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

[TMM]

Why is everyone throwing around the "Cheap $15 meter" excuse for a $180 meter? You can buy cheaper meters which do actually stand a chance in hell of meeting their ratings. Price isn't an excuse in this price range.

Ethics of improper CAT ratings aside, you aren't going to find a meter which has the same features for any where near the same price. If you can overlook the pitfalls suddenly $180 doesn't look so bad when you consider that a 40000+ count datalogging meter like the Fluke 289 will set you back $540 and a power multimeter like the Gossen Metrahit Energy will hit you (no pun intended) for an eye watering $1200.

The current going price for a UT71E from a chinese seller on ebay is US$169.90 including international shipping and there is a Australian seller who has them for US$147.22 including national postage.

[Wytnucls]

BTW the safety issues with this meter has been known and published here for years.
https://www.eevblog.com/forum/testgear/a-list-of-multimeters-that-do-not-appear-to-meet-their-claimed-safety-specs/
and a quick search for UT71E would have brought up almost all the issues with it too. I was actually very surprised to see one in Dave's hands for a review after its history here.

It is old news. There are better UNI-T meters now, but they don't do Watts anymore, which Dave was after. I don't know what is written on the box, but the manual clearly stipulate that test voltage must fall between 50V and 250V AC only.
The best and cheapest meter for that type of measurement including energy, would be a second-hand 310,000 count Gossen 29S, with 0.1uW resolution, which usually sells for less than 400 Euros in Germany.

[Hootch]

Bought my UT71C from Reichelt, same fuses as yours, it came with some kind of compliance note, don't remember what it was about.

Yes indeed, my dmm is also from Reichelt. As you mentioned, there was a note. After a little search, I found it finally:

"Manual amendment
In order to meet new CE Standards, the following amendments are made:

Alligator Clamp - Removed

Safety Measuring Cable - Removed"

Nothing about fuses...

[allikat]

Re the Uni-61E:

BS 1362 specifies sand-filled ceramic-bodied cylindrical fuses, 1" (25.4 mm) in length, with two metallic end caps of 1/4" (6.3 mm) diameter and roughly 1/5" long. The standard specifies breaking time versus current characteristics only for 3 A or 13 A fuses.

For 3 A fuses: 0.02–80 s at 9 A, < 0.1 s at 20 A and < 0.03 s at 30 A.
For 13 A fuses: 1–400 s at 30 A, 0.1–20 s at 50 A and 0.01–0.2 s at 100 A.

Source: https://en.wikipedia.org/wiki/AC_power_plugs_and_sockets_-_British_and_related_types

Even if those fuses meet spec, are they good enough protection for your meter?  I mean, they claim to be BS1362 fuses, but are the wrong colour!

BS 1362:1973, allows any fuse rating up to 13 A, with 3 A (coloured red) and 13 A (coloured brown) as the preferred (but not mandated) values when used in a plug. All other ratings are to be coloured black.

Source: Same wiki page as above.

I know how they do these meters for the price, the same way a lot of junk electronics companies do. By bloody lying and using parts off the back of whatever's passing at the time.

By the way, the CE marking usually means "China Export", and it's an intentionally similar to the CE European conformity marking.

[AF6LJ]

Having seen those Uni-T meters at hamfests I was really wondering how good they were, now I need some Eye Bleach. GAAACHH What a pile of damp Cat Crap.
I guess I will still be looking for a new decent meter, my Beckman (Labeled as such and not Wavetek) RMS-225 I was given by Beckman to evaluate 24 years ago will have to fill the bill for some time before I retire it.

It is amazing the crap they pass off as test equipment these days.

[Monkeh]

Even if those fuses meet spec, are they good enough protection for your meter?  I mean, they claim to be BS1362 fuses, but are the wrong colour!

BS 1362:1973, allows any fuse rating up to 13 A, with 3 A (coloured red) and 13 A (coloured brown) as the preferred (but not mandated) values when used in a plug. All other ratings are to be coloured black.

Source: Same wiki page as above.

And the 1A and 10A fuses shown in the UT61E are.. black!

By the way, the CE marking usually means "China Export", and it's an intentionally similar to the CE European conformity marking.

No, it means nothing of the sort.

[Wytnucls]

The fact that the chip is 20k count but the meter claims to be 40k count is pretty damming. Fake parts too... So unnecessary, there are quality low cost parts from Chinese manufacturers that will do the job, no need for a fake label on them.

Dave, maybe you should look at getting some equipment for testing installation at the rated voltages that manufacturers claim. See if you can get the magic smoke out.

