Fixing an old Voltcraft VC 650

[Cold-1]

Hi,

I have an older Voltcraft VC 650 Multimeter. I can turn it on, the display works, as does switching all the measuring positions and ranges. Also the menu works ok. But it seams, all the input sockets are dead. No matter what I try to do with the probes, the meter stays at the initial values. I have already checked the fuses, they seam ok.

The PCB looks pretty messy, there are a lot of hand made additions.

Is there any way to just point at a component and say "that's the culprit"?

Regards

[retiredcaps]

I'll try to help here.  Let me ask some questions.

1) If you try to measure a 9V battery, what does the VC650 show?
2) With nothing connected and the VC650 set to resistance, what does it show?
3) Connect a 1k ohm resistor (or any resistor), what does the VC650 show?
4) Set the VC650 to resistance mode.  If you short out the leads, what does the VC650 show?
5) Did the VC650 just stop working or was it subjected to high voltage (say 500V or more)?
6) What happened to the fuse on the mA fuse?  It looks rather odd?  Is that aluminium foil?

PS. Most modern multimeters only use fuses for protection of the current range.  That is, the multimeter can still measure voltage and resistance with blown/bad fuses.

[Cold-1]

Hi, thanks for the fast reply. Ok, I try to answer the questions:

1) It just shows "0L V" (This is shown after initialisation).
2) It shows "0.L MO"
3) Same as 2)
4) Same as 2)
5) I can't say. I know I lent it to some one at that time (it's been a while) and whet it got returned, it was in the current state.
6) Yes, thats foil around it. It got worn a bit when I took it out to measure it.

[Lightages]

Wow! That thing is a real mess. Terrible construction and bodges everywhere. It is not surprising it has failed. You need to check each component that is connected to the volts/ohm jack. ne of the resistors or a PTC has gone open, or there is a cold solder joint. If you have the option, save your time and buy something better.

[retiredcaps]

7) I forgot to ask what the VC650 shows on the lcd when set to DCV with no probes attached?

Regarding 5 and 6, if the person who borrowed it tried to measure voltage while the meter and probes were inserted into the mA jack, I can see the main IC getting fried because the mA fuse was hacked with aluminium foil!  Fuses are supposed to blow if too much current to protect the user and the meter.

I found a schematic of the VC650.  The quality isn't great, but it is better than nothing.  I'm still learning how to read one so it will take a while for me to digest all the pages.  More test points/checks to come later tonight or tomorrow depending on my time.

http://elektrotanya.com/voltcraft_dmm_vc_650_sch.pdf/download.html

It also includes a Bill of Materials.

[retiredcaps]

The schematic for the VC650 suggests it uses an Intersil 7135 IC.

http://www.intersil.com/content/dam/Intersil/documents/icl7/icl7135.pdf

8] Can you verify this is the correct IC?  It is likely underneath the lcd.  I count a 28 pin IC on the pcb backside.

9) If this IC is blown/bad, replacements are available on ebay for reasonable prices.  $3.29 USD + free ship.

http://www.ebay.com/itm/171555349966

I have no experience or financial affiliation with the seller.  The 7135 may be available in your country at similar prices from reputable distributors like digikey, mouser, farnell, etc.  It is $5.21 USD on digikey and mouser.

[retiredcaps]

Some more tests.

10) The 7135 datasheet shows pins 9 (IN LO) and 10 (IN HI).
a) Put your VC650 into DCV.
b) Use the VC650 to measure 5V DC (like standby power from an ATX PSU or some wall wart ac/dc adapter - it doesn't have to be exactly 5V, just something close).
c) Report VC650 lcd display result.
d) Using another multimeter, measure the DC voltage across pins 9 and 10 of the 7135.  The black probe should go on pin 9 and the red probe should be on pin 10.  Keep the VC650 connected to the 5V DC source during this test.
e) Report your reading for pin 9 to 10.

11) Measure input impedance of the VC650 when it is in DCV mode.
a) Put your VC650 into DCV.
b) Using another DMM, put it into resistance mode.
c) Put other DMM red probe into the V/ohm jack of the VC650.
d) Put other DMM black probe into the COM jack of the VC650.
e) Report the other DMM reading in ohms.

12) Measure the voltage of the VC650 when it is in ohms mode.
a) Put your VC650 into resistance mode.
b) Using another DMM, put it into DCV.
c) Put other DMM red probe into the V/ohm jack of the VC650.
d) Put other DMM black probe into the COM jack of the VC650.
e) Report the other DMM reading in DCV.

[ModemHead]

Agree, something seems open.  This awesome work of art has a two-coil latching relay right at the inputs.      Sheet metal logic is definitely a possible failure point. Next one is the big 10Meg resistor right beside the relay.  Then a couple of 4051 mux chips, U4 and U5.  If the bottom end of the divider was not grounded through U4, you'd get an OL for practically any input.

As suggested, troubleshoot in DCV mode with known input, and try to trace it along through the divider and on to the ADC.

[gilbenl]

This doesn't look right (red circle). Might be the photo though.

