UNI-T UT61E Multimeter teardown photos.

[Kakalas]

Hello,

Maybe someone could help me to solve this problem... 
I have a multimeter UT61E for some time, I use it a little, it's almost new. Recently I noticed that UT61E began to show a false VDC (when nothing is connected shows about -1.15 VDC). I tried to carry out calibration according to the guide (link below), but no result, since the margin of error is quite large and goes out of calibration range. Would anyone have any ideas how to fix it?

http://gushh.net/blog/ut61e-calibration/

[retiredcaps]

Recently I noticed that UT61E began to show a false VDC (when nothing is connected shows about -1.15 VDC).

Questions.

1) Have you tried a new battery?
2) If you short out the probes with the meter on DCV, does it show 0.0?
3) Post clear focused pictures of your meter so we can see the pcb (both sides) and look for any obvious damage.
4) Does ACV work?
5) Does resistance work if you measure a 1k ohm resistor?

[MrSlack]

Mine did all sorts of weird stuff when the battery went flat. First point of call on these I think.

[joeqsmith]

I am interested in knowing more about how people who have recommended changing the meter have validated their changes.

Just a little thought on this....   I ran both the 61D/E and they failed with a basic ESD pulse.    I also ran a UT181A that failed with a single ESD event.  The UT181A has PTC, MOVs and secondary clamps.       

On the flip side, I have ran a few meters that have survived not only ESD but some much higher energy transients that had no MOVs.   The last one I ran was the UT10A which not only survived the basic ESD but went on to survive a 3KV 100us FWHH with a 2 ohm source with the meter set in all modes. 

In the case of the UT181A, the cost of the meter (and my own interest) drove me to have a look at the design in more detail.   In this case the layout has a few problems that needed to be addressed.  A little Dremel, X-acto and dope.    And no, adding a clamp to the output of the regulator was not one of them.   

With the recommendation to add MOVs, I am curious what the goal is?  Safety, robustness?  I would like to know how this was validated?   Do we even know if there was ANY improvement?  Worse, did adding them make the meter less safe?      I can see someone getting a false sense of added safety.   

It could very well be that UNI-T actually tested the meter with and without the additional parts and had the same performance and decided to save some money.  I have no idea.   

[hgg]

I have ordered two of these :

http://www.tme.eu/gb/details/siov-s20k625/tht-varistors/epcos/b72220s0621k101/

In what circumstances adding these MOVs will make the multimeter less safe?

[joeqsmith]

Do you feel sprinkling in a few parts will make it more safe or robust?  If you do and in fact it has no effect then that in of itself would make it less safe.   How will you know if there is an improvement?

I have ordered two of these :

In what circumstances adding these MOVs will make the multimeter less safe?

[hgg]

How will you know if there is an improvement?

I guess you would not know and you'd better not find out the hard way.
I will send it to you to throw some kVs and see what happens...   

But seriously, if there is a transient and the MOVs are working, wouldn't they clamp it down
safely?   Before them there are two PTCs as well. 

[joeqsmith]

I don't know about the safety side at all.   I just don't throw that sort of energy at them.  People recommending to add parts should have some data based on fact.   I gave up on Santa Claus and the Tooth Fairy years ago.   

The 181A fully populated with all of it's fancy protection failed with just one hit of a very low ESD event.  Had they removed the parts and saved that money, it would have failed just the same..   On the flip side, had I received the meter without the protection circuitry populated and then correctly guessed what the parts were and added them, it would have still failed.     The layout is important and can't be ignored. 

What you have suggested may not be a bad idea.  If we forget about safety, I could run your meter to see if there was an improvement in robustness but you may not like the end results.   What's that meter cost?  When I made changes to the UT181A, I could have just said I fixed it or just reapplied a single ESD transient and left it at that.   I thought it was better to throw everything I could at it and leave no question the meter's robustness had indeed been improved.
 

How will you know if there is an improvement?

I guess you would not know and you'd better not find out the hard way.
I will send it to you to throw some kVs and see what happens...   

But seriously, if there is a transient and the MOVs are working, wouldn't they clamp it down
safely?   Before them there are two PTCs as well. 

[hgg]

Do you have the specs of the UT181a MOVs it came with?

[joeqsmith]

Yes.  It was all covered in that second half of the video starting about 8 minutes into it.   

Do you have the specs of the UT181a MOVs it came with?

[hgg]

Layout design is key of course but from what manufacturer were the 07D751K MOVs?
There are many Chinese replicas around with the same part numbers.  Maybe they
were not so fast as their datasheet claims.  After the protection modification, you should
have used the same MOVs to see if it survive.  (If they were not damaged of course).
Did you use the same MOVs?

[joeqsmith]

Comparing the UT61E's shown and the one 5ky had sent, they are a lot different.   Note the place for 4 MOVs.   I guess even within one revision of the meter you can't be sure what you are getting.   Roll the dice..

[hgg]

I guess even within one revision of the meter you can't be sure what you are getting.   Roll the dice..

True...

My UT61E has a different layout.


The isolation PCB gaps between the PTCs are also missing...


Same MOVs and survived after the mod.  Clearly poor design.
Both the 181A and the UT61E do not have proper HV isolation slots.
Maybe that's why they fail.

[joeqsmith]

Yours looks  like the others I have seen posted.     Even if I wanted to do an A/B compare, getting meters that are the same appears to be a problem.     

