| Author |
 Topic  |
|
|
e5frog
Sweden
5 Posts |
 Posted - 11/16/2016 : 14:26:45
|
I was trying to read and program with an adapter for the 27C160 on the dual powered Willem 5.0C, didn't work very well and I have been trying to find the problem.
I have tried two different 12V PSU:s, didn't see any problems when measuring voltage. Even installed an earlier version 0.97ja on another computer and ran that. Tried swapping the parallel-cable, no difference.
Problem is programming works fine until $8000 - did so on the 27C160 - when working the best and the same with a 28F020. When programming I get an error that the chip isn't the same as data. In the last case it reported $44 for chip and $55 for data.
Erasing the 28F020 is flakey, needs to be done more than once to give a completely empty chip.
When reading back already programmed 27C160 I get errors (plugging chip in and verifying with the file that is programmed) I noticed areas $400-$7FF and $C00-FFF etc where messed up with other contents that it was supposed to.
Have resoldered the adapter but as the 28F020 gives an error it seems the programmer itself is defective in some way.
Any tips?
EDIT:
Measured the ESR of all the electrolytic caps, all OK, swapped the 7805 just to see if there's a difference. Sadly no difference.
I usually use a Windows 98 computer that I basically only use for eprom programming, when it didn't work there I moved to the Windows XP computer that has also been successfully working.
What next - replace all IC:s?
EDIT again:
Seems I got parts from $1410 at position $9010 - it's not consistent seems an address pin is floating or something. |
Edited by - e5frog on 11/16/2016 16:22:48
|
|
| Reply #1
e5frog
Sweden
5 Posts |
Posted - 11/17/2016 : 15:40:38
|
I started desoldering some chips, the CD4069 didn't test OK - it tested bad, CD4503 and a CD4015 tested OK... I'll probably desolder and check all IC:s before assembling it again, ordered a full set anyway except the one by the voltage selection.
Haven't decided if I should put sockets there or not, it makes swapping easier but the sockets themselves can cause problems with poor connection after a few years...
|
 |
|
| Reply #2
e5frog
Sweden
5 Posts |
Posted - 11/18/2016 : 06:06:24
|
Found a tube of CD4069 - seems my tester disapproves of all these - even though they are NOS. Something wrong with the software apparently - they all tested fine when tested as the pin compatible 74*04.
I did got a bad test on a 4015 as well so in any case one chip was defective.
Have placed sockets for the eight IC:s - awaiting delivery of new chips.
Not sure if I should swap the MC34063A as well, have tested the electrolytic capacitors, they all seem fine.
While at it I also bought a proper TEXTOOL socket (232-1287-00-0602J) instead of the knockoff "TEXTDOL" - I noticed the contact surfaces were worn.
If it doesn't work after that there's more parts to be changed... |
|
|
| Reply #3
e5frog
Sweden
5 Posts |
Posted - 11/22/2016 : 11:15:12
|
No luck with this.
All logic IC:s test OK, reading a working 28F020 (erased and programmed with no problems in the TOP2004 programmer) in the soldered PLCC32 socket or using adapter in the ZIF-socket doesn't matter, it's the same faulty result with a repeating pattern as if there's some problem with the addresses. It even starts with the first byte wrong and then there are some proper ones before the loop starts.
Kind of out of ideas...
I thoroughly checked the DIP-switch as well, no problem there.
Parallel cable?
I'll try the other computer and different cable. |
Edited by - e5frog on 11/22/2016 11:19:39 |
|
|
| |
Topic |
|
Problem, programming fails at $8000
Forum Locked
