Which Digital Multimeter?

[G0HZU_JMR]

You might not, but if the need arose, at least you're aware of the risks, and could select an appropriate instrument.

Many folk on here will have cheap DMMs that say "1000V CAT III" on the front, and would have no hesitation using it when the need arises. Why wouldn't they?

I'd like to think that I will never ever open the consumer unit in my garage but I agree that some people will try and act as an amateur electrician and use a cheapo meter on a consumer unit.

I'd argue that they shouldn't be measuring inside a consumer unit in the first place especially if they are naïve enough to use a cheap Chinese meter for stuff like this.

In my case, I avoid measuring mains voltages (inside equipment) as much as possible. I see it as an unnecessary risk. Too easy to slip with a probe or connect the probes up to the wrong test points and I always like to see equipment designed such that all 230V connections are insulated or hidden from access. Even when repairing equipment that has issues with a PSU I will try and avoid having to measure the primary side voltage. I find that it is rarely necessary to do this. Obviously others will disagree on this but my take on this is that for the majority of home/hobby electronics you are taking unnecessary risks when trying to probe mains voltage test locations regardless of how safe the actual meter is. Even a perfectly safe and indestructible meter doesn't remove all the risks to the user or the equipment.

Originally Posted by Craig Sawyers

Going back to an earlier question regarding the formality of the CE mark itself, this is the logo definition

And as I said, no tolerance is given so it has to be perfect not even an atoms size difference, so unmakeable.

[Heatercathodeshort]

It all depends on how you are going to use it. For radio and television servicing I have the trusty AV0 8 used with the digital meter on the left. I can't remember what I paid for it but I doubt if it was more than £20.
It is totally reliable, accurate and rugged.

The centre meter was purchased from Lidl stores about a year or so ago for £9. Again very rugged and it's home is my toolbox. It gets bashed about, mostly used at the museum at Dulwich.

It also does good service at the local truck garage where I work part time.

For general service work you certainly don't need to spend more than £20 on a meter. Then if it gets dropped, blown up, [likely, especially with television receiver servicing no matter how careful you are] driven over, lost or stolen it costs little to replace.

Regardless of price my digital meters give identical accurate readings.

Just a thought, John.

[McMurdo]

John makes a good point about accidents. I used to swagger about with a fluke in my mobile toolkit and one day went to sort out a garage door opener as a freebie for my uncle. At the top of the ladder I leant my meter on the top of the motor box whilst doing a bit of probing. Unfortunately it slipped off and fell on the garage floor, smashing into a thousand pieces like the chandelier scene from Only Fools and Horses. 'Oh dear, dear' said my uncle, 'hope that wasn't a really expensive one'.

An expensive favour indeed.

[Craig Sawyers]

Originally Posted by merlinmaxwell

And as I said, no tolerance is given so it has to be perfect not even an atoms size difference, so unmakeable.

We're slipping off topic here, but no font definition has an associated tolerance, and the CE logo definition is no different in that regard. It should not be regarded as an engineering drawing, because it isn't. Font or logo resolution (ie tolerance) is determined by the printing process alone.

The only requirement, apart from the logo definition, is that it has to be at least 5mm high.

[Biggles]

My meter at work has a thick rubber protective surround that the meter is press fitted into and has saved it on many occasions from knocks and even the odd drop onto a hard surface. Worth every penny, especially if you work outdoors in a mobile environment.

Originally Posted by Craig Sawyers

It should not be regarded as an engineering drawing, because it isn't. Font or logo resolution (ie tolerance) is determined by the printing process alone.

My last word here, if I had a printing press that only resolved to 6mm would two dots do?

[TonyDuell]

It's odd... I like high-end test gear, Tek 'scope, HP Logic analysers, etc. I even collect old high-end test gear (the more obscure the better!).

But I have had several problems with Fluke meters in the past. And I am not convinced that all these have been solved on more recent Fluke instruments. Alas the documetation for more modern ones is much poorer than that for the older ones so while I can see design problems (not safety-related I hasten to add) in older ones, I don't know about the same issues in modern ones. They may be fine...

For the moment I am using one of the cheaper Uni-T ones. I deliberately bought one with few features in that I released the range switch ratchet was the same on all models in the range, so adding features (frequency, temperature,...) deleted some resistance and current ranges that I would find more useful. I am a little worried about the safety of it. While much of my work involves low voltages (5V or so), albeit at high currents (5V, 100A PSUs are not uncommon in the machines I work on), I do work on mains and higher voltages. While I am not likely to probe around inside my consumer unit when it is energised, I do work on the mains side of SMPSUs, and the energy stored in those capacitors could cause a spectacular failure.