I was worried when you removed that jumper, could have cleared calibration data or something. Seems okay though.

No need for the "one hung low" either, but otherwise an interesting and enlightening tear down.

It pays to read previous posts. Post #12 will tell you that Dave was wrong, the count of the Cyrustek chip on the 71E is actually 40,000, not 20,000.

[Monkeh]

By the way, the CE marking usually means "China Export", and it's an intentionally similar to the CE European conformity marking.

That's an urban myth, similar to the one about the Japanese town of Usa.

You are correct about the fuses though.

The fuses are black. How is he correct?

[G7PSK]

All this talk about the uni T meters has got me looking at other meters as I was thinking of getting another meter, so does anyone know anything about the Megger AVO410. I was thinking of getting one.
http://www.tester.co.uk/megger-avo410-digita-multimeter-cativ-600v

Or am I doomed to getting a yellow peril (Fluke)

[Wytnucls]

Perhaps not relevant to the thread, but your Megger is an Appa 73 meter. You're doomed.

[Lightages]

Which also appears to be an Ideal meter.

[TMM]

The fact that the chip is 20k count but the meter claims to be 40k count is pretty damming.

Dave made an oopsy - he pulled up the datasheet for the ES51922 instead of the ES51966 which is actually what is in the meter. It's a 44000/440000 count meter chip http://www.cyrustek.com.tw/spec/ES51966.pdf
There is however a catch - on resistance and current (at least when set to amps) the resolution does seem to be lower than 40000 because i've noticed that it has a tendency to snap to certain increments. It is perhaps operating in 4000 or 4400 actual counts with some averaging going on.

[lapm]

Edit relating 20.000 count crap...

Also if they lie about cat ratings, what else do they lie about? Or what else hidden treasures that meter has hidden inside it...

[Monkeh]

Ok now im seriously curious how they get 40.000 counts out of 20.000 count chip set...

You should really read the last few posts.

[Grapsus]

I don't get it why so many people think that Dave was somehow unfair with this meter and Uni-T in general. What is all this justification stuff ? Are you trying to make yourself feel better about having wasted 180$ and bought a crappy and dangerous meter ? There is a lot of alternatives : 1W resolution mains power meters can be bought for 20$ in any PC shop and for 150$ you can have a used 20000 counts Fluke.

The PCB is a horrible mess, it has every possible 5$-meter cost reduction trick except for a few precision parts to meet the announced precision. Look at every photo posted here, there is silkscreen for a spark gap "SG1" but it is always missing. Aside from cost reduction it looks like the designers just don't give a shit about their product, a screw hole in a 10A track, seriously ?!

Yes, from a single teardown like this we can deduce that Uni-T is a company that reduces costs, plays tricks on its customers and puts them in danger by claiming unmet security standards. Therefore all their products should be considered as dangerous crap until proven otherwise. And if they don't like this negative bias they are welcome to make good quality stuff for several years as others have done to earn their reputation.

[TMM]

I don't get it why so many people think that Dave was somehow unfair with this meter and Uni-T in general. What is all this justification stuff ? Are you trying to make yourself feel better about having wasted 180$ and bought a crappy and dangerous meter ? There is a lot of alternatives : 1W resolution mains power meters can be bought for 20$ in any PC shop and for 150$ you can have a used 20000 counts Fluke.

The ~$170 E variant is a bit of a gimmick but the cheaper versions without the power measurement have some merit. You aren't going to find a 40000 count datalogging meter with USB for $130 from anyone else. Hell most of the other manufacturers charge you that just for the USB cable.

I don't know why people find this so hard to understand. It's like all the car manufacturers selling V8 sports cars for $50,000. Then some chinese company comes along and sells a V8 sports car that is just as powerful for $7000 but the catch is that it has no seatbelts and the fit and finish is a bit naff. Suddenly everyone is like "Oh my God! you can't sell a car without seatbelts!". But who cares, it's $7000. Install them yourself if you are so inclined.

If all you want is a 20000 count multimeter without any bells and whistles then yeah buy an old second hand 87...

[Wytnucls]

On the 71D, 10,000 data memory, accessible on the meter or the PC. 30,000 on my Gossen 30M, but I have to buy the Gossen cable and software to see any of it (350$). That's after you paid 900$ for the meter, 60$ for the bag and 140$ for the 5V wall wart, otherwise you have 16 hours of battery life. Oh, it's only CAT II 600V, with 100mA max. And it comes with a bodged cap on the DC to DC converter.

