| Author |
 Topic  |
|
|
ScienceofSpock
USA
18 Posts |
 Posted - 04/06/2015 : 14:26:38
|
Howdy y'all,
I purchased a Willem PCB5.0E on ebay a while ago, and have used it successfully to burn cartridge and kernel ROMS for some of my older computers. I now have a need to burn a custom Kickstart ROM for an Amiga, so I did a bit of research, purchased a ADP-054 and a handful of AM27C400 EPROMS.
I trimmed the extra 4 pins on the ADP-054 and set my dip switches according to the diagram for 27C400 in the software (011011010101) and J1 on the ADP-054 is set to 1-2, and I believe the adapter is working (at least in some form) as I can read the original Kickstart ROM with it, and blank tests for the new AM27C400s show that they are indeed blank.
The problem I have is that I cannot seem to burn them properly. I have tried twice, and both times it fails at just over 50% with the following error message: "Error at 0x045c01 Chip = 0x00 buffer = 0x06"
I am not sure what I am doing wrong, as I cannot seem to find any instructions aside from a few diagrams on this site here: http://www.mcumall.com/forum/topic.asp?TOPIC_ID=4729
Are there any tutorials for burning a 27C400 with this setup? What am I missing here?
Thanks,
SoS |
|
| Reply #1
ScienceofSpock
USA
18 Posts |
Posted - 04/06/2015 : 20:54:05
|
I did some more research, and it looks like it could have been a power issue. I was powering it using USB. I switched to a 12v 5A power supply, and it got all the way to 100%, but still failed. I dumped the data from the chip and compared it with the actual ROM file, and it's got bad data all over the place. It's like 99% correct, but every so often, there is a bad bit. There is some strangeness to this as well.
The error reported was this:
"Error at 0x000389 Chip = 0x19 buffer = 0x11"
Except that at 0x000389 taken from the chip dump, the value is 0x11, not 0x19, and this matches the ROM file. The actual first error is located at 0x00016F and the value is 34 in the chip dump, and 24 in the ROM file.
The other thing I have noticed is that all the errors have one digit correct. The first error I just mentioned is a good example. It's a 34 where it should be a 24. The next 4 errors in order look like this:
Loc: ROM value CHIP Value
0x0001A1 50 D0
0x00024C 48 68
0x0002CC 48 4A
0x00032C 01 B1
Every mismatch in the 2 files has 1 correct digit. I may be reading too much into that, but I don't know, it seems too coincidental.
I have since burned another chip using the real power supply and it too went all the way to 100% and still failed with an incorrect error location. The 2 chips do not have the same errors, however, the "1 correct digit" observation stands with both chips.
For reference, here are my settings:
Device select: 27C400
tWP set at 140.0 uS (Shows 240 when burning)
tWC set at 20.0 uS (Shows 40 when burning)
I'm not sure what is going wrong. Any ideas?
Thanks,
SoS |
 |
|
| Reply #2
ZLM
2937 Posts |
Posted - 04/07/2015 : 09:37:42
|
try:
1. increase the chip erase time. This is to avoid chips with marginal blank data.
2. check your VPP is correct. 12.7V
3. check your chip, adapter pin contacting. Especially on data pins.
|
|
|
| Reply #3
ScienceofSpock
USA
18 Posts |
Posted - 04/07/2015 : 11:21:13
|
quote:
Originally posted by ZLM
try:
1. increase the chip erase time. This is to avoid chips with marginal blank data.
These are UV erasable and all of them passed the blank test before burning. Does chip erase time apply to UV erasable EPROMS, and if so, where do I set it and what should I set it to?
quote:
Originally posted by ZLM
2. check your VPP is correct. 12.7V
Where should I check that?
quote:
Originally posted by ZLM
3. check your chip, adapter pin contacting. Especially on data pins.
It's ZIF connectors all around. The chip sits in a ZIF on the adapter and the adapter sits in a ZIF on the burner. Everything LOOKS like it's connected fine and there isn't any visible corrosion.
Sorry if these sound like stupid questions, neither device came with any instructions or manuals, and I have found ZERO documentation for either of them online. I'm pretty much muddling through this blind.
I can post pics of my setup if that would help.
Thanks,
SoS |
|
|
| Reply #4
ZLM
2937 Posts |
Posted - 04/07/2015 : 21:45:41
|
Yes. erase time applys to the UV eraseable chip. Example set the erase time to 30 minutes.
Use multi-meter to check your VPP is correct. 12.7V
This PCB5.0 is old willem programmer. It has no CPU on programmer board. The programming staybility always depending on the computer performance.
|
|
|
| Reply #5
ScienceofSpock
USA
18 Posts |
Posted - 04/08/2015 : 20:50:53
|
quote:
Originally posted by ZLM
Yes. erase time applys to the UV eraseable chip. Example set the erase time to 30 minutes.
I assumed you meant this was a setting on the programmer (which is why I was confused). I do not have an eraser yet, but I have one on order.
quote:
Originally posted by ZLM
Use multi-meter to check your VPP is correct. 12.7V
Right, I get that, but *where* do I test that? Can I check it without having to burn a EPROM?
quote:
Originally posted by ZLM
This PCB5.0 is old willem programmer. It has no CPU on programmer board. The programming staybility always depending on the computer performance.
I had not considered this. My computer is a relatively powerful, very stable Win7 box, but it HAS been on for over a month. Perhaps a reboot is in order. |
|
|
| Reply #6
ScienceofSpock
USA
18 Posts |
Posted - 04/10/2015 : 18:42:21
|
So I tested using an old gateway laptop with a built-in parallel port, and I get the same issue.
Is there any documentation for this adapter anywhere? How do people get anything done with these things? |
|
|
| Reply #7
ScienceofSpock
USA
18 Posts |
Posted - 04/11/2015 : 15:29:05
|
My eraser arrived today. After running 3 chips through it for about 40 mins, all 3 chips programmed and verified properly! I guess I'll just have to start erasing every chip before attempting to program them. Still very strange. I have purchased and programmed both 28C256 and 28C512 chips without issue, and without having to erase them first.
Thanks for the help, y'all! |
|
|
| Reply #8
ScienceofSpock
USA
18 Posts |
Posted - 04/18/2015 : 19:05:00
|
So, I programmed the wrong ROM image to my 27C400, and had to burn another one. I'm getting the same issue again, only this time, I'm erasing the EPROMS for 50 mins before programming. All dip switches are set properly, adapter seated in the ZIF socket snuggly, but every single time, it stops at 49% with an error.
After exhaustive searching, I finally figured out how to test VPP, and now, I'm only getting 11.58v. I get this regardless of whether I'm using USB or a wall wart. I don't know what it was measuring before because I didn't know how to test it, but I was getting something within tolerance because I burned 3 working ROM chips (the ROM file was the wrong version, but the chips still functioned in my Amiga)
How the heck do I fix the VPP? Is it going to mysteriously drift again?
Thanks,
SoS |
|
|
| Reply #9
ZLM
2937 Posts |
|
| Reply #10
ScienceofSpock
USA
18 Posts |
Posted - 04/20/2015 : 17:59:40
|
I'm getting 4.3v at A16.
Edit: I also noticed that some of the AM27C400 chips I have, the twP and twC changes from 140uS and 20uS respectively to like 340uS and 60uS when I start the programming process. Is this indicative of something I should be aware of or is this a normal thing? |
Edited by - ScienceofSpock on 04/20/2015 18:06:36 |
|
|
| Reply #11
ZLM
2937 Posts |
Posted - 04/25/2015 : 23:02:56
|
You A16 is fine. Check A17 too, see if it has good voltage while reading.
While reading, the A17 of chip should change the voltage from 0 to ~5V at 50%. |
|
|
| Reply #12
ScienceofSpock
USA
18 Posts |
Posted - 04/26/2015 : 12:33:52
|
quote:
Originally posted by ZLM
Check A17 too, see if it has good voltage while reading.
While reading, the A17 of chip should change the voltage from 0 to ~5V at 50%.
I manually tested A17 using the Test H/W tab, and it goes to 4.3V when turned on.
During a read, however, A17 changes voltage several times:
% ---- V
------------
0% --- 0.0V
15% -- 0.25V
25% -- 2.85V
50% -- 0.25V
60% -- 0.50V
80% -- 2.85V
90% -- 3.0V
Percentage numbers are close but not exact. The readings are the same whether measured from the ZIF42 on the adapter or the ZIF32 on the programmer.
For the ZIF32 I was testing Pins 16(G) and 30(A17), and on the ZIF42 I was testing pins 31(G) and 33(A17).
|
|
|
| Reply #13
ScienceofSpock
USA
18 Posts |
Posted - 04/26/2015 : 12:52:19
|
| Here is something I don't get: If the AM27C400 CHIP has A17 on Pin 1, how is the programmer/scriptdapter supposed to program it properly, if A17 is on a completely different pin? It's on Pin 30 on the programmer, and Pin 33 on the ADP-054. What am I missing here? |
|
|
| Reply #14
ZLM
2937 Posts |
Posted - 04/26/2015 : 18:18:18
|
Your tested pin is not A17. The 27C400 has highest address line is A17. So, it only changes voltage to high when reading passes 50% point.
The programmer uses A18 mapped to A17 on chip. That because the programer used A0 to toggle the hi-low bytes.
see:
http://www.mcumall.com/download/WillemProg/16BitEPROM_sch.pdf |
|
|
| Reply #15
ScienceofSpock
USA
18 Posts |
Posted - 04/26/2015 : 19:52:55
|
quote:
Originally posted by ZLM
Your tested pin is not A17. The 27C400 has highest address line is A17. So, it only changes voltage to high when reading passes 50% point.
The programmer uses A18 mapped to A17 on chip. That because the programer used A0 to toggle the hi-low bytes.
see:
http://www.mcumall.com/download/WillemProg/16BitEPROM_sch.pdf
That's the exact PDF I was looking at that shows A17 at pin 33. This is what is so confusing. If that's not where A17 is, then why is it labeled A17 in the schematic?
Checking on Pin 13 of the socket (Pin 12 on the chip) and Pin 5 of the socket (Pin 1 of the chip) stays at about 0.5V for the entire read.
If I go to the Test H/W tab and turn on A17, pin 30 on the programmer socket goes to 4.3V, however, Pin 5 on the ADAPTER socket does not.
If these aren't the correct pins, Which are? how do I correctly discern which pins I should be testing when you say "Check pin A17"? Is there a chart somewhere? I know you're trying to help me, and I really appreciate it, but I am not an expert. I took 3 semesters of EE, but that was 25 years ago. If you could just go ahead and assume I know nothing (Which is basically true), I think it would help immensely.
Thanks,
SoS |
|
|
| Reply #16
ZLM
2937 Posts |
Posted - 04/26/2015 : 23:26:02
|
Check the A18 pin. The pin 31 of the programmer ZIF. It should be connected to the chip pin 1 (A17).
That pin should change from 0 to 5V on the 50% point when you reading the chip 27C400. |
|
|
| Reply #17
ScienceofSpock
USA
18 Posts |
Posted - 04/27/2015 : 17:20:42
|
quote:
Originally posted by ZLM
Check the A18 pin. The pin 31 of the programmer ZIF. It should be connected to the chip pin 1 (A17).
That pin should change from 0 to 5V on the 50% point when you reading the chip 27C400.
I have 0V at A18 on the programmer, on the Test tab and during chip reading. |
|
|
| Reply #18
ZLM
2937 Posts |
Posted - 04/27/2015 : 17:55:34
|
That is the problem.
Check your DIP switch setting. The last two bits switch should be OFF and ON. (DIP bit 11 = OFF, bit 12 = ON). |
|
|
| Reply #19
ScienceofSpock
USA
18 Posts |
Posted - 04/27/2015 : 19:23:03
|
quote:
Originally posted by ZLM
That is the problem.
Check your DIP switch setting. The last two bits switch should be OFF and ON. (DIP bit 11 = OFF, bit 12 = ON).
They are. Here's a pic of my programmer, with the dip switches set for the 27C400 //kgreene.com/misc/willem.jpg
|
|
|
| Reply #20
ZLM
2937 Posts |
Posted - 04/27/2015 : 20:06:52
|
Make sure your DIP switch bit 12 is working good.
Why your J3 on lower two pins? That set the A18 to the Pin 1 of ZIF. I think it should on the top two pins or removed. J3 connect the A18 to the pin 1.
Also, if you still can not get the A18 signal. Check CD4015 chip, U6 pin 11, that is the pin where the A18 comes from.
See: http://www.mcumall.com/forum/topic.asp?TOPIC_ID=2028 |
|
|
| Reply #21
ScienceofSpock
USA
18 Posts |
Posted - 04/27/2015 : 21:18:47
|
quote:
Originally posted by ZLM
Make sure your DIP switch bit 12 is working good.
Why your J3 on lower two pins? That set the A18 to the Pin 1 of ZIF. I think it should on the top two pins or removed. J3 connect the A18 to the pin 1.
Also, if you still can not get the A18 signal. Check CD4015 chip, U6 pin 11, that is the pin where the A18 comes from.
See: http://www.mcumall.com/forum/topic.asp?TOPIC_ID=2028
J3 is on the lower 2 pins because that's what the DIP diagram for the 27C400 shows. Is that not J3 in the diagram?
Edit: I tried programming a chip with J3 on the top 2 pins and it failed immediately. Same thing with J3 completely removed. |
Edited by - ScienceofSpock on 04/27/2015 21:44:28 |
|
|
| Reply #22
ZLM
2937 Posts |
Posted - 04/28/2015 : 16:48:31
|
| Check CD4015 chip, U6 pin 11, that is the pin where the A18 comes from |
|
|
| Reply #23
ScienceofSpock
USA
18 Posts |
Posted - 04/28/2015 : 17:29:10
|
quote:
Originally posted by ZLM
Check CD4015 chip, U6 pin 11, that is the pin where the A18 comes from
Are you certain that schematic applies to my programmer? It's labeled as PCB3B, mine is a PCB5.0E. U6 is an ON 14503B, not a CD4015. The datasheets show different pinouts and different logic for those chips, but I'm not knowledgeable enough to know 100%.
Thanks,
Keith |
|
|
| Reply #24
ZLM
2937 Posts |
Posted - 04/28/2015 : 19:28:52
|
There are three CD4015s on board. You need to find out which one is U6 of original schematic.
If you are unable to find it out, then you need a new programmer. |
|
|
| Reply #25
ScienceofSpock
USA
18 Posts |
Posted - 04/29/2015 : 17:25:30
|
quote:
Originally posted by ZLM
There are three CD4015s on board. You need to find out which one is U6 of original schematic.
If you are unable to find it out, thn you need a new programmer.
I could probably figure it out, if there was an accurate schematic for this board *somewhere*, but that seems not to be the case.
There ARE 3 HAF4015BT chips on the board, at U8, U9 and U10, but I have no idea which one controls A18.
I will tell you one thing, this is probably the LAST Willem product I buy. It died after programming only a handful of chips, and there is virtually no info or real support for it that I can find. Not even a real manual. Well, hindsight is 20/20.
If I can figure out which HAF4015BT is the one controlling A18, I will make an attempt to replace the chip, but my SMD de/soldering skills are only OK at best.
Anyone have any recommendations for an EPROM programmer for retro-computing stuff (27CXXX and 28CXXX mostly)?
Thanks,
Keith |
|
|
| Reply #26
ZLM
2937 Posts |
Posted - 04/29/2015 : 19:19:27
|
| manually tested A18 using the Test H/W tab, check all three CD4015 chips pin 11, you should be able to find out A18 pin. |
|
|
| Reply #27
ScienceofSpock
USA
18 Posts |
Posted - 04/29/2015 : 19:50:47
|
quote:
Originally posted by ZLM
manually tested A18 using the Test H/W tab, check all three CD4015 chips pin 11, you should be able to find out A18 pin.
Except if it's bad and providing no voltage like it seems to be, wouldn't all 3 of them read 0V (Assuming the other chips don't also have something to do with A18)? |
|
|
| Reply #28
ZLM
2937 Posts |
Posted - 04/30/2015 : 10:25:27
|
Those CD4015s provide all address signals: A0-A23.
See schematic for all pins for addresses. |
|
|
| |
Topic |
|
Burning AM27C400 with Willem PCB5.0E and ADP-054