So I have been thinking of getting something 'better'. Any suggestions for a portable battery-powered DMM with (a) proper documentation (at least a circuit diagram and theory of operaton), (b) spares are available (I don't care about spares for a £40 instrument, I do for one costing 10 times that) and (c) proper adherence to safety standards. Oh and (d) a fast-reacting continuity beeper. The Uni-T gets this right It's an analogue comparator across the probes which are energised by a constant current source, so it triggers very quickly, those meters which use the ADC to do a conversion and then see if the digital value is below a threshold -- including some Flukes -- are useless to me.

[mhennessy]

a. I've yet to see a modern meter that comes with schematics - at any price point. But the same can be said about pretty much all modern test gear; it's simply not meant to be repaired like the old days. Board-level maintenance at best. Like all other electronic products today.

b. I know that Fluke keep a good range of spares - see the exploded diagrams in the manuals. I've never had reason to check others, but have read on other forums that Keysight regard hand-held DMMs - even the expensive ones - as completely non-servicable.

c. Look for TUV/UL/CSA.

d. Most decent meters do this. Since Dave Jones appeared on the scene, just about every YouTube review of a multimeter features some furious mashing together of the test leads. The term for this is "latching", referring to the fact there's a monostable that ensures the sounder works for a minimum period of time. I've yet to find a Fluke that doesn't (perhaps some of the older ones were less good, but I've just gone looking - right back to my 8021B - but they all are latching and very responsive.

Dave Jones is working on his own multimeter design, and IIRC it will be open-source hardware. That's perhaps the only way you could satisfy all 4 of your requirements.

[TonyDuell]

(a) I don't care for a cheap instrument, which is likely to use standard parts anyway. But something costing several hundred pounds I do want to be able to repair. I have never trusted 'board level repairs' and never will....

(b) That is one thing that put me off Fluke. I had an LCD fail (leakage, darkening at the corners) and no spares were available. In fact they were most unhelpful. I agree that Keysight (were Agilent) handheld DMMs were not supposed to be repaired, their bench instruments did have real service manuals until quite recently (if not still).

(c) Provided said marks are being used honestly. We've all seen plenty of things with BS
numbers on them that clearly don't meet the appropriate specifications, I can believe it's the same for other such marks.

(d) There are several issues here. One is the minimum time of contact required for the thing to beep. Another is how long it beeps for (too long on a momentary contact would
drive me mad). And yet another is the delay between contact and beep. In other words if a 50ms contact will trigger the beep, it may beep for a minimum of 100ms (monostable tme) but there might be a 0.5s latency between those 2 events (that's perhaps extreme). Now try clipping one probe onto a pin at one end of a complex wiring harness and stroking the other probe down other connectors to find where it goes (something I do a lot of). That latency will drive you mad.

The only problem with a homemade meter is _knowing_ it will meet the safety requirements. Otherwise I would have made my own by now....

[evingar]

Originally Posted by TonyDuell

The only problem with a homemade meter is _knowing_ it will meet the safety requirements. Otherwise I would have made my own by now....

I would assume (if it's like other open source equipment I've seen) that a range of people will take up the opportunity to manufacture to a high standard. If the design is anywhere near contemporary (i.e not the size of house brick), it will almost certainly use IC(s) that are beyond the average home users ability to assemble or replace (BGA's, micro fine pins, solder pads underneath etc).

The advantage of open source is the availability of a circuit diagram, this will at least mean some element of repair can be attempted. You can also be sure that Dave will vet the resulting efforts, as he won't want his name dragged through the mud.

Long discussion about it here

http://www.eevblog.com/forum/project...e-open-source/

A quick scan through and most comment seems to be about the firmware rather than the hardware i.e the capability to customise the meter. I for one aren't that bothered about that, I just want the thing to read Volts, Amps and resistance accurately - and safety .

[mhennessy]

Originally Posted by TonyDuell

(a) I don't care for a cheap instrument, which is likely to use standard parts anyway. But something costing several hundred pounds I do want to be able to repair. I have never trusted 'board level repairs' and never will....

(b) That is one thing that put me off Fluke. I had an LCD fail (leakage, darkening at the corners) and no spares were available. In fact they were most unhelpful. I agree that Keysight (were Agilent) handheld DMMs were not supposed to be repaired, their bench instruments did have real service manuals until quite recently (if not still).

Take a look at the Keysight website - you'll find that service manuals do not contain schematics. It was like that for the last few years of Agilent; one by one, older products (with detailed manuals) got replaced by ones that didn't. A shame, but that's the reality. Test gear - at any price - is simply catching up with the rest of the electronics world - it's becoming disposable.

