| Author |
 Topic  |
|
|
Wrathchild
United Kingdom
8 Posts |
 Posted - 04/06/2005 : 15:50:25
|
Hi,
I received the Enhanced Willem Board after buying it from Ebay
and have just had the chance to get familiar with it.
Using the 0.97g software (under W2K) and setting the board
to PCB3, I've managed to burn a AM27C128 rom successfully,
but have not had any joy with an AMD 29F010-120JI.
I've set the DIPs correctly, device LED set to W27C and
selected the following in the software:
Device = 29F010 and then unchecked (5555,2AAA...)
The chip ID is reported corrcetly, 0x01=AMD, 0x20=Am29F010.
After running the erase, and re-reading the chip, the buffer
reads repeats of 0x10,0x20,0x00,0x00. Because of this, attempting
to write a ROM image fails on the first byte. Can anyone
help with finding the right settings?
In the 0.97ja, when you select 29F010, there is no option
for the (5555,2AAA...) - so should this version be ignored
for use on the Enhanced Willem board?
Many thanks,
Mark
|
|
| Reply #1
Wrathchild
United Kingdom
8 Posts |
Posted - 04/07/2005 : 04:24:48
|
Forgot to mention, this is a PLCC32 pin device and so
I'm using the socket mid-right of the main Zif socket. |
 |
|
| Reply #2
Wrathchild
United Kingdom
8 Posts |
Posted - 04/08/2005 : 14:33:40
|
Another followup - using an AM29F040-120JI,
same thing happens - except the repeated 4 byte pattern
is 0x01,0xA4,0x01,0x01 - 0xA4 being the ID for AM29F040.
Found this though, setup the Dips when in PCB3 mode, then
switch to Willem. For me, this make it easier to understand.
Select AT29 LED using the button (after removing the safety
jumper) and the chip can then be erased.
However, if you load an image and start programming then
it gives an error on the first byte :
Error at 0x000000, chip=0xFF, buffer=0x01.
But, if you set the LED back to 'normal' then programming
and verifyinh works?!
Still - this trick doesn't seem to work for the AM29F010,
so I'll keep trying. |
|
|
| Reply #3
usb
138 Posts |
Posted - 04/12/2005 : 13:16:24
|
All AT29FXXX use same command set to program/erase chip.
So, if it is working for 29f040 then it should work for 29f010.
The only difference is the address line setting is different. So, you may need test your address line if OK. |
Edited by - usb on 04/12/2005 13:16:46 |
|
|
| Reply #4
Wrathchild
United Kingdom
8 Posts |
Posted - 04/13/2005 : 07:45:17
|
OK, pins tests and the following was seen:
All data and the majority or address lines 4.9V
A9 = 3.4V
A11 = 2.5V
A13 = 3.9V
A15 = 3.4V
I don't have the expertise myself to track down where
things are going wrong but do have people who can
look for me - any hints on where to start.
Also, I'm a little puzzled as to how the 27C128 worked
as it would be using those address lines affected?
My initial confusion over the settings is that the
'grid' of supported processors lists Am29F010 in a
separate section (Atmel flash memory) to 29F010, which
is in 'Flash Memory'. This grid bears close relation
to the list available when selecting the chip in the
software, but in this case its not clear - perhaps the
suppliers of these boards should provide a list of
supported chips against the configuration to use, both
in software and on the board? I've not found a site/doc
with that sort of info, does anyone know of any?
Thanks,
Mark
|
|
|
| Reply #5
usb
138 Posts |
Posted - 04/19/2005 : 14:35:47
|
| Your chip may be "Hardware Protected". You need unprotect the chip if this is the case. Lots of BIOS chips are protected from mothermoard factory. |
Edited by - usb on 04/19/2005 14:36:16 |
|
|
| Reply #6
Wrathchild
United Kingdom
8 Posts |
Posted - 04/20/2005 : 04:59:45
|
The 29F010's have this? Can they be unlocked through the
Willem programmer using some other software?
