Author Topic: Help with Fluke 77 II repair  (Read 6258 times)

0 Members and 1 Guest are viewing this topic.

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Help with Fluke 77 II repair
« on: December 19, 2014, 10:56:08 am »
Hello there,

I have a Fluke 77 II in perfect aesthetic condition, but it has a defect that renders it unusable.  The service manual is available with full schematics :

http://exodus.poly.edu/~kurt/manuals/manuals/Fluke/FLUKE%2077,%2075,%2073,%2070,%2023,%2021%20Series%20II%20Service.pdf

The defect appears in the following manner :
 - Turn on the DMM
 - Self-test screen with all segments appears
 - Self-test screen goes away and normal operation starts (this is true for any mode)
 - After a few seconds (at most 10) the screen stops updating and you cannot do anything except power-off, the mode selector has no effect
 - This "corrupted" state persists for a few minutes after power-off and even battery pull.

I've tried to follow the troubleshooting guide in the service manual even though it is not obvious if my defect is 2, 4 or 5.
I've checked for obvious stuff, like shorted parts and pins on U1 and U2. The crystal oscillator works as expected even after the corruption.

I don't understand what keeps the corrupted state after the battery pull, it is not the caps between Vss and Vdd, since they're all empty after power off and the corruption persists. Maybe it is some other cap or the temperature of some component ? Do you think that U1 or U2 may be fried ? any other ideas ?

Thank you,

Edit: corrected a typo
« Last Edit: December 20, 2014, 02:07:27 am by Grapsus »
 

Offline retiredcaps

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: Help with Fluke 77 II repair
« Reply #1 on: December 19, 2014, 06:45:34 pm »
- After a few seconds (at most 10) the screen stops updating and you can do anything except power-off, the mode selector has no effect
Can you clarify the above? Did you mean that you cannot do anything except power-off?

In addition, can you give us some background of this meter?  Is this yours?  Did this problem just start happening on its own?  Is pcb nice and clean?

Quote
Maybe it is some other cap or the temperature of some component ?
Some suggestions.

1) Measure the current consumption of the 77.  It should be around 250 to 300 uA for most functions.
2) If current draw is significantly more, then there could be a component that is running hot and shutdowns causing the "corruption".  After a cool down period, then it works again.
3) Post a clear focused picture of the pcb (front and back).
4) Excavatoree said in one post that sometimes the soldering of U1 is not very good and can cause problems.   Maybe take a 10x jewelers loupe and take a very close look.

http://www.eevblog.com/forum/testgear/fluke-77-strange-demise/15/
« Last Edit: December 19, 2014, 07:01:50 pm by retiredcaps »
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #2 on: December 19, 2014, 08:14:44 pm »
Hello, thank you for your reply.

- After a few seconds (at most 10) the screen stops updating and you can do anything except power-off, the mode selector has no effect
Can you clarify the above? Did you mean that you cannot do anything except power-off?

Exactly, changing mode has no effect, only the OFF position turns the screen off. If I understand the schematic correctly, the rotary switch has one face which is "hardware" ie. it cuts the battery in the OFF position and the other face is purely "software" because it is tied to IO pins on U1.

In addition, can you give us some background of this meter?  Is this yours?  Did this problem just start happening on its own?  Is pcb nice and clean?

The meter is not mine since I was 2 years old when it was produced in 92  ;D I purchased it locally from a guy who is a retired test equipment technician. He said the meter has never been serviced. I was aware of the defect but went for it anyway because the price was good and the meter is in perfect condition, it looks like it has been barely used and comes with the original probes.

The PCB is clean except for some flux residue here and there.

Quote
Maybe it is some other cap or the temperature of some component ?
Some suggestions.

1) Measure the current consumption of the 77.  It should be around 250 to 300 uA for most functions.
2) If current draw is significantly more, then there could be a component that is running hot and shutdowns causing the "corruption".  After a cool down period, then it works again.
3) Post a clear focused picture of the pcb (front and back).
4) Excavatoree said in one post that sometimes the soldering of U1 is not very good and can cause problems.   Maybe take a 10x jewelers loupe and take a very close look.

http://www.eevblog.com/forum/testgear/fluke-77-strange-demise/15/

I will post good pictures when I get home.
I can confirm that U1 looks hand soldered and that the solder quantity is uneven among pads. I have a hot air station so I will be able to re-solder it completely if necessary.
 

Offline ModemHead

  • Frequent Contributor
  • **
  • Posts: 686
  • Country: us
  • No user-serviceable parts inside.
    • Mr. ModemHead
Re: Help with Fluke 77 II repair
« Reply #3 on: December 20, 2014, 01:51:36 am »
That's an odd failure. I would have suggested checking the clock, but I see you've already done that.