[Cold-1]

Wow, you guys are great...

@gilbenl: This is just a bad angle of the photo. The resistor is solderd ok.

@retiredcaps: Yes, this is the right IC.

I will do all the other tests when I get home from work. Have a long day in front of me...

Regards

[retiredcaps]

For test 11, your other DMM will have to measure at least 12 or 13M ohm.  The cheap $5 830s only measure up to 2M ohm.

[Cold-1]

Ok, here are the measurements for 11) and 12):

11) Very interesting: One DMM (a Voltcraft 820) shows 7.4 M ohm, my Fluke 87V shows 0.9 M ohm
12) Again, Voltcraft and Fluke different: For the former, it changes from 124mV to 126mV, for the latter it changes from 63mV to 76mV

Edit: I just measured some reference resistors, and they are correctly identified on both meters.

[retiredcaps]

Test 13.

13) Based on your results for 11, here is test 13.
a) I have circled what I think is R77 in the schematic.  It should be a 10M ohm resistor (brown, black, black, green, blue = 10M ohm 0.25%).
b) I suggest desoldering one leg and verifying out of circuit it is 10M ohm with the Fluke 87V.

PS. Your Voltcraft 820 is a rebadged Uni-Trend multimeter.
PPS. Don't forget about test 7 and 10.

[retiredcaps]

14) When the VC650 is in DCV function, does it start autoranging from the mV or V?  I can't seem to find an user manual and pictures of the VC 650 front are very small on the Internet?

[Cold-1]

Hi,

7) It says "0.L V"
10c) "0.L V"
10e) "1.3mV"
13) Fluke says 10M ohm
14) When turned on, it starts in VAC mode. When swithing to DC, it autoranges at V, not mV

I have ordered 2 ICs for 5 bucks. If the one on the meter is really fried, I will replace it and at the same time put in a socket, if there's enough clearing from the display.

[retiredcaps]

So let me summarize what has been reported.  R = reported.

R1) VC650 displays 0L on DCV with no probes attached.  It also displayed 0L measuring a 5V source.  Both are clearly incorrect.
R2) VC650 displays 0L on resistance when measuring a resistor and with probes shorted.  Both are clearly incorrect.
R3) The voltage reading across IN LO and IN HI is reported at 1.3mV for a 5VDC source while the lcd displays 0L.  This is clearly incorrect.
R4) The input impedance of the VC650 when set to DCV is reported as 7.4M or 0.9M ohm depending on which meter is used.  This likely incorrect.  I expected a reading like 10 or 11.11M ohm.
R5) R77 in the schematic is a 10M ohm resistor and when measured out of circuit, it is 10M ohm.  This is correct.

I would like the results of a few more checks despite Cold-1 ordering two 7135 ICs (Note 1).

15) Can you measure the voltage between pins 9 and 10 without any probes plugged in and VC650 set to DCV?

16) I marked up the schematic and would like to see the voltage measurements at the following points a, b,c, d, e, f and g all with respect to ground.  For this test, please set the VC650 to DCV and connect it a 3V DC source (due to Note 3).  Again, please report the reading on pins 9 to 10 again for the 3V source.

Note 1: When I searched yesterday on ebay, the lowest price I could find was 3.29 USD.  I was surprised that Cold-1 found 2 for $5 and when I did another search today, I found the auction for $2.48 USD.

Note 2: I can't find a decent picture of the VC650 or user manual so I have attached a picture of the VC670 to show it doesn't have the traditional rotary switch.

http://clx.freeshell.org/pics/multimetres/voltcraft_vc670.jpg

http://translate.google.ca/translate?hl=en&sl=de&u=http://www.kappenberg.com/pages/wandler/gat073.htm&prev=search

Note3: Despite using a 4.5 digit 7135 IC, the VC650 is a 4000 count multimeter according to

http://home.arcor.de/bernd_kunze/digiscop.htm

Note 4: I finally found the original manufacturer of this multimeter despite clues in the schematic staring at me right in the face.  It is a Wens 50.  Wens is a Korean company.

According to company history, the Wens 50 series was developed and manufactured in Oct 1998 and is CE and GS (German Standard) certified.

http://www.wens.co.kr/sh_content_eng.html?page_gr=3&page_id=3

http://wens.en.ec21.com/WENS_50_SERIES_Concept_Multimeter--83354_83355.html

[Cold-1]

Ahhhrg! Stupid me... I measured the wrong IC lane.

The correct value for 3) is "160mV". For 15) it is starts at "-2mV" and goes up to "0.7mV" in a roughly 2 second time period.
Also concerning Note1: I did order them from AliExpress since the cheap offers from eBay are the same Chinese merchants.

Now for 16):
a) 3.2V
b) 0.423V
c) 420mV
d) 63mV
e) 138.6mV
f) 137.6mV
g) = f)

I don't know if it matters, but for measuring R22 (e and f) I had to remove the LCD.

[ModemHead]

Now for 16):
a) 3.2V
b) 0.423V
c) 420mV
d) 63mV
e) 138.6mV
f) 137.6mV
g) = f)

I don't know if it matters, but for measuring R22 (e and f) I had to remove the LCD.

