Defiant MSH902

[Beardyman]

My Defiant MSH902 finally got a dose of electrons today. It's been an adventure which has generated so much paperwork & information. As far as I know there are no circuit diagrams for this model, even after contacting the Co-op archives. They have a few bits & pieces but nothing for the MSH902. I did finally find out what the manually adjustable inductors do, its a form of sensitivity adjustment. After having checked, double checked everything in sight it was time to plug in & switch on. No bang (good), no magic smoke escaping (even better), filaments coming up, HT rising nicely. After about a minute the set came to life, tuned in a few stations, Radio 4, Radio Solent, Radio 5 Sport. Good tone too. Both MW & LW working ok, switched to pickup input & I could hear arcing! Switch off quick! The multi way switch is something else, 12 layers & 3 positions. I had taken this out to clean it a while back but I obviously didn't do as thorough a job as I thought. A few cotton buds & IPA later it was all ok. This set switches the HT off to the RF end of things when the GRAM input is selected. I'll upload some pics later when I'm on the other PC!

[boater sam]

If the switch takes off all the HT I would put a resistor across it to feed just a little HT to avoid cathode poisoning.
HT on wavechange switches is a problem causer unless its on a completely separate wafer.

[Beardyman]

This set has been giving me problems since my last post. One was that the audio would gradually fade out but not completely & the other was a rustling noise in the background on both wavebands throughout its tuning range. The fading problem was, I discovered, caused by...me! I'd put a 51K resistor in place of a 500K from pin2 to pin 6 of the FC4, most other radios that use the FC4 have around 50K, some 100K but mine originally had 500K!?! Still, that got rid of the fading issue. The rustling has been another matter altogether. I must confess that in my ignorance I put it down to sheer age, the set is after all, nearly 90 years old. I decided to do some research across several forums to see if others had experienced the same issue. Lo! It came to pass they had but seemed more prevalent in America. After much searching, going square eyed & reading far too much clap trap the answer would appear to have its roots in mica capacitors. I've always been led to believe they are amongst the most reliable but it would appear that time is catching up with them. They suffer from a number of issues, one of which is called silver migration. This can be a cause of the rustling, popping & f*rting noises. I also stumbled across a paper written in 1960(!) about their failure modes & causes. Link here: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=2ahUKEwjU9b7999vnAhWVEMAKHVTWDtQQFjAAegQIARAB&url=https%3A%2F%2Fescies.org%2Fdownload%2FwebDocumentFile%3Fid%3D62188&usg=AOvVaw04lQQ5ao10e6UqL2TWJLvf
It's a PDF & makes interesting reading. I replaced the capacitor in the Defiant, the only one I had left alone by the way , applied some power & once it had warmed through (takes about 30-45 seconds) it was like listening to a different set. Now, that particular fault could have gone unnoticed (and to some extent it did) for a very long time but it's a testament to the original design & build that it still functioned in an acceptable way. This set is probably one of my favourites & I have no idea why, it's nothing special to look at but does have a certain presence. Have a good day one & all.

[ppppenguin]

Well done. Just goes to show that although we think we can trust low value and silver mica capacitors it's not a universal truth.

That IEE paper is a certainly a comprehensive account of problems with silver mica capacitors.

[Mike Watterson]

It's usually only a problem if there is a significant voltage across them. Some periods of USA models suffer badly. Often they tune IFT coilss or are on RF or LO circuits across coils so the problem doesn't arise.

[Beardyman]

Quick update.
It developed yet another fault, the audio would fade slightly & become distorted after about 4-5 hours. On the GRAM input this didn't happen which was a relief! Eventually I found the problem, more by accident than anything else. Chassis out on the bench running, switched between MW/LW and back then noticed some arcing/tracking on the monster of a multi way switch, between the metal work of the switch & the paxolin panels holding the leaf switches. Closer examination revealed charring right on the edge nearest the metal work. After about an hour of swabbing with cotton buds & acetone to remove any residue that may be there I ground off the charred paxolin with a dremel like tool. This was then given three over coats of conformal coating. When powered back up the familiar background hissing was almost none existent. I thought I'd done some damage somewhere! When tuned to my pantry TX it was working fine, better than it ever has. Left on all day & no problems at all. I've since found out via the Radiomuseum website that there are two versions of this set, one for AC only (mine) & an AC/DC version, same chassis, same model number but different line up of valves & tuning dial. I'm still working on the final diagrams for this set, once I 've finished & I'm happy with them I'll upload them.