DIY Fluke 5700A calibrator
I've started new project, second 5700A series MFC from random dead parts.
My current board list / condition:
- Front panel, A1/A2 = have complete new front panel assembly
- A5/A6 = no plan to install
- A3 = repairs needed, missing K13 and MOV
- A4 = burn damage, repair needed
- A7 = 5700A version, drift 400ppm/day
- A8 = 5700A version, untested, had open heater on hybrid
- A9 = 5720A version, good
- A10 = 5700A verison, good
- A11 = have two 5700A version board, untested
- A12 = have three 5700A version board, one should be good
- A13 = 5720A version, untested
- A14/A15 = no plan to install, will not have 1kV/2.2A range
- A16 = 5700A version
- A17 = 5720A version Rev.B, good
- A18 = 5720A version, fixed, good
- A19 = unknown, on the way in shipping
- A20 = 5700A version, good, will upgrade to 5720A
- A21 = rear board, unknown, on the way in shipping
- A22 = transformer , unknown
Missing chassis cover and rear metal parts + shields. Will DIY.
Wow, an ambitious project.
Am I interrupting? If it is, will erase later.
No need erase, I'll be posting updates as the progress go, with posts mostly updated module by module.
Actually my first "Hulk" that you can see on background on top of 4142B is also "DIY" built, and happy running 24/7 fully complete since end of 2018.
Since my article is not secret here anymore, worklog about Hulk is here. It need reorganize and rewrite many things though, so just a "draft".
A8 - Switch matrix , 5700A board
Board made in 1988. Oldie 5700 series I model. Bought board as broken.
Hmm, I wonder what those labels may mean? Should be nothing, right?
Pretty heater for precision DC amplifier circuit on alumina hybrid module. Looks OK?
Hermetic OP177EZ from PMI in ceramic DIP8 for DC amplifier circuit. This hybrid works to generate 220mV and 2.2V ranges from 22V range, delivered from DAC A11.
Let's check heater resistance at the hybrid circuit.
Nooooo....it's open! Good heater should read around 40-ish Ω.
Aha! I think problem is pin crack on the corner, not touching ceramic pad anymore :(
Now it's fixed. Will test board once I get motherboards and power ready. Should be alright, no worries! Maybe swap some relays later.
My main 5720 died. Probably caused by earthquake Taiwan had April 18. Was doing some test yesterday, and found out there is no resistance output on any of the 10ohm - 100 Meg range. Took it apart, found this:
I guess now I have to use spare board from this 5700A "Hulk-2" project and look for new 5720A A10 :(
@TiN THY888 said that probably the 10 Ohm resistor (in Z1/1R22?) exploded because of over power. He come across this before and replace the resistor with two 20 Ohm PWW in parallel.
Strange, how would over power happen. 5720A was connected to 3458A during last days.
pipelie last edited by
@TiN latching relay is sensitive to vibration, the relay probably went wrong during the earthquake, shaking likely toggled a relay a small amount.
I guess you probably right.
Anyhow, I did a livestream on YouTube about Hulk-2, 5700A.
Troubleshooting A3 outguard PCBA, checking transformer and blowing up fuses.. :)
Some A4 repair pics.
Mains switch traces on layer 3 and 4 exploded, causing FR4 to become conductive and delamination of the PCB.
To fix this I drilled the damaged part of PCB out and redid wiring of broken traces. Ordered Epoxy from Digikey to secure everything permanently afterwards.
Now the more hidden but more serious problem - 2 pins on High-voltage 900VAC winding from transformer show 12.5 ohms to low voltage 44VAC winding. This is because internal short and damage on inside layers 3 and 4 in middle of PCB. I've used dremel to make a cave for access to damaged areas:
Now need to restore connections and restore shielding.
This is outguard power supply module, provide +5, +12, +35 and +75V power to CPU, VFD and rear PCB.
No much problem here, just broken L8. I replaced all electrolytic capacitor and all CC resistor as preventive maintenance.
This is interface board that have RS232, GPIB logic and rear amplifier ports + phase lock/phase in.
Calibrator will NOT work without this PCB (hang on boot process, rear PCB DUART fail message).
Full size photo are available on my site.
THY888 last edited by
Mains switch wiring after cutting out charred board patch, to prevent leakage and future problem.
And cave work in two spots where traces and charred PCB were milled out out:
Once done will fill all with epoxy and solder copper foil on top to restore shielding.
Digikey parts arrived today. Among others I've ordered single-component heat-cure epoxy to fill the carved parts of PCB material. It worked quite nicely for first ever try. :)
Also some other random parts, including brand new Sunon Maglev 92mm fans. I want to make this second calibrator quieter than first one, since it will not have high voltage A14/A15 boards I guess it could run lower speed fan as well.
Part 2 stream is out, we got first power on :)
Received today REF/DAC A11 module upgrade for my second 5700 "Hulk-2" calibrator.
This is true 5720A version, with LTFLU-1ACH (44 week 2005 + 48 week 2004) and magic hermetic Fluke seal 893334 resistor.
Board don't have physical damage, but have faint sharpie scribble "Artifact fail, Diag error code". However seller sell as "Used, all test pass". We shall test that if any problem.
No rewords or mods, like on early 5700 boards.
Fragile ceramic Hybrid module is ok, no crack or damage.
Fluke 8842A ADC used as null-meter.
Magic reference to generate ultra-stable 6 and 13V.
I just need to find Fluke A14 board + transformer to make this calibrator complete :D. Otherwise it works without 100V/1100V and 2.2A ranges.
magic hermetic Fluke seal 893334 resistor
Which one is this?
Nice photo to say at least.
Last photo shows it. Resistor glued to the back of the LTFLU REF/DAC HR.
Older 5700A have different network, P/N 761338
Are those hermetic?
I thought true hermetic should be glass-dumet encapsulated, or metal-glass-kovar seal.
I think these are hermetic. It is ceramic substrate with quartz glass cemented on top. Cement is very hard, and I see no dependence from humidity on resistance. I've cracked few resistor of this style from 8842A and when quartz is damaged (element exposed to air) there is impact from humidity. Cheaper blue resistor network with passivation are "less" hermetic too.
Ceramic and quartz glass are hermetic, but cement? I doubted.
One other thing declared by manufacturers as hermetic, apart from metal/glass/Kovar, is the ceramic package, such as VPR247Z and 1446Q used by 8508A.
lymex last edited by lymex
Ok, I found an evidence of 732C example supporting it is hermetic:
This is the thin film resistive networks in replace of the old 5 hermetic PWWs. It seems to have the same ceramic substrate/glass top, and the file name suggest it is hermetic.
An inside photo of 732B by mytek: