Fluke 8842A Blank Display

[omegaone]

I recently bought 8842a from ebay. When arrived the display is blank but can hear the relays clicking.
Pin 14 & 16 near transformer reading 4.1 v ac. By ribbon cable near vfd reading 4 v ac.

Any help would be appreciated.

[TiN]

Broken glass?

[coromonadalix]

No needs to multi thread


You have service manuals available,   check supply lines, contacts   ..

Problems with the transformer solders,  check this first  as you threaded here
https://www.eevblog.com/forum/repair/yet-another-fluke-8842a-transformer-solder-fail/

schematics
https://elektrotanya.com/fluke_8842a_digital_multimeter.pdf/download.html

Other threads
https://www.eevblog.com/forum/repair/vfd-wanted-for-fluke-8842a/
https://www.eevblog.com/forum/testgear/fluke-8840a/
https://www.eevblog.com/forum/repair/fluke-8840a-teardown-and-repair/

[omegaone]

No physical damage on VFD except top left corner has spot of discoloration.Picture attached.

[coromonadalix]

as TIN wrote 

Check if the VFD is physically ok, is someone before you had a peek in it or tried to do some job ...

You have voltage for the filament and voltage(s) for the display driver ic

And the transformer is known to do cold joints ..

Check your page 9-12
+30vdc,  the 7824 is 6vdc boosted with an zenner to give you +30vdc
-30vdc,   the 7924 is 6vdc boosted with an zenner to give you -30vdc
-6.2vdc
+5vdc
+15vdc
-15vdc
+7.5vdc
-8.2vdc

And the final 4.5 volts ac for the filaments, has you said it was there

As you see tons of voltages ... check for bulged or leaking capacitors

You have the frequency ref signal going to u202,  50hz or 60 hz ??

[TiN]

Before checking joints and voltages, worth to check if glass itself is ok (no vacuum lost).

[omegaone]

Table 6-23  Power Supply Voltages

+ 5 Vdc        read   + 4.88 Vdc
+7.5 Vdc      read   + 7,25 Vdc
+15 Vdc       read   +14.76 Vdc
+30 Vdc       read   + 29.4  Vdc
-8.2 Vdc       read   -  7.64  Vdc
-15  Vdc       read   - 15.05 Vdc
-30  Vdc       read   - 29.8  Vdc

- 6.2  Vdc   read    - 6.1 Vdc

No leaky or bulged capacitors.

Unable to check reference signal to u202 (I do not have a scope at this time) but assuming frequency should be 60 Hz)

[coromonadalix]

I think i saw somewhere the flat cables had problems on some models not necessarily your  but worth to check, check if the display driver ic is supplied, we can assume the power supplies are good.

On page 261   Something near the main cpu is bad, clock crystal, bad contacts, flat cable problems,  all the ic's between the display and the main cpu (ic202), check the reset line if it not stuck at low (c204 is bad) it must be 5v  to run,

U212 is the display driver, and its 7 ic's around him.

The ac sense line is going to u202 pin 34, 50 or 60 hz, maybe it change the display refresh speed ??
Or an corrupted U222 (2732 EEprom) and or the other U220 (x2804/2816)

Check if the ic's are well socketed, no unbend pins under the chip,   happened to me in the past, the pin was bent and made little contact on the socket (not inserted), pull them and re-seat them ?

[omegaone]

Thank you. I am getting an oscilloscope and will start with the  In-Guard micro Chip U202-2 as per your initial suggestion as I have not completed this step yet. And I will follow through each suggestions and post the results.

As of now, I was simply checking ic u217 and u215. ( 30 Vdc ) One of them as soon as I touched the contact pin, affected the relay and the relay does not click anymore. (relay used to click as soon as power button pushed in,  K301 or K401)

[lowimpedance]

 You will need a new display, the one in the unit is 'FUBAR" !. The white patch indicates the internal space is no longer under vacuum.
It should be a blackish silver color for the 'getter', white indicates vacuum is lost.
 Good news though its most likely the rest of the instrument is probably okay, the bad of course is trying to find a replacement.

[TiN]

Yep, somehow I did not display photo before. It's good only as pretty piece of glass art now I have same on my 8842A, but I don't intend to repair the unit, just a parts mule.

[omegaone]

I kind of suspected that and thanks for your input.

If anyone happen to know where to post to source for a VFD please let me know.

[Ordinaryman1971]

When you have those white spots on VFD that means that the air got inside and it's dead. You can still use it over GPIB, the meter is probably fully functional.

[lowimpedance]

When you have those white spots on VFD that means that the air got inside and it's dead. You can still use it over GPIB, the meter is probably fully functional.

huh..... did I not say that 3 posts up  . BTW this also applies to valves in general ie white getter= no or soft vacuum. Simply put air has got in !.
But yeah use the GPIB to test functionality, (was this an option or always fitted to the '42' ??).

[mzacharias]

as TIN wrote 

Check if the VFD is physically ok, is someone before you had a peek in it or tried to do some job ...

You have voltage for the filament and voltage(s) for the display driver ic

And the transformer is known to do cold joints ..

Check your page 9-12
+30vdc,  the 7824 is 6vdc boosted with an zenner to give you +30vdc
-30vdc,   the 7924 is 6vdc boosted with an zenner to give you -30vdc
-6.2vdc
+5vdc
+15vdc
-15vdc
+7.5vdc
-8.2vdc

And the final 4.5 volts ac for the filaments, has you said it was there

As you see tons of voltages ... check for bulged or leaking capacitors

You have the frequency ref signal going to u202,  50hz or 60 hz ??

This is true regarding the bad solder at the transformer, and the connections may look fine. Re-solder them all.

Some may not stand out as belonging to the transformer. Be careful not to miss any.

My 8840A (same meter basically) had intermittent problems powering up, and although this is not your symptom, it's certainly worth a try.

This would be a standard FL tube arrangement, with a filament voltage of perhaps 3-8 volts AC or DC, there will be a -30 Volt supply, and the display driver IC will have a (probably) 5 volt supply.

FL tubes are kind of notorious for developing bad solder connections, especially at the filaments.

[omegaone]

Thank you.
Yes I did re solder the supply voltage 14,15,16 and checked the voltage up to the ribbon cable. (4.0 Vac at ribbon cable)

The offset DC voltage coming from U217 and U215 at 30 Vdc checked to be ok.

Now as few members have already stated the VFD possibly lost the vacuum and not functioning. I am looking for a replacement.

I am also continuing the trouble shooting as per the manual, to check for In Guard micro controller U202-2. The signal should be 8Mhz sine wave with approximately 3.5 V peak to peak. I have yet to confirm this with a scope. (at this point I only know the pin 2 and ground pin 11 at U202 only showing 0.3 of ac at 60Hz volt when measured with a voltmeter)

The meter relays ( k301/ K401 ) used to click when power button is pushed, now they do not. I also need to check these relays.

If anyone has any input please share, as I intend to troubleshoot this meter and post detail description of all that happened .

[pelule]

Have you measured the voltages/signals at the VFD pins directly. If there is a legal signal (multiplexed, grids & anodes) and if the voltage at the filaments are correct - there is just one possible cause for a blank VFD display - a damaged VFD.
See also: https://www.instructables.com/id/A-Simple-Driver-for-VFD-Displays/
/PeLuLe

[tooki]

I kind of suspected that and thanks for your input.

If anyone happen to know where to post to source for a VFD please let me know.

Take a look at https://www.eevblog.com/forum/repair/fluke-45-vfd-display-broken/

In there, someone shows an 8840A with an LED replacement screen.

[coromonadalix]

As i wrote on the thread, the Intel 8279-5 can do  led displays, there some schematics floating around the web,  it just need to be tried easier said than done  loll

Some nice infos and timings : http://ece-research.unm.edu/jimp/310/slides/8086_IO3.html

And some specs and app notes too :  http://www.idc-online.com/technical_references/pdfs/electronic_engineering/Keyboard_and_display_interface_using_intel_8279_microprocessor.pdf

[omegaone]

Good information, Thank you.

My first option is to try and replace the VFD. The main reason is, the reason for getting this particular meter is for precision and accuracy within a budget!

Replacing to and LCD would be and option 2 and I rather not pursue it at this time.

I am still going through the board checking the required items. My scope is not fully up and running. The In-Guard microcomputer clock at pin 2 and ground pin 11 should be 8Mhz approximately 3.5 Vac peak to peak as per the manual. (see attached pages) When using a voltmeter, I am reading 0.30 Vac only.

[lowimpedance]

 What's the bandwidth of the voltmeter you are using ?. If the signal is  8 MHz  then it will be beyond the the bandwidth of most voltmeters. Fire up the Oscilloscope.
 As for the Led mod, sure its relatively easy to do , first you need to change/bodge in the driver Chip's watching out for pin out differences. Then you need a simple supply off the in guard supply rail to isolate switching noise from the main analogue supplies. Next you need to make a suitable display drop in with annunciators which is the harder bit !.
When doing that make sure you keep the average current to around 70 mA or so to suit the selected Led brightness. Its a matrix style display .
And make sure you are careful with the display ground return connection location if you don't want to upset  the A/D.
 All in all best to look for a parts donor and avoid the alternative display mods as in my opinion this meter is not worth the effort ,unless you are keen to learn something or have no other alternative as I did and I also thought how hard could it be to do it cheaply surely not  .

[omegaone]

I borrowed a Fluke 289. Not look at the bandwidth range, it did read about 60 Hz but it must be in error.Yes do realize doing a mod is no a desireable option.
There is a possibility Ordinaryman1971  might be able to get me a VFD, waiting to replace it.

[coromonadalix]

That's why i hate VFD's  they are $#%#@^ to get when they break, long live leds display  loll

[lowimpedance]

long live leds display  loll

Amen to that.

[omegaone]

"A big Thank you" to Ordinaryman 1971 who was able to find a replacement VFD and tested it before sending it to me all the way from US to Canada.
VFD replaced.  The display flashes but goes off at this point!