EEVblog #712 - Uni-T UT71E Multimeter (Why Uni-T Meters Suck)

[TheNewLab]

I am surprised at so many pages on this DMM review. I am not sure to be amused or perturbed about all the flaming. Ya, Dave, even  jumped into the fray!

So, I am dropping my 2 cents in. I've been watching Dave, since he was doing this in his garage. This review is just his style..Fluke is always the gold standard. I have no idea when Gossen got into the Electrical test gear market. Their light meters have been around as far back as the ...1960's?? They are great meters.
Their 3 color light meters were great and many pros swear by their reliability.

The UNI-T is Chinese made. Talk inside China say the company wants to compete in the same market with Fluke. Obviously not yet. Remember The grab bag episode where Dave received an analog multimeter? I believe it was made by the Beijing Research..blah blah! HA! HA! loved it.

Regarding CAT ratings. many companies are still behind the curve on the new standards issued recently. And certification? UL is the only outfit what is independent and tests the product to earn the UL trademark  I wish the EU would put some teeth behind the CE certification. Someone to cross check UL.

Now for what I would like to see!!

1.) with battery compartment closed drop it a few times! better yet slam it on the floor a few times! Let's see if the battery compartment pops open..
2.) Try out some Chinese fake Ceramic fuses and pump say 15 to 20 amps through that baby and see what happens. Maybe this one is not such a good idea. how about 8 to 10 amps and just let it simmer.
3.) How about sending it over to Martin Lorton and he can freeze it then bake the DMM!  I could not stop laughing when he did that test early on...I love that guy!

I would guess Dave, that with that review, the meter will just get tossed into some dusty box, or in a dumpster.. So why not have some fun?
- Extech did a video slamming their DMM into a wall.
-The Lightware  company filled up their camera case with Hasselblad equipment and drove a truck over the case repeatedly.

Finally, one thing I would like to see is someone actually adjusted the trimmer pots on a multimeter with those cheap ones to see if it could be brought into a tighter spec, then see if it stays there. When I look at the inside of such multimeter. I have this incredible urge to just experiment with each trimmer and see what each one actually adjusts.

OK, that is my personal rant about the UNI-T 71E review!

[TheNewLab]

Oh, Perhaps I should add I am clearly an amateur hobbyist. That obvious?  I do have a kind of twisted, mischievous streak in me.
I also have the Uni-T  D71D. I am very happy with it. the precision is way beyond what I need. It matches my two Amprobe meters. also way more accurate than what I need. 

[f4eru]

>> I have this incredible urge to just experiment with each trimmer and see what each one actually adjusts.

These are uncal pots ))
Each and every one of these pots will put the meter out of cal, coz it's just barely in spec with no aging/ temp/whatever.

[Rick Law]

May be because I am not an engineer, what I see is the UniT meters being built as an OEM platform:
- Glass fuse sitting on HRC fuse foot-print
- MOV marking and component holes
...

So what they were making is a platform for an OEM-buyers to refine this OEM manufacturer's meter into the OEM-Buyer's own.  Say I am the DMM maker seeking OEM, it is a platform for me to specific one with MOV, with HRC fuse, with this goodie or that goodie to differentiate my XX71 meter and make it more suitable for my selected market; whereas, Joe elsewhere may specify the most basic for BigCrap71Extreme model for his market.

It obviously worked some, as reported in this forum, some other makers do put they brand name (and enhancements) over what would otherwise be a bare UniT meter.  But it obviously didn't work well enough.  It fell so far below target that they need to market it under their own name, which OEM manufacturer are generally reluctant to do - to compete with their own customers in general is a no no.  As "an OEM platform" also explains their somewhat crazy product line-up.  Perhaps 71ABCD and E were not expected to be sold under the same brand name.  Instead, the 71A under brand name X might have been competing with 71B under a different brand name.

If indeed they begun life as viewing themselves as OEM manufacturer, it follows that they prepared their manufacturing (finishing) for those who would have specified enhancements that qualifies the meter for certain labels (CE, whatever) per OEM customer's enhancements.  But when they started selling it under their own brand, they decided to "just enable all" the labels - which of course is not a good thing to do.

While it worked (as OEM manufacturer) to some degree, they don't have a big enough OEM manufacturing market;  Yet they are not quite prepared to be on their own as their own brand name.  The more they sell it under their own brand, the less they are a desirable as OEM manufacturer.  They were at a catch 22.  Now they reached a point where perhaps they are no longer look at as OEM manufacturer but a DMM maker that also OEM's.

So, they have to alter their behavior - start earning a name for themselves under their own name plate.  Clean up their own product lines, begin marketing themselves...  most of all, decide which niche they want fit in (or dominate).

