Not bored with the Boards.



  • I have bought a plastic box from Maplin to put it in . I hope it is big enough
  • Re ZX Power supply.

    It may be that the main interest for me in doing this is figuring out how to mount it all safely .The big lump of metal holding the coils has to be secured and the PCB part also has to be mounted in the case properly . The case is big enough so the question of how to secure it all and stop it "flopping" about inside the case is probably my main area of interest at the moment.A continuity test shows the windings are OK .

    I should say that mounting PCB s is an area of interest for me as I have a USB keyboard with a wobbly insecurely mounted PCB USB bit . So the ideas from this could be reused later on this other keyboard project
  • Your ZX Spectrum Power Supply

    First a question: do you still have the original case? Only I would like to know the rating please.

    Second: That one is probability better off in the bin.

    In the picture that shows the top of the PCB and the transformer, you can see four black plastic pieces with metal tags that go through vertically. These are suppose to be part of the plastic of the transformer frame, they have all broken off (that's why the transformer is loose).
    As a result, there are no longer any terminals left on the transformer to solder the solid core wires of the windings to. Let alone anywhere to solder any flexible internal wires to.
    The wires of the windings should not be left to float around, it degrades the performance of the transformer. Also the enamel insulation cannot take very much movement before the thin insulation rubs off
    So it cannot meet the requirements of the regulation that requires double insulated items to have all mains conductors secured so they cannot touch the low voltage output wiring and vice-versa.
    And as this type of transformer is designed to be PCB mounted, there are no other properly designed mounting points.

  • Hi Mark

    Thanks for the reply :This is all there is so no other parts or info available. The caps are 16v 1000uf and 16v 2200uk. There is no there is no other writing about this on it

    it looked "worth a go". This supply came from a "pile of debris" I must have picked up years ago in a job lot.. And yes I had begun to think that I may have "bitten off" a bit more than I could chew" on this one : . I could spend a bit to get parts needed if this is an option . I may well be unrepairable however: your opinion on this would be welcome . I can always leave it as it is ..

    Someone has "had a go" at it before . There is writing scratched on to the PCB tracking "Back +" and "Back -" I think it says Thanks
  • The reason I asked about the rating is it looks like a 1.2A ZX81 PSU.

    Although apparently the first ZX Spectrum issue 1 16k byte machines also were supplied with 1.2A PSUs.

  • Hi Mark

    Thanks for the info

  • ZX Power supplies

    On You Tube "Joules per Coulomb " has done a helpful video on fixing the different types of ZX Spectrum Power Supply . I have recently "found" my old Speccy power supplies which I had not used much as the Opus Discovery did the same "power supply" job for the Spectrum .Looking at these old supplies now and comparing them with today's electronic gear these supplies look very "heavyweight" indeed . I thought I would do a "survey" of their state , to find out which of them work OK and which need repair . This will enable me to get more information on the different types , the parts needed for repair , like the capacitors types used ( which can also be replaced at the same time if I get a complete list I can do a "one off order" ) So the initial aim is to get a complete list of the parts needed to repair them , and all of this to be done at a "sedate" pace as there is no need to rush , as these supplies have been sitting in my loft for over 20 years : this project is for "edification only"
  • ZX Power Supply no 2zx%20Spectrum%20power%20supplies%20001_zps5orlajz0.jpg
  • Re above pic (No 2 48k power supply in my survey!)

    I will call this No2 . From the five 48k power supplies I looked at today there were several "states" of repair : 1) Two of them powered up a 48k Spectrum OK : 2) One was in bits (see the one in pieces well above) ; 3) One of them as per in the pic above (No 2) gave a multi meter reading of 9.7 v but did not power up a 48k Spectrum . I resoldered the wiring to the PCB on this one ,reversing it so that the striped wire was on the negative side of the PCB and I also put a "proper" ZX Spectrum type power plug on it. But that was not enough . More work needed on this one. At least it shows some signs of life on a multimeter. 4) And one , the last one ,showed nothing at all it was dead.
  • Power connectorszx%20Spectrum%20power%20supplies%20004_zpsk5omdfp6.jpg
  • Re wrong plud on power supply No 2

    Yes originally I had been "winging it" with the wrong (see short connector in pic above) power plug (off a calculator) . I put the longer type on but it made no difference . The Speccy still did not power up
  • No 2 Power Supply again

    This one had a 16v 4700uf capacitor
  • The PCB area on No 2 Power Supplyzx%20Spectrum%20power%20supplies%20006_zpsbqfsctns.jpg
  • No 3 works OKzx%20Spectrum%20power%20supplies%20007_zpss1gn5wmp.jpg
  • Re no 3

    One of the 2 that works OK but the yellow unidentifiable capacitor gives of a buzz . Quite loud
  • No 4 works OKzx%20Spectrum%20power%20supplies%20015%20-%20Copy_zpsbdkq9siq.jpg
  • Re above pic of No 4 Power Supply

    This one has 2 16v 2200uf capacitors . It gave a voltage reading of 13.53 volts with a multimeter and powers up a 48k Spectrum OK
  • The clever casezx%20Spectrum%20power%20supplies%20013_zpskyema02w.jpg
  • Re above pic

