Philips PM3331 oscilloscope delay line replacement

[xackus]

I am repairing a Philips PM3331 oscilloscope (a variant of PM3335). I acquired it already disassembled, and it's missing the delay line cable and some 50 Ω coax cables.
All the PCBs look fine, but it likely needs some capacitors replaced.

I need a replacement for the 54 cm delay line cable. I have a picture of it in another Philips scope.
Can I replace it with two 75 Ω coax cables of the same length?

The 50 Ω coax connectors seem to be obsolete. I can't find anything about them.
Am I correct in thinking that it wouldn't significantly hurt the performance of the scope (it's 40 MHz) if I just used regular pin header connectors (soldered to 50 Ω coax cables)?
The delay line cable doesn't use coaxial connectors either.

[Alex Eisenhut]

It's hard to tell from a picture but those coax connectors could be d-sub coax contacts.
https://www.mouser.com/datasheet/2/18/amphenol_d-sub_inserts-1157297.pdf

The problem with delay lines is that they're usually made with special dielectric to slow them down so you don't need so much of it. And they need to be real good so as not to distort the displayed signal.The main job is to delay the vertical long enough for the trigger and unblanking to work.

But at 40MHz, just for laughs, try a sata cable

https://www.ebay.ca/itm/24-inch-SATA-3-0-III-Serial-SATA-HDD-Hard-Drive-Data-Cable-Long-Wire/131243331625

It is two coaxes in there after all. But I don't think they're that tightly specified for impedance, it's like +-15% or something. But the price is right.

[xackus]

It's hard to tell from a picture but those coax connectors could be d-sub coax contacts.
https://www.mouser.com/datasheet/2/18/amphenol_d-sub_inserts-1157297.pdf

Looks like you are correct. But these connectors are bloody expensive!

The problem with delay lines is that they're usually made with special dielectric to slow them down so you don't need so much of it. And they need to be real good so as not to distort the displayed signal.The main job is to delay the vertical long enough for the trigger and unblanking to work.

I found in the manual that the delay should be at least 15 ns. This would imply a velocity factor of 0.12. That seems really low. I would need 2-3 meters of regular coax for that.
Looking at the diagrams, the delay shouldn't matter in digital sampling mode.

But at 40MHz, just for laughs, try a sata cable

https://www.ebay.ca/itm/24-inch-SATA-3-0-III-Serial-SATA-HDD-Hard-Drive-Data-Cable-Long-Wire/131243331625

It is two coaxes in there after all. But I don't think they're that tightly specified for impedance, it's like +-15% or something. But the price is right.

I have some sata cables lying around. I will try them out haha.

Thanks for the insight! I'll be sure to post my results.

[Alex Eisenhut]

Here's one

https://www.ebay.ca/itm/New-ITT-Intercontinental-Socket-Contact-Gold-D-Sub-Coax-DM5374-5001/283812674889

But only one... Maybe you can "contact" (hahahahaha  ) the seller for more?

The delay line is so you can see the leading edge of the waveform that caused the trigger, if you just bypass the delay line your scope would still work but you won't see the trigger event on the display.

If you change the impedance of the cable, even at 40MHz, for that length of cable, I'd say you'd start seeing some amplitude inaccuracies, but nothing grotesque.

edit: found this

https://www.ebay.ca/itm/DB-Female-Pin-D-SUB-Signal-Connector-White-Teflon-crimp-for-RG174-RG316-RF-cable/172896096974

That's the ticket