Author Topic: UNI-T (UNI-Trend) UT90C review.....I killed it!  (Read 31938 times)

MJLorton

  • Administrator
  • Hero Member
  • *****
  • Posts: 816
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #45 on: May 21, 2012, 03:14:53 AM »
PTC1: 1K
PTC2: 567 ohm...out of circuit
PTC3: 589 ohm...out of circuit
R26: 100 ohm...out of circuit
R31 SMD: 10 M ohm...checked several times...

Okay, I'm using this datasheet as a reference.

www.ic-fortune.com/upload/Download/FS9721_LP3-DS-20_EN.pdf

With the above, I was using the datasheet's pcb designators (page 19) as described by the various diagrams/figures, but it looks like the UT90C is using a variation.

R31 is good as your 10M ohm resistor.  As long as R26 matches the color code for a 100 ohm resistor, it is good.

PTC1 at 1K is good.  PTC2 and PTC3 are lower than what I expected (1K) unless they are in series, but I don't think the low values are the cause of the UT90C not working.  If one of the PTCs were completely open, then it might explain why the UT90C is not working.

Let me think about this for a few more days.  Unless I spot something obvious, I think the FS9721-LP3 might be bad.

I tried going to that link but I get a warning that it's an "attack page"....and might steal my data....odd...

Thanks again for your efforts. I thought it might be something in the main IC....but I would have thought that if that had gone faulty the meter would not display anything "intelligent" at all. I presume it can have a partial failure?

Cheers,
Martin.
Play, discover, learn and enjoy! (and don't be scared to make mistakes along the way!)

retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #46 on: May 21, 2012, 11:52:24 PM »
I tried going to that link but I get a warning that it's an "attack page"....and might steal my data....odd...

Thanks again for your efforts. I thought it might be something in the main IC....but I would have thought that if that had gone faulty the meter would not display anything "intelligent" at all. I presume it can have a partial failure?

I'll study the datasheet some more.  I do have a "cheap" multimeter here that I can take apart and study to see what is going on and that may help with your UT90C.

I also got the same virus warning, but I checked the pdf with multiple AV software and it is clean.  I tried to attach the pdf and a zipped copy of it, but the file can only be 300KB and the zipped is 1.1MB.


retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #47 on: June 03, 2012, 12:05:39 AM »
I found this while looking for some info

http://www.amazon.co.uk/VOLTCRAFT-Voltcraft-VC290-DMM/dp/B003A5UGBA

I found this part of the description particularly LOL

"The device is protected with comprehensive safeguards against misuse."

MJLorton

  • Administrator
  • Hero Member
  • *****
  • Posts: 816
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #48 on: June 04, 2012, 06:32:25 AM »
I found this while looking for some info

http://www.amazon.co.uk/VOLTCRAFT-Voltcraft-VC290-DMM/dp/B003A5UGBA

I found this part of the description particularly LOL

"The device is protected with comprehensive safeguards against misuse."
I love it!! I read the entire advert....the "English" is certainly entertaining....
Play, discover, learn and enjoy! (and don't be scared to make mistakes along the way!)

retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #49 on: December 29, 2012, 03:45:58 AM »
Hey Martin,

I was thinking about your UT90C again and thought you might be inspired by one member's recent blog site.  His site is full or repairs and pictures.  Highly recommended.

http://mrmodemhead.com/

MJLorton

  • Administrator
  • Hero Member
  • *****
  • Posts: 816
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #50 on: December 29, 2012, 07:43:52 AM »
Hey Martin,

I was thinking about your UT90C again and thought you might be inspired by one member's recent blog site.  His site is full or repairs and pictures.  Highly recommended.

http://mrmodemhead.com/

Brilliant site Mr Sharp Eyes! I've linked to it from my channel page and hope to make some time to go through it.
I'll drop Mr Modemhead a message too.

Thanks.
Play, discover, learn and enjoy! (and don't be scared to make mistakes along the way!)

ModemHead

  • Newbie
  • *
  • Posts: 16
    • Mr. ModemHead
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #51 on: December 29, 2012, 09:31:42 AM »
Wow. A whole thread about a broken multimeter and somehow I missed it...   :-[

retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #52 on: December 29, 2012, 05:47:58 PM »
Martin,

I may have more test suggestions for you just to confirm that the FSP 9721 on the UT90C is indeed fried. 

I have an UEI multimeter that is not working properly and it uses the FSP 9711 chip.  I think the two are close enough to make comparable readings on certain pins.  It is an on/off project for me as I discover new testing techniques or learn more about how the chip works in general.

ModemHead

  • Newbie
  • *
  • Posts: 16
    • Mr. ModemHead
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #53 on: December 29, 2012, 08:55:15 PM »
Just thought I would throw in my 2 cents worth...

I looked at the datasheet 'common use' circuit to see how the Hz measurement function was implemented. There is no guarantee that the Uni-T engineers followed this exactly, but in the absence of a more accurate schematic, it's all we have to go on. In any case I would not expect the component designators to match.



For Hz mode, J9, J6, and J8 would be jumpered by the function switch, providing an AC-coupled path through C15 straight to the ADP input of the DMM IC.  Q1, Q2, and the 1K PTC are supposed to provide over-voltage protection for Hz, Ohm, Cap, and Diode functions.

Base-emitter junctions of transistors function quite reliably as zener diodes when reverse biased, especially when the base-collector junction is taken out of the equation by shorting it out. Note that Q1 and Q2 are in this configuration and are back-to-back, in series.  The B-E reverse breakdown voltage is probably about 5V, and the B-E forward voltage is around 0.7V as usual.  So regardless of the polarity of the input voltage, the drop across the two Qs should never exceed about 5.7V.  A bidirectional voltage 'clamp'.

That means that point 'X' in the diagram should be limited to +/- 5.7V.  If a higher voltage is applied, it will be dropped across the PTC thermistor, and current will increase through the PTC and the clamp transistors (green path).  As the voltage drop grows, the thermistor will heat up and its resistance will increase, limiting the current and dissipating the excess energy.  That's how it should work, but something went wrong.

Q1 and Q2 are switched out for voltage measurements, so even if they failed short it should not affect voltage functions.  About the only scenario I can come up with is that the PTC opened up permanently.  Or, given that the Hz function feeds the ADP input almost directly, the DMM IC is fried. :(

retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #54 on: December 30, 2012, 03:02:53 AM »
Q1 and Q2 are switched out for voltage measurements, so even if they failed short it should not affect voltage functions.  About the only scenario I can come up with is that the PTC opened up permanently.  Or, given that the Hz function feeds the ADP input almost directly, the DMM IC is fried. :(

Hi ModemHead,

Thanks for giving a play by play on the schematic and input protection.  Interpreting schematics is one area I'm weak in and appreciate the full detailed explanation.

I had Martin pull and test all the 3 pin transistors and they all tested fine.

The PTCs seem to test okay?  Martin reported

PTC1: 1K
PTC2: 567 ohm...out of circuit
PTC3: 589 ohm...out of circuit

I have no idea what a good PTC should measure other than relying on the FS9721 datasheet values (that is 1K).  So PTC2 and PTC3 could be bad or they could be used together in series (giving it a 1K value)?

For my own learning, I decided to measure the DC voltage on pins 26 (TENM) and 24 (ONEM).  I found two points on my UEI board that fed into the above pins making it easier to measure.

Since I don't have a variable DC power supply, I got 4 semi-dead 9V batteries and noted the following:

9V battery voltage     TENM     ONEM
8.08v                  80.7mv   0.0
4.90v                  52.6mv   0.0
3.76v                  37.5mv   0.0
2.466v                 228mv    228mv

Now my UEI display is stuck on "all segments" test, but what I'm trying to show with the above testing is that the input voltage divider is working by measuring pins TENM and ONEM.  My UEI has two PTCs.

So if Martin does this same test, he can verify that the input section is working up to the FSP9721 (implying that the PTC works)?