None of those readings are near what would be expected.  With a 3.2V input, the voltage divider should end up in 1/10 mode.  Unloaded, the voltage at point b) would be 320mV, but with loading from another 10Meg DMM, it should read about 290mV.  The rest of the points all the way up to the ADC should be about the same value, maybe a little less as the series impedances add up.  e) and f) are at R23 not R22.

You can verify what mode the divider is in by checking U4 pins 9, 10, and 11. And verify that U4 pin 3 is really 0.0V (ground).

[Cold-1]

e) and f) are at R23 not R22

Yep, I ment R23 ;-)

Pin 3 on U4 is kind of 0V, it's  0.1mV. As for Pin 9-11: only Pin 9 seams to be high (0.7V). Pins 10 and 11 are around 6mV.

I have done a measurement of c) and d) again, and I realised, I got off by two pins (counting is an art...). Here are the correct values: c) = 428mV and d) = 147mV

I'm starting to loose hope. Maybe it's best to just trash the thing, although it'd give me a heartbleed... (I just don't like throwing stuff away that potentially could work)

[Lightages]

I'm starting to loose hope. Maybe it's best to just trash the thing, although it'd give me a heartbleed... (I just don't like throwing stuff away that potentially could work)

It is very commendable that you don't like to throw things away and I feel the same way. There are times though that the effort is not worth the result. In this case the result could be unreliable and if it has failed once, how will it fail the next time without warning. Could the failure the next time contribute to wrong readings which would then lead you apply the wrong voltage to a circuit, or believe you are going to be touching with safe voltages? The level of confidence you could have in the multimeter after it is fixed would be very low.

I also really am impressed with the time that retiredcaps has taken to help you. There is a time to quit at some point. For $50 you can have a brand new and safer meter to work with. It will be much more reliable.

I can recommend this:
http://www.ebay.com/itm/Uni-T-UT139C-True-RMS-Digital-Multimeter-with-Temperature-NCV-Backlight-/171213085325

[ModemHead]

0.7V is too low to be considered logic high for CMOS.  Based on the schematic, a 000 control input for U4 would select the 40M ohms range.  Not right at all.  If the state of this control bus is wrong, then we're chasing wild geese with the other test points.

The control bus originates from an octal latch (U2). Testing this will require testing the logic state of its outputs while selecting different functions, and trying to deduce if they're right or wrong.

[retiredcaps]

I'm starting to loose hope.

I wouldn't lose hope because

a) there is a schematic - try finding that for the Fluke 87V!
b) most of the parts in the multimeter are "standard" parts and easily and cheapily available
c) I still have some ideas and test points, but I can't guarantee they will work out and may lead to dead ends
d) there is a bill of materials so there is no guessing as to the values and identity of potentially bad parts
e) it only seems hopeless because of the timezone differences and other priorities that interfere

PS. If you give up all hope, check how much it would send the VC 650 to USA using the lowest/cheapest cost shipping and PM me.

[Cold-1]

@ModemHead:Thanx for the help. I measured U2 in the initial state (AC Volt) and then switched to DC Volt. All pins are 0V except the following:
ACV: Pin 6 = 0.75V / Pin 7 = 1.36V / Pin 18 = 1.36V / Pin 19 = 0.9V / Pin 20 = 0.49V
DCV: Pin 18 = 1.02V / Pin 19 = 0.74V / Pin 20 = 0.16V

I also measured U1 for fun and it shows the exact same result for both modes.

@retiredcaps: Thank you for the heads up and all the time you invest in this. I really appreciate it!

I know what most of the parts do by them self and could replace them without a problem. But sadly, I don't know how to read schematics, so I'm totally blind in this regard :-(

[retiredcaps]

17) Can you provide pictures of the back of the front case where a traditional rotary switch would be?  I want to see how those rubber buttons make contact to the pcb.

edit: typos

[ModemHead]

@ModemHead:Thanx for the help. I measured U2 in the initial state (AC Volt) and then switched to DC Volt. All pins are 0V except the following:
ACV: Pin 6 = 0.75V / Pin 7 = 1.36V / Pin 18 = 1.36V / Pin 19 = 0.9V / Pin 20 = 0.49V
DCV: Pin 18 = 1.02V / Pin 19 = 0.74V / Pin 20 = 0.16V

I didn't ask you to check the power rails because the processor is running, but I should have, my mistake.  U2P20 is connected to Vcc and should be about 5V, but you have indicated 0.5V or less.  It would be good to verify the power supply before going further.

With the power off, first verify that the COM input jack is connected directly to the battery negative terminal with a continuity check.  The schematic indicates that it should be.  Then use the COM input jack as your ground.

There are many locations on the schematic labeled "Vcc" and "Vss".  There is also a rail labeled "Vcc(CPU)" which is also 5.0V, but does not get turned off by the auto-power-off feature.  Try to find a convenient spot for each and measure the voltage with respect to ground.  Careful with the probes not to short anything.