Poof goes the 7805...

[Jim McClanahan]

Actually it wasn't really that dramatic.

I hadn't used the PAL-1 in a while until tinkering with the keypad routine the last few days. One thing I noticed was that if I just left it running, after 10 or 15 minutes the LEDs would blank out. It was doing it consistently enough to bother me so I reseated all the chips.

I wrote a simple BASIC program with two nested loops that counted from 1 to 1000 and display the two numbers and their product. It got up to 17 in the outer loop before it gave me some garbage, an OK, and then some more garbage and was locked up.

I reset it and things seemed to work fine again.

Intermittent failures are the worst. :-(

I'm powering it from a regulated and metered supply. This afternoon when I turned it on, it went to 2.5 A and the voltage dropped down to around 2 volts--the over-current protection had kicked in. I powered it off and removed the 2nd RIOT board (for no particular reason) and things powered on fine.

I typed in the same BASIC program and it didn't even make it all the way through the inner loop once before it locked up and showed drawing 2.5 amps again.

There are a couple of things I tend to be suspicious of.

One is the ribbon cable jumper from the board to the expansion bus. (If those aren't crimped with a great deal of precision, they'll short. And I've also seen stray "strands" from the stranded conductor get free somehow before and lead to issues.) I pulled that and it worked for a few seconds before hitting overcurrent again.

At this point I'm suspecting the 1st RIOT (because some people here have had troubles in the past), the EPROM (because I've had 27C64s short on me before), or the 7805 regulator (because--well because it's the regulator and it runs hot).

I pulled the RIOT and EPROM and had the same problem. Now it was pretty much a hard failure.

I measured the output of the 7805 and saw basically no voltage. I measured the input to the 7802 and saw around 0.6 volts--like a junction that was shorting it to ground. (And the cables I was using added enough resistance to give me the higher voltage reading at the supply itself.)

So I cut the 7805 out, used some micro clip leads to connect to the power supply, cranked the supply down to 4.9 volts, and powered things up.

So far it is working fine. I have the same bit of BASIC running and it has made it through the inner loop over twenty times now.

I think I'll probably just jumper things so the board can be powered by 5 volts. I'd rather have the heat in the power supply than on the board and I've had enough regulators fail that having a crowbar circuit is reassuring.

Thanks,

Jim W4JBM

[Ronny Ribeiro]

Hi Jim,

Thank you for your report.

Before I even received my PAL-1 I was planning to use it with my own home made add-ons, as I am doing now.

So I didn't even bother to install the 7805 (some readers of this blog were already complaining about overheating issues) and instead installed a tiny, pin compatible 3A capable DC-DC converter.

I have the system motherboard fully populated running with some other home made electronics attached to the second RIOT and everything is very stable.

It seems ditching the regulator entirely for an external, overcurrent protected regulated powersupply may not be a bad idea at all.

Best Regards,

Ronny.

--
You received this message because you are subscribed to the Google Groups "PAL 6502 computer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pal6502+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pal6502/0cf6f46b-2346-44da-ba99-462be5ccc05bn%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[GN Liu]

Thanks Jim for sharing the new troubleshooting process!