Fluke 5700a repair

[veedub565]

After posting on a few different and random threads I thought it might be best to create my own Fluke 5700a repair thread. I've already had some hints and tips from TiN. I've already made a few mistakes and done some things wrong, but no harm seems to have been done.

So, I have acquired a Fluke 5700a physically and cosmetically in good shape, but with some faults

Date of last cal 2013

I guard rear terminal post snapped off

ETIME = 7422640 (14yrs)

FATALITY =
1/20/92    10:21:03   Fault    1822   5220  No Longer Connected
7/13/28    13:45:13   Fault    1824   5205 No Longer Connected
7/18/28    13:30:00   Fault    1824   5205 No Longer Connected
3/18/95    08:54:07   Fault    1602   Unexpected NSA From Inguard
3/18/95    08:54:17   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:54:27   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:54:37   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:54:47   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:54:57   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:55:07   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:55:17   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/95    08:55:27   Fault    4012   Sequencer Timed Out Waiting For Inguard
12/20/00    12:36:08   Fault    1602   Unexpected NSA From Inguard
12/20/00    12:36:18   Fault    4012   Sequencer Timed Out Waiting For Inguard
3/18/13    10:38:27   Fault    232   5725 Illegal/Unexecutable Command
3/18/13    10:39:26   Fault    232   5725 Illegal/Unexecutable Command

I ran the self diag and it came up with a few errors (there was an error with the A12 card, but this was resolved by fitting a spare card.
Yes I know the cal values will be incorrect now. I have access to a cal lab and I may be able to arrange to have this unit re-certified IF I get it repaired.
Yes I know swapping cards willy nilly without first recapping and smoke testing the PSU is NOT they way to do it. That's a good way to blow up a good card. Too late, it is done, and I appear to have been quite lucky.

A8: Relay Fault (3819)
A15: DC HV Amp Offset Fault (3107)
A5: ODB Output Attenuation Fault (3914)
A5: 10DB Output Attenuation Fault (3915)
A5: 20DB Output Attenuation Fault (3916)
A5: 30DB Output Attenuation Fault (3917)
A5: 40DB Output Attenuation Fault (3918)



to check the base hardware I ran self diag without A5/A6 and A14/A15/A16 cards and only reported fault was
A8: Relay Fault (3819)

[veedub565]

First order of business now is to replace ALL electrolytics as they will be at the end of their useful life. Also to replace the carbon composite resistors, especially the 1W ones which are to be replaced with 3W ceramic. I'm also going to replace the fans with some Sunon ones.

How do I know which is the correct lead pitch for the capacitors ? some appear to be 5mm, some 10mm, some in between. And there are 4 or 5 different options when ordering new ones

Also a good idea to remove all the cards and give the chassis a good clean and remove all the dust inside.

I would also like to be able to get the cal report from it, but I'm still having some difficulty with this.

-> *IDN?
<- FLUKE,5700A,5185014,HA*

-> CAL_RPT?
 ! VI_ERROR_TMO: A timeout occurred
Visa ErrorCode: 0xBFFF0015 (-1073807339)

<- FLUKE,5700A,5185014,HA*
-> cal_rpt? check
<- "«x0A»«x0A»«x0A»«x0A»«x20»«x20»«x20»«x20»------------------------------------------------------------------------«x0A»«x20»«x20»«x20»«x20»JOHN«x20»FLUKE«x20»MFG.«x20»CO.,«x20»INC.«x20»«x20»«x20»«x20»«x20»S/N«x20»5185014«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»5700A«x20»CALIBRATION«x20»CHECK«x0A»«x20»«x20»«x20»«x20»PRINTED«x20»ON«x20»29.05.08«x20»AT«x20»15:46:54«x20»5700A«x20»«x20»S/N«x20»5185014«x0A»«x20»«x20»«x20»«x20»--------------------«x0A»«x20»«x20»«x20»«x20»------------------------------------------------------------------------«x0A»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»MODULES«x20»PRESENT«x0A»«x20»«x20»«x20»«x20»------------------------------------------------------------------------«x0A»«x20»«x20»«x20»«x20»Software«x20»Revision«x20»HA*«x0A»«x20»«x20»«x20»«x20»Switching«x20»Matrix«x0A»«x20»«x20»«x20»«x20»DC«x20»Volt«x20»Module«x20»«x20»«x0A»«x20»«x20»«x20»«x20»AC«x20»Volt«x20»Module«x20»«x20»«x0A»«x20»«x20»«x20»«x20»220V«x20»Module«x20»«x20»«x20»«x20»«x20»«x0A»«x20»«x20»«x20»«x20»1100V/2A«x20»Module«x20»«x0A»«x20»«x20»«x20»«x20»Current«x20»Module«x20»«x20»«x0A»«x20»«x20»«x20»«x20»Ohms«x20»Module«x20»«x20»«x20»«x20»«x20»«x0A»«x20»«x20»«x20»«x20»Wideband«x20»Module«x20»«x0A»«x20»«x20»«x20»«x20»Hires«x20»Osc«x20»Module«x0A»«x20»«x20»«x20»«x20»Rear«x20»Panel«x20»«x20»«x20»«x20»«x20»«x20»«x0A»«x20»«x20»«x20»«x20»------------------------------------------------------------------------«x0A»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»«x20»ARTIFACT«x20»CALIBRATION«x20»DATES«x20»AND«x20»TEMPERATURES«x0A»«x20»«x20»«x20»«x20»-------------------------------------------


I'm not sure what I'm doing wrong here. I'm not familiar with Python unfortunatly, and I'm not sure I'm sending the correct commands and the response seems to be a bit rubbish 

[Kleinstein]

For the capacitors there may be an info in the parts list if provided in the service manual.
Alternatively measure the real thing.  If it is off a little (e.g. 6 mm VS 1/4 inch) this can be sill acceptable. Chances are there are not that many different types actually used. So if you don't find tham all in the measurement run - same values are likly same spacing / though not sure).

With the carbon composite resistors one may have to be carefull in the high voltage and higher frequency parts. Wire wounds usually have more inductance and may not have the same withstand voltage. Higher resistance values (e.g. > 100 K) can be tricky as wire wound. Sometimes the resistor case  / rating in choosen because of the peak voltage and not just power.

[veedub565]

For the capacitors there may be an info in the parts list if provided in the service manual.
Alternatively measure the real thing.  If it is off a little (e.g. 6 mm VS 1/4 inch) this can be sill acceptable. Chances are there are not that many different types actually used. So if you don't find tham all in the measurement run - same values are likly same spacing / though not sure).

With the carbon composite resistors one may have to be carefull in the high voltage and higher frequency parts. Wire wounds usually have more inductance and may not have the same withstand voltage. Higher resistance values (e.g. > 100 K) can be tricky as wire wound. Sometimes the resistor case  / rating in choosen because of the peak voltage and not just power.

I had a look at the parts list in the manual, it provides all the values, tolerances etc, and even part numbers (although finding data on 30yr old part numbers is hard to say the least) I have started physically measuring the pitch, mostly the bigger ones are 10mm and the smaller ones are 5mm but there are a few in between. I guess it doesn't really matter too much, even Fluke have stretched the legs out on some. I just want to try and make it neat and tidy. It's more important for the bigger ones that the pitch is correct.

I'm going to use Carbon Film to replace the old Carbon composite resistors as this appears to be what Fluke have used in the series II version. For the higher power values (3W) Fluke list ceramic, but the pictures I've seen show a thick film resistor being used, so I'll go for them (https://www.digikey.com.au/product-detail/en/tt-electronics-irc/GS-3-100-2203-F-LF/989-1203-1-ND/2408023)

[veedub565]

Well picking the right replacement caps is proving to be a right PITA. The part numbers themselves contain no information at all relating to the lead pitch. So I've been physically measuring the pitch and there appears to be 4 different flavours, 2.5mm, 5mm, 7.5mm, and 10mm. Here's an example of a problem I just discovered

There are 11 22uF 35v electrolytics in the whole unit, I measured one and it had a 5mm pitch. Great I'll order 11 of those.. . Not so fast! some of them are a 5mm pitch and some of them are a 2.5mm pitch, I wouldn't be surprised to also see one with a 7.5mm pitch.

So I've either got to physically measure, and log the pitch of every single electrolytic in the entire unit. So that I know for a given value how many of each pitch I need to order. Either that or order 11 of each flavour in order to cover all eventualities.

[bdunham7]

So I've either got to physically measure, and log the pitch of every single electrolytic in the entire unit.

I'll be looking forward to reading more!

As for ordering parts, I'm going through a less-desirable 5101B right now and it is the first time I'm doing a complete overhaul vs repair, so I'm doing the same thing and have the same issue.  What I'm doing is pulling each individual board and making a list/project file for it so that I can order the parts in groups.  Mouser lets you save 'projects' like this so I can just order a parts group by board number.  And yes, you have to measure all the dimensions of each capacitor and it is a huge PITA.  Its like separate groups designed each board and they weren't on speaking terms.  Mine has a plague of tantalum decoupling caps and they have used at least 10 different types throughout the unit. 

[veedub565]

Ah I'm glad it's not just me then. Just after I posted I came to a similar conclusion, that I'm going to have to break it down and make a list of parts for each separate board. Treat each board as a mini project all it's own. Haven't even started on the Tants yet !

I always knew this was going to be a long term project, still could be worse at least this one is complete and in decent shape overall.

5101B looks an interesting bit of kit worth saving

[TiN]

Response is correct, but crappy VISA you used shows ASCII data as hex values. I'd suggest using Raspberry Pi with one of the images (like one recently posted by e61_phil in metrology section) to capture reports over GPIB.

Don't forget to replace carbon composition resistors, especially on 14,15,16,17,18 boards .

I'd also ditch boards 5 and 6 out of calibrators until you get everything else working. A8 error means most likely unhappy tests with some stuck relay somewhere (not necessarily on A8 )

[veedub565]

Thanks TiN, I've got a Raspberry Pi to try this with, although I'm no programmer.

ALL carbon composite resistors are going to be replaced. Actually very little evidence of burning on the A18 board, even with 14yrs operation. I tested the 220k 1W CC resistors and they are all approx 220k except for one which is more like 250k. Maybe a new board was fitted at some point. Anyway they are all going to be replaced and the 1W ones upgraded with 3W thick film 100ppm

Yes tracking down stuck relay is going to take some time. I already removed A5/6 and still get the relay error. I'm going to work through this sequence (attached) and see where I get the error

[veedub565]

Its like separate groups designed each board and they weren't on speaking terms.

I'm beginning to wonder if each board had separate groups working on it who also didn't talk to one another.

A17 board -
C8,C13 = Tantalum 22uF, 25V, 20%, T356G226M025AS
C67-70 = Tantalum 22uF, 25V, 20%, T361B226M025AS

So identical value, voltage, tolerance tantalum caps. Except 2 of them are 356 series and 4 of them are 361 series.... on the same friggin board ? 

Pitch is the same, physical size is the same.... you have to wonder if there is some specific reason why the 356 series was chosen for one part of the circuit. And the 361 series was chosen for another part of the circuit. And if that's the case,  you have to wonder if it's wise to blanket change all the tantalum capacitors, unless you replace like for like with the exact same series and not just the same value.

[bdunham7]

You can try to look at the circuit to see how the tantalum cap is used, but AFAIK they are all non-critical decoupling types of applications.  By non-critical, I don't mean you can use just anything, but I think any similar tantalum should be fine.  They use a lot of the less common values--instead of 33,47,68, you see 39, 51, etc.  I'm just using a close value--and I think up or down is OK, so for a 39, I have no problem with a 33 or a  47--and as high a voltage as I can get, either 35 or 50 volts depending on the size.  I'm still going to need at least 7 different part numbers just for the tantalums.  Then there are the weird value polypropylene film caps...

These are sophisticated instruments, but that doesn't mean every design choice was a stroke of genius.  The power supplies alone should be prove that--the 5100 series PSU boards are prone to going up in flames.  Here's one design and component that I think takes the cake---it is a simple switch that takes a resistor in and out of an oscillator circuit on the AC oscillator board.  They could have used anything, but this is what they chose:

[veedub565]

They do just look like decoupling caps. I found where they are in the circuit, and interestingly C8,C13 have TANT marked on the schematic where C67- 70 don't. Although it says TA for both on the part list. I wonder if some are tantalum polymer and some are straight tantalum.

I attached a picture from the schematic showing both, and an extract from the part list also showing both.

[veedub565]

First parts order placed £164 worth of electrolytic capacitors, metal film and precision wire wound resistors, and a couple of Sunon fans. Not a cheap project ! and still the original faults to fix after this.

Still making my mind up about the tants, if any were going to fail they would have done by now. And they don't dry up and age like electrolytics.

[TiN]

Still making my mind up about the tants, if any were going to fail they would have done by now. And they don't dry up and age like electrolytics.

They probably will be just fine, but I wouldn't be able to sleep at night or go to work at same time knowing that at home there is $20k calibrator running 24/7 and if $3 tantalum cap starts fire, nobody will be there to save it. You going thru all trouble to repair unit, why stop half-way? 

Btw, you asked about extenders for boards.

Here is the Gerber file for it. Provided AS IS.

Layers stackup:

TOP - L1 top external layer
INT1 = L2 1st inner layer
INT2 = L3 2nd inner layer
BOTTOM = L4 bottom external layer

[veedub565]

Yes that is a good point, going to all the trouble to refurbish it why stop half way. I'm sure I read something about replacing a few zener diodes in one of your blogs, I can't find where I read it now though.

Many thanks for the gerber file, I have made an order for these with JLC PCB. I think minimum order was 5 so I have some leftover if anybody else comes across this and needs one. Trying to fault find without the extender card would be very difficult.

[TiN]

Funny enough, most of boards I fixed without use of extenders. I got them only when I was troubleshooting wideband assembly. Perhaps having no case helped back then 

[veedub565]

Yes I guess it must be possible to do without the extender cards, they will make life easier though. I think because the cards are close together, even without the case it makes it difficult to get a probe on the measurement points.

I will try and get some pictures of this project soon.

[veedub565]

So a small update on the 5700a repair.

All boards removed, I've cleaned out the chassis, wow it was full of so much dust and crap! looks much better now it's clean. I took the opportunity to check the A3 motherboard relays, all check out ok.

New Sunon maglev fans fitted,  these run much quieter than the originals.

So far all capacitors and carbon comp resistors replaced on A21,A20,A19,A18. ALL of the carbon comp resistors had gone up in value, most of them had gone outside of tolerance. So well worth replacing them. It was interesting to note that the new parts were both higher in spec than the originals, and in many cases smaller size. Technology marches on!

The GPIB connector works, but has a bent pin so must be replaced. I'm having some difficulty sourcing the correct replacement part though. I've salvaged a couple from old equipment but the pins are too short.

The "I Guard" terminal on the rear is snapped off. I was going to replace this, but when I removed the A21 board I found that the rear terminals, although fitted, are not actually connected up to anything. So a bit strange, but at least I don't need to bother repairing this now.


It was powered up with A3+A4 + A19+A20 + A1+A2 Front panel + A21 +A18, I got the expected hang on error “Guard crossing”. Next task is to check the voltage rails. I'm expecting things to be a bit tighter in tolerance than before.

[veedub565]

Two resistors, both the same value, both the same power rating, big difference in size

[picburner]

I was going to replace this, but when I removed the A21 board I found that the rear terminals, although fitted, are not actually connected up to anything.

Even in my F5700A the rear terminals are connected to nothing, it probably depends on the various options of the instrument.

[veedub565]

I was going to replace this, but when I removed the A21 board I found that the rear terminals, although fitted, are not actually connected up to anything.

Even in my F5700A the rear terminals are connected to nothing, it probably depends on the various options of the instrument.

That's useful to know, so looks like all of them came with the rear terminals, but only wired up as an option.

[veedub565]

I need to sort out some image hosting, keep hitting the maximum upload size here.

[pcwrangler]

Two resistors, both the same value, both the same power rating, big difference in size

Always amazed at the progress of some components. Depending on the project I sometimes take the opportunity to increase power rating for reliability/longevity since space permits. Nice repair/restoration btw. Watching for pics.

[YetAnotherTechie]

Two resistors, both the same value, both the same power rating, big difference in size

Yeah, i've seen those, but just because progress has made 1/8W resistors able to withstand higher temperatures, and be relabeled as 1/4W doesn't make it right. Imagine the case were the old resistor was dissipating 0.15W, maybe its surface temperature would be 50ºC, on the new one, i wouldn't be surprised if the surface temperature was >100ºC

[veedub565]

Two resistors, both the same value, both the same power rating, big difference in size

Yeah, i've seen those, but just because progress has made 1/8W resistors able to withstand higher temperatures, and be relabeled as 1/4W doesn't make it right. Imagine the case were the old resistor was dissipating 0.15W, maybe its surface temperature would be 50ºC, on the new one, i wouldn't be surprised if the surface temperature was >100ºC

Well I did wonder I must admit, it doesn't look right. I ordered replacement parts as per OEM spec though, and that's what turned up. I ordered Metal film in this case to replace Carbon Composite.

It's a bit difficult, if the parts list says 1/4W and you order 1/4W and then it turns up a fraction the size of the original part. It's a bit difficult to try to second guess everything.

[TiN]

veedub565
Upload to my FTP server, I'll give you web-page with all links automatically. I still have some 180GB on storage available, should be enough 

I meant to make a huge article about all 5700-related shenanigans (and photos, I have lot of gigabytes of photos of all different 5700/5720 and even 5730 variants), but easier to say than done.

That's useful to know, so looks like all of them came with the rear terminals, but only wired up as an option.

Not all of them, but most. And yes, you could order rear terminal option which means Fluke would disconnect front terminal cable and route it to rear instead. There is little different guarding as well in that configuration. The procedure is briefly outlined in SM, as user can do that swap too.

Yes, when I replaced resistors I went for bigger size instead of matching BOM to the letter. E.g. in many cases I've replaced 1/4 or 1/2W resistor with 2W or so, given other specs match or exceed. Also I liked to use MF resistors instead of thin/thick film, where possible, except high power resistors.

[veedub565]

Looks like I'm going to have to re-order some of these resistors. At least they aren't too expensive.

[veedub565]

So after measuring the CC resistors from the A18 board there seems to be a general pattern. And I would imagine its the same for all the boards.

All measurements are approx

0.125w =
0.25w = 2.3x6.5mm
0.5w = 3.5x9.5mm
0.75w = (don't think there are any)
1w = 5.7x14.5
2w = 7.9x17.7mm


So I'm just going to have to re-order the resistors, and pick a size close to the originals, that obviously matches or exceeds the original wattage rating.


Just the A18 board is a lot of work, a lot of measuring component sizes, measuring the pitch, picking correct spec etc. Quite a bit to think about really. It looks nice with all new electrolytics fitted though

[MegaVolt]

Also to replace the carbon composite resistors, especially the 1W ones which are to be replaced with 3W ceramic.

Can you tell me which resistors you consider to be carbon? The ones in the picture are of different sizes, I think they are wire-wound.
And judging from other Fluke instruments, some of them are used in precision circuits and are specially selected.

[Kleinstein]

The larger resistor in the size comparison is a typical small carbon comp resistor. The smaller one is a film based resistor, likely metal film.

The carbon composite resistors usually have the brown plastic body with smooth surface (no wire visible). The come in different sizes but usually very similar optics.
The carbon resistors usually are with 5,10 or 20% tolerance - hardly ever better than 5%. Precision wire would are often better than 5% (except for low values).  The component list should state the type of resistor they are. They would not replace a corbon with a precision wire wound - a power wire wound or metal film maybe. The carbon resistors were never meant to be stable. It was know from the start they are poor stability.

[picburner]

Board A18 for example: all resistors circled in red are carbon composite resistors.

[MegaVolt]

Board A18 for example: all resistors circled in red are carbon composite resistors.

Why do you think so?
According to the documents these are IRC BWH resistors. Wirewound.
https://www.tedss.com/LearnMore/Failsafe-Moulded-Wirewound-Resistors

They are also used in the Fluke 5440 in the voltage divider circuits (after the special selection). That is, they can provide accuracy and matching.

Where did you get the idea that these resistors are carbon?

[picburner]

Why do you think so? According to the documents these are IRC BWH resistors. Wirewound.

Simple: the value of resistance.
The resistors in the board go up to 560Kohm while those of your link reach 2.4Kohm max

[MegaVolt]

The resistors in the board go up to 560Kohm while those of your link reach 2.4Kohm max

Do you have a datasheet for these resistors?

So far I see in the Fluke 5440:

The resistors are in the output divider and provide 5ppm instability. These can't be carbon resistors. Fluke are not stupid.

Their resistance is 99kOhm and 18kOhm

[MegaVolt]

The ratings on the schematic are different from what is on the board. This may be the difference between the A and B version.
The ratings on the board are really less than 2.4kOhm.
I'm confused... how do I distinguish one resistor from another?

[picburner]

I dissected a resistor of this type that I had in the junk drawer.

1: insulating body (bakelite or ceramic)
2: resistive material (mixture of coal, talc and other binding compounds).

[MegaVolt]

I dissected a resistor of this type that I had in the junk drawer.

What is the brand of these resistors? Perhaps a datasheet?

[Kleinstein]

It us rather difficutl to so if they are carbon composite or the linked high power wire wound ones with overmold.
The resistor value gives a hint as the wirewound are low resistance only. The molded carbon compasite type was quite common.

Anyway the wire wound version linked is not a precision type, but one to work at relatively high temperature and thus high power for the given size. It may be a good choice for those resistors that get quite hot. There may be a few in those case actually be such wire wounds, but is hard to tell from the look.

For the precision it would not make that much difference (except that the cabon ones usually have a negative TC (some -200 ppm/K) while the wire would are more likely to have a postive TC like + 100 ppm/K.  So in theory one could test (usually out of circuit): postive TC is likely carbon and negative TC is not carbon.

[MegaVolt]

Anyway the wire wound version linked is not a precision type, but one to work at relatively high temperature and thus high power for the given size.

The Fluke 5440 uses them for an accurate output divider. They promise 5ppm.

[MegaVolt]

What is the brand of these resistors? Perhaps a datasheet?

The carbon resistors are from Allen-Bradley. In order to distinguish them you need to read the BOM sheet carefully.

[MegaVolt]

Datasheet

[veedub565]

Board A18 for example: all resistors circled in red are carbon composite resistors.

Why do you think so?
According to the documents these are IRC BWH resistors. Wirewound.
https://www.tedss.com/LearnMore/Failsafe-Moulded-Wirewound-Resistors

They are also used in the Fluke 5440 in the voltage divider circuits (after the special selection). That is, they can provide accuracy and matching.

Where did you get the idea that these resistors are carbon?

because it says so in the service manual parts list

[MegaVolt]

because it says so in the service manual parts list

Yes. I already figured it out. Externally, the same resistors can be both good wire resistors. They can also be bad carbon ones. And you can tell by their rating if they are greater than 2.4 kOhms.

[TiN]

These can't be carbon resistors. Fluke are not stupid.

Fluke came to senses and in newer A18 boards they replace all CC resistors to better types already. This is also confirmed by comparing old 5700A BOM and new 5720A BOM.

Here's how newer (year 2001) board look like:



Here how old one looked like during repair, when I cut all carbonized FR4 out and before install and air-wire all broken connections on 4 layers.



 

[MegaVolt]

Everything I said about the 5440 and the divider in it was wrong. I was looking at the wrong board. The divider is made correctly with anobtanium resistors.

[veedub565]

The new extender boards arrived, I had them made up by JLCPCB and they look pretty good (I went for blue)




It's a shame they don't actually fit the connectors 

[veedub565]

Images won't embed for some stupid reason. Here you go

https://flic.kr/p/2mn7aQo

https://flic.kr/p/2mmYusc

https://flic.kr/p/2mn3n5s

[TiN]

Oopsie, sorry about that 

Forgot about that, I used Panduit DINs and they are not same as usual ones from Digikey.

You can file the excessive PCB allow connector fit.

[veedub565]

Doh! I used Panduit ones too (100-964-053) same as Fluke used.

Yes I wondered if I could file the PCB down a little to make them fit. I didn't want to risk damage any internal tracks which may be there. Looks like it should be ok though

[TiN]

You can cut freely, nothing to worry about. Edge plane, including inner layers not even connected to anything. 

[veedub565]

A18 is finally completed, woohoo. All electrolytic capacitors replaced with new, higher spec ones. All carbon comp resistors replaced with properly sized metal film replacements (thick film for the 220k 3W ones). Only slight irritation is some of the smaller caps I ordered with 5mm pitch, actually arrived with a 2.5mm pitch and the legs bent out. So they won't sit flat to the PCB 

[/url]20210909_152224 by Ben Tocknell, on Flickr[/img]

On the solder side you can see some small burning of the PCB, I got to this one in time before more damage was done.

[/url]20210909_152247 by Ben Tocknell, on Flickr[/img]


Still can't get the images to embed, hopefully you can see them via the link.  I'm using the bbcode and "insert image" button from Flickr, it should work.

[picburner]

You have forgotten four carbon composite resistors, the ones highlighted in blue.

[veedub565]

No, I haven't forgotten them. Those are Carbon Film resistors not Carbon Composite, according to the parts list in the service manual anyway.

[veedub565]

Well it's one step forwards and two steps back with this at the moment. The replacement GPIB connector I ordered arrived, I couldn't source the OEM part so I used an alternative. This came with the wire latch which had to be removed. So that is all done and fitted now.

The I/O board is a bit of a PITA to remove, so before I refitted it I took the opportunity to test the relays. This unit has a relay fault somewhere!. Now the service manual says...

"Check individual relays by connecting an external power supply (a typical 9V battery
works well) across the set coil to place the relay in the set position or the reset coil to
place the relay in the reset position."

I used this method to test the relays on the motherboard, and it worked ok. So I applied the same method to testing the relays on the I/O board (can you see where this is going yet?)

K8 checked out ok, so I moved on to test K7...bang!!. A visual inspection shows it has blown the sh1t out of the relay driver U16, literally blown a hole in the middle of it. Looking at the schematic I'm not sure how this is possible?

So I need to source and replace U16, maybe U14 too as a precaution. Yet more money on parts! I'm reluctant to even attempt this again on other relays, it seems bad advice really. Repairing the I/O board is not too bad compared to some other boards in this unit.



By the way, attaching a 9v battery to test the relay on the A18 board did not work, it didn't activate the relay. Some investigations shows diode CR244 to  be s/c both ways, ho hum. So the question, is this a result of connecting a 9v battery across the relay coil with the wrong polarity... or was it like this to begin with. One of the original faults was error 3107 which relates to the DC HV amplifier. And this relay is related to that part of the circuit. I've either stumbled across one of the original faults here.... or blown something else up by attaching a battery like FLuke said to do.

[TiN]

Well in battle between common sense and manual, common sense wins  . Perhaps back in 80ies 9V batteries were not as advanced and powerful? I used SMU with current limit 100mA and 5V voltage (no idea why manual recommends 9V for 5V relay test). But most of the times I wouldn't bother testing relays and just swap the suspect ones (based on failing function/mode and circuit location). Obviously it takes longer time to figure out topology and how everything is related.

This thread reminds me of Hulk-3 project, which still have dead A18 before anything else can happen with it. 

[veedub565]

Yes I placed too much faith in the service manual I think here. I agree it is strange why they recommend 9v battery for 5v relay test. Your method sounds better, using SMU 5v with 100mA current limit. Less chance to do any damage that way. Using 9v battery seems quite a nice way to blow things up. Unfortunately I'm only learning this now 

I think I will forget about testing relays for now. It may not even be a relay that is faulty, could be a faulty relay driver or some other related problem. Like you said learning topology may be best, see what function doesn't work and track it back.

Lot's of bedtime reading to learn topology, I might have to make some large scale printouts.

[picburner]

In my 5700a I had an error 3409 (Phase lock loop fault) and I also checked all the relays of the A13 board without taking them apart.
I used a 5V smu with current limitation of course and I also paid attention to its polarity to avoid any damage to the connected relays driver and other circuit on the board.
The relays all worked well but their driver, a ucn5801a, showed no sign of life: none of its 8 sections worked.
For further confirmation I disassembled it and did a vector test with my eprom programmer: device completely dead.
Once the ucn5801a was replaced with a new one, the error disappeared.
We must be very careful when making tests of this type: it is absolutely necessary to avoid creating more failures than those already present, otherwise the repair bill goes up....
P.S. common sense is often better than what is written in the service manual (also for the resistors I showed you before: they are all carbon compounds, trust me!)

[veedub565]

I neglected to pay attention to polarity here which is probably what has blown the driver on A21 board. At least this mistake is repairable, but yes it puts up the repair bill.

The service manual lists those resistors as carbon film,  I looked up the part number listed (EB4741) though and yes it appears to be composite. Maybe Fluke made a typo, wouldn't be the first time that's happened.


This is certainly a learning experience !

[TiN]

Yep, I swapped those vertical resistors too. Round bobbin resistor = shoot on sight 

[picburner]

In my A18 board I have not changed any capacitors or resistors.
All are perfectly within their respective tolerances, I measured all one by one (even if on board) with my ESI Videobrige 2160 rcl bridge.
Yet the construction date of the instrument dates back to 1990 based on what is written on the integrated circuits.
Probably the instrument had a few working hours but I will never be able to know for sure because the CR2477 battery was completely empty then the ETIME? command returned an unreal value: 134214657.
Probably, in few years, I will have to replace all this components me too if I don't sell the instrument meantime....

[TiN]

picburner is brave man . I have my MFCs 24/7 unattended, so peace of sleep is paramount here 

ETIME? may be reset once PROM is formatted/replaced/corrupted. So it's not a protected counter to rely on.

[veedub565]

Another update

A18 board is now completed (again) carbon comp resistors I missed last time have now been replaced with suitable metal film. A7, A8, A10, A15, A16, A17, A19, A20, A21 board are also all completed with new caps and resistors. WOW there is some dust and dirt on these boards! So all have been cleaned. 

Broken GPIB connector on A21 board was replaced. I couldn't find the exact Amphenol part, so an alternative was used. This one had the bailey lock, but it was possible to drill out the rivets holding it and then left with normal connector. Blown relay driver was also replaced.

Backup battery on CPU board measures 3v, date/time are correct, but it may be worthwhile replacing anyway as I don't know when it was last replaced.



Unit was reassembled with following boards - A21, A20, A19, A18, A17 (and of course motherboard)unit powers up and works as expected with only limited boards attached. I made some voltage measurements for A19, A18, A17 boards (results attached) As you can see most of these are within tolerance, but a couple are out. In particular +5LHR and -5LHR warrant further investigation. Ripple on both rails is is 2.66v p-p (50hz)

I can see why it is a good idea to test these supply rails first before connecting any other boards. A bad supply rail could damage the other boards. This is why swapping cards willy nilly from a working unit into a faulty unit is a nice way to blow up working boards.

[TiN]

I didn't bother replacing coin battery on A20, it's not powering anything critical.

I like to use scope to poke onto voltage rails, to see for ripple/noise/etc. Also I remember that some labels are misleading a bit, with test point on non-rectified secondary voltages (where you will see "ripple") versus regulated DC rails.

[veedub565]

Yes the labels are quite misleading, for example +5LHR and -5LHR are actually meant to be 12v not 5v like you would think. In this case they are measuring more like 18v which is quite long way from what it should be. 2.6v ripple is within spec for the positive rail (3v) and slightly out for the negative rail (2v). Seems odd the negative rail would have a tighter spec for this than the positive rail.