[Lightages]

There is a real point of contention here when we look at the UT71E. It is clearly, obviously, intended to be used on high energy circuits as it has an attachment and function that are required to be connected to a high energy circuit in order to use them. The design and execution of the meter is clearly, obviously, not up to the task in a safe manner. On this point alone, the UT71E is a piece of junk that is not suited to its task and could cause someone serious injury.


The majority of Uni-T meters are cheaply built and have less than minimal input protection. Some of the latest multimeters have much improved input protection and I am comfortable recommending the UT139C at least.

[Grapsus]

I don't know why people find this so hard to understand. It's like all the car manufacturers selling V8 sports cars for $50,000. Then some chinese company comes along and sells a V8 sports car that is just as powerful for $7000 but the catch is that it has no seatbelts and the fit and finish is a bit naff. Suddenly everyone is like "Oh my God! you can't sell a car without seatbelts!". But who cares, it's $7000. Install them yourself if you are so inclined.

Well in civilized countries it is a crime to knowingly sell cars that don't meet the required safety standards.

[Yago]

I don't get it why so many people think that Dave was somehow unfair with this meter and Uni-T in general. What is all this justification stuff ? Are you trying to make yourself feel better about having wasted 180$ and bought a crappy and dangerous meter ? There is a lot of alternatives : 1W resolution mains power meters can be bought for 20$ in any PC shop and for 150$ you can have a used 20000 counts Fluke.

The ~$170 E variant is a bit of a gimmick but the cheaper versions without the power measurement have some merit. You aren't going to find a 40000 count datalogging meter with USB for $130 from anyone else. Hell most of the other manufacturers charge you that just for the USB cable.

I don't know why people find this so hard to understand. It's like all the car manufacturers selling V8 sports cars for $50,000. Then some chinese company comes along and sells a V8 sports car that is just as powerful for $7000 but the catch is that it has no seatbelts and the fit and finish is a bit naff. Suddenly everyone is like "Oh my God! you can't sell a car without seatbelts!". But who cares, it's $7000. Install them yourself if you are so inclined.

If all you want is a 20000 count multimeter without any bells and whistles then yeah buy an old second hand 87...

How would you make that meter meet the CAT ratings?
A new PCB design and quality components?
How much would all that time, effort and parts cost in the end?
Be easier to buy a quality meter and paint Uni-t on the front, but then why would you.

[c4757p]

I don't know why people find this so hard to understand. It's like all the car manufacturers selling V8 sports cars for $50,000. Then some chinese company comes along and sells a V8 sports car that is just as powerful for $7000 but the catch is that it has no seatbelts and the fit and finish is a bit naff. Suddenly everyone is like "Oh my God! you can't sell a car without seatbelts!". But who cares, it's $7000. Install them yourself if you are so inclined.

I think you're saying this seriously, but many of us would consider such a statement to have a high likelihood of being satirical...

[TMM]

So we are all agreed that its a moral dilemma that the incorrect cat rating text is printed on the front panel of the meter and in the literature and that the meter doesn't actually represent horrible value for money? Good, I thought so too.

There is a real point of contention here when we look at the UT71E. It is clearly, obviously, intended to be used on high energy circuits as it has an attachment and function that are required to be connected to a high energy circuit in order to use them. The design and execution of the meter is clearly, obviously, not up to the task in a safe manner. On this point alone, the UT71E is a piece of junk that is not suited to its task and could cause someone serious injury

There is a disclaimer that the mains attachment is only to be used up to 250vac. Also to probe a household mains outlet only requires cat II. Whether or not the meter would pass cat II is a secondary matter but it certainly does not need to be catIV

[Yago]

It is criminal and life threatening, can't put a price on the latter.
So it represents the poorest of value for money.

[Quarlo Klobrigney]

I haven't read through all 9 pages yet because the teardown of the extremely DANGEROUS PIECE OF SHIT led me straight away to a case I read about where an electrician and a fire official were killed due to arc  flash.
They died because of Chinese shit manufacturing with safety as a no-thought. The meter the electrician used to check all 3 phases exploded and the result as I said...
Please read this trade journal regarding this story. NEVER use cheap meters across a circuit that can literally vapourize you in an instant. Make sure that it is of the proper CAT rating for the job and from a reputable manufacturer with real approvals. Sorry if this has been said and linked before, but I can't see someone getting hurt by equipment like this.
the link: http://ecmweb.com/arc-flash/case-deadly-arc-flash

[boffin]

I'm guessing Dave ended up with a Chinese knock-off of a Chinese cheap brand.  The PCB is likely just a copy, but the components/fuses are the cheapest knockoff they could find.