What matters, therefore, is what the companies will do to support their products. I know you've had a bad experience with Fluke, and I'm sure this will come as no consolation, but you appear to be in the minority. E.g. they had a problem with a leaking "supercap" in the 189/287/289 meters - if you're affected, send them the meter and they'll fix it FOC under their lifetime warranty. By "fix", they replace the main PCB. You get a brand new meter with a cal certificate for free. Under the terms of their warranty, only the original purchaser applies, but in practice they've been doing it for anyone who asks. Pretty good. There will come a time when they no longer have the parts available to do that, but for now, what more can you ask for...

Keysight have a meter that gives wildly inaccurate readings in the face of conducted HF. They gave affected customers 2 options - have an upgraded model now, or wait for a couple of months, and we'll ship you the same one, but with the fix implemented (probably a redesigned PCB to accommodate the mods). And if you ask, they'll give you a USB data lead for free as well. Folk receiving the new meters haven't been asked to send the old ones back...

Putting myself in their shoes, I can see why a manufacturer really wouldn't want to encourage their customers to poke around inside their products in today's highly litigious world, where such an act could render an item less safe and cause reputational damage.

So, much as I can empathise with the sentiment, the world has changed.

(d) There are several issues here. One is the minimum time of contact required for the thing to beep. Another is how long it beeps for (too long on a momentary contact would
drive me mad). And yet another is the delay between contact and beep. In other words if a 50ms contact will trigger the beep, it may beep for a minimum of 100ms (monostable tme) but there might be a 0.5s latency between those 2 events (that's perhaps extreme). Now try clipping one probe onto a pin at one end of a complex wiring harness and stroking the other probe down other connectors to find where it goes (something I do a lot of). That latency will drive you mad.

You forgot the values of resistance needed to trigger it. It's not always included. Having just checked a random handful of Fluke models, this is <25 ohms if quoted.

The response time seems to range from 10us to 1ms. I can't say I've ever noticed a difference between these models. The cheap models I mentioned earlier (the RS-14 and Bside ADM02) are non-latching, and those are conspicuously poor by comparison. But a lot better than the mechanical buzzer in my first multimeter (an Altai HM-102BZ).

The only problem with a homemade meter is _knowing_ it will meet the safety requirements. Otherwise I would have made my own by now....

It will be available as a built, tested and calibrated product - just one that happens to be OS.

[julie_m]

Originally Posted by Biggles

My meter at work has a thick rubber protective surround that the meter is press fitted into and has saved it on many occasions from knocks and even the odd drop onto a hard surface. Worth every penny, especially if you work outdoors in a mobile environment.

Seconded! If you have spent good money on a piece of equipment you expect to last forever, then it's a false economy not to invest a little in protecting it.

And a rubber protector used with a "Chinese Cheapie" provides an extra layer of insulation / containment, which can hardly be a bad thing.

[Craig Sawyers]

Originally Posted by merlinmaxwell

My last word here, if I had a printing press that only resolved to 6mm would two dots do?

How many angels can we get dancing on the head of this particular pin?

[G0HZU_JMR]

I googled the OP's "Duratool Digital Multimeter" and those meters look to be similar to the ones given away free (as in a freebie perk) by some of our test equipment suppliers. Over the years they offered various tools like this to our company, sometimes it would be a DMM and other times it would be some form of mini toolkit. All you had to do to get a free DMM was hire or buy some decent RF test equipment from them.

Obviously, no one was allowed to use these cheapo meters for company based work and I'd like to think that most of them went straight in the bin. They would be OK for casual low voltage, low power stuff but you would have to be brave or desperate to use one of these for high voltage, high power testing.

I have several DMMs here and one cheapo Korean DMM (TMK brand) that gets used for low voltage stuff. See below for an image and I bought this meter many years ago at a rally for about £1. But of the decent branded meters I own, none of them is perfect or ideal. Each one has its good points and bad points and I try and select the best one for each task. The one that is by far the best in terms of specs and cost (when new) is the one I use least because it has by far the worst human factors of all of my meters. But it can do things the others can't so I do use it now and again. For what it's worth I still use my old Maplin Gold DMM for most tasks but I also use a Fluke 45 bench meter quite often.

Note that I've never managed to damage or break a DMM either here or at work. I guess it would be different if I used cheapo meters all the time but you only have to spend a few minutes with a cheapo meter like my little TMK DMM to realise that the build quality is pretty dire. I'd expect the rotary switch to wear out or become intermittent if I used this meter a lot. I tend to leave it on DC volts pretty much all of the time meaning the rotary switch stays in one place all the time as it is an auto ranging meter. I wouldn't want to use this meter for high voltage or high power stuff. eg I wouldn't want to go near mains voltages with it or any high DC voltages.

[julie_m]

I think there is a bit of snobbery going on here!

