I've also used a carefully selected Vellaman one for a long long time as the Henry's one was a bit naff. Off putting even down to a truly nasty plastic yellow case. I'll wonder now about selling the Velleman or sticking it in a draw some where for if. I've had that for a long long time and can only remember changing the battery once. I don't use them daily though. More long periods untouched and then several days with lots of use.

Both of the models I have mentioned are just short of £100 less VAT discounted. I've started doing voucher searches before I actually buy which knocked a bit of that. They are both industrial heavy duty units. The extech is the 520K. K for kit because it comes with a full set of leads and connections for the ends. They are good leads but as is often the case longer than needed really. I'd say the extra few quid for the kit is well worth the money. Large and small croc's and the usual probes.

I noticed Draper but on looking wondered about the price I saw it for. Extech also do some +/- 0.03% but I thought what on earth do I need that for now. I noticed and old latching sliding switch Fluke on ebay recently. If the right one that may be better than +/- 0.1 % on all ranges. I used one at work for lots of years. A job change eventually resulted in a newer model.

John

–

[Neil WyattModerator @neilwyatt]

Much as I like to watch a needle swinging to and fro, the sad truth is that even a budget DMM knocks the socks off my high-end analogue meters for accuracy and not loading the circuit under test.

My only use for an analogue meter now is a signal that changes too slowly for a scope and too fast for a DMM.

Neil

[AjohnwParticipant @ajohnw51620]

LOL Brain problems – of course I meant thermocouple. No idea why thermistor came out. I have a stainless thermocouple probe I can dip in most things and get really hot unlike the wire ones. Also a surface probe. That's why I wanted one with a standard thermocouple socket. Lots come with wire thermocouple that is pre wired to a none standard plug.

Some DMM have a bar graph to give an analogue meter type effect. Ancient big AVO meters with mirror scales percentage wise were pretty accurate. More so than some cheap DMM's. They also had the advantage of averaging ripple that is on all dc to some extent. They sometimes have a better frequency response as well. True RMS dmm's help a bit with that aspect. The fact that analogue meters average can be pretty useful at times. Things like pulsed dc especially.

John

–

[Anonymous]

Posted by Neil Wyatt on 21/07/2016 21:08:28:

Much as I like to watch a needle swinging to and fro, the sad truth is that even a budget DMM knocks the socks off my high-end analogue meters for accuracy and not loading the circuit under test.

My only use for an analogue meter now is a signal that changes too slowly for a scope and too fast for a DMM.

Oh dear, and I've just bought a secondhand Avo Mk8.

At least a moving coil meter doesn't inject noise into the circuit under test.

Andrew

[John Stevenson 1Participant @johnstevenson1]

Somewhere I have 4? big single meters in nice wooden cases.

Real class stuff as supplied and used by Ericsson's telephones some years ago.

Are they worth digging out and getting details off or are they just destined for the skip ?

[AjohnwParticipant @ajohnw51620]

I don't think a decent DVM injects any noise Andrew. There are various types about and some of the accuracy figures may seem a bit weird. Plus N digits for instance when digital aught to be +/- 1 bit.

I suspect few people realise how accurate some of the old AVO's were. For normal measurements they didn't match up to early Fluke DMM's used in many electronic labs of all types when they were widely used. The company seems to have relaxed it's spec's a lot over the years on some models probably to sell more to a wider market. When they first appeared in the UK they were cowboys as far as data loggers were concerned and I don't think that they could match Solatron for DVM's either. Pretty expensive stuff once it gets down to accurate uV levels.

John

–

[davebParticipant @daveb17630]

Posted by Neil Wyatt on 21/07/2016 21:08:28:

Much as I like to watch a needle swinging to and fro, the sad truth is that even a budget DMM knocks the socks off my high-end analogue meters for accuracy and not loading the circuit under test.

My only use for an analogue meter now is a signal that changes too slowly for a scope and too fast for a DMM.

Neil

Absolutely! This is the reason we used analogue meters for testing and digital meters for measuring.

Dave

[Neil WyattModerator @neilwyatt]

Posted by Ajohnw on 21/07/2016 21:57:23:

Ancient big AVO meters with mirror scales percentage wise were pretty accurate.

Rose tinted glasses. My Dad had two AVOs, 1% of FSD on current and 2% of FSD for voltage.