You're right about the switch. The bottom is function code (digital), the top is input configuration (analog).

The closest problem I've ever had is a 75-II that kept getting hung in the boot-up sequence intermittently. I re-flowed each pin on U1 to clear up the problem. That was a couple years ago, it hasn't malfunctioned since.

When I was trouble-shooting that problem, I used a very low-tech method called "poke it with a stick".  :)  I noted a change in the display when I pressed on various parts of the PCB with a plastic tool.  I never did see the fault visually.
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #4 on: February 15, 2015, 12:22:54 pm »
Hello again,

Sorry I left this multimeter for a while. Here are the pictures.

We can now be sure it's not the soldering job on U1 or U2 : I completely desoldered them with a hot air station, cleaned everything and soldered them back, same problem  |O

Also, I tried to see if the corrupted state of the multimeter is kept between power cycles by the temperature of some component and it looks like it's U2 (LCD controller). If the multimeter starts correctly and I keep freezing U2 with an air duster upside down, it keeps working. I managed to keep it running in this way for 5 minutes while it usually hangs after 10 seconds max. Do you think U2 has some kind of fault ?

Thank you,
 

Offline retiredcaps

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: Help with Fluke 77 II repair
« Reply #5 on: February 15, 2015, 05:10:49 pm »
Do you think U2 has some kind of fault ?
If U2 is getting hot, U2 could be internally shorted or the surrounding components are shorted.

If U2 is bad, getting a replacement would be difficult and probably not economically feasible.

However, if the surrounding components are shorted, then you might have a chance of repairing it.  For example, you might want to check the capacitors to see if they are shorted?

Candidates like C13, C4, C10, C2?
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #6 on: February 16, 2015, 02:17:06 am »
Nothing is getting really hot on the board : it always draws around 250 uA whether the glitch already occurred or not. But it is a fact that if I keep U2 very cold it keeps running, I tried it several times and it's not randomness, without doing anything the meter stops working after 10 seconds and with the air duster I managed to keep it running for 5 minutes and the glitch came soon after I stopped cooling the IC.

I tried removing all decoupling caps (C13, C4, C10, C14) also C2 which generates an initial reset signal and the meter behaves exactly the same, it works for several seconds then glitches.

I'm more and more inclined to think that U2 has some kind of subtle defect. The communication between U1 and U2 is not documented but I tried to understand how it works. You have two 4-bit busses C0-3 (probably stands for Control) and D0-4 (data ?) then there is a CLK signal and finally NDAV.
  - CLK just looks like a cleaned up digital version of the 32 kHz crystal signal and it works even after the glitch, so no worries here.
  - The data bus looks unidirectional from U1 to U2, I think it's dedicated to A/D samples, after the glitch I still see data spit by U1 on this bus.
  - NDAV is maybe "new data available" ? because it is driven at roughly 25 Hz by U1 and 25 Hz is the update rate of the Bar Graph so I think U1 pulses this line each time is has a new sample. This line also functions after the glitch.
  - Finally the control bus is probably data from U2 to U1 (or maybe it is bidirectional ?) in order to tell U1 to switch to a different range. This bus has a lot of activity before the glitch and nothing after.
 

Offline retiredcaps

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: Help with Fluke 77 II repair
« Reply #7 on: February 16, 2015, 11:29:52 am »
You did more troubleshooting then I could have done because I don't own a scope.

At this point, I don't have any more ideas.  :-//
 

Offline ModemHead

  • Frequent Contributor
  • **
  • Posts: 686
  • Country: us
  • No user-serviceable parts inside.
    • Mr. ModemHead
Re: Help with Fluke 77 II repair
« Reply #8 on: February 16, 2015, 11:54:40 pm »
It sounds like you've done a bang-up good job trouble-shooting the problem. Thanks for documenting it, I learned something reading your comments.  In my experience the large ASICs usually either work completely or fail completely.  This does indeed sound like a subtle defect in U2.  Here's a really dumb idea, wonder if U2 would still work without glitching, on a slightly reduced supply voltage?
 

Offline DmitryL

  • Regular Contributor
  • *
  • Posts: 235
  • Country: gb
Re: Help with Fluke 77 II repair
« Reply #9 on: February 17, 2015, 12:22:31 am »
Hello there,
at renders it unusable.  The service manual is available with full schematics :
The defect appears in the following manner :
 - Turn on the DMM
 - Self-test screen with all segments appears
 - Self-test screen goes away and normal operation starts (this is true for any mode)
 - After a few seconds (at most 10) the screen stops updating and you cannot do anything except power-off, the mode selector has no effect
 - This "corrupted" state persists for a few minutes after power-off and even battery pull.