Martin, to test this (probably 15 to 30 minutes of your time)

1) Use your variable power supply to supply DC voltage to the input jacks of the UT90C.
2) Use a working multimeter to measure the DC voltage at pins TENM and ONEM.
3) The black probe on the working multimeter goes to the UT90C COM jack for ground.
4) The red probe goes on TENM and ONEM (or some physically bigger component that feeds these pins to make probing easier - looking at your photo, I can clearly see traces leaving pins 26 and 24).
« Last Edit: December 30, 2012, 03:13:34 AM by retiredcaps »

ModemHead

  • Newbie
  • *
  • Posts: 16
    • Mr. ModemHead
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #55 on: December 30, 2012, 09:22:49 AM »
Sorry, I think I blew the margins in this thread with my diagram...

PTCs: Those values look OK to me.  The actual PTC resistance generally isn't important because it is very small with respect to the impedance of the rest of the circuit (1K vs 10Meg+).  Another way of saying that is that the normal current through the PTC is so small that the voltage drop on the PTC is negligible.

Speaking of high impedances, notice that the TENM pin on the chip is connected through a 10Meg resistor.  This is the 'top' end of the voltage divider.  Using internal CMOS switches, the micro-controller in the chip (under control of the auto-ranging algorithm) will tie this point to one of the other precision resistors, which you can see are grounded when the meter is in voltage mode.  That makes the 'bottom' of the voltage divider, which is capable of dividing by 10, 100, 1000, and 10,000.

Depending on how the divider is configured, adding the 10Meg impedance of another DMM at the TENM pin can actually affect the voltage there significantly.  So voltages measured there may not make sense at first glance.  Observer effect...

Note that in your measurements, TENM and ONEM were the same voltage for 2.466V input. That means the divider was in 1/10 mode.  I assume the divider was in 1/100 for the higher voltages (HUNK would be in use instead of ONEM.)  As the ratio goes up, the bottom resistor gets smaller and your 10Meg DMM has less effect.

Your test plan is valid, but I think the easiest place to verify all input circuitry (inc. PTCs) would be to locate point 'X' on the circuit board.  It should be on side of the 10Meg input resistor not connected to the DMM IC.  And it should reflect exactly (1:1) whatever voltage is applied to input jacks.  The voltage divider resistors can be verified separately with the power off if necessary.

Hope this helps...

retiredcaps

  • Full Member
  • ***
  • Posts: 248
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #56 on: December 30, 2012, 04:06:58 PM »
Note that in your measurements, TENM and ONEM were the same voltage for 2.466V input. That means the divider was in 1/10 mode.  I assume the divider was in 1/100 for the higher voltages (HUNK would be in use instead of ONEM.)
So for my own learning again, I went and measured the HUNK

9V battery voltage     TENM     ONEM     HUNK
8.08v                  80.7mv   0.0      80.0mv
4.90v                  52.6mv   0.0      52.4mv
3.76v                  37.5mv   0.0      36.7mv
2.466v                 228mv    228mv    0.0

The "4.90v" battery above measures around 5.431V today. 

It is obviously dead because the top and the bottom are both bulging.  The expiry date on it is Mar 2004.  I think I got it from buying an used multimeter which is why I always check the battery of any electronic device before I use it.

ModemHead

  • Newbie
  • *
  • Posts: 16
    • Mr. ModemHead
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #57 on: December 31, 2012, 12:36:03 AM »
So for my own learning again, I went and measured the HUNK
With those measurements, I think you have proven that the micro-controller is running and performing its auto-ranging task properly.  Which implies that it is getting good information from the A/D converter.

MJLorton

  • Administrator
  • Hero Member
  • *****
  • Posts: 816
Re: UNI-T (UNI-Trend) UT90C review.....I killed it!
« Reply #58 on: January 02, 2013, 06:14:38 AM »
Brilliant input gents...thanks very much.

I'm just busy on my DC load project and hope to have time after that to have a fiddle with the UT90C.

Cheers.
Play, discover, learn and enjoy! (and don't be scared to make mistakes along the way!)