Neil

[Neil WyattModerator @neilwyatt]

Just for comparison a £17 Draper DMM is 1% +/-2 digits for current and 0.5% +/-2 digits for voltage.

Bear in mind that at, say, 1/3 scale the AVO 8 can be 6% out on voltage and be in spec.

But I'm sure yours will look lovely on the shelf Andrew

Neil

[AjohnwParticipant @ajohnw51620]

Those figures don't match their catalogue Neil

**LINK**

My AVO used to get checked regularly Neil along with a lot of others. The error used to be added to a label on it on one range. The scales are organised to get round the low deflection errors.

Andrew does have a point after a fashion. Analogue multimeters are arranged to measure with a certain resistance per volt so the input resistance varies according to the range that is set. DMM's have a constant input resistance usually 10 meg ohms what ever range they are in. That makes them much more prone to picking up noise even via the leads. However some meters clearly take steps to get round that. On modern day electronics the current drawn by an AVO when it's measuring voltage is often too high as it interferes with the reading. The old way of getting round that problem was to use bridge circuits adjusted so that there was no current drawn when the voltage reading was taken.

Clearly we all need to use £17 quid Drapers as some Flukes don't meet that spec.

John

–

Edited By Ajohnw on 22/07/2016 09:32:50

[Gordon WParticipant @gordonw]

I've got digital and analogue meters, the accuracy as such does not bother me much because I don't need it. Nowadays I find I'm using a cheapo moving coil nearly all the time, far more use trying to sort out dodgy wiring on the car. Wiggle a wire and if the needle moves start checking that wire. Just horses for courses.

[Neil WyattModerator @neilwyatt]

Well, the figures I gave agree with the manual http://www.drapertools.com/products/manuals/60792ins.pdf

Fluke's meter cleaning wipes cost more than a cheap multimeter!

Their low accuracy models (3%) seem to be the auto-ranging ones, 0.5% is pretty much the de-facto standard for cheap DMMs; even the A2D inside an AVR chip has a basic accuracy far better than that so I don't doubt that it's easily achievable for cheap consumer items.

Truth is, hobbyists got by fine on analogue meters inferior to the AVO8 (I have an Eagle K1400), and I very much doubt anyone outside a lab really needs the accuracy digital meters offer.

But I last used an Avo 8 nearly forty years ago, so what do I know?

Neil

[SandgrounderParticipant @sandgrounder]

I usually use a DMM but the very high sensitivity can sometimes be a disadvantage, once when measuring the voltage across a 12V bank of NiFe cells one of the leads slipped off a terminal and onto the wooden frame holding the cells and the DMM read 4V, the cells have an live outer casing and when in the wooden frame which was probably a bit damp but not wet allowed sufficient current to pass, it's something to remember especially when working on low voltage circuits, a car switch in the off position with a leakage of say 5 megohms which would be insignificant at 12V would show up approx' 8V when measured with a 10 megohm input resistance DMM instead of the expected 0V.

John

[MuzzerParticipant @muzzer]

The idea that a hand held(?) DMM with flying leads could meaningfully measure anything to 0.1% or better is fanciful. I suspect the published "accuracy" numbers are just lazily calculated from the number of bits in the ADC. 12-bit would be 0.024%, 11-bit would be 0.05% and 10-bit would be 0.1% in this mentality.

Even for a high quality bench instrument or embedded ADC, getting the last bits of a 11 of 12 bit ADC to behave is pretty tough. As for calibration of these meters, who here actually gets their DMMs calibrated each year and what possible hobby activity would require that?

Back to the original question, you don't need a "proper" yellow thermocouple socket as long as the connections are at the same temperature and of similar metals. But same question – what possible requirement is there for deadly accuracy – and how would you know if it was correct anyway?

I also have an Avo 8 (a fairly "modern" MkV) and although (because?) it's like a house brick, I tend to use it in the garage.

Worth bearing in mind that many DMMs have an analogue bar scale as well as a digital display. Not quite as good as a moving coil meter perhaps, but in some cases quite useful.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Andrew Johnston on 21/07/2016 22:03:35:

Posted by Neil Wyatt on 21/07/2016 21:08:28:.

…

Oh dear, and I've just bought a secondhand Avo Mk8.

…

Andrew

The worst feature of the Avo8 is the special 15V battery it uses. My older model wants a BLR121. I thought they were unobtainium these days but I see Maplin's still have them at £12.99 each.