Once I had  a very similar problem with Fluke 77. I just replaced 32Khz qaurtz oscillator; works like a charm since.
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #10 on: February 17, 2015, 01:34:43 am »
Once I had  a very similar problem with Fluke 77. I just replaced 32Khz qaurtz oscillator; works like a charm since.

Thank you for the advice, unfortunately I have already tried another quartz and it makes no difference  :(

Here's a really dumb idea, wonder if U2 would still work without glitching, on a slightly reduced supply voltage?

That's an interesting idea, I will give it a try, but just powering the meter with a lower voltage won't achieve this. If I understand the schematic correctly, the battery is attached between Vss and Vdd and U1 has some kind of built-in regulator that maintains GND-Vdd=3.3V by adjusting GND. Hence I will have to lift the Vdd pin on U2 and tie it to an adjustable power supply. If it works it would constitute a pretty ugly fix, but at this stage it's worth a try.

Regarding the communication protocol between U1 and U2 would it interest you people that I reverse engineer it further ? It looks like the same ADC "AP 75" is used in a lot of meters, even the III series use the same signals, so it may be interesting to know how this communication works in detail. I don't have a logic analyzer, but since the clock is only 32 kHz I can attach the signals to GPIOs on a linux dev board and gather the data with a small script.
 

Offline ModemHead

  • Frequent Contributor
  • **
  • Posts: 686
  • Country: us
  • No user-serviceable parts inside.
    • Mr. ModemHead
Re: Help with Fluke 77 II repair
« Reply #11 on: February 17, 2015, 01:49:05 am »
That's an interesting idea, I will give it a try, but just powering the meter with a lower voltage won't achieve this. If I understand the schematic correctly, the battery is attached between Vss and Vdd and U1 has some kind of built-in regulator that maintains GND-Vdd=3.3V by adjusting GND. Hence I will have to lift the Vdd pin on U2 and tie it to an adjustable power supply. If it works it would constitute a pretty ugly fix, but at this stage it's worth a try.
Sorry I should have elaborated a bit, but yes I was just thinking maybe a Schottky diode in line with Vdd if you could find a suitable break point to avoid affecting anything else but U2.  If the current draw doesn't go up, obviously the thermal fault is on some kind of hairy edge, maybe a few hundred millivolts will give it a margin.

Regarding the communication protocol between U1 and U2 would it interest you people that I reverse engineer it further ?
I suspect the audience may be ummm... limited.  I'd be curious of course, I've never gone quite that far into how they work.  If I perceive a digital-domain fault, I usually give up pretty quickly.   :D
 

Offline retiredcaps

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: Help with Fluke 77 II repair
« Reply #12 on: February 17, 2015, 10:48:38 am »
Regarding the communication protocol between U1 and U2 would it interest you people that I reverse engineer it further ?
I would definitely be interested in seeing your progress.  I'm studying all aspects of how multimeters work and this would help for sure.
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #13 on: February 17, 2015, 02:26:05 pm »
Sorry I should have elaborated a bit, but yes I was just thinking maybe a Schottky diode in line with Vdd if you could find a suitable break point to avoid affecting anything else but U2.  If the current draw doesn't go up, obviously the thermal fault is on some kind of hairy edge, maybe a few hundred millivolts will give it a margin.

It was a brilliant idea! There are 4 pins on U2 tied to Vdd: 8, 20, 21 and 38. I lifted all of them. Found that 8 and 38 are internally connected, so it left me with 3 points to try with Vdd minus something. Normal Vdd to ground value seems to be 3.1 Volts. One after another I connected the lifted pins to a lab power supply and played with various voltages. Pins 20 and 21 had no effect on the glitch, but 8+38 did delay it significantly! I had to go as low as 2.65 V in order for the glitch to stop occurring completely. At 2.40 V the CPU went in brown-out and started to display crazy stuff. Hence there is a sweet spot around 2.55 V with 100 millivolts of safety margin.

By the way I think this latest experiment confirms that the fault is inside U2, since cutting the power of the lifted pins for a while removes the fault and the meter restarts correctly while U1 and all the other circuitry remained powered.

Finally I decided to implement a fix with a diode. It's been on for half an hour and still working :D It's not ideal and I'll not count too much on this meter to work at all times but it's better than throw it. It somehow reminds the recent Dave's struggle with the intercom, I will have the same problem of trust, only in my case I can live with it. Also my bodge only concerns the micro-controller and not the ADC so it doesn't affect the measures taken by the meter. Maybe one day I'll find a cheap broken Fluke 77/75/73/70/23/21 and salvage the U2 chip from it.

Thank you all for the advice. I will still try to reverse the communication protocol out of curiosity.
 

