Digital multimeter accuracy

[Caecilius]

I've got Velleman DVM 68 digital multimeter that I paid GBP 9.99 for about ten years ago in a special offer at Maplins. It's does what I've needed it for, although I've never really trusted its CAT III

600v rating so I've not used it on the mains.

As one does, I've thought it's time to get a better one. My excuse being that I want something that I can measure mains voltage with.

So I've looked at Fluke, which seems to be the DMM of choice according to various reviews, and I had my eye on a Fluke 115 which costs about GBP 140.

Then I looked at the claimed accuracy. Taking AC volts (which is generally the worst in terms of accuracy), my el-cheapo claims +/-

0.8% or 3 digits, and the Fluke claims +/- 1% or 3 digits. In other words on paper the Fluke's accuracy seems to be worse!

1% is 2.4 volts at mains voltage. I'd have expected something like

0.1% given the price difference. I wonder if my DMM manual is just telling porkies, or if it really is just as good in terms of accuracy as a modern entry-level Fluke. Maybe I got an unexpected bargain at Maplins all those years ago.

[tabbypurr]

Fluke gets you a given set of functions/specs for a much higher price. Industrial users like them because they're robust & reliable. If you don't need that one can get far cheaper.

If you're after accuracy, try a Datron 1061. Very nice.

NT

[Theo Markettos]

The Fluke is a true RMS meter, the cheapo one isn't. That means if your mains is anything off a pure sinusoid the cheapo one will read garbage, while the Fluke will give you the actual RMS of the waveform that it is seeing. Given that measuring RMS is trickier than just taking peak/sqrt(2), I wouldn't be surprised that the Fluke claims lower accuracy.

I wouldn't believe a cheap meter on AC for this reason.

Theo

[Cursitor Doom]

I recently bought a Megger DCM330 fork multimeter which specifies DC accuracy at up to 1000V of +/-1%+2 digits; AC accuracy same range +/-1.5%

+5 digits.

Perhaps your cheapie meter is exaggerating its accuracy figures. After all, it will not have any rep as such to lose if such claims are exposed as false?

[Archibald Tarquin Blenkinsopp]

What level of accuracy do you want and why?

If you plug your 240V appliance in does it matter if you are feeding it with 220, 230 or 240V?

I get through meters on a regular basis, so I stopped buying Flukes years back. [Liquid contamination not overVolts BTW]

I just buy cheap, Volts to 600AC and a 0-10A ac current range. Apart from that a DC 20mA range and some kind of squeaker on the Ohms range fits the bill.

The question you should ask is what will I use it for, then worry about the accuracy.

If you are really enthusiastic about accuracy, traceability and the opportunity to argue the toss with your local electricity supplier, there are any number of places offering a calibration service. Just read the certificate with care!!

AB

[Brian-Gaff]

Many people seem to want one that can read the display. After all, aside from a blind person, there are times when the ability to get at the right place to see it is not going to be possible when measuring in complex large equipment. Does anyone make a talking meter these days. After all Talking clocks and watches seem to abound and be given away as novelties, so its obviously not a problem of cost. Brian

[The Natural Philosopher]

longer test leads sort that out..

[Caecilius]

I don't think longer leads will make it talk.

[tabbypurr]

Maybe I'm being ignorant, but should someone that can't see be putting metal probes in powered up electrical equipment?

[Cursitor Doom]

Ah, you're being discriminatory. ;-)

[whisky-dave]

But perhaps not enough to make it chreap enough for those that want it.

There's things like sample and hold. Which could be better.

I think so, although expensive, seems like most have been discontinued to stop blind people using them as that was a serious safety concern. Thre's few other tools that are sold for the blind. Angle grinder SDS drills, hammers. I think where there's interfaces such as USB mthere maybe option but expensive.

Maybe I'l suggest it as a studetn project.,

but with voltages, they vary quite quicly with time. what will the voltage be at the first stroke 9.03V 1/3 sec later 9.04..,. back to 9.01 next.

It's do-able but not very useful.

[Martin Brown]

I have an old Fluke and to a large extent what you are paying for is a unit that will survive industrial levels of daily use. I also have several small sacrificial meters of dubious Chinese manufacturer. They don't survive being dropped and stood on anything like as well.

Although not precisely calibrated they are all more capable than the Fluke in terms of functions and features as well as being smaller.

The chipsets for DVMs are all much of a muchness and you are mainly paying for better calibration stability and quality of build.

Rapid do some own brand ones that aren't bad if you only want basic DVM functionality. These days it is hard to avoid getting frequency, capacitance and transistor testing included as well.

[John Rumm]

One of the main differentiators, is the quality of the design and implementation of the input protection. Many cheaper meters will not even be marked with their input protection category.

Flukes will normally be properly designed with shrouding, air gaps, spark points, HRC fuse protection, MOVs etc. Much of which is invisible to casual inspection from outside, but adds substantially to the cost to do right.

Not much of an issue if you are taking readings on low energy equipment, but more significant if you are planning on taking readings on mains equipment in high energy situations.

[Caecilius]

This is very true. Also some cheap meters basically lie about their input protection. My el-cheapo says CAT III 600v / CAT II 1000v. Having looked inside it and seen the simple glass fuse, lack of any cut-outs, and gnats-whisker track spacing on the PCB, there's no way I'd trust it on the mains.

[Cursitor Doom]

I fully agree. I'm so cynical about cheap (sub 10 quid) Chinese meters and their absurd CAT ratings I won't use them for *anything* let alone mains testing. Anything high energy I'll use a Megger/Avo. Some middle-of- the-road Chinese multimeters in the over 50 quid price range are generally fine for electronics testing, though. In fact many professional electronics engineers use them, and Chinese oscilloscopes too for that matter.

[Andy Burns]

Not cheap

can send readings via bluetooth to a mobile phone app which can speak the result, of course a touch screen interface on the smartphone probably isn't very friendly if you're blind either ...

[rick]

My AVO digital meter died ... and I looked around for replacement, I'm not using it every day ... so don't need the robust build (& price of a Fluke) I opted for an Aidtek VC97 ........ liked the large display, the fact was almost identical to Fluke 17B and it came with a temperature probe .. and it was well priced.

Also first meter I have had with built in frequency meter & Transistor tester.

Had it over 2 years and would recommend it without reservation, great meter for ~£20

[harry]

Hah. RMS is the easiest thing of all to measure. Any instrument will measure RMS regardless of wave form. It is the "fall back"position. It is the only thing needed to be known in 99.999% ofcases.

[Martin Brown]

Only if you can safely assume that the signal is a pure sine wave.

Clueless as ever. True RMS is moderately tricky to calculate depending on how complicated the waveform being measured actually is.

[John Rumm]

Is there really no beginning to your knowledge?

Funny how you have to pay significantly more for true RMS meters then isn't it.

or not...