4

I've seen a similar question to asked here but but I'm not sure it fully addresses the issue I'm having. I'm using a PSU the same as this one from the 128K +2 Spectrum 128k +2 PSU PSU underneath

According to this site, this PSU should be compatible with my rubber key 48k (issue 4A). However when I turn on the PSU it makes a disconcerting buzzing noise - it sounds just like a tattooist's needle! The voltage of the dc jack measures around 12.5 volts on the multimeter. I've read that these PSUs are often overvoltage until a load is placed on them. When plugging this PSU into my rubber key speccy, checking the contacts around the DC input on the board gives a reading of between 11v-12v. I'm concerned this could damage the speccy? Opening the PSU to inspect it I can't see any leaking capacitors or any obvious signs of damage. Anyone have any ideas?

EDIT Here's a link to a video showing the buzzing PSU

1
  • Is that the PSU that has a switch or not?
    – tofro
    Mar 24, 2022 at 16:49

2 Answers 2

4

Buzzing can have two basic reasons:

One (luckily the unlikely one) is electrical current jumping an air gap in the PSU causing intermittent shorts through an arc. It's unlikely you could have measured a stable output voltage in this case, but it's still worth checking carefully. This case is dangerous and a fire and electrical hazard!

The other (more likely one) is just a mechanical annoyance: The metal sheets or the copper windings (or some other parts) in the transformer have losened over time and start vibrating with the AC frequency. In this case, check for loose parts (espcially the PCB on top of the transformer) and try and fasten them. If this doesn't help, check the transformer. It's normally held together with long screws you might need to tighten. If that doesn't help either, you can try a generous spraying of the sheet packet and windings of the transformer with special isolating lacquer

example lacquer.

This is not quaranteed to help, but might (glues the single components of the transformer together and stops the vibrations). Make sure you get a product that is certified as an isolating lacquer spray for electric components - don't use any old paint you might have!

With regards to the voltage:

The linear regulator in the ZX Spectrum can easily "eat" the 12.5Volts. No reason to bother, it will simply transform the rest into heat. If you do bother, however, you can replace the voltage regulator by a switching one, like a Traco

1
  • thank you for this it's been really helpful in the troubleshooting. I can't find anything that's obviously loose to cause the buzzing. the isolating lacquer sounds like it might be a good option... going to post a video lower down which may help
    – iamkl00t
    Mar 27, 2022 at 21:24
3

Welcome to the Retrocomputing Stack Exchange. First of all, if you have any reason at all to suspect the PSU, it's not advisable to connect it to the machine until further tests.

Buzzing could be just due to 50hz vibrations corresponding to the line frequency in the mounting or inside the transformer itself and is not necessarily critical, albeit annoying. You can maybe feel the vibrations if you put your hand on it. Maybe a mounting can be tightened, maybe there is nothing to do about it.

Buzzing can also be caused by arcing due to fx a loose connection on the primary side, which is worse (from a safety standpoint). The sound probably wont be as regular and you might be able to smell ozone in that case. Try to power it up outside its case (being careful not to touch anything on the primary side or to put it on something conductive) - it may help to locate the origin of the noise. Posting an image of the insides could assist people commenting on the issue as well.

I have not serviced this particular version of the Sinclair PSUs, but if they are like the ones I have they only contain the transformer, rectifier diodes and a smoothing capacitor. In that case an over voltage failure mode of the PSU is unlikely. If it contains active regulation, that's a failure point that could fail with over voltage - but as I recall, the regulation is ultimately done with a 7805 inside the Spectrum itself (that regulator can still fail with age, but that's another story as long as the focus is the PSU itself).

It is normal that a voltage measurement without the anticipated load is higher than the nominal voltage. 12-15V for a 9V supply sounds quite normal. But if you loaded it with say around 1A, 12V sounds a little on the high side - but the Spectrum's regulator should be able to deal with it. (Of course, the higher the voltage it has to dissipate the hotter it will get and the more strain is put on it and the shorter its lifespan may ultimately be.)

TL;DR - If there is no active regulation in the PSU, the voltage coming from it is likely fine. (That doesn't mean that it may not need service, fx degraded capacitors could lead to excessive ripple and noise/stability issues, but in general that it is not a condition that will damage the connected device.)

4
  • 1
    Wow what a thoroughly well explained and thought out post. Thank you very much indeed you've given me lots of food for thought there. It may be a week or two until I get a chance to try these suggestions out, but I'll return here once I've done that. Thanks again for your help, and for the welcome :)
    – iamkl00t
    Mar 24, 2022 at 21:05
  • 1
    Overvoltage failure modes of a mains transformer are not that uncommon. The mechanism is shorting and burning loops in the primary.
    – fraxinus
    Mar 25, 2022 at 17:47
  • This can certainly happen but is quite uncommon in my experience (never had a case of that myself in the last 10 years of servicing 8-bit machines), whereas pass elements in active regulation failing with corresponding over voltage failure is definitely not uncommon (had multiple cases of that in the same duration). YMMV. But that's talking odds - either way, the PSU should always be tested.
    – Retrograde
    Mar 25, 2022 at 20:20
  • Thank you i've tried your suggestion and added a video to my original post, showing the inside of the PSU as well as the buzzing noise. I rolled up a piece of paper and put it to my ear, moving side to side over the PSU to gauge the exact origin of the buzzing. It seemed to be coming exactly from the centre rather than either side. With regard to the ozone smell, I don't really know what ozone smells like but there is a slightly funky smell coming from the PSU when it's plugged in. I'd describe it as a cross between the smell of mineral oil and new trainers... weird smell.
    – iamkl00t
    Mar 27, 2022 at 22:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .