DigElec model 824 Eprom Programmer service manual and help sought.

[Paceguy]

I picked up a DigElec 824 portable eprom programmer yesterday. It powers on but the display doesn't light up. I popped it open and noticed that there are a couple of electro caps that have bulged. I would like to get the power supply up to spec before going any further. Would anyone in this group happen to have any service info showing rail voltages and test points? It would be very much appreciated if someone could help me out.

[m k]

Microcontroller.net has few photos, but maybe you want few of your own.

How much is a battery voltage?

TTL chips have voltage pins, 1488 drivers also.

What types are TO-220 etc. components around power out connector?

Among capacitors are coils, all are possibly for different voltages.

[Paceguy]

Good news! I managed to get it up and running. Appears to have been a bad connection. The display is nice and bright, but it sometime dims a bit. I'll have to find out what causes this dimmong.

[Paceguy]

The 824 has a built in UV lamp and slide-out drawer for erasing eproms. Mine is missing the entire lamp assembly but has the slide out drawer. Does anyone know how I could add a UV lamp? It would have to be low profile and fit inside the 824. Approx. 6 inches long and 2.5 inches wide.

[m k]

Check UV-LEDs, correct wave length is available.
They are fast, but I've only tried close contact distances.

I read that the programmer has a removable software pack, is it mandatory or an add-on?

[Paceguy]

The plugin is mandatory for operation.

[Paceguy]

Would it be possible to mount 20-30 or so of the UV LED's onto a piece of perf-board the length and width of the slide in tray (drawer)and mount it about over the tray (drawer) by a couple of mm's? Do you know what wave length I would need ?

[madao]

only user Manual

but better than nothing :-)

[m k]

I can't see any reasons why discrete LEDs couldn't do it.
But can't think you need that much of them either.

wavelength is stated to be as 253.7 nm, so 260 nm is just fine.
Search UVC disinfection LEDs.
My LEDs are like those with six way circuit boards and double +/- pads.

[Paceguy]

How many of these 260nm LED's would you suggest that I use in the drawer? Would one or two be enough?

[m k]

No idea, you need to do few tests.

With one LED and one chip, do distance and angle tests.
First straight above and then a bit aside, offsite/angled distance power from the straight above power is most likely not linear.

The angle from where the chip is still erased in comfortable time is safe, with two offsite/angled LEDs side by side and a chip in between you have around a double erasing power.
Now you have a density value for LED mat above freely positioned chips, if chips are forced to ordered position you can of course do with fewer LEDs.

[Paceguy]

Thank You m k, now to figure out which wattage of uv LED's are required.

[helius]

To erase EPROMs you need adequate intensity (µW power per square cm) and even illumination. You should use a fluorescent bulb, they are cheap and provide enough power.
The type used in this sort of eraser is the G4T5, which costs around $5 in single quantities. It can provide up to 10,000 µW/cm² of UVC, enough to erase an EPROM in 5 to 10 minutes.

There isn't a single commercial EPROM eraser that uses LEDs: check the prices of 260 nm LEDs and you may understand why. Their radiation pattern is also all wrong: you need to have diffuse illumination to light up the drawer, this isn't something that can be safely aimed at a chip with your Mk.I eyeballs. Unlike visible light devices, it is not practical to place lenses on the light source to correct the beam pattern with UV.

[m k]

One manual has
The UV wavelength of 253.7nm at initial exposure levels of 4000uW/sq cm ensures complete erasure
in typically 15 minutes or less but the light output reduces with use causing the time to increase.

Unfortunately few light powers are stated as is needed here.
But if 1W LED is 50% UV light you can guestimate that 120 degrees cone can be pretty big and still function.
500/4 sqcm is around 12 cm wide circle and so LED is more than 6 cm above.
Unknown is how tilted far side memory cells under far side chip window are and how much light they get.

Maybe a sledge for chips is more practical.
Back in the day with tube light the ledge was V-shaped and a small tube was only few cm away.