Looks like +5LHR and -5LHR are derived from "5 LH AC1" and "5 LH AC2". So effectively +5LHR/-5LHR come straight from the transformer, via a little rectification and smoothing. So there doesn't appear to be any possibility to adjust the voltage, it is what it is. Strange then it's out of spec.

I use a scope to check for ripple too, haven't checked all of them yet.


One other thing I noticed, is that on my A16 board C32, C34, C31, C33 electrolytic capacitors are all Orange drop capacitors instead. It looks factory fit, so I've left them.

[TiN]

Be careful not to blow up scope, as many rails have isolated grounds so you can easily catch some hundreds volts pk-pk between test points.

[veedub565]

Yes I'm wary of things like this with floating earths. Sometimes best not to connect the scope ground at all, or even disconnect earth from the scope mains plug. Either that or I'll have to invest in a differential probe.

[alm]

Please don't disconnect the scope from ground and then go poking in an instrument that contains lethal voltages. Does the 5700A service manual contain resuscitation instructions like manuals of their previous calibrators?

If you don't have a real differential probe, the safe thing to do is use two probes on two channels in subtract mode and leave off their ground leads. Common mode rejection is not great, though.

[veedub565]

Please don't disconnect the scope from ground and then go poking in an instrument that contains lethal voltages. Does the 5700A service manual contain resuscitation instructions like manuals of their previous calibrators?

If you don't have a real differential probe, the safe thing to do is use two probes on two channels in subtract mode and leave off their ground leads. Common mode rejection is not great, though.

No it's not the best plan, although we have done it on occasion at work for certain instruments (can't remember the reasons why). Biggest issue is that the BNC's on the scope, or metal parts of the probe itself. could float at hundreds of volts potential relative to an actual earth. Fluke do still put the resus instructions in the manual.

Like you said, CMRR isn't great with two probes. And a differential probe may end up costing more than the scope (TDS3054)

[TiN]

HV Probe like EEVBlog HVP70 is perfect match for jobs like troubleshooting +/-44V or higher 5700A supply rails

[veedub565]

I got some information out of this unit finally (attached). Formatting is not the best, but readable. Unit was left to warm up for 24hrs, ambient conditions not tightly controlled but pretty constant 21deg and 58%.

Many results are ok, some have drifted a long way.

[veedub565]

Anyway, update on progress

All boards are now completed with new electrolytic capacitors, carbon comps replaced with suitable metal film.

DC voltages check out ok except:-
                           +5LHR and -5LHR unregulated supply is slightly high. Since this is derived straight from the xformer I'm not sure anything can be done to correct this.
                            +44S / -44S and +17S / -17S regulated rails slightly high

I would be interested to see what others measure on these voltage rails, and be able to compare to working unit. (I did have access to a working unit, but not anymore unfortunately. )

Awaiting a new differential probe so as to safely check AC ripple p-p

Self Diagnostic routine throws up the same errors as before, I tested this with limited cards fitted as follows:-
A20+A19+A18+A17+A16+A7+A8+A11+A12+A13 = A8 Relay Fault (four times)
A20+A19+A18+A17+A16+A7+A8+A11+A12+A13+A14+A9+A10 = no further errors
A20+A19+A18+A17+A16+A7+A8+A11+A12+A13+A14+A9+A10+A15 = HV Offset Fault (3107)
A20+A19+A18+A17+A16+A7+A8+A11+A12+A13+A14+A9+A10+A15+A5+A6 = 40db Attenuator fault (3918)

I did a few tests using a Datron 1281 (last calibrated by Fluke July 2020) *edit* I forgot to zero the DMM, updated measurement values below

5700A                 1281
10.000000VDC = 9.9986185VDC
9.999150ohm =   10.380749ohm (2w comp off) 10.023993 (2w comp on)



So plenty of work still to do to get this calibrator back working properly.

[ManateeMafia]

Here are some measurements that I made on a 5700A. The repair is incomplete, hence the missing values.
The ripple measurements are comparative and made with my multimeter. I just wanted to make sure the replacement parts didn't make things worse.

I have found that making these spreadsheets are easier than writing them down.

Good luck on your progress. You will find these are high maintenance devices. Relays will be your biggest problem, especially the ones behind the front panel. They show up when you run artifact calibration and the 1 ohm fails.

[veedub565]

Thanks, I appreciate you sharing that for me. Always helpful to have something to compare to in these cases.

I just re-measured the voltage rails on my A18 board and they are now well within spec. +/-5LHR now measures 14.6VDC and 0.2VAC (RMS) both using handheld Fluke DMM. 

A few possibilities
Maybe it needed some time to burn in after having all the new parts fitted.
When I made the original measurements only A20,A19,A18,A17 cards were fitted. This time it is fully equipped. Maybe the measurements are meant to be taken under load with the other cards fitted.
Or maybe I just got the wrong pins originally 

Well at least it seems to be in spec now


And yes, I can see this will be high maintenance. Nice to have though if I can fix it up. Tracking down the dead relay (or relay driver) is going to be tough. Like TiN said, learning topology is key here to narrow down the search. Replacing the relays behind the front panel might be worthwhile anyway if I can source replacements, I'm sure I saw a video where they were replaced.  A lot of these old relays are obsolete and hard to find now.

[TiN]

Maybe it needed some time to burn in after having all the new parts fitted.

No, 15 minute "burn in" is enough, SD and calcheck should not give out any errors without burningins..

When I made the original measurements only A20,A19,A18,A17 cards were fitted. This time it is fully equipped. Maybe the measurements are meant to be taken under load with the other cards fitted.

You went waaay too fast again.  Stop, remove all unnecessary cards (HV and AC) and test DC boards first and calibrator DC output.
Configuration should be
A20+A19+A18+A17+A7+A8+A9+A10+A11, nothing else.

Because :

5700A                 1281
10.000000VDC = 9.9986185VDC

Means its unhappy and broken. With original EEPROM (given you didn't format/erase it on A20) 5700 should never be that much off 10V.
Also forget about 2wire in resistance measurement for now, use only 4w mode. Resistors are fixed values here (unlike 55XX) from fixed resistors, so should be close to calibrator value as well, unless some relays in path are bad.

Well at least it seems to be in spec now

Wait, what? What is in spec? You still have a path ahead in repairs...

I got some information out of this unit finally (attached)

Cal check report is useless without actually running calcheck. You can see all shifts reported as 0.000 ppm. Meaning actual Calcheck procedure was never ran or aborted with failure. It takes bit under an hour to complete whole calcheck. No external connections needed.

[veedub565]

No, 15 minute "burn in" is enough, SD and calcheck should not give out any errors without burningins..

I'm talking about DC voltages on the A18 board test points here not SD or calcheck. +/-5LHR rails were measured around 18v after capacitor and resistor change, and now they are 14.6v. I was speculating that maybe the new components had burned in after I originally made the measurements. Either that or I got the wrong test points.

Wait, what? What is in spec? You still have a path ahead in repairs...

The spec for the DC voltage rails on A18 board. +/-5LHR rails are now 14.6v which is inside the spec or 12v +/-4v

Cal check report is useless without actually running calcheck. You can see all shifts reported as 0.000 ppm. Meaning actual Calcheck procedure was never ran or aborted with failure. It takes bit under an hour to complete whole calcheck. No external connections needed.

I noticed that after I posted, I did run calcheck, maybe I didn't leave it long enough to complete fully, or it aborted with failure.

You went waaay too fast again. 

I still have much to learn, I will go back to A20+A19+A18+A17+A7+A8+A9+A10+A11 configuration and take it from there.

[veedub565]

I don't think I left it long enough last time before interrupting CAL_CHK. Here is the completed CAL_RPT_CHECK. Looks like DCV section works, but long way out of specification as expected.

[TiN]

How I would start checking relay paths for DCV output from A11. Also good idea to look at DC output on a scope, if there is any oscillation or noise there. Main boards to test are A8, A11 and mainboard. Without DCV working properly all other ranges and functions will be out of whack.

[veedub565]

So there is a fault here then and not just out of calibration?

Reading the manual, almost everything is derived from the DAC in some way. Makes sense if that is out of whack then everything else will be.

[Kleinstein]

It looks like there is some change in the offset to the DAC. Depending on what was changes in the repairs so far, this may be normal and just need a new calibration. However the change is quite large (like 0.5 mV) and thus more than I would expect even from a rather drastic change of parts (e.g. new OPs).
It could be still a faulty old calibration / zero adjustment.

I may be worth to check the DAC a bit more in detail. e.g. compart the DAC output for some -10 V to 10 V in 0.5 V steps to a good DMM. Another point would be to check the ouput stability at a low voltage (e.g. some 10 mV, but still in the 10 / 20 V range) for some 1 hour or so. Also check for ripply / osciallation on the ouput (the more critical setting here would be some 5-7 V).

[veedub565]

No major change in parts. Only capacitors and carbon composite resistors. And I don't think the DAC card has many of those.

I've got a Datron 1281 which is pretty accurate for testing output voltages.

I will try as suggested, and also have a run through the A11 troubleshooting guide to see if that flags anything up.

 The A11 card generates 22v and 11v, then A8 card divides it for 220mV and 2.2mV I think. The fact all of these ranges are a long way out would suggest  A11 is best starting point.

[TiN]

The A11 card generates 22v and 11v, then A8 card divides it for 220mV and 2.2mV I think. The fact all of these ranges are a long way out would suggest  A11 is best starting point.

Yes, but bad relay on A8 could also load/offset A11 output as well. So main focus would be A11 and A8.

[veedub565]

Well it's been a little while, work and other project have got in the way of progress on this. Anyway I've worked through the troubleshooting for the A11 DAC (attached)

I was able to work through all the troubleshooting steps with no issues, until the last one (step 16) Set the output to 10V, measure the voltage at TP8, then at the output high binding post. And the two results should be within 10ppm of each other

Problem is the voltage at TP8 is constantly drifting down. In the time it take me to switch the probes from TP8 to the binding post the output has drifted by 100ppm anyway. For example:

After 2hrs warmup = 9.9980882v
After 24hrs = 9.9973883v

I did make the measurement anyway after 24hrs
TP8 = 9.9973883v
Binding post = 9.9974139v

So this doesn't look correct at all. Not only is it not within 10ppm. The output voltage should be stable not drifitng. 

I went back and rechecked step 2 but with higher resolution this time. I measured 13.0805634 and the last 3 digits were also drifitng around a little ending up at 13.0805714. I mean it's still within the specifications of the test, but still.

[Kleinstein]

For the 13.x V  raw refrence is can be difficult to measure. Meters that use a divider at the input may not be stable enough for this.  If the meter needs to use the 100 V the meters drift may be main part.
It may still be good enough to compare the 13.x V and the nominally 10 V output. This would be mainly measuring the DAC gain.

[veedub565]

I'm using a Datron 1281 on 8.5 digit resolution for the long range measurements. If it's only a couple of decimal places required then I use a Fluke handheld DMM

I did notice the HR5 can was knocked loose allowing air in and heat to escape. I reseated that and it seemed to improve the 13v ref measurement. Didn't make any difference to the output slowly drifting down though.

[TiN]

You have the plot of the drift or CSV? Did you check that all heaters on DAC board hybrids are good and not open (should read ~42 ohm or so between heater pins).

[veedub565]

Heaters measure ok (HR5 and HR6). I don't have a plot of the drift yet, if there's some code for doing this with the Datron 1281 that would be great, if not I'll have to learn how to do it. Or manually take some readings.

I've also made a start checking the relay paths. K1,2,3,4,8 all seem ok on the A11 DAC and function as per the table in the schematic. One thing I have noticed though, when pressing "operate" I hear the relays make a positive click, but I also hear a strange fluttering sound, almost sounds like a moth fluttering kind of sound. It's only brief, and I'm not sure if it's normal or not. It could be the sound of a failing relay.

I can hear it definitely isn't coming from the A11 DAC board. I removed the A10, A9, and A7 boards and still hear the sound. So it must be coming from the A8 board or possibly motherboard.

[veedub565]

Another update on this, long term project and I believe I've found a faulty relay.

So with A20+A19+A18+A17+A7+A8+A9+A10+A11 cards installed, running the self diagnostics revealed no faults (besides the usual complaints about missing cards) However I still seemed to have a problem with my DC voltages tens of thousands of % out of spec and drifiting. Trying to nail the culprit here is tricky. Even trying to identify the faulty card was hard. So I added in the A12 card so that I could learn a bit more and see if I could identify any functions not working at all that may help.

Well as soon as I added the A12 card then I got 4 A8 relay faults flag up. SO what has changed here. Well the A12 card adds some ACV functions, 2.2v and 22v. A8 then divides these down to make 2.2mV, 22mV, and 220mV ranges.

Going through the  troubleshooting revealed no problems until the last step
"Check the mV Dividers. Set the 5700A to 200 mV at 1 kHz, operate. Using a DMM
measure the ac voltage at the cathode of CR6"
Setting the calibrator to 200mV 1khz caused it to trip out to standby mode. In fact, setting any of the AC mV ranges caused it to trip out. 2.2v and 22v worked ok. So there is a problem here!

I investigated the divider circuits, and used the tables in the A8 schematic to trace through the relay paths. And I found the A8 relay K15 is stuck in the "SET" position. this should be in the "RESET" position for all the AC mV ranges. It could still be a faulty driver IC, so I removed the A8 card and manually applied 5v to the K15 relay reset coil... nothing, nada, no click. So it appears I have a faulty relay, probably the contacts have stuck in the "SET" position.

This is bad because K15 and K33 both connect to the out/sense low line, but not at the same time. With this fault it allows both relays to connect to out/sense low at the same time.  Maybe this is contributing to the DCV volt problems, maybe not, but it certainly looks like a faulty relay here and it needs replacing.


Two questions
1. Whats the best way to remove the relay without damaging the pcb ? I don't think solder wick is going to work here. A hot air gun may work... or it could just lift the pads. I'm aware this is through plated and theres a risk of pulling pads from the opposite side of the board.

2. where is a good source of replacement relays ? they are obsolete and out of stock now from most mainstream suppliers. There are quite a few sellers on ebay, but there's always a risk of buying fake or factory reject parts.
 

[TiN]

1. I used solder wick, worked well with good care. Best would be desoldering iron with pump. Worst is hot air gun, as 99.99% of them would ruin the board.

2. No good source. DS4E's are ebay-sourced, I did buy number of them and after a sniff test used most promising ones. DS2E's are available from Digikey DS2E's also obsolete  ..

[picburner]

For DS2E series relays you can find something here.

[veedub565]

I had a quick go with wick, but difficult to get all the solder out especially from component side. I've got a Hakko FR803, but I agree the hot air gun risks ruining the PCB. The foil on this PCB already looks crinkled and fragile in places. I guess this is a good excuse to invest in a vacuum desoldering station 

Thanks for the link to relays, might buy a couple of those, works out £15.70 each with shipping though. I've not found the DS4 ones anywhere except ebay at £3 + free shipping to £17 + £19 shipping each 

[veedub565]

This could be an interesting source of replacement relays https://cephosrelays.com/

**UPDATE**
These are £17.83 each, minimum order qty is 10, and £15 shipping (to UK)

So £193 for 10 replacement DS4 relays 

[veedub565]

Well bought a vacuum desoldering gun which made short work of removing the K15 relay. I was able to test the relay out of circuit and found that it sets and resets just fine. So I refitted that, and tried replacing the driver IC just in case there was a problem there, but still no joy.

I've made a spreadsheet and started logging all the relay test results from A8 Switch Matrix (attached) and made a couple of interesting discoverys.

ACV - K15 is always SET when it should be RESET. I know the relay physically works because I tested it, and the relay driver was replaced.

DCV - K23 is always RESET when it should be SET. In ACV it is in the correct SET position. So this proves the relay is physically able to switch and that the relay driver is working.


This is very strange because there's nothing wrong with these relays, they aren't sticking, or the driver IC's. For some reason the 5700 is intentionally setting the relays to these "wrong" positions. Either the relay matrix diagram is wrong, or the 5700 logic has gone wrong in it's old age. there's no reason at all K23 should be in the RESET position for DCV except for 1100V mode with external amplifier. 

[Kleinstein]

Control signals often use some shift registers like HC595 to ouput the large number of bits. These are somewhat error prone from ringing on the clock signal and in case of the xx595 also from delays between chips. So exchanging one of the shift registers could also effect the signal integrety. At least the clock is worth checking.

[veedub565]

The 5700 uses a software controlled 82C55 with 4028 decoder. This strobes the data into several 5801's, which then set/reset the relays. A8 K15 and K23 are consistently set to the wrong state, I'd have thought a dodgy clock would be a bit more random. It's definitely worth checking though.

For a 5801 with 4 relays, each with a set & reset coil I imagine it would be strobing in 10101010 to set them all to "set" state. To get 3"set" and 1 "reset" it would have to look something like 10011010. K15 and K23 are set by 2 separate 5801's (one of which I already replaced) so I'm discounting relay or 5801 fault..... it's not a random error, more like a deliberate act,  They've gotta be receiving incorrect data from the 82C55, which in turn gets it from software. Or the manual is wrong.   

[Kleinstein]

A wrong manual is a very real possiblity. One may be able to check at least 1 connection to see if the direction is correct.

A broken chip that gives out the wrong possiblity is rare - though I have heard of such an example (some old Comodore ISA GPIB cards that could be fixed with an external inverter when they were broken in the usual way).

[picburner]

I have had a completely dead ucn5801 driver in the F5700 but, as Kleinstein rightly said, it is unlikely that a faulty IC will cause the relays to operate in the exact opposite way.
On the other hand, I found an interrupted track in an internal layer of the A13 pcb that drove me crazy before being discovered!
If you find it useful and your eprom programmer allows it, I made an adapter to test the 5801's by vector test.
Since I bought the spare parts in China, through this adapter I checked the devices before inserting them in the pcb!

[veedub565]

That must have been a nightmare fault finding, and it's an interrupted track somewhere.

At this point I'm not even sure which card has the fault on it. Only A8, A11, A17, A18, A19, A20, A21 fitted at the moment (and of course motherboard). Main suspects are A8 and A20 board. I can't see anything else affecting the relay switching.

Slim possibility it may be A21 causing a problem. This is the card which I blew the UCN5801 on using a 9v battery to test the relays (before I knew better). I did replace the IC, but maybe there is some other problem still. I say this because K23 relates to the B SENS LO line which is only used with the external amplifier. Maybe the 5700 thinks it has an amplifier or or something attached, and incorrectly setting the relays. That theory doesn't really fit though because even if 5725 amplifier was fitted K23 is still only reset for 1100v range. It may be worth double checking that card though.

Also way back when before I knew better. I did try swapping over the A8 card from a working 5700 (sadly no longer available to me) and still got the same A8 relay errors. So May not even be a problem with the A8 card.

[veedub565]

Well it's been a while since the last update. I've done some pretty exhaustive testing on the A8 card. And coupled with the fact I did previously test this card in a working unit with no errors. All leads me to believe the A8 card is fine, nothing wrong with it.

K15 and K23 in the A8 card are indeed set wrong, but there's nothing wrong with the relays or drivers. And the card works correctly in another unit anyway.

So it looks likely the problem is with the A11 DAC card. I followed the troubleshooting guide for A11 and there were no problems. I checked out all the A11 relays switch correctly, and they do. So it's a bit of a mystery now as to what the problem is with this A11 DAC. Quite how or why the A11 DAC card causes 4x A8 relay errors is a mystery.

There can't  be that much wrong with it else it wouldn't pass all the troubleshooting steps. Gotta be something simple causing  all these problems. 

[TiN]

I'd check motherboard connections ;-) I had dodgy A3 (motherboard backplane) with open connection between cards, had similar shenanigans too, "good boards" were no good with this A3

[veedub565]

Thanks Tin, I thought I'd fixed it by replacing A11 U31 but I was a bit premature in my celebration.

I don't think A3 is the problem, I was able to try my A11 card in a working 5700a and got the same fault. I know this isn't good practise, but it 100% proves it is the A11 DAC card that is the problem here somehow 

Troubleshooting procedure reveals no fails, so core functions seem to be ok. Something very strange and probably very simple has gone wrong. It might be worth me checking the motherboard connections on the A11 DAC card though. Maybe same problem you had on A3 has happened to my A11.

[veedub565]

A wrong manual is a very real possiblity.

I think this may be the case, or at least open to interpretation.. The truth table that shows relay positions for the A8 switch matrix shows K15 as Reset for (AC) 2.2mV, 22mV, and 220mV.

According to the theory of operation though, all K15 does is switch between external and internal sense. I checked this and confirmed K15 is in the set position at all times. Unless external sense is selected on the front panel, and then K15 goes to Reset.

There's no reason for K15 to be reset just because (AC) 2.2mV,22mV, 220mV ranges have been selected. So I may have been chasing a bit of a red herring here trying to work out why A11 was causing A8 not to set K15 correctly. This makes more sense as I just can't see how that could ever really happen anyway.