The CAT ratings are based on contrived, worst-case scenarios. Somebody who is continually making measurements on high-energy circuits is going to have a non-negligible probability of encountering similar circumstances in real life.

Unless there are especially nasty transients present -- which is less likely on a peaceful suburban housing estate, than in a factory environment with heavy machinery switching on and off suddenly -- a separation of 10 mm. from mains is adequate, and PCB traces carrying mains need only 2 mm. separation from anything non-user-accessible. Air at normal temperature and pressure needs a voltage stress of at least 1000 V / mm. to become conductive, and most plastics need more than that.

It's possible for a high-voltage transient on the mains to arc across the gap in a blown fuse, with catastrophic consequences. But for someone who is mainly making measurements for a short time at a time, downstream of several fuses, an MCB and an RCD, it's far more probable for them just to get a nasty surprise from a charged capacitor. And even that is probably a smaller risk than you take travelling to work and back each day. What meter is the best to protect you from drivers on the ring road?

A cheap meter will be fine for occasional measurements on mains circuits, unless you are extremely unlucky or just careless. However, for such measurements I would not be happy without a decent pair of probes. They would have to sufficiently-sized have guard rings to keep my fingers from slipping down and contacting the tip, and ideally double-layer insulated test leads to make any damage self-evident.

If I was making measurements on high-energy circuits all the time, doing it for a living and with a family to feed, then I might well be more exacting with my requirements. And similarly if I was designing test equipment that could only ever be as accurate as the instruments used for calibrating it in the first place. But for repairing radios, TVs and other electrical appliances at home, I am rather more easily satisfied.

[Diabolical Artificer]

Some good point's here, well raised. I guess meter's are like cars or guitar's, some folk get attached to them and call them affectionate names, for other folk they want all the bells and whistles yet for other's they're just a meter.

For me the Fluke 25 is the mutt's nuts, well protected with HRC fuses etc, well built, can handle being dropped off the bench or the odd 300v on the wrong range setting. It's affordable at around £30, has a good manual with schematic and it has autoranging, relative and hold function's and a "analog bar graph" which is useful. I'm not telling you what her name is. : )

A.

For domestic mains use would not a mains plug fuse do?

[Radio Wrangler]

As Mark has said, the world has changed.

Once upon a time there was rapid progress going on in test equipment and the likes of Tek, HP, Fluke and R&S were comfortable about giving full service data with their equipment. This info could be used to copy products, but by the time copies come out, the main manufacturers will have moved on to something the user would really find better.

There is no longer that pace of progress in basic instruments. The new stuff has a few twiddled features and cosmetics, but little to make it something users really have to get in order to fight their competitors. So plenty of low cost instruments have sprung up and the big test equipment firms are scared of their stuff being copied. Service info has vanished - they don't want to help the companies eating at their margins. HP used to publish a monthly journal giving lots of info on how the new stuff worked and was designed. They stopped that back in the nineties. Even the designer's names, photos and short bio's were grist for headhunters.

I'm not sure that stopping the availability of service info slows the cheapo companies down at all. They can just buy one and take it apart. That's another reason why hire companies cover their stuff in 'calibration voided if broken' seals. The cat's been out of the bag for a long time for basic instruments. There is no secret sauce... not any longer. They feel they have to protect their designs like the crown jewels, but it honestly isn't working.

The well known names of the established and trusted firms still have one thing to offer their customers. Trust. The really cheap firms cut any corners they can... see that video on track clearance on the freebie DVMs. The cheapish firms still can't resist cutting some corners.

Occasionally, the big firms have cut corners and made something best avoided, sometimes it was something no-one knew about at the time. Sometimes they buy in stuff from cheapish firms with their logo added. A big name isn't a 100% guarantee of no regrets.

Maybe the best bargains are from the cheapish firms who haven't cut any corners. The only way to know is to open one or watch Dave Jones' EEVBlog.

I have a nice pair of test leads from RS with HRC fuses in the probe ends for chunky work. But what happens if the fuse in the 'cold' lead opens first?

David

[Craig Sawyers]

Originally Posted by Radio Wrangler

I'm not sure that stopping the availability of service info slows the cheapo companies down at all. They can just buy one and take it apart.
David

A year or so ago I had a long meeting with the CEO of a global company who made seismic monitoring equipment. A few years ago they were trying to penetrate the Chinese market.

During a fact finding trip, he was taken to a major outfit who's sole business was reverse engineering non-Chinese products, and selling information to other Chinese companies whose business was copying products. In one of the rooms there were 200 people who's sole job was to decompile embedded software, FPGA code, crack software security features etc. There were many such rooms devoted to mechanical drawings, electronic schematics, design notes etc etc.

Chinese plagiarisation is big business.

Craig