This should be possible using the the DriverLINX Port I/O Driver,
so do actual or example sources/programs exist for driving the
Willem Programmer (and compatible with the Enhanced board).
Thanks,
Mark |
|
|
| Reply #7
usb
138 Posts |
Posted - 04/24/2005 : 14:50:59
|
quote:
Originally posted by Wrathchild
The 29F010's have this? Can they be unlocked through the
Willem programmer using some other software?
This should be possible using the the DriverLINX Port I/O Driver,
so do actual or example sources/programs exist for driving the
Willem Programmer (and compatible with the Enhanced board).
Thanks,
Mark
The Willem programmer is unable to unprotect the AM29F010 chip currently. That because the programmer has limited I/O lines. To unlock the AM29F0X0, ther are extra Vpp control lines are needed.The latest software said it is able to unprotect 29F040, but not 29F010 chip. Also, it need modify the programmer circuit.
|
|
|
| Reply #8
Wrathchild
United Kingdom
8 Posts |
Posted - 04/25/2005 : 03:26:08
|
Latest update:
Appears the chips are OK, I'm using a FlashCart for the Atari 8-bit micros (www.atarimax.com) and this erases/writes to the chips OK.
After checking the circuit board out with a friend of mine, we
unsoldered (but couldn't remove) the ZIF socket, then soldered it
back and after this the ID and erase now work when AT29 is selected.
The only problem now is that the programming fails at
address $8001. This I can see is down to the data intended
for that page being overwritten from $0000. E.g. both pages
begin with the value $F6 and so this remains the same,
however address $0001 should be $61 and $8001 should be $57.
When read back, the chip shows $41 in both addresses, which
makes sense. As the address line involved here is A15, it
indicates a problem with that. On the 4015, the output is
OK at 4.9V but the combination of the DIP switches and the
diode drop this down to the 3.45 seen. This could indicate a
problem with the diode but this checks out OK on the meter
or could be some kind of short.
Any other ideas - I'm pretty close now :)
Regards,
Mark
|
|
|
| Reply #9
usb
138 Posts |
Posted - 04/25/2005 : 12:06:42
|
Yes. EZoSat is correct.
Do not select AT29XX setting and write the data to 29F010. They are the different size of flash. The result will be overlaped address line.
Use 29F010 setting.
|
|
|
| Reply #10
Wrathchild
United Kingdom
8 Posts |
|
| Reply #11
usb
138 Posts |
Posted - 04/29/2005 : 11:54:14
|
When you programming the 29F010, the device type LED on Enhanced willem should stay on Normal position. Which is the first LED of the device type.
Other wise your address lines will be diverted to 29C256 address line. The device type AT29 position is only for AT29C256 type only. |
|
|
| Reply #12
Wrathchild
United Kingdom
8 Posts |
Posted - 02/15/2006 : 16:36:04
|
(also posted on the Willem forum)
I've just received a Tomsad 4.9a board and comparing it to the Enhanced Willem I think the problem maybe down to the lack of the jumper set on the enhanced which is positioned just to the right of the zif socket? The willem application (EpromM51) highlights which jumpers to use so I guess without this on the enhanced board it does not correctly access or power the chip? On the Tomsad board by not using the jumpers resulted in the chip not being able to be identified but with the jumpers in place the Am29f010 and Am29f040 chips Id'd correctly. Can anyone confirm if the enhanced board is therefore flawed in this respect or if there is something else one should do in order to get the board/chips to behave correctly?
Thanks,
Mark |
|
|
| Reply #13
Lightguy
USA
4 Posts |
Posted - 11/26/2007 : 21:24:04
|
| I have ran into the same problem your having. I have ended up using my Logical Devices e-prom burner to erase and burn these chips. I hope someone comes up with a solution for the Willem. I'm using the dual power version. |
|
|
| |
Topic |
|
Can't erase Am29F010
Forum Locked