If UniT is a child, I would use the word "encourageable" to describe them.  I would also tell them "honesty is the glue that make value sticks to the brand".

[miguelvp]

On my UNI-T UT71B my MELF resistors are straight and they used the proper footprint ones. Also the clearance to the shunt is larger than the E.

Seems all the boards are different, but mine is very similar to the 71D

[TMM]

2.) Try out some Chinese fake Ceramic fuses and pump say 15 to 20 amps through that baby and see what happens.

Nothing exciting will happen, it will just go open circuit. Same goes for a glass fuse.

When things get exciting is when you use one to put a stop to hundreds or thousands of amps with thousands of volts behind it. The result is that the fuse cannot put a stop to current flowing and you get a big boom as the insides of your multimeter burn up.
http://en.wikipedia.org/wiki/Breaking_capacity

[BradC]

Nothing exciting will happen, it will just go open circuit. Same goes for a glass fuse.

When things get exciting is when you use one to put a stop to hundreds or thousands of amps with thousands of volts behind it.

I wonder if you could do some fuse tests with a TIG welder with HF start. The fuse would blow, then the HF start would chuck a couple of thousand volts through it to generate a plasma which should draw a couple of hundred amps.

[TMM]

Daves already done multimeter destruction before 


A test with longer surge duration would be fun though

[Wytnucls]

I noticed all these amateur high voltage/current tests are always performed on cheap dubious quality meters. Never on Fluke or equivalent equipment.
Ideally, the well documented IEC CAT III 1000V test should be replicated to show how a meter with a UL listing passes the test, while a lesser meter doesn't.

[Maister]

Where did you buy it?

My girlfriend bought it from www.reichelt.de (Germany) (X-Mas present for me )

I had a hunch that it was from Reichelt. I wonder if they are the only ones selling these GS + TÜV marked meters or if other German shops also have them. Or put another way: Are they a special version made for Reichelt? I also wonder why they aren't branded differently or at least as another series to differentiate them from the lesser versions.

You are right. It Comes with a "CE - www.reichelt.de"-sticker on May be a Special Version for Reichelt.

[DanielS]

Nothing exciting will happen, it will just go open circuit. Same goes for a glass fuse.

When things get exciting is when you use one to put a stop to hundreds or thousands of amps with thousands of volts behind it. The result is that the fuse cannot put a stop to current flowing and you get a big boom as the insides of your multimeter burn up.

But there should not be thousands of volts at the inputs of a multimeter only rated for 600Vac/1kVdc. Thousands of volts would fry any meter not intended for that range and from there, anything goes. It takes only 60-80V to maintain an arc once it has been struck - that's why standard breakers and many switches are only rated for 24-30VDC, so there really is no need for thousands of volts. If you want to make air explode, you can use a low-power, high-voltage, high-frequency transformer to create arcs and then hook up a 400V 470µF cap via a small inductor to avoid extinguishing the arc to dump current in the arc... that was one of the loudest electrically-generated bangs I had ever created.

Anything other than a nearly perfect vacuum will arc over if you give it enough voltage and blow up if you give it enough current.

[m100]

I was looking today on E bay and saw quite a few 87's that were supposedly like new with protection film still on etc the probes did not look right and they were asking a third of the price as a buy now that other used were selling at auction for.

Not sure what price that equates to but current retail (RS / Farnell) is GBP400 +VAT   I picked a brand new boxed one on Ebay UK about three years ago for GBP228 delivered.  Nice meter, pity about the arse about face AC/DC current selection

It costs very little to nothing extra to properly engineer something.   To engineer to a deliberately degraded quality and mark otherwise IS CRIMINAL.   The directors,  accountants (and maybe the engineers)  should be behind bars for a very long stretch.

IMHO kit that doesn't meet, and has no possibility of meeting its claimed ratings should be crushed.   It's not protectionism, it's not colourism aka Fluke yellow trademarks, it's ultimately saving lives somewhere and until the Chinese and importers and retailers (like Amazon etc) realise this then they will continue to sell substandard kit and get away with it. 

The 'China Export' mark claimed further up this thread to be an urban myth exists, we've all seen it, it's not a CE mark as the characters have the wrong form and spacing, but I wouldn't rely on any of those markings nor the CAT marking on anything but a well known brand with a long established reputation.   (By the way I've no objection in general to Chinese kit and have a number of power supplies, a hot air rework station, very low voltage test kit and a Rigol 1052)   

[G7PSK]

There seems to be quite few Flukes from China at under £60-00 I cannot find the seller who had several as new Fluke meters still with screen protector on at £98-00 each either they have all gone or the add has been taken down.
Personaly I have only come across 2 deaths from electrocution one was a bulk tipper driver who raised his buck into overhead 11000 volt lines and the other was someone welding inside a silo he was killed by 86 volt OC from a welding plant. I have searched through the HSE web site and cannot find a case of someone getting killed or injured by an exploding DMM, I am thinking that this might be the kind of event that ocurs more in the US due to the way the distribution system works there, the pole tranformers are set up so that they are auto transformers using the same line for earth and neutral for both HT and LT sides.

