jumper settings for 27c512

By: JEDI-77 - 2008-01-11 13:44:25

Hi there, I have a Willem PCB45c eprom burner, and am currently spending a lot of time on this forum, reading and learning about it. I intend to write bin files onto either 27c256 or 27c512 EPROMs for my Nissan ECU. I was hoping someone could check my jumper settings in the pics below. I have highlighted the jumper locations with red boxes.

Essentially, I am trying to achieve the following:

VCC = 6.2v (removed jumper 8)
VPP = 12.75v

This is the jumper info that came with the burner...


These are my jumper settings..... are they correct? (please ignore the position of the dip switches as they were not positioned correctly for the 27c256 / 512 eproms when I took the photo).


In my reading, I learnt that to blank check correctly, you should set the VCC to 3.6v... I think the PCB45c can do this, and I think this function is associated with the area I have highlighted in the green box... Is it as simple as just removing the jumper where it reads "open=3v6"??

Also, I need to buy a plug pack to use this burner as I dont want to use the USB power supply.... is a 9VDC, 300-500mA plug back suitable??

Finally, can you get UV erasable 27c256 or 512 eproms?? and if so, are they named differently? If I read correctly, the willem burner will not erase these chips so I need to use a UV eraser?

Many thanks for your help...

Mark

Re: jumper settings for 27c512

By: DL - 2008-01-11 14:52:22

Hi Mark,
Quote:
These are my jumper settings..... are