Offline retiredcaps

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: Help with Fluke 77 II repair
« Reply #14 on: February 17, 2015, 02:44:53 pm »
It was a brilliant idea!
I am glad you got it working and thanks for sharing the pictorial results.
 

Offline ModemHead

  • Frequent Contributor
  • **
  • Posts: 686
  • Country: us
  • No user-serviceable parts inside.
    • Mr. ModemHead
Re: Help with Fluke 77 II repair
« Reply #15 on: February 17, 2015, 03:45:13 pm »
Nice work on the implementation.   :-+   Yeah, obviously you can't depend on it, but hey isn't it fun to "win" and keep a nice piece out of the "bin".
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #16 on: February 18, 2015, 01:11:40 am »
A few days ago I asked for a quote to replace U2 on one of those parts supply sites. Look at what they replied, this is hilarious :

Quote
Hi Alexis,
Thank you for using sierraic
We have received your Request for Quote, and we are happy to quote you the following:
Part Number Requested: 659656
Part Number Quoted: 659656
Manufacturer Quoted: FLUKE
Quantity Requested: 1
Quantity Quoted: 1
Price Quoted: $335 EACH

Do you think they are planning on buying a brand new meter and tear it apart in order to get the IC ?  :-DD
 

Offline Altizachen

  • Newbie
  • Posts: 3
  • Country: ru
Re: Help with Fluke 77 II repair
« Reply #17 on: February 17, 2018, 04:22:50 am »
Hello. I bought a flawed Fluke 23 series II on eBay.
Symptoms of 100% are the same as those of the author. The device after switching on
worked 10 seconds, then everything like Grapsus. Current Consumption
about 280 mkA. If you cool the U2 chip, the device worked
while there was cooling. I thought, if the device reacts to cooling,
then it can also react to heating. I deleted the display, manufactured
from the cardboard shield with a hole in size U2, so that the hot air
got only on the case of the chip and heated the microcircuit for 1.5 minutes
a termofan with a temperature of 250 C. After that, the device began to work for 42 seconds.
Then I repeated this procedure for another 3 minutes and 290 C. Multimeter
began to work normally. The current consumption became 250 - 260 mkA.
« Last Edit: February 19, 2018, 12:24:17 am by Altizachen »
 
The following users thanked this post: ModemHead, lowimpedance

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #18 on: February 17, 2018, 04:59:48 am »
That's weird, I try heating mine since I got nothing to loose. Even with lowered voltages I still experience crashes from time to time.
 

Offline mos6502

  • Frequent Contributor
  • **
  • Posts: 531
  • Country: aq
Re: Help with Fluke 77 II repair
« Reply #19 on: February 17, 2018, 05:08:02 am »
Sounds like moisture getting into the chip ... Altizachen, try if you can get it to malfunction again by soaking the chip in water  :D
for(;;);
 

Offline Altizachen

  • Newbie
  • Posts: 3
  • Country: ru
Re: Help with Fluke 77 II repair
« Reply #20 on: February 18, 2018, 09:18:09 pm »
Yes, it's funny, but the device works confidently.On the photo shield after the procedure.
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #21 on: February 18, 2018, 10:10:47 pm »
I confirm that this worked on my Fluke 77 too !
I just blasted the chip package with 400°C air from a rework station for about 2 minutes.
The meter is now glitch free. I even removed the voltage lowering diode and never had the display frozen.

Not sure why this works, but since replacing U2 is not an option, this is a great trick.
 
The following users thanked this post: ModemHead, lowimpedance

Online helius

  • Super Contributor
  • ***
  • Posts: 2462
  • Country: us
Re: Help with Fluke 77 II repair
« Reply #22 on: February 18, 2018, 11:06:35 pm »
Not sure why this works, but since replacing U2 is not an option, this is a great trick.
I'm not a solid state expert, but my guess would be that the die attach solder began to migrate into layers that formed a junction and this treatment reflowed it.
 

Offline Grapsus

  • Regular Contributor
  • *
  • Posts: 223
  • Country: fr
Re: Help with Fluke 77 II repair
« Reply #23 on: February 21, 2018, 07:13:18 am »
I'm impressed ! I made myself use this meter for a few days to check it for glitches and it's been working flawlessly, so the fix seems permanent  :-DMM

Also the Fluke 77 is actually a very cool DDM. It's small and light. Readings are very stable and inspire confidence. There's something neat about the minimalism of it. The only thing that's really annoying compared to higher end series is the slow continuity tester.
 

Offline Altizachen

  • Newbie
  • Posts: 3
  • Country: ru
Re: Help with Fluke 77 II repair
« Reply #24 on: February 21, 2018, 10:58:32 pm »
I completely share your opinion. I also really love the 70 series. I have 6 of them.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf