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

[Lightages]

Current measurements use an almost 0 resistance to measure the current and are placed in series with the device being tested. Some meters might have 100ohms for a resistance, or more. So the voltage drop inside the meter is very low, even with 8000V being the source. So a transient of tens of times that voltage, or even hundreds, would be needed to make a big voltage drop in the multimeter. So the fuse only needs to be rated for the actual voltage use of the meter. Good meters have fuses rated to break 10kA.

If you connect your meter incorrectly and think you are using the meter to measure voltage and you have it connected for current, then the fuse blows and the meter isn't connected anymore. This is a very transient thing and you will have almost zero chance of making the first grave error in use and also face an 8000V+ transient on what you are connecting to.

[Wytnucls]

CEM has at least 7 meters in my database with UL certifications, which I could verify. It is not common yet, but it will be soon as more manufacturers get shut out of some markets. Another incentive is the presence of Fluke, now that they produce cheap and safe meters in that part of the world.
Small manufacturers with low value meters may not change their ways, if people keep buying their products, without any concern for safety.

[Wytnucls]

The PTC will do its job of clamping the voltage above that when required.
The MOV is not a requirement, but an additional protection, to back up the PTC.

Actually this is no quite right.

If I remember correctly, the MOV is acting faster than the PTC, relieving some of the high voltage, while the PTC is busy heating up to eventually clamp the current flowing into the meter. That's what I meant as back up.

[WackyGerman]

There is no scenario where you will get high voltage transients on the current ranges.

...

If you measure current in series within a mains circuit, with high voltage transients present, the voltage will be low across the meter and the fuse will rupture if the current goes high for long enough (doubtful with short transients).

But after the fuse ruptures, whether because of a transient spike in current or an ordinary continuous non-transient overcurrent situation, or some combination, won't the entire voltage of any subsequent transient be present across the fuse?  If not, where is the transient voltage going?

I'm certainly not an expert on CAT ratings or testing, but I wonder, if you expect 8000V transients to be present on your mains, how can you be sure they won't ever be imposed across a meter measuring current through those mains?

Well in a case of a transient while measuring the transient also will be present across the fuse .

The 8000 V is the pretended voltage of the IEC 61010 that a multimeter has to withstand without danger the operator of the multimeter in CAT IV 600 V . Well that is not far-fetched , it is a summary of experiences from experts . Here is an example : Some electricians from a overhauling company for subways measured the voltage on the bar where the railcars of the subways are getting the feeding for control reasons . After some time one and another multimeter suddenly died . They are really wondering about that because there is roundabout 800 v dc on the rail and that should not be the problem . But the problem are the transients , they reached 9000 V and they killed the multimeters after a short time

[Lightages]

One thing that sticks out to me is that if you really want to be sure, as sure as you can be, that the manufacturer is building meters to the right standards, then you have to check a few things.

#1. Do they have a valid and current ISO900X approval and no violations? This is to ensure that they actually make things the same way every time and don't change their methods without a notification and paper trail.

#2. Do they have a proper third party certification, like UL, to show that their designs and construction methods actually meet the ratings they assert? Do they have a clause in the procedures, as listed with their ISO certification, that demand a new third party testing anytime a new part or change in design is implemented anytime it is different from the originally tested configuration?

#3. Can you find and read a record of their ISO certification and current status?

#4. Can you find a record of their third party testing/certification for compliance to the standard they mark on their meter?

#5. Is the third party certifying lab also ISO certified and can you find their records?

As you can see, the average consumer does not have the resources nor time, and probably not the education, to know what to look for or even where to look. This is assuming an educated consumer. What about all the "noobs" and people who assume too many things? This is where Dave's review and the rantings and opinions of others can help. We try to educate. It is up to the individual to assume their own risk. It is not up to the company selling them something to assume that the buyer knows the risk and understands that the product they have purchased is labelled with lies that could kill them.

In the ideal world, everyone would treat everyone with respect and help ensure each others safety.

In the real world, companies take your money and hope you don't complain or find out they have taken you money based on lies. When some companies in China started putting melamine in baby food to artificially boost the protein content test results, it didn't seem to matter to them that they were literally killing babies to make a bigger profit. The occasional injury or death from wrongly marked multimeters in a  few million users world wide probably means nothing in the big scheme of their profits. Most people probably get these injuries or problems and don't know what happened, or have no recourse.

[Lightages]

Well in a case of a transient while measuring the transient also will be present across the fuse .

