HP 4192A LF Impedance Analyzer - Line filter replacement

[MichaelPI]

Hello all,

I have got a broken HP 4192A LF Impedance Analyzer. After some trouble-shooting it became clear, that the line filter must be the culprit. I can measure basically 0 ohms between the line and neutral.

HP used an EMI filter made by Corcom with the part number 9135-0084.

L = 2x(7+7) mH
R = 10 MOhm
Cx = 3x270 nF
Cy = 2x2.8 nF

Unfortunately I could not find the datasheet or any further information. I want to replace the filter with a new one (no new old stock, because it is unclear, if the capacitors already degraded from usage or just storage). But I could not really find a direct replacement.

The distance between the mounting holes is approx. 75 mm, the filter dimensions without the mounting plates are approx. 63 mm length, width approx. 50 mm and a depth (thickness) of approx. 40 mm. Smaller is as well perfectly fine (except the distance of the mounting holes). The connection is done via 3 input wires (L, N, earth) and two output wires (L, N). If possible, I would stick to a solution with wires (faston connectors could be possible as well).

HP specifies a 1.25 AT or 0.6 AT fuse, dependent on the line voltage setting, hence I think a filter capable to handle >2 A should be fine.

The only EMI filter I came across so far is the FN2090-3/FN2090-4/FN2090-6 by Schaffner (ideally with order code -07 for the wired variant), that should fit mechanically in the HP. The values of the filter components do unfortunately not really match and it got another pair of Y capacitors.

Does anyone has another suggestion as a replacement or maybe could give a couple of hints regarding the proposed selection? Is there something else to watch out (current rating of more than two times the fuse rating is fine)?

Thanks.

 

[MichaelPI]

Some more pictures added.

[squadchannel]

maybe CORCOM F2120C, but not listed in TE.
te site listed datasheet for F2120E, but the filter circuit is different. no X capacitor in the last stage.

I looked at the CORCOM catalog. same circuit, similar size: 3VR3 and 3ER3.

catolog:
https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=1654001_CORCOM_PRODUCT_GUIDE_EEA_EEB&DocType=Catalog%20Section&DocLang=English&DocFormat=pdf&PartCntxt=6609000-9

[MichaelPI]

Thank you for the proposals. I managed to disassemble the back cover of the unit, without disassembly it is basically impossible to access the line filter. It is probably one of the less service friendly components of the overall unit. I had to disengage the line switch assembly, remove a couple of connectors, unscrew the VCO BNC connectors and labelled all of them.

Anyway, I will probably go ahead with the repair by removing the line filter and connecting the mains voltage directly to the fuse for testing purposes and see, if the unit is functional without it, before I'm going to buy a replacement filter.

So far my plan is (if the unit is functional) to replace the line filter and two or three (can't remember exactly) RIFA safety capacitors (cracked) and maybe the electrolytic capacitors but I am not really sure, if it makes sense to do so.





 

[MichaelPI]

Short update: I replaced the line filter from CORCOM with one from Delta Electronics (05DRCW5). The replacement filter fits perfectly into the unit. There were only two downsides: The inductance of the common mode choke is considerable lower compared to the original filter made by CORCOM and the cable length was shorter (I had to connect a part of the original piece of wire from the CORCOM filter as an extension to the wires of the DELTA filter, not very nice, but it will do it) - the cables of the new filter are stiffer as the ones from the original one, adding the ferrite cores on the input was a bit of the challenge and I messed up the bending of the brown cable... but anyway.
I added a picture of the power supply section after replacement.

[MichaelPI]

In addition I replaced the Y2 RIFA capacitors with WIMA Y2, but honestly I guess that was not really necessary, as none of them showed any signs of cracks in the case after close inspection.

[MichaelPI]

Fortunately, after removing the shorted line filter the unit is powering up again (I checked it already without a line filter) and showing plausible measurement results (within a certain frequency range). The unfortunate part of it is, there seems to be another failure somewhere in the analog section. The self-test passes without any trouble, but at higher frequencies (above 18 kHz for an inducter and 2 kHz for a capacitor (56 nF)) the unit shows UCL - something in the signal path must saturate. Before replacing the line filter and cleaning up the unit (toothbrush and vacuum cleaner) this issue was not there or I did not notice, hence the trouble shooting has to go on... suggestions are welcome.