If anybody is able to confirm on their 5700a that A8 K15 should be set all the time (unless ext sense selected) then that would be great. Only on special rare occasion do I get the opportunity to test with another working 5700a


It still makes not much sense really because I still have the problem A11 causes 4x A8 Relay errors (3x A8 relay error and 1x A8 MBD relay error). A8 is 100% tested and known good. A11 tested in a working 5700 produces the same errors, so also 100% A11 is the problem.

It would be nice to know what the 5700a is doing when it steps through the self diag. What is it attempting to do when it flags these errors. And what is it expecting to see. We know A8 is working correctly but possibly DAC is not seeing what it expects and self diag thinks A8 is the problem. In reality whatever it expects to see is not there or incorrect for some other reason (not A8 Relay problem)... Argh 5700a shennanigans !

[veedub565]

Another update for this long running repair job.

I haven't had much luck disassembling the firmware to see what's happening while running self diag. Instead I ran the self diag and when it flagged up the A8 Relay error I checked the position of all the A8 & A11 relays. Every relay on A8 is in the "set" position except for K32 which is "reset". What this means is that "PA COM" is switched to the "RCL" line

On A11 K5 and and K6 are "set" this means the "RCL" line is switched to the + input of the ADC amp, and "REFCAL" is switched to the - input.

I checked the configuration of A11 U32 and it's configured so that "ADC COM SEL" is enabled.

So at the point where I get the first A8 error, it seems that the ADC is looking at "PA COM" on the "RCL" line and comparing it to "ADC COM" on the "REFCAL" line. And for some reason it doesn't like what it see's and thinks there is an A8 relay switching problem. We know the A8 Switch Matrix is perfectly fine as it works in another unit, so the problem is with A11.

The voltage across the inputs to the ADC amp is 0.003v and I checked that the ADC amp has the correct gain of x11 and that this voltage gets to the input to the ADC ok. It may be coincidence, but when set to 10VDC output I get 9.997VDC... 0.003v out! I'm not quite sure what this means (if anything)


I know the ADC must be working because, well, lots of other things would fail if it wasn't. One thing I did notice is that the ADC reference should be 6.4v +/-0.2v and I measured it at 6.20v it's just borderline in/out of specification. I wonder, if the reference is slightly out of spec, the ADC isn't going to work properly, isn't going to self calibrate or zero properly. Possibly (and I haven't checked this) the ADC reference is 0.003v outside the max spec (6.203v).

I don't know though, I mean a spec of +/-0.2v is pretty large, presumably the A11 DAC is calibrated with the ADC reference at whatever voltage it's at within that range. Changing the zener diodes that control the ADC Reference voltage could throw the calibration totally out and still not solve the problem.


**Edit**
One more thing, resistors R88,89,90 on A11 DAC could have an effect on the ADC voltage reference. These resistors were changed from the original carbon comp to metal film. I checked they were all exactly the right value, but the calibration would have been performed with the old carbon comp resistors in place at whatever value they had drifted to. Replacing these resistors may have altered the reference voltage and thrown the cal out resulting in the output voltage error, totally unrelated to the A8 relay error. I do still have the original resistors so I could try putting them back in and see if it makes a difference. May have been best to leave these critical value resistors alone.

[Kleinstein]

The absolute value of the reference votlage should not effect the ADC function very much and the calibration should take care of this.
The calibration should be using the main DAC reference to calibrate the ADC's full scale.
Especially a change in the reference should not effect the zero point / offset.

The resistors R88,R89,R90 at least in the plan (PDF file) I have are low tolerance precision resistors that do effect the ADC. They should not be carbon composite, but high quality types. There are some versions that look like carbon composite but are quite a bit better.
Wrong resistors could really upset the ADC offset. Chances are there is an calibration compensation step for this, but the range my be limited and wrong resistors may have more effect on the ADC. 

[veedub565]

Thanks that's interesting to learn. I verified 13v reference at 13.02v nice and stable. Also 6.5v reference is good too.

It's possible then that replacing these resistors is what has caused the big drift in output voltage. I'll try putting the original ones back in and see what happens. Originally when I first got this, it did output 10.0000v (but still with A8 relay error) after resistor and cap change then output is 9.997v. I assumed a bad zero cal due to A8 relay error, but there is a chance it is the change in these resistor values, especially if they can really upset the ADC offset. I did replace with modern low tolerance metal film, but still values could be slightly different.

[veedub565]

Argh it seems I have kept all the old resistors except those 3 

The new ones I put in all measure dead on 20.0k and one 50k


Honestly I even contacted Fluke. even though it's not officially supported they offered to look at it £750 investigation charge. And then diagnostic/repair time on top of that assuming they can even fix it and have the parts. And then calibration on top of that. Repair costs could easily spiral out of control, and easily spend £750+ and they could still say can't fix it... I'd pay to have it repaired, but I'm not signing a blank cheque for maybe or maybe not fix it

[Kleinstein]

The 3 resistors work togeter. So the individual resistors do not matter, but just the combined value of R88*R90/R89. So it is only a single value to trim. So there is a chance one could try trimming the value with an added parallel resistor (e.g. 10 M range) to one of the 20 K ones.
This way one could at least check what effect the resistor would have.

PS: I had a look at the service manual description of the ADC. The AZ capacitor C95 is supposed to compensate for an offset error of the DAC and this would include the part of R88,R89,R90. So unless the resistors are far off (e.g. >> 1 %), they should not have a significant effect on the result.

[veedub565]

Well I have tried a few different values in parallel and series with A11 R90 to see the effect. I was able to get the reference voltage at U27B pin 1 almost exactly 6.4v. And also I was able to make it a long way out of spec (5.90v). In all cases, the output voltage did not change at all, even after zero cal.

So yes it does appear the absolute value of the reference voltage does not affect things very much. Changing the value of R90 also has very little effect.


Back to the drawing board for me then

• When I first got this the output was 10.0000v, and I guess most operators don't perform the zero cal. Then it developed A8 relay fault, but output was still good. Then I performed zero cal and output is now bad. So whatever problem the DAC has is causing bad zero cal. Although it seems to think zero cal has performed successfully.

• It also seems to think it has an A8 relay problem, but 100% fault is on the A11 DAC

• output at 10VDC is out by 0.003v and also input to ADC amp is 0.003v

• At the point during self diag where it throws A8 Relay error, it appears to be comparing "PA COM" to "ADC COM", both of these should be 0v (when measured it is 0.003v) without getting into the firmware It's difficult to know what it is expecting to see here

• Following the A11 troubleshooting steps shows no problems, also no A11 self diag errors. So whatever the fault is has evaded both troubleshooting and self diag.

• If A12 is removed then I don't get the A8 relay error, only when A12 is fitted do I get the A8 relay error.  Yet both A12 and A8 have been tested in a working 5700a and both are fine, problem is with A11.

Some or none of these things may be significant, trying to piece it all together to even guess where to look 

[Kleinstein]

I would expect the zero cal step to be repeated from time to time, especially as part of any calibration, but also on a few more extra occasions.
There is a slight chance the old calibration was done without the extra zero step - hard to tell, were and when this was done. With a broken instrument there may be a more sloppy adjustment in an attempt to fix the problem.

A difference between PA COM and ADC COM could be a contact problem with a card connector or a fault in ground current compensation. The two may have to come together. So when the ground current compensation is working well a bad contact may not cause an error, and if the connection is very good it may not trigger an error even with poor current compensation.
 
It the 3 mV difference measurend with the calibrator (e.g. some bug mode?) or with an external DMM ?
The ADC part could have some offset from the amplifiers (2xOP227 in the amplifier stage and LT1056 as input buffer). This offset would be subtracted digitally and should normally not reach 3 mV, as the LT1056 is only after the gain.

[veedub565]

The zero cal should be performed regularly, but from experience engineers don't often do this like they should.

I did look at the solder joints for the card connector, and all looked ok, but I will try giving the connectors a clean with some IPA.  And also take a look at the ground current compensation circuit, maybe something odd going on there.

The 3mV difference, I set the calibrator to 10VDC output and measured at the output terminals (with a DVM) 9.997V. So it's reading 0.003V too low. Then I ran the self diag routine,  at the point where it fails A8 relay error I measured the input to the ADC AMP (common to TP12) I see 0.000V on the - input (REFCAL) and 0.003V on the + input (RCL) so a 0.003v difference, which seemed a suspicious coincidence. I'm guessing the self diag routine isn't expecting to see any difference at all here, and any problem then A8 relay error.. but that is just pure speculation

[Kleinstein]

For the 3 mV error for the output, is this more more like a 3 mV offset of a 300 ppm error in the scale factor. A measurement at some 1 V or 0 V should give the answer to this.

[veedub565]

For the 3 mV error for the output, is this more more like a 3 mV offset of a 300 ppm error in the scale factor. A measurement at some 1 V or 0 V should give the answer to this.

I'm back at work for rest of the week now so not much time to investigate too deeply at the moment, but I did make a few measurements to help answer above. All measurements taken using calibrated Datron 1281 at 6.5digit resolution, and after zero calibration had been completed.

220mV Range
0V = 0.2889v

2.2V Range
1V = 0.9999v
2V = 1.9997v

11V Range
3V = 2.9995v
4V = 3.9992v
5V = 4.9990v
6V = 5.9988v
7V = 6.9986v
8V = 7.9984v
9V = 8.9981v
10V = 9.9979v


So 0V output, even after zero cal shows some 0.2889v error. And then at 10V error is 0.0021v.


Among other things suggested, I'd also like to investigate the duty cycle circuit and do the math to see what output voltage should be. I'd be interested to see if duty cycle math works out at 10.0000v or 9.9979v. It would be interesting to see if DAC is setting the correct voltage, but somehow between DAC output and terminal posts it's going wrong. Or if DAC is setting the wrong voltage to start with.

If DAC is setting output to be 9.9979v (@10v) and same 9.9979v is measured at terminal posts, then not much wrong with DAC output. And problem may be with ADC side, and maybe reading it's own output incorrectly. .. It's a theory and something to investigate anyway.


Part of troubleshooting guide for the duty cycle says the following

"Check OUT1 from U6. Connect an oscilloscope to U6 pin 16 (common to TP1). Set
the 5700A to 6.5V dc, operate, and set the oscilloscope to 2V/div at 2 ms/div. The
oscilloscope should display a TTL-level square wave with approximately a 50% duty
cycle.
"

I measured 48% duty cycle.... but no spec only approximate so I thought that's ok.... OUT1 is only for course adjustment, so maybe 48% is close enough to 50%. Then OUT2 is fine adjustment. Still, worth investigating and to see how the maths works out

[Kleinstein]

The error mainly looks like an error in the scale factor. The 220 mV is however way off. This could be the divider 220 mV setting itself, but also the part used for the internal 5700 cal steps to measure the divider and offsets.

48% PWM should be close enough to 50%, as the absolute values of the reference voltage has quite some tolerance.

[veedub565]

The 220mV divider is on A8 which works ok, I'd say in that case more likely problem with part used for the internal 5700 cal steps to measure the divider and offsets. ADC ic itself must be ok otherwise lots of other things would fail self diag.

Maybe have a look at the parts A11 plays in self cal and zero steps.

[veedub565]

Another update on the progress of this repair

I've methodically tested each diode/zener and marked them off as tested. I've also been methodically removing each transistor from circuit and testing them one by one. The 9th one I puled tested bad !

Q23 The Peak Atlas DCA component tester identifies this as a faulty component. I checked with a DVM and the gate is open circuit from the source & drain. I considered that maybe the Atlas DCA can't identify this part properly. So I pulled a couple more with the same part number and it correctly identified those as a N Channel JFET, and identified the Gate terminal.

So it seems Q23 is indeed faulty. I'm not sure what effect this would have on the circuit, or how it would produce the faults I'm seeing, but it can't be good. So I now need to source a suitable replacement part, and see if this cures at least one of the problems I've got (after many false promises I'm not holding a celebration party just yet)

I did already do a check of all the transistors (one of the first things I checked months ago), and checked for shorts. usually they fail short circuit, it's a bit unusual for this to have failed open circuit.

[Ugur]

Hello,
Can you check U2 in the A11?

[veedub565]

Yes of course, U2 Neg Offset Circuit and Shunt Linearity Control.  Is there anything in particular you want me to check ?

I need to replace Q23 first before I can do any measurements. I was going to replace U2 actually before I found Q23 was faulty. 

[veedub565]

Q23  I checked with a DVM and the gate is open circuit from the source & drain.

I must have made some mistake yesterday, I rechecked this morning with the DVM and read the following

Drain to Gate = 168ohm
Source to Gate = 258ohm
Drain to Source = 320ohm

it does still read faulty on the component tester

[Kleinstein]

At least in the plan I have Q23 is a MOSFET and not a JFET. The gate should thus be isolated. With an open gate the drain source reading can be rather random, depending on the gate charge.

[veedub565]

On my schematic I have Q23 as J2086, but the parts list shows it as J2907-TR3 and the part I pulled out is marked J2907 (this is on the A11 DAC card)

[Kleinstein]

OK- the part numbers repeat on different PCBs.  In the DAC part Q23/Q24 are used as diodes. In this use the JFET type is not that critical.
It looks like for protection between different nominal grounds. At least for testing it should be OK to use another low leakage diode there (e.g. BAV199 or transistor BE junction) - for a 1st test even 1N4007 could do.

Used as a diode the part could actually even be a BJT, but doubt the current part is a 2N2907.

[veedub565]

Thanks for the explanation, I've got a 2N3819 JFET handy would that work? it's only 25v not 60v but if it's not that critical might be ok.

I'm hoping this solves the problem I have with this card, but I can't really see how it would. Especially if it's just there for protection. This one appears to have gone low resistance Gate to Source/Drain so acting more like a resistor than a JFET, maybe having some strange effect. Anyway it can't be good and needs to be replaced. 

[Kleinstein]

The 25 V rating should not be a problem, as it is used as back to back diodes, so never more than 1 V reverse voltage and likely relevant with less than 100 mV.
The 2N3812 may not be very robust, but should work.

[veedub565]

Great I'll give it a try

[Ugur]

Q23 part code J203 N-Channel Jfet.
You can also take out the Q24 and check to be sure of the part, it's the same part.
By the way, it just occurred to me.
There are two large relays just behind the front panel of the device, these relays can be removed from the board after opening their tabs.
Check the resistance of the contact terminals of this relay. (in open and closed state)
should be zero ohms.
Remove the cover and clean the contact ends and common ends with fine sandpaper, then clean with oily contact spray and try again.
This could be causing the error you are experiencing.

[veedub565]

Well Q23 was definitely faulty, replaced with another JFET, but it hasn't made any difference at all to the fault 

Back to checking out every single part one at a time, you never know I may find another failed FET somewhere. 

[Ugur]

Hello,
The problem may not be just the semiconductor parts.
You have to be patient, yes it is not easy but you can do it if you want.

Now can you tell me what errors did you get as a result of the diag test?
Can you do the reset test?

First of all, can you remove the relays behind the front panel that I mentioned in my previous message and check the contact resistors on and off?

[veedub565]

So I have some good news finally

I found that CR5 is short circuit. I should have spotted this far sooner, it's connected back to back with another diode so when I measured it in circuit I assumed the low reading was because of this. I lifted a leg today to test it and found it short circuit.

This would mean that SCOM and ADC COM were shorted together, which would really upset the calibrator !

I put a replacement diode in (1N4148) and it has fixed the rather misleading A8 Relay error.


The voltage output is still way off though, even after warmup and zero cal it still outputs 9.997V instead of 10.00000v. A good DAC card in this calibrator produces a good 10.000v so we know there is still a problem with this DAC card !

Bit strange in the past couple of days I've found CR5 and Q23 are both faulty. They are both in the same area, and both arranged as voltage clamps/limiter. I wonder if some spike or surge got on the line. Maybe someone put a voltage up where they shouldn't.

[SometimestheDragonWins]

I think i can dig up my parts list i ordered from digikey when i replaced all the caps on the 5700a i used to have if you think it would be handy. Something like 400 bucks in caps but all from the same place.

[Ugur]

Maybe you just need a adjustment calibration.

[veedub565]

Maybe you just need a adjustment calibration.

Yes this is what I've been thinking. I think it was Kleinstein or TiN said that a 5700 should never be that far out of spec. Thing is though I've refitted all the cards back into the unit and it now passes it's self diagnostics with no errors at all. So there can't be that much wrong with it now.

If someone had done a bad cal in the past, like tried to cal without fixing the fault. Or tried to cal it using the wrong reference standards... maybe that could put it a long way out of spec. The UKAS lab at work might do a cal on it for me if I ask them nicely. I don't really want to go to all that trouble though only for it to fail.

[veedub565]

I think i can dig up my parts list i ordered from digikey when i replaced all the caps on the 5700a i used to have if you think it would be handy. Something like 400 bucks in caps but all from the same place.

I did replace all the electrolytics some time ago, but yes it might be handy to have a list. I've not done all the tants yet (I'm living dangerously) I wanted to make sure I could get it fixed first before spending too many hundreds on new parts

[Kleinstein]

From the plan it looks like the reference is part of the DAC card  (not sure about this). In this case it could still be a question of the calibration (from numbers from a cal in defect state).

The self test can do a lot of checks, but there are also limitations. So there are some possible defects the self test can not find (HW limits) or that are just not included in the software, as there are way to many ways things can go wrong and they did not consider this as a probably failure mode.

After the repair I would wonder if the zero and offsets are still correct - I would expect the described defect diode to possibly effect the zero in some ranges. The other point with a broken DAC would be that the linearity could be off - which could cause the ACAL procedure for the higher lower ranges to be possibly be wrong, but it should hardly effect the 10 V point.

[veedub565]

Yes the reference is part of the DAC card. It uses a very stable 13v reference voltage, which it can then divide or multiply as required to generate the different outputs.

I did perform a zero cal following the repair, didn't make any difference. The faulty diode seemed to be functioning as +/- voltage protection. When it went short it would have shorted two of the commons together that shouldn't have been.

I would speculate that the ADC part is working ok, when CALCHK is performed it is correctly measuring the output and saying, hey this is 1,000% drift from the cal data I have stored. It can read it's own output and it can see that it's a long way out. I've been unable to find anything wrong with this part of the circuit, or the null amplifier. So chances are the zero cal works properly too.

 I know there's nothing wrong with the 13v reference as I've checked it. Maybe something wrong with the DAC itself, it's got the right data, it's got a good 13v reference, but for some reason the output is still incorrect.

[Kleinstein]

If the reference and DAC are changes one would also need different calibration constants. The references are stable at some 13 V, but individual units may be different quite a bit, maybe up to 1% (AFAIK no official specs for the LTFLU reference).

I would not expect much to go wrong with the DAC itself to cause such a large error in the scale factor.  The DAC gain is more or less following the PWM ratio with only a small part (some 2% each - which is surprisingly much)  from the fine part and neg offset The fine part should be checked during selftest ACAL.

In theory there could be some trouble from the linearity correction part:  e.g. U38 or Q4 in the DAC not working may reach about the order of the observed error of -3 mV at 10 V nominal out.

[veedub565]

Thanks I'll investigate these area's more closely

[veedub565]

I've been having a bit of a poke around the linearity control area

U2 and U38 both replaced with new parts
Q4,5,6,7 all tested working
VR7 tested working

One slight difference is Z2 which says 14k and 11k on the schematic, but I measure it as 14k and 14k (even with U38 removed). it seems rather too precise to be a fault, possibly an error in the schematic

[Ugur]

Z2
14 Kohms from 1 to 2 pin
11 Kohms from 2 to 3 pin

Legs 2 and 3 are connected to legs 2 and 6 of U38.

Maybe there is a revision as you said!

[veedub565]

So a small update on this

Having got my 5700a to pass all it's diagnostics tests. I was lucky enough to be able to take it into the UKAS lab at work and perform an artifact cal. I left it switched on for 24hrs to warm up and stabilise. I had access to all the correct (calibrated) 10v, 10k, 1ohm standards.

The 10v part went through with no issues, and a small change of around 7ppm (which doesn't seem to account for the large PPM this instrument is out by)
The 10k part also went through, but with a rather larger change of 137ppm

The 1ohm part failed though with the following error message - "Couldn't Ext Cal 1 Ohm (Ext Meas) (2582)".

[veedub565]

Had another couple of tries at this today and got the same error "Couldn't Ext Cal 1 Ohm (Ext Meas) (2582)" and also "Couldn't Ext Cal 1.9 Ohhm (Int Meas) (2587)"

I also ran the Cal_Chk report again, now that I've got it to pass it's self diagnostic routine, results attached.

Quite interesting to note a couple of things

1. It doesn't look like this unit has been calibrated since 2006, that's a lot of years for this to have drifted a looong way out of spec. Plus recent change of resistors and capacitors and trauma of being opened up and worked on.

2. The full scale shift on 11v range is -193.91ppm or  -0.002133v. Well a shift of -0.002133v corresponds nicely to the 9.997v I'm reading when the unit is set to 10v. When running the artifact cal with a lab standard, it showed a reference adjustment of 6.5v=7.2ppm and 13v=5.6ppm ?


 I'm beginning to think this just needs a good calibration now... except it fails artifact cal on the 1ohm part, and I've tried it 3 or 4 times now, in a lab, with the proper calibrated lab standards.