[Fungus]

I have searched through the HSE web site and cannot find a case of someone getting killed or injured by an exploding DMM,

I posted an example earlier in the thread: http://ecmweb.com/arc-flash/case-deadly-arc-flash

[AG6QR]

Nothing exciting will happen, it will just go open circuit. Same goes for a glass fuse.

When things get exciting is when you use one to put a stop to hundreds or thousands of amps with thousands of volts behind it. The result is that the fuse cannot put a stop to current flowing and you get a big boom as the insides of your multimeter burn up.

But there should not be thousands of volts at the inputs of a multimeter only rated for 600Vac/1kVdc. Thousands of volts would fry any meter not intended for that range and from there, anything goes.

That's the point of CAT ratings.  Transients DO occur, and a CAT rated meter must be designed to handle them.  A 600V CAT IV rated meter is required to be tested to withstand 8000V transients.

http://www.testequipmentdepot.com/application-notes/pdf/safety/abcs-of-multimeter-safety_an.PDF

[G7PSK]

I have searched through the HSE web site and cannot find a case of someone getting killed or injured by an exploding DMM,

I posted an example earlier in the thread: http://ecmweb.com/arc-flash/case-deadly-arc-flash

I was referring to UK incidences of arc flash there seems quite a lot in the US very often with thunder storms in the story but I cannot find any in the UK government archives that was triggered by a test meter.
http://www.hse.gov.uk/

[Wytnucls]

Industrial electrical fatalities (USA):
http://esfi.org/index.cfm/page/Workplace-Electrical-Injury-and-Fatality-Statistics,-2003-2010/cdid/12396/pid/10272

[MBY]

If Uni-T had official ties outside China, that would expose them to lawsuits for wrongful death or injuries attributable to failure to meet their safety rating claims if there has been any.

Well they sort of do. Uni-T meters are sold under the "Tenma" brand through Farnell/Element14 for example. If anyone knows if the Tenma variants are built safer I'd be interested to know.

I have checked the insides of my Voltcraft (Conrad) VC-940 a few times. I bought it maybe 4-5 years ago. Since Uni-Ts letter scheme is so confusing, I'm not sure, but I think this is the same meter as shown in Daves video (AC power measurement, 40,000 counts but no data logging [but working RS232 interface]). My unit is clearly better inside. By "better" I don't mean "good". No HRC fuses no MOVs (or only small MOVs), but no stupid screw holes in high current path (or any path I know of) and a proper TRMS circuit in a metal can. And only one 10-turn trimmer and differences in board layout.

I share the opinion that some Uni-T meters is quite good, but the 71-series is not among them. My VC 940 is prone to freezes, have terrible battery life (since I use it mainly for AC power measurement with logging I have hacked in a isolated DC-supply replacing the battery) and confusing to use in the menus.

As a AC power meter with True RMS it is quite decent (but, as Dave points out, it is only suitable for higher loads and when accuracy may not be the most important thing) and stable, as long as the load is "nice". Loads with terrible power factor or fast, inductive changes, the meter clearly has some EMC issues and can freeze.

When I think about it, this is my second VC-940. The first one was DOA.

But I very much like my VC-820, an Uni-T UT60E I believe. A terrific little meter with optic RS232 and a battery that lasts for months or years (or weeks, if you run a datalogging experiment with the meter constantly on).

Another one is my "VC-1008" (yeah, all three is from Conrad) DMM/oscilloscope which of course is a joke as far as "oscilloscope" goes, but it does one thing very well (and thats what its for): oscilloscope function on all (well, except ohms and cap range) ranges, including 10A-range. And this is mainly what I use it for, current measurements in the time domain.

So, mixed feeling about Uni-T. They are cheap, and you get what you pay for, but sometimes, just sometimes this is also exactly what you need. I really value my UT-60E/VC-820 (in contrast with Dave I really value PC-connectivity in my DMMs. Especially RS232 and with a documented protocol [no matter how retarded or overcomplicated. Uni-T meters actually sends strings with a "bitmap" relating to the display segments. Awful, but documented, so you can make your own software]).

[open loop]

Further to my post on page 4.

I have made some measurements on the Tenma 72-7730A (Farnel version of the Uni-T U71A)

This is what I had found when looking at current measurement:

When the selector switch was at VDC the consumption was 4.5mA,
At VAC position the consumption was at 9.0mA,
At the ohms position it was at 6 to 7mA
When measuring temperature it was at 10mA
The back light in low mode seemed to add 5mA

Now I let it go to low power/auto shut off mode and then made the following measurements:

When the selector to VDC and the meter going to sleep consumed 0.578 mA
When the selector was on temperature it was using 1.167 mA

As we can see the battery life would not be great therefore we only use re-chargeable batteries in these meters. Supposing we had a 560 mA/h Duracell and a current of about 5.8 mA (assuming that we would be mostly measuring voltage and resistance)

This gives a battery life of about 97 hours so Wytnucls was spot on with 100 hours a few posts later. He also pointed out that the mains on ohms test would cause no drama, thanks for pointing that out. Anyway I never really liked this meter (it was bought before I joined the department I now work in).

I think Uni-T could save money by doing two things.

1 Do the front end of their meters properly using the quality components. This would mean only one BOM and the quality front end parts then would be bought in quantity and therefore cheaper. This would also stop there being several versions of one product.
2. Cut the number of meters in a range just have a couple in each price bracket as I can only imagine it would be way too confusing choosing a meter, like going to the restaurant with a 10 page menu for just the food.

But more importantly to save reputation rather than money.

DROP THE FAKE CAT MARKINGS  and actually test the products they produce to prove it meets the CAT rating.

[MBY]

Dave,

We have the Tenma branded version of these meters at Work, don't ask why - long story. Next time you are in the lab check out the battery consumption of these meters and then do a Blab rant about the battery consumption.

These meters are not terrible but not great either, ok if you are doing low energy work. (Next breath) I like to see your mains on the "ohms range" test.

One non-bad thing I can say about my Uni-T meters are that they DO survive the "230VAC on ohms test". My VC-940 (UT-71E(?)) is "tested" several times. Every time I forgot to turn off the auto power off function while in power meter mode, the simplest way to bring it back is to turn it off and on again, sweeping through ohms, mV and V ranges. It does squeak while mains AC on the inputs in ohms mode, but it has so far survived without problems. I also do this when it freezes.

VC-940 and Amprobe PM55A is the only two meters I ever saw freezing. Actually, in that regard, the Amprobe PM55A is much worse. I even have to "overvolt" the damn thing, squeezing two CR2025 batteries in series instead of one CR2032. That keeps it stable and does not affect it in other ways so far (apart from killing the battery economy...). 

[eCloud]

Dave shows why Uni-T multimeters generally suck in this teardown (and a little bit of a review) of the $180 UT71E 0.025% accuracy multimeter with power measurement capability.

For all interested in the power measurement capabilities of the UT71-E, I have attached some interesting pictures.
7.5 Watts going into my inverter circuit and (tada ....)
724 Watts coming out.

The UT71E measures 295V while only about 45 V are there.

By the way: The load is purely resistive. Cos Phi is also off.

[G7PSK]

Wow Ecloud you have cracked it with the help of UNI T. You have proved free energy from over unity. Joking of course but perhaps this sort of thing leads to some people being convinced they have.

[Wytnucls]

Dave shows why Uni-T multimeters generally suck in this teardown (and a little bit of a review) of the $180 UT71E 0.025% accuracy multimeter with power measurement capability.

For all interested in the power measurement capabilities of the UT71-E, I have attached some interesting pictures.
7.5 Watts going into my inverter circuit and (tada ....)
724 Watts coming out.

The UT71E measures 295V while only about 45 V are there.

By the way: The load is purely resistive. Cos Phi is also off.

I don't know why people try to operate their equipment outside of the manufacturer's specifications. 50V AC is the minimum input voltage.
Does it give normal W readings when you plug your 220V kettle in the adapter? If not, the meter is obviously faulty.
Furthermore, there is no mention in the manual of using the W feature without the power adaptor, so was most likely developed for mains voltage testing.

[asp]

So my UT71E (V20051120-6) looks different ( take a look at the bottom side ). The traces look better spaced from what I see and there is no trip pot at the top of the fuse either.

I think I ordered this directly from China (via ebay) some time ago.

[Ericho]

I share the opinion that some Uni-T meters is quite good, but the 71-series is not among them. My VC 940 is prone to freezes, have terrible battery life (since I use it mainly for AC power measurement with logging I have hacked in a isolated DC-supply replacing the battery) and confusing to use in the menus.

Yea, my VC940 freezes from time to time as well