My ZX81's stopped working. :-(

[stephen_usher]

I brought it down from the loft to give it a check-out before the 40th anniversary in March, plugged it in and switched it on without the RAM pack and it worked for a few seconds and then the output signal disappeared.

Swapping the vLA81 ULA replacement out and putting the original ULA at least got a picture, but without an end of frame so the TV couldn't get vertical hold.

I'm going to have to get the oscilloscope down now to try to work out what's happening, but I think the vLA81 must be toast. I burnt my finger on one component just "north" of the CPLD. I'm not sure if that's a MOSFET. It was extremely hot.

[NormanDunbar]

Capacitors fail after long periods of being unused.
That is my entire hardware knowledge!

Cheers,
Norm.

[stephen_usher]

There aren't really many electrolytics on the ZX81 board and they're reliable 1970s/early 80s ones anyway. Most of them are ceramic caps.

The voltages on the board are fine, about 4.85V on the +5V. The regulator's getting a bit warm as the PSU seems to be supplying 11V rather than 9V.

Given that it mostly works with the original ULA but not at all with the vLA81 suggests the latter has a major issue. The original ULA issue may be caused by the contacts on the socket being opened up by the pins on the vLA81 and hence the pins not making good contact.

[Chr$]

There aren't really many electrolytics on the ZX81 board and they're reliable 1970s/early 80s ones anyway. Most of them are ceramic caps.

The voltages on the board are fine, about 4.85V on the +5V. The regulator's getting a bit warm as the PSU seems to be supplying 11V rather than 9V.

Given that it mostly works with the original ULA but not at all with the vLA81 suggests the latter has a major issue. The original ULA issue may be caused by the contacts on the socket being opened up by the pins on the vLA81 and hence the pins not making good contact.

I've always found the VLA's 100% reliable and I know Charlie makes sure each one works before sending it out. Has the VLA seen many hours of use? I hope it is something trivial like the socket causing grief! I assume the original ULA is a 184 type without the back porch (is that another term for end of frame?), hence the pic issue?

The ZX81 is 40, amazing. I actually have one that is 40 in a couple of weeks - at least the latest chip on it is. I think it's the earliest known ZX81 component-wise. 8102 ULA, 8106 Mod box, 1980 dated Z80 and RAM. And it's in amazing condition too... nothing ever been on the exp port for example. And it came with a ZX80 PSU. Quite proud of that one!

[Pr0f]

It may be the ceramic resonator that's failed after all these years, and without a clock signal - some components like the CPU don't like to operate in a static condition - only the CMOS Z80's - the one in the ZX81 is an NMOS version.

Can you pin point the component getting hot?

[bwinkel67]

I brought it down from the loft to give it a check-out before the 40th anniversary in March, plugged it in and switched it on without the RAM pack and it worked for a few seconds and then the output signal disappeared.

Swapping the vLA81 ULA replacement out and putting the original ULA at least got a picture, but without an end of frame so the TV couldn't get vertical hold.

I'm going to have to get the oscilloscope down now to try to work out what's happening, but I think the vLA81 must be toast. I burnt my finger on one component just "north" of the CPLD. I'm not sure if that's a MOSFET. It was extremely hot.

In the meantime, you can get your ZX81 game fix with the http://ZXSimulator.orgfree.com/ it never fails...well, unless you find one of my many hidden bugs (not hidden on purpose)

Of course, if you need the real-hardware fix, I think running it on a BBQL will be the next closest thing. It doesn't need extra speed or memory.

viewtopic.php?f=3&t=3193&start=130#p36955

[Cristian]

The voltages on the board are fine, about 4.85V on the +5V.

This sounds wrong. The 7805 regulator output must be 5V. I suggest a double check: the PSU or the 7805 may be faulty.

[Pr0f]

Assuming of course the multi-meter is good

4.85 should be enough, but does seem a little low - I've seen 4.9-5.05 volts out of the various 7805's I've used in the past. The lower voltage may be an indication of a load on the supply - particularly if things are getting hot.

[stephen_usher]

I've always found the VLA's 100% reliable and I know Charlie makes sure each one works before sending it out. Has the VLA seen many hours of use? I hope it is something trivial like the socket causing grief! I assume the original ULA is a 184 type without the back porch (is that another term for end of frame?), hence the pic issue?

Nope. It's producing the end of line pulse but there's no gap at the end of the frame. This also means that essentially the CPU isn't actually running programs, merely running the screen as it's continuously outputting scan lines.

I've got some new sockets on their way and should arrive tomorrow. They're turned pin so the vLA81 won't damage them, unlike the original.

[Chr$]

I've always found the VLA's 100% reliable and I know Charlie makes sure each one works before sending it out. Has the VLA seen many hours of use? I hope it is something trivial like the socket causing grief! I assume the original ULA is a 184 type without the back porch (is that another term for end of frame?), hence the pic issue?

Nope. It's producing the end of line pulse but there's no gap at the end of the frame. This also means that essentially the CPU isn't actually running programs, merely running the screen as it's continuously outputting scan lines.

I've got some new sockets on their way and should arrive tomorrow. They're turned pin so the vLA81 won't damage them, unlike the original.

I have a load of ZX81 spares if you need anything after changes the socket(s).