At the start of that video, I had shown what the ESD transient looked like after a few people had asked.    It's pretty fast.    Funny that the UT10A with it's one PTC survived so well compared with so many of their much more costly meters.   Again, roll the dice..

[hgg]

Funny that the UT10A with it's one PTC survived so well compared with so many of their much more costly meters

Indeed. 
There must be an explanation though. 
How did it survive with just the PTC which is just for over current  protection?   

[joeqsmith]

Oh, I have an explanation.   When you roll the dice, there is a certain amount of luck involved!      

[smithnerd]

Comparing the UT61E's shown and the one 5ky had sent, they are a lot different.   Note the place for 4 MOVs.   I guess even within one revision of the meter you can't be sure what you are getting.   Roll the dice..

Those 4 MOVs are populated in the German version (which is down rated to CATIII 300V, CATII 600V). I think any UT61E manufactured after 2015 shares the same PCB design as the 'safer' German one.

Image of the German version from earlier in this thread:

[hgg]

...and below is how I received my UT61E 'Special Edition'...



C8b is a capacitor on top of another and flux residue all over the place!   
Am I the only one with a special edition ?   

[joeqsmith]

5KY's was rated for CAT III 1000V, CAT IV 600V without the MOVs.   It must be a safer version than the German one with the MOVs!   Besides your meter having the MOVs, at a quick glance:

PCB is marked different
Metal can
Cap next to metal can is mounted differently
Slits in board near can are different
Added pot near can
Blue pot is backwards
Q5/6 are larger package
Fuses are sized differently (smaller parts used on higher CAT rated meter)

Those 4 MOVs are populated in the German version (which is down rated to CATIII 300V, CATII 600V). I think any UT61E manufactured after 2015 shares the same PCB design as the 'safer' German one.

Image of the German version from earlier in this thread:

[Hydrawerk]

OK, so UNI-T is able to produce a good multimeter, but you cannot recognise it at the shop.  Oh, wait, you can buy an Amprobe meter, that is UNI-T design but with proper fuses, MOVs and PTCs. 

[joeqsmith]

Maybe we could modify AMPROBE's AM61E or Fluke's 61E+.   No wait,  those would actually be fairly robust.   Best to stick with the UT61E.

OK, so UNI-T is able to produce a good multimeter, but you cannot recognise it at the shop.  Oh, wait, you can buy an Amprobe meter, that is UNI-T design but with proper fuses, MOVs and PTCs. 

[Kakalas]

Recently I noticed that UT61E began to show a false VDC (when nothing is connected shows about -1.15 VDC).

Questions.

1) Have you tried a new battery?
2) If you short out the probes with the meter on DCV, does it show 0.0?
3) Post clear focused pictures of your meter so we can see the pcb (both sides) and look for any obvious damage.
4) Does ACV work?
5) Does resistance work if you measure a 1k ohm resistor?

Hi,
-> 1) Yes, I have tried a few new batteries, meter VDC readings are almost the same.
-> 2) No, it doens't, see pictures bellow.
-> 3) See pictures bellow, sorry for poor quality, for now I only can make photos with phone camera. I do not see any visible damage.
-> 4) Yes AC works correctly, see pictures bellow
-> 5) Yes resistance works correctly, see pictures bellow.

I also noticed:
Resistance -> buzzer function, no sound if contacts are shorted
Resistance -> diode function doesn't work correctly, see pictures bellow (shows ~1,6 V when nothing connected, if contacts shorted buzzer sound apears  )
Capacity measurements works correctly.
Frequency (AC) measurements works correctly.
Current uA, mA, A measurements works correctly.
 





higher quality:
http://postimg.org/gallery/3igu9z1le/

[ModemHead]

@Kakalas, you probably should start your own dedicated thread in the Repair section to avoid stepping on other discussions here.

Connect another DMM in parallel with the UT61E, with both in VDC mode.  If the working meter indicates a "real" voltage present at the UT61E's input, then current is leaking into the front end through some fault path back to the power supply.

Inspect closely for obvious stuff first: PCB contamination, wayward solder bits, carbonization from unfortunate over-voltage events.  If nothing turns up, then you will have to trouble-shoot to find the source (or sink, given that the offset seems to be negative) of the fault current.  Note that the inputs of the DMM ASIC are suspect due to possible internal failure of the chip.

[Hydrawerk]

At school I found an UT61E that has some strange behaviour in capacitance mode. When nothing is connected, there is no bargraph. When I short the input, a bargraph appears and starts moving while the displayed value is OL. This is weird.
Strange behaviour starts at 0:43.

[mariush]

It's normal.

In resistance mode, you're measuring the resistance of the wire and the banana plugs, the value seems normal for that wire thickness.

When switching to capacitance you get the default base capacitance of around 2-300 nF.
Capacitance measurements are done by the multimeter by sending short pulses of voltage to capacitor and measuring the difference between pulses. A bigger capacitor charges slower.. With a short (the wire) you're basically ever so slightly heating up the wire with each pulse and the multimeter stays at OL (overload, or no capacitor) because there's no charge on the wire, no difference for the meter to measure.
 
You'll see the same behavior with a large capacitor (let's say 3300uF 16v) connected there, the bar at the bottom will pulsate several times at least before you get the value shown on screen.