The 8000 V is the pretended voltage of the IEC 61010 that a multimeter has to withstand without danger the operator of the multimeter in CAT IV 600 V . Well that is not far-fetched , it is a summary of experiences from experts . Here is an example : Some electricians from a overhauling company for subways measured the voltage on the bar where the railcars of the subways are getting the feeding for control reasons . After some time one and another multimeter suddenly died . They are really wondering about that because there is roundabout 800 v dc on the rail and that should not be the problem . But the problem are the transients , they reached 9000 V and they killed the multimeters after a short time

The meters died and they were safe, Why? They did not have their meters connected incorrectly on the current jacks and tried to measure voltage. I think this validates my statement that you are HIGHLY unlikely to be make the error of connecting your meter for current and then mistakenly try to measure for voltage and then also have an 8000V transient at the same time.

Unfortunately you can not make something 100% safe. There are always risks and everyone needs to know the risks involved. Education should have told the workers that the rails were connected to inductive motors with huge capacity and that the voltage could spike from the inductive kickback. I am sure the designers of the railway system knew this, why didn't the workers? That was bad education and a bad application of test equipment due to the bad education.

[Lightages]

The PTC will do its job of clamping the voltage above that when required.
The MOV is not a requirement, but an additional protection, to back up the PTC.

Actually this is no quite right.

If I remember correctly, the MOV is acting faster than the PTC, relieving some of the high voltage, while the PTC is busy heating up to eventually clamp the current flowing into the meter. That's what I meant as back up.

PTCs do not clamp voltage, period. That is the problem with your statement. PTCs limit current.

[WackyGerman]

One thing that sticks out to me is that if you really want to be sure, as sure as you can be, that the manufacturer is building meters to the right standards, then you have to check a few things.

#1. Do they have a valid and current ISO900X approval and no violations? This is to ensure that they actually make things the same way every time and don't change their methods without a notification and paper trail.

#2. Do they have a proper third party certification, like UL, to show that their designs and construction methods actually meet the ratings they assert? Do they have a clause in the procedures, as listed with their ISO certification, that demand a new third party testing anytime a new part or change in design is implemented anytime it is different from the originally tested configuration?

#3. Can you find and read a record of their ISO certification and current status?

#4. Can you find a record of their third party testing/certification for compliance to the standard they mark on their meter?

#5. Is the third party certifying lab also ISO certified and can you find their records?

As you can see, the average consumer does not have the resources nor time, and probably not the education, to know what to look for or even where to look. This is assuming an educated consumer. What about all the "noobs" and people who assume too many things? This is where Dave's review and the rantings and opinions of others can help. We try to educate. It is up to the individual to assume their own risk. It is not up to the company selling them something to assume that the buyer knows the risk and understands that the product they have purchased is labelled with lies that could kill them.

In the ideal world, everyone would treat everyone with respect and help ensure each others safety.

In the real world, companies take your money and hope you don't complain or find out they have taken you money based on lies. When some companies in China started putting melamine in baby food to artificially boost the protein content test results, it didn't seem to matter to them that they will literally killing babies to make a bigger profit. The occasional injury or death from wrongly marked multimeters in a  few million users world wide probably means nothing in the big scheme of their profits. Most people probably get these injuries or problems and don't know what happened, or have no recourse.

Yes that s the problem . The government accepts the CE self declaration as a legal certificate but it is a carte blanche for the company to sell their products without quality control . The main thing is profit for the company and taxes for the states and they don t really care about the customers . We just have to shut up and consume . 

[necessaryevil]

Well, doesn't make a false CE clarification make a manufacturer vulnerable to lawsuit? Especially when something goes wrong? They can't say they sold a device which is not intented for the european market, and they can't claim they didn't know.

Yes, this is just in theory, of course I know CE sucks.

I have always been taught to buy tools of good quality. I did not always listen, but some stuff you have to find out for yourself. An expensive multimeter isn't really expensive if it lasts for 20 years, doesn't it?

That being said I am a huge hypocrite for owning a cheap ass multimeter. But it is on my wishlist!

[Wytnucls]

The PTC will do its job of clamping the voltage above that when required.
The MOV is not a requirement, but an additional protection, to back up the PTC.

Actually this is no quite right.

If I remember correctly, the MOV is acting faster than the PTC, relieving some of the high voltage, while the PTC is busy heating up to eventually clamp the current flowing into the meter. That's what I meant as back up.

PTCs do not clamp voltage, period. That is the problem with your statement. PTCs limit current.

As the 1KOhm PTC resistance increases as it heats up, the voltage drop across it increases too, so it does clamp the voltage seen by the circuit in the DMM, limiting arcing possibility.

[Lightages]

The PTC will do its job of clamping the voltage above that when required.
The MOV is not a requirement, but an additional protection, to back up the PTC.

Actually this is no quite right.

If I remember correctly, the MOV is acting faster than the PTC, relieving some of the high voltage, while the PTC is busy heating up to eventually clamp the current flowing into the meter. That's what I meant as back up.

PTCs do not clamp voltage, period. That is the problem with your statement. PTCs limit current.

As the 1KOhm PTC resistance increases as it heats up, the voltage drop across it increases too, so it does clamp the voltage seen by the circuit in the DMM.

Sorry, this is a wrong way of thinking about it. The voltage drop across it increases only if it is part of series of other circuit and a current flows sufficiently to cause it to heat up.  For example, my UT61E had a PTC in the circuit, but it still allowed the 2500V to pass directly to the selector switch and it fried the main IC. There was not enough current there to cause the PTC to change its value at all.  It did not "clamp" any voltage in this case. It does not "clamp" the voltage, as you said. This could give people the wrong idea of what a PTC is. It is a resistor that varies its resistance quickly with temperature. If a PTC is not connected to anything and it is heated up, its resistance goes up. This change is independent of voltage as it is not connected to anything. This is a fundamental.

[WackyGerman]

maybe the 2500 V will arc over to the com rail on the board because of the 1 or 2 mm clearance and then some current will flow and the ptc heats up and will increase the resistance . But then the multimeter is barbequed and fit only for a knacker s yard . It s a good example for planned obsolescence combined with the shitty input jacks 

[Wytnucls]

The idea is not to protect the multimeter, but to protect the user. If no current flows through the PTC, serious arcing is not likely. IC damage is not a concern. A Fluke may not function properly either after an overvoltage.

[WackyGerman]

Well I am sure at 2500 V a fluke will survive this . In most Fluke I know there are ptc s and movs and when the transient will occur the movs will decrease the resistance and so current will flow and the ptc heats up and increase the resistance . The energy is divided by the movs and the ptc and the chance that the fluke will survive is much higher than a uni t that will definitively be scrapped at 2500 V , also at 1000 V . Look at youtube . There is a video where a brymen 869 is tested with 2000 V and it survived without any problems . The 869 has 2 ptc s and 3 movs and not whimpy ones

[Wytnucls]

Fluke 87 damage, after accidental overvoltage (1200V):
He doesn't say if the iC was affected.

Just an update, got the service manual from Fluke. I was really suprised they have them for download.
After a thorough evaluation of the front end of the meter, I'm replacing the fusible resistor R1, both varistors RV1 and RV2, the thermistor RT1, and a cap C1 that was damaged by the varistors when they blew. Basically the entire front end protection except for the spark gap. It appears to still be a spark gap.
I bought the parts directly from fluke through thier estore. Also picked up an extra set of fuses. I thought $40 for those original replacement parts wasn't bad considering what it might cost to send it in for repair.

[Lightages]

For the record, I have put 5000V through a Uni-T UT71E, a UT61E, a Brymen BM257, a BM869, a BM525, a Digitek DT2843R, and a Fluke 27/FM. All but the UT61E survived.

P.S. I should also say that this was with very limited current, like 5ma.

[Wytnucls]

Was it on all selections or just the voltage range?

[Lightages]

Hmmm, I don't remember for all of them. I will have to repeat my tests on video! 

Actually I don't think I will. I don't want to potentially destroy all those meters on a test they aren't rated to survive.

[Wytnucls]

Hmmm, I don't remember for all of them. I will have to repeat my tests on video! 

Actually I don't think I will. I don't want to potentially destroy all those meters on a test they aren't rated to survive.

Do it on the 71E. You won't cry if it dies... 

[Lightages]

You obviously missed the story earlier in the thread of the electrician and some people standing close who died because of a falsely rated meter. Or the story of the electrical workers who had meters dying on them mysteriously because they hadn't received proper training.Not all countries have rules about house owners not touching their electrical distribution boxes.

[gildasd]

You obviously missed the story earlier in the thread of the electrician and some people standing close who died because of a falsely rated meter. Or the story of the electrical workers who had meters dying on them mysteriously because they hadn't received proper training.Not all countries have rules about house owners not touching their electrical distribution boxes.

Here in Belgium, I can do anything I want downstream of the meter.
But to get the installation accepted, I need to document as I go (bills, pictures and plans) and be on my toes the day of the inspection.
What worries me is that DIY superstores sell 5€ to do said work. A novice could easely short the phase before the box (240v loadsa amps in my house) and the unprotected earth...
But that being said, I find that the other end of the spectrum, like Oz, where you need a diploma to rewire a plug is ridiculous.

[Quarlo Klobrigney]

Never underestimate stupid, or at least ignorance. I am still in the process of refurbing my home, and I was tearing out a wall which had only a light switch (in use) and a 2 prong receptacle which never worked. When I got most of the wall apart, I saw a rather thick cable feeding the light switch, then exit to a smaller cable to the outlet. I assumed it was a feed to the switch, back to the light, with a carry to the outlet. Not exactly the best way, but it would work. When I opened up the mess it was all too apparent, someone wired the receptacle in parallel with the switch.
Another story of the same house, after removing a breaker with a visible trace of a cable to it, I was ready to cut the cable. A surprise "hit" me as the cable was still hot. I followed the cable branches back still further, and found that it was fed by another breaker on the same phase, in the ON position. A loop circuit. Needless to say I check & double check never trusting what should or shouldn't  be, as it CAN be.

So, never say never, sometimes ignorance or sheer stupidity can be laughable or deadly.

[gildasd]

Never underestimate stupid, or at least ignorance. wavy lines wavy lines
wavy lines wavy linesAnother story of the same house, after removing a breaker with a visible trace of a cable to it, I was ready to cut the cable. A surprise "hit" me as the cable was still hot. I followed the cable branches back still further, and found that it was fed by another breaker on the same phase, in the ON position. A loop circuit. Needless to say I check & double check never trusting what should or shouldn't  be, as it CAN be.

So, never say never, sometimes ignorance or sheer stupidity can be laughable or deadly.

I found something similar renovating this house.
While finishing up the electricity and getting ready to plaster, I was pulling out old deactivated cables and WHITE FLASH BANG found an active that should not have been.
So out comes my trusty UNI-T, all the mains are cut at the source, a long wire is pulled between the offender and the electrical box.
And it's beeping in way too many places, in the old stuff and my new box.
Eventually, I trace it down to a series of connection boxes that are meant to feed the push button activated lights in the entrance, stairs and landing.
And to the plugs in said places.
And to the lights and plugs of the four bedrooms.
And to the lights and plugs of the bathroom.
And to the lights of the kitchen.
AND TO THE BLOODY OUTSIDE GARAGE.
ON ONE 0.8mm² wire. And that's only the phase, I have yet to discover were the neutral is hidden...
There was probably more stuff connected to it that we already had cut out.
The house was wired up like this 30 years ago by a pro (there were still his stickers and his details on the house documents) and no inspector ever noticed...
(The same inspectors that failed an inspection on the previous house for having about 1cm of nearly visible (but fully isolated, safe etc) wiring on the top/behind of the fuse box.)
So, I spent the next 8 hours making this thing "safeish" and disconnecting as much of as possible.
Some "experts" here might say "you should have done X and Y". Yeah we did, dying in a shower of sparks is not my plan, but we tested one of the two wires and it lead to a set of fuses, we disconnected them both and marked the wires as "out". But one of the two cables bi-passed the fuse in a rather sneaky way and ended up connected to the rest of the stuff. 

It's a pain in the arse, but it vindicates my approach the fitting all now wiring and fuses, fuses boxes etc in the house. TRUST NO ONE.

On a side note, the internet connection over the electrical wiring has improved dramatically (in the process to of being replaced by a router and CAT6 cabling).

To get back on subject, if you are "just doing continuity testing" on a "professionally installed electrical wiring" and the fuses are all "off" for that area, you might still have a surprise that can blow a 5$ DMM to kingdom come.

[Paul Moir]

You obviously missed the story earlier in the thread of the electrician and some people standing close who died because of a falsely rated meter...

I presume you mean the article on ecmweb?  I know it seems pedantic but the details are important, especially if you're reading a one sided story like this one.  The article describes the multimeter as advertised as an "industrial multimeter" with no IEC rating at all.  For all we know it could have been a pre-CAT meter.  The meter-induced arc flash root of this accident seems to be more with the electrician's ignorance of those ratings than the meter itself, and focusing on the meter rather than the CAT ratings misdirects the lesson.

[Lightages]

The point of all of this is not one specific incident and the details. The point is that education saves lives and stopping companies from cheating on their ratings and specifications can save lives. I don't understand why people argue against education and chastising companies for false advertising.