I don't use my Avo much because it sticks in it's leather case and when it is out the thing is inconveniently big and heavy. The 10A AC range is useful though.

Cheers,

Dave

[Neil WyattModerator @neilwyatt]

Posted by Muzzer on 22/07/2016 10:47:03:

The idea that a hand held(?) DMM with flying leads could meaningfully measure anything to 0.1% or better is fanciful. I suspect the published "accuracy" numbers are just lazily calculated from the number of bits in the ADC. 12-bit would be 0.024%, 11-bit would be 0.05% and 10-bit would be 0.1% in this mentality.

The figures I found quoted ranged from 0.5 to 2% +/- a digit or two, which I'm happy to believe for a 10-bit DAC.

Neil

[Neil WyattModerator @neilwyatt]

Posted by Muzzer on 22/07/2016 10:47:03:

The idea that a hand held(?) DMM with flying leads could meaningfully measure anything to 0.1% or better is fanciful. I suspect the published "accuracy" numbers are just lazily calculated from the number of bits in the ADC. 12-bit would be 0.024%, 11-bit would be 0.05% and 10-bit would be 0.1% in this mentality.

The figures I found quoted ranged from 0.5 to 2% +/- a digit or two, which I'm happy to believe for a 10-bit DAC.

Neil

[MuzzerParticipant @muzzer]

Yes, that's a bit more realistic. Apart from the resolution, you have to allow for the tolerances, ageing and temperature coefficients of the reference, ADC, various potchains etc – and the actual calibration itself. And that's before you think about noise.

You can get bootleg BLR121 batteries for £5 on ebay.

[AjohnwParticipant @ajohnw51620]

I need a proper socket 'cause I don't want to cut the plugs of my probes Muzzer so if I buy a meter where the socket is available I may as well do that. A number do come with an adapter. The one I did use had the socket built in.

There are meters about that will measure to insane accuracies. Here is one for instance, long gone now. They mention PPM.

**LINK**

I've used something similar on a data logger with extremely long leads on it. It gave accurate readings and reasonably scan times. This is why I mentioned that Fluke have been cowboys – they did offer a data logger. Looked good spec wise but was entirely hopeless when leads were added. A very common problem with scanning with A/D's of any type. On things like micro's they spec what can be done theoretically. Getting the precision some of them offer isn't that simple especially when scanning.

If you want to see how accurate A/D's can be look around at dual slope integration. They can achieve 16bit at base scale easily. I'd guess that Fluke used this initially when they sort of took over the world in electronics areas.

The spec's vary in content as well. This is for the one I bought

It gives a temperature range for one and accuracy varies in the same mode. As would be expected. In some ways it's worse than some 2000 count DMM's but it's a 4,000 count so really that depends on what is being measured. It's common for digit error to be larger on high count ones. It's true RMS so another bit in the way. More bits add errors so that wins in some respects and looses in others. !0A isn't too good really. That range usually uses a separate shunt. Difficult to fit one in a small meter. I have a 100mv / amp one and it's about 4in long and heavily built to minimise heating effects. I used to need that sort of thing at home.

I've never had a DMM fail on calibration at work. New ones are usually bought when some one starts a new job so in some case that has been well over 10 years. As far as I know they only check them. I've seen AVO's fail because some one bent the needle / seriously overloaded it.

I've not used one with a bar graph either. The IEEE got excited about them years ago especially for technicians so that they could see that they were not measuring a steady voltage. No idea how good they are but where I worked all technicians were provided with one more or less as soon as this cropped up.

John

–

[David lawrence 3Participant @davidlawrence3]

35 years ago the repair shop I worked in bought a very expensive £ 300.00 Gould advance dvm. to measure 350 volts dc ( 1000v setting) on some electronic photographic studio flash as we thought the Mk 8 avo was a bit old. Second time I used it the board arched over and blew up, sent the unit back for replacement under warranty. They would not replace or repair it, they said abuse. We never bought any advance gear again, the boss got out the old avo and we carried on. 30 years on I was still using analogue meters measuring high voltage and fet voltmeters for small audio circuits. Digital is fine when you need to know that exact voltage but a quickly moving voltage can seen fine on an old meter scale.

[Anonymous]

I understand that some of the formal tests for safety qualification of the telecoms equipment mandates that moving coil meters are used in the test set up.

Andrew

[Author]

Posts