Marconi 2024 SGen SMPS faulty

[onewatt]

Hi group,
Is anyone familiar with the above siggen? I have downloaded the manual but the power unit appears to be different from the circuit included with the manual. the PU is a Coutant lambda HVI-60-47A.
I have Vy low supplies and the transformer output is being held down. If I remove the diode arrays and regulators the tx supply comes back up and the tx doesnt get hot at any time, although its getting warmer now there's some voltage on its output. I've checked all the diodes and caps in the out train and all appear good. the 1st diode array ia a bit of a strange one and i've not been able to track a data sheet, its a C8275 and under diode test gives 0.2 across each diode, but given both are the same I'm assuming its ok.
Advice sought please.

Regards
Chris

[dragonser]

HI,
I'm not familiar with the sig gen, but in general I wonder if a regulator or crowbar device has gone short circuit or partly short circuit ? I know that sometimes the diodes / rectifiers have to be a high speed version as the ac from the switch mode transformer is much higher than 50 hz .
Hopefully you will be able to track down what is causing the problem.
regards Peter B

[Radio Wrangler]

Have you tried disconnecting the outputs to the rest of the sig gen? The first step is to be sure whether the sig gen is loading the PSU with something like a ceramic capacitor gone to short-circuit, or whether the PSU itself is the problem.

We have quite a few of these sig gens at work. There was a lot of them released by the government with ovened frequency standard, high power output (+23dBm) and fast pulse modulator options.

David

[onewatt]

Hi,
Thanks for the replies.

The power supply is removed from the siggen and being powered up on the bench remote from the rest of the unit.
It appears to load the transformer as soon as the first diode array is reinserted.
Can anyone identify the C8275 diode array? I've not managed to find any spec?
ATB
Chris

[karesz*]

Chris,
you have to test an SMPS electrolyt under WV for leakagide-current and check for ESR on switching(similar) HF _a simple capacitance measuring isn`t enough. As I see it on your pic; such caps are possible some humpbacked_to change...
Some times are rectifier OK in LV tester, but faild iunder Wvoltage.
Greetings-Karl

[Radio Wrangler]

I'd try connecting in a bench power supply to the output of the rectifier. See if it sees a short. Remove the rectifier and see if it still sees a short. This ought to lead you to the root cause.

Probably all the capacitors have been stewed at elevated temperature for a long time. Were it mine, I might treat it as a rebuild job and swap the lot.

David

[onewatt]

Hi Karl, David,
Thank you for suggestions.

Your right regarding a better test for the caps. All I did Was remove all and tested out of circuit. Non are blown and they don't get hot, but it's certainly not given that they are good.
Yes David I agree it's worthwhile rebuilding and replacing all caps which I will do and I hadn't thought of substituting a DC supply for the rectifier which is what I'm just about to do.
Thanks guys, I'll let you know what happens!
ATB
Chris

[G0HZU_JMR]

We have quite a few of these sig gens at work. There was a lot of them released by the government with ovened frequency standard, high power output (+23dBm) and fast pulse modulator options.

Yes, I bought three of them with this spec a few years ago. There was a chap in Leicester selling a lot of them really cheaply if you haggled hard enough. I kept one and sold the others on.

The one I kept only had 120hrs runtime and looks brand new inside and out.
These generators are normally quite reliable and we have loads of them at work. The ones that have failed were usually because someone picked it up by one of the plastic 'ear' handles. These break easily and the generator then hits the floor. So I've seen quite a few of them with broken handles at the front and signs of impact damage to the case and connectors. Sometimes the display fails and this may also be from dropping the sig gen. The front 'ear' handles are just plastic and they become weak and brittle with age.

I've got the service manual here and I may still have the contact details of the guy you need to phone in the states if you lose the level 2 password. The default is 123456 but if it gets changed by the user (and then forgotten) you have to phone the states and get the master override (also six digit) key so you can reset the level 2 password to a known number. This is what I had to do with mine and I've reset the password to the default 123456.

[onewatt]

This one has the same options as well so who knows, it too, may have been from the same place.
Regarding the caps in the supply, those in the primary circuit are only 250v rating and look as if some have already been replaced so I'm going to blitz the caps and replace all.
The line in circuit has around 300v at the primary side of the Tx so I'm going to fit some better rated caps, but could do with some assistance with the ESR.
Having looked at Farnell, there's a lot of options to choose from, so I wondered if its possible to suggest a ballpark ESR value that should do the job? Secondary V's are 5, 12, 24vdc and the rating again looks a bit tight on the caps.
Am I right in assuming that the ESR on the secondary side can be the same as the primary?
Thanks all,
Chris

[dave cox]

Primary side caps don't need good ESR, they only handle 100Hz, although these seem to suffer far less than the secondary. The switch side ones are different - best to try and match the original spec of the ones you take out, if you can find the spec. Usually somewhat lower ESR (than original) is OK ...

dc