    The 4 plastic protrusions poiting upwards do 2 things : 1)They secure the PCB on the bottom and 2) support the metal lump with the windings . This is what I need to replicate if I am to attempt to repair power supply no 1 . But there are other bits missing on this No 1 supply as well which may make it difficult to do
  • No 5 did not work at allzx%20Spectrum%20power%20supplies%20019_zpss0gfoau5.jpg
  • Re no 5

    Years ago I put this in a bigger case and Araldited it all in . It appears to not be working at all
  • Conclusions on 48k Spectrum Power Supplies

    There was some rusting on some of the metal plates : this wasnot serious but I may treat it anyway with anti rust.Also why the loud buzz (from the yellow capacitor with no markings) on No 3 Power Supply . Is buzzing a sign of incipient failure? The "near miss" was no 2 power supply : I checked the polarity of the plug again: it was OK . The voltage was about 9.5 to 10.2v with some variation . It gives off a normal low level buzz and appears to be working at least in some respects but not enough to power up a 48K Spectrum.

    No 1 needs mounting in a case and refabricating it to secure it properly: a complete rebuild is needed really

    No 5 shows no signs of life . This one was mounted into a bigger case years ago but is not working now . It does not give a voltage reading on the connector.

    No 3 and 4 were good . So I know know the good ones. So the "survey" was worth the effort.

    The capacitors found were 1) 16v 4700uf ,2) 16v 1000uf ,3) 16v2200uf : so 3 indentifiable types . There was also an unidentifiable one with no markings on No 3 Power Supply which had a loud buzz but at least it worked .

  • Further retesting on No 5 Power Supply

    I checked it again and it was found to be working . The extension reel I was using earlier had probably cut off the power without me realizing so it shows that checking this over again does pay dividends . So only 2 of the 5 power supplies need further attention and one of these is partially working . Not a bad result
  • No 5 was an "ad hoc" repair but it works . Won't do it again thoughzx%20Spectrum%20power%20supplies%20020_zpsnyfa25zj.jpg
  • Re above pic

    "Ad hoc" repairs put together without too much thought do sometimes work OK but they do have disadvantages : namely , in this particular case , it cannot be taken apart very easily without busting up the repair itself . the Araldite would have to be carefully removed and the "Araldited in" bolts would have to be unbolted . So not to be contemplated unless absolutely necessary . So this is not the solution I propose for the bits that make up no 1 Power Supply (see pic above)

    This time I will try to replicate the support structure that the original Sinclair case provided for the "bits and pieces" that make up the internals of the power supply . The 4 plastic vertically aligned plastic pillars can be replicated by using dowelling screwed into a support plate .The aim would be to make the new case function in a similar way to the Sinclair originals There is the question of whether or not the missing plastic "shrouding" is crucial or not . I am not sure about its function and will take another look at what Mark says about it.

  • Mark refers to the "transformer frame" which implies it is just a support structure ,and says "wires should not be left to float around" : And also reference is made to enamel insulation can wear off and to "PCB mounting" I think PCB mounting refers to the fact it sits over the 4 plastic supporting pillars and is therefore mounted above the PCB.

    So there is some useful ifo. there on how to proceed ,. I think I will take my time on doing the next bit on this.

    But what about the No 2 supply? This is the one that looks as though it should work but does not actually power up a 48K Spectrum ? And it DOES give a multimeter reading as well as an nice low level buzzing sound Well I have fitted a new powerconnector on the lead. Could that be the problem?. Or maybe the capacitor is underperforming? Or is it another problem? Food for thought.
  • Back to Power Supply No 1 the " severely broken ZX Spectrum or possibly a ZX81 Power Supply

    Well ,wiring up issues aside , I think I have figured out how to put it back together in a new Maplin plastic box so that it is easy to take apart again (important if I make an error and need to rework it)

    1) I have cut a piece of "footplate" style right angled black plastic to size so it fits in the base of a Maplin box and can be screwed down into the base of this box.

    2) The right angle flange on this plastic can be used to screw in a block of wood (screwing through the right angled flange) to act as a base for the doweled pillar supports which will secure both the PCB and the coil block.

    3) I have planed down the dowel so I know the PCB's holes are OK for the dowel to go through.So the dowels are ready to be used.

    4) Next the dowel pillars are glued into 4 holes to be drilled into the wooden base secured to the plastic.

    5) The PCB can then be slid onto the doweled supports to rest at the base on top of the wooden block

    6) The metal plates with the coils can then be supported on the top of the wooden pillars the height of which need adjustment to make sure the "lump" is held securely between the pillar tops and the underneath of the top side of the Maplin box.

    7) Then I have to figure out where the wires go and solder it up which should be interesting!

    So that is the theory of the rebuild fabrication part . Now I need to "mug up" on how the primary and secondary coils relate to the PCB and the coils on top so the wiring is done correctly
  • The dowels in action on ZX power supply no 1joystick%20extenders%20006_zpslbibhvz6.jpg
  • Re above pic zx power supply no 1

    In fact this design has problems, wood takes up a lot more space than plastic so this will not go in the case I got for it : too tall
Sign In or Register to comment.