Watching one of the XDev video's and their calibrator was even further out than mine is, reading 9.6v when set to 10v. So also kind of makes me think mine just needs a good cal... if I could get it to pass the 1 ohm. 

[veedub565]

Eugh dirty relay contacts. The contacts inside the relay aren't any better, worse if anything. This isn't going to help matters !

I switched the relays around and did the artifact cal again. It still failed 1ohm, but for 10v and 10k the ppm changes were drastically different, like hundreds of ppm. So probably these relays need changing.

Potter and Brumfield (now TE) R10 E6297-3 relays, custom made for Fluke I think, gonna be hard to get hold of.

[TheDefpom]

I have had good success removing the relay covers and cleaning them with deoxit gold on a stip of paper placed between the contacts and worked backwards and forwards a dozen times, I have done this a few times now on different equipment and it has worked every time, and continues to work to this day (they haven't failed again yet).

[veedub565]

Another update on the long running repair of this Fluke 5700a, and I've resorted to sending it to Fluke UK for investigation.

After the latest issue with it failing the 1ohm and 1.9ohm part of artifact calibration, I checked out the A9 Ohms cal and A10 Ohms main boards. I couldn't find anything wrong though, and to be honest was a bit lost at this point for where to look.

I contacted Fluke and surprisingly they agreed to look at it for me. I'm surprised firstly that they would even deal with me as a private customer not a business. And secondly these units are 2yrs out of support now. They quoted me a standard investigation fee, I won't say what it was but it was hundreds of £££. And any technician time to repair a fault would be added on top of that. I was a bit worried this may spiral out of control !. However as a minimum for the cost of the investigation fee they would test all the boards and be able to tell me what was good, and what was faulty.

Anyway I dropped it off, and a couple of weeks later they contacted me to say all boards tested. 1ohm fail caused by A3 Motherboard, 1.9ohm fail caused by A8 Switch Matrix. Additionally the 3 big relays behind the front panel were swapped out with new ones, and no change was observed so the original ones were refitted. They did not have any replacement boards available, and seemed to think there wasn't much further they could progress with this.

I did try and buy some spare relays from them, but it seems even Fluke is having difficulty getting hold of them. So I accepted their findings, and surprisingly Fluke dropped the investigation fee and just charged me for the technician time which was substantially less.

I have to say the service from Fluke has been excellent, and something a certain other large test equipment manufacturer could learn from. The only thing I was a bit bummed about was I sent the unit in to them in a proper FLuke Calibration box, with handles,and the proper foam inserts. As standard practise they binned it, and returned the unit to me in a large generic box, with no handles (making it awkward to manage) and (badly) packed with expanding foam 


So armed with this new information that A3 and A8 are the source of the latest faults, I did a bit of fault finding on A8. lo and behold CR13 is short circuit. Clearly wasn't causing any self test errors, but perhaps contributing to this new fault in some way ?. Looking at the schematic it's related in some way to "Out Lo", "Out/Sense Lo", "PA Com", "OSC Sense Lo", "Int Sense Lo"  . Looks like CR12 and CR13 are protection diodes limiting the voltage to +/-0.6v so CR13 short, I'm not sure what effect that would have. I've ordered a new part anyway, so will try it and see.   

 

[veedub565]

Another small update. Replacing CR13 on the A8 Switch Matrix didn't fix anything, but I didn't really expect it too. It was faulty and needed replacing anyway.

I took a look at the A3 Motherboard, and I found some burnt out tracks. This isn't good news, means a lot more work now to cut out the burnt bits and repair tracks. From what I can work out the following tracks have burnt open

K13 Pin 11 to K6 pins 6/11
K13 Pin 9 to K7 pins 13/4 (this trace also goes to K6 & K8 pins 13/4)

Also burnt out areas at J201 (A8 Switch Matrix) B FB and J611(HV Control) also B FB

Bit tricky to work out what's happened, but seems to relate to SCOM, OSC SENS LO, OSC LO GND, and OUT LO (front panel binding post), and B FB line.

The diodes that I've found to be S/C on A11, and A8 seem to be for voltage limiting / protection diodes.

Trying to piece together what's happened to this unit, maybe someone put too much voltage or current up the OUT LO binding post.... or maybe something attached like external amplifier unit because B FB line seems to be related to that function.

[veedub565]

Well it's been a while, but I'm still working on this on and off. I've got so many other projects on a the go and work commitments.

After stripping the 5700 down to remove the A3 motherboard, I salvaged an old LED light panel to use as a kind of lightbox. It's great to be able to put the board on and see right through it, and see where the damage is.

Some of the damage was under the motherboard relays, so I removed these to be able to see more clearly, and also I was able to test them out of circuit to verify they work. Is there a datasheet or pinout for the custom COTO relays as I'd like to test these too.

I also tested RV1 Varistor and it measured 5ohms, which seems a bit low. I tried applying some voltage to it, and found it was already drawing 500mA at only 2v so I think this is probably faulty. Also the two tracks leading to this component are both burnt out! presumably any standard 22v varistor replacement would work here.

Finally got around to being brave and getting the dremel out and cutting into the A3 motherboard PCB to cut  out the bad bits. I've never done this before and I was rather nervous, but hey it's broken anyway so nothing to lose right. I was quite pleased with how it went and has turned out. Now to epoxy repair the PCB and then the really hard part of linking all the broken tracks and getting it right !

[veedub565]

Ok so I have another update for this long running Fluke 5700a repair. I've been a bit sidetracked with other projects lately, mostly I've been working on a HP 8341B and setting up my own 8510C VNA system.

Anyway, the 5700A. I think last update I'd sent it to Fluke UK who tested all the boards and reported back that A8 Switch matrix and A3 Motherboard had a problem. Problem with A8 turned out to be a couple of blown protection diodes, which I replaced. And the motherboard turned out to have some burnt out tracks. 

I finally got around to cutting out the burnt tracks on the motherboard and soldering in some wire links. It looks like the tracks from RV1 to K7 and K13 had burnt out, RV1 itself was blown, the "B FB" track had burnt out in 3 places between J611 High Voltage Control and J201 Switch Matrix. And "PA Com" had burnt out between JP1 E16 and J611 High Voltage Control. I tested all the A3 motherboard diodes and relays and all tested ok.

I've no idea what could have caused all this damage, I can only suspect something external was put up somewhere it shouldn't have gone. especially as Fluke confirmed all the other cards were ok.


So the latest update, After I did all the wire links I fitted minimum amount of cards for it to work (DCV only) and tested the output.... still 9.997v Hmm, I performed a "zero", now previously this would perform straight away as soon as the instrument was switched on. Now however it says unable to perform zero calibration. Hmm, that's different, so i left it switched on for 30mins to warm up and tried again. And it went through ok. I suspect with PA COM open circuit it wasn't referencing anything properly, probably just seeing some fixed value. now it's seeing true value and takes time to warm up and stabilise.

I tested the output again... still 9.997v Damn !

Then I remembered that the DAC card in this unit isn't the original one. The original DAC card was faulty and swapped with a working DAC from another (broken) unit. In hindsight this was a bad move, and I should have repaired the original. Anyway, it probably means that the original calibration data is a long way off and pretty worthless now.

I decided to do a rough and ready DCV calibration using a Time Electronics DCV source. I know this isn't really good enough source, heck I didn't even have the case fitted to the Fluke, it was still open. I just wanted to see if it did actually adjust the output to be more accurate, or at least did something.

DC source was verified at a stable 10.0003v using Datron 1281 DMM. Fluke 5700a DCV calibration completed succesfully, biggest reported change was some 5,900% on 220mV range (this concurs with the thousands of % error I saw in the self test report).

 I tested the output..... and it's now 10.0001048v

Thats much better,  error from 10v is now 0.0001v instead of 0.003v. It's not perfect, but remember this is a very quick and dirty cal.

So this is a positive development, Fluke say all the other cards are ok, but I think I should still double check the High Voltage card.

I'm lucky I have access to a UKAS calibration lab, so I am able to perform a proper cal on this now against primary standards. Fingers crossed I now have a working FLuke 5700A Calibrator !

[veedub565]

Bah ! well I thought it was all fixed 

24hrs after I did the ACAL I ran Cal Check Report and found that the 220v DC range was some 300% of spec, other ranges didn't look toooo bad, although 220mV range is 60% of 12 month spec, that's not good either.. I ran Cal Check again after 48hrs and 220v DC range was now some 800% of (12 month) spec. Additionally some other ranges had shifted quite a bit too now.

I measured the DC outputs with 8.5 digit DMM.
1,000v = 1000.00v
10v = 10.000001v
1v = 1.000013v
100v= 98.7v

Clearly a problem with 220v range


So, a bit further along but still not happy. Going to be hard now because there's no obvious fault, it's essentially working but drifting out of spec.  Suspects at the moment are the switch matrix and power amp assy. Following the troubleshooting doesn't reveal a problem with either 

Also wondering if I left it long enough to warm up prior to conducting ACAL. Could be 11V reference wasn't up to temp and stable enough, and alot of other DC ranges depend upon that. Any small drift of the 11V ref would be magnified by the power amp assy. Just a thought

[veedub565]

Time for another update

I'm not sure how this happened, or how I missed it before. Anyway, K17 on the A8 Switch Matrix was fitted backwards, i.e. it was reset when it should have been set and visa-versa.  Not sure what effect this would have had, but it can't be good.

I refitted it the correct way, and made some checks to confirm it was operating as it should be. I was hoping that might be the final piece of the puzzle... HAHA no chance ! I now get a new error when performing artifact cal or cal_chk

 "unable to cal fine tune (14bit) DAC (2520)"


There's very little information in the manual about the 14bit DAC, and no information at all about error (2520). The output of the 14bit DAC is checked during the self test and this passes ok, so one would assume this is working properly.

Speaking to the team at XDevs even they aren't really sure on this one, but  it seems this is some issue with the AC calibration constants. Given all the repair work on this unit, and the fact it has a replacement A11 DAC card fitted it's not surprising if the original factory cal constants are causing a problem. Next step is to backup the CALROM and format to factory default.

Interestingly I was speaking to an ex Fluke engineer about this, he remembers working on 5700A and has apparently seen this error before. He told me that replacing the A11 DAC card fixed the problem. Interesting because A11 was already replaced on this unit, shame I don't have a spare working A11 DAC to try, or the original one to put back in.

[veedub565]

Picture here from October 2021 shows A8 switch Matrix, and you can just about see K17 top right of picture... and it is backwards compared to every other picture I've seen of A8. Pretty sure at that point I had only just replaced all the CC resistors. I mean that is backwards right ? because this new error has only occurred after I switched it around.

https://www.flickr.com/photos/193832816@N04/51648147365/