Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: simono71 on June 08, 2010, 07:24:43 PM
-
If this is in the wrong place, please move it accordingly.
I was hoping someone might be able to guide me in what I am doing wrong.
I have a Willem pcb 5.0 eprom programmer with the 16bit eprom board.
I have successfully read my 1.3 kickstart rom from one of my amiga 500's so it seems my programmer board is at least somewhat functional.
The problem arises when I am trying to burn this image to a 27C400 so I can replace the 1.2 kick rom in my other A500.
When I insert the 27C400 and do a blank check all is well. However as soon as I hit the program button it immediately returns saying error: at 0x00000, chip xFF, buffer x11. Obviously the chip is not writing properly.
I have tried using external power, and usb power all to no avail. Either I got sold bum eprom chips (I bought 4 and have the same result with all of them), or I have done something wrong. I have checked and double checked all the jumpers/DIPs and they all seem right to me, but I am a newbie at burning rom chips.
Hopefully someone can shed any light on what is or is not happening,
Simon
-
what 16 bits suppot are you using ? left one or right one ??
Mine work reading both .. but only burning on the left one ...(just under 27c160/322 inscription)
-
The left one (which is annoying because it is a pin hole type socket and is a lot harder to put a chip into than a standard dip type socket such as the right one.)
-
i know, i'm using the same with 12v psu and 0.97d11 software...
-
Have you tried UV erasing them? I had same issues (but I got test blank errors) but when I made a 10min UV clean... everything worked like a charm
-
I tested all of them for being blank (they were sold that way), and got no errors. I don't have a UV eraser right now unfortunately.
The chip fails immediately on attempting to write, doesn't even get past the first byte so I was wondering if the power wasn't enough or one of the jumpers was set incorrectly, but they all seem fine, and I am using the external power supply.
As mentioned above they all test fine as blank, and I don't have an eraser so not much else I can do there.
I would perhaps wonder if my programmer was bad, but it seems to read the rom correctly, so at least part of it is working. I don't have any other types of blank eproms so I have no way of doing a test write just to verify that the programmer works at all.
I guess its possible I got sold a bum batch of eproms :(
-
you can buy some cheap chinese uv eprom eraser on ebay, i'm using one since some months and working well ;)
-
Can you still get eproms for the 68000? Or is it a little harder today?
-
The 27C400 is a 512k device and 16 bit Kickstart 1.2 and 1.3 are only 256k. In order for that eprom to program properly and work you have to load the binary image file twice. Once at 0x00000-0x3FFFF and again to fill the other half of the eprom at 0x40000-0xFFFFF (It's been a long time but I think this is correct, anyone?). Use the buffer editor or viewer to make sure the data has been loaded correctly. You could also use a 27C200 for 256k ROM images as I recall. Check the BYTE order using the buffer editor as well and make sure you don't have to have BYTE SWAP enabled on your burner. You REALLY need a UV eraser as you will no doubt end up trying a few times before you get the process and your programmer figured out. Good Luck!
-
The rom image I have comes from an existing 1.3 rom I had. When you read it into the buffer with the device set to a 27c400 it reads it in twice anyway, so the image ends up being repeated automatically.
If you take the raw image from an existing rom you read, then there is no need to byte swap, as the bytes are also pre-swapped.
I intend to buy a rom eraser, but right now if I can't figure out how to get a blank rom to burn I will probably save my money.
It is possible my board is bad too. Saw someone on a Willem forum with the exact same issue as mine but with a different rom chip, but unfortunately there was no resolution.
On the forum that I mentioned the general consensus seemed to be the guy did not have enough voltage to the Vpp pin. However I have checked mine with a multimeter and it reads 12.7v. The specs calls for it to be 12.5v +/- .25v so I should be within spec, and I am using a 12v external power supply, not USB, which appears to be a common problem. I know people are burning these EPROMS using my setup, because I checked on forums before I bought to make sure I was getting the right thing.
Unless it is something really obvious, I guess its either a bad board or a bad set of chips :(
-
Not that anyone is still interested by this point, but going through the complete self test on my willem is making me think either my cable or board is bad.
All the address lines are outputting correctly when using the toggle switches, and so is Vcc and Vpp. However I can't get any reading on the data output pins. I assuming that other people when using the self-test tab and enabling data output lines get a close to 5v reading. I am getting 0v across the board on all data out lines and the oe enable line which I am sure is my problem. Guess now I need to do a continuity check on my parallel cable, or assume my programmer is busted.
-
Ok, hopefully this helps someone out but it seems to be working now. I did break the cardinal rule by changing two things at once, so I am not exactly sure which of the two was the magic ticket, but I have my suspicions...
Basically I did some more research on the Willem board, and found people setting their bios settings to EPP (mine was set to ECP). So basically I swapped out the parallel cable for a different and shorter one, and also changed the parallel port setting to EPP. One or both of those things enabled me to burn my eprom and have it validate correctly. Haven't tried putting it back in the A500 yet, but it is a big step in the right direction, and hopefully that helps someone else too.
-
Ok, hopefully this helps someone out but it seems to be working now. I did break the cardinal rule by changing two things at once, so I am not exactly sure which of the two was the magic ticket, but I have my suspicions...
Basically I did some more research on the Willem board, and found people setting their bios settings to EPP (mine was set to ECP). So basically I swapped out the parallel cable for a different and shorter one, and also changed the parallel port setting to EPP. One or both of those things enabled me to burn my eprom and have it validate correctly. Haven't tried putting it back in the A500 yet, but it is a big step in the right direction, and hopefully that helps someone else too.
I would say the EPP/ECP setting did the trick. ECP has issues leaving the strobe line in undetermined state or in some locked state so if you want to control it through software it causes problems. And worse of all, that's the default setting on most parallel ports.
I had other issues with programming SRAMs/EPROMs-- the power management screwing things up and i/o ports not being accessible due to some user access control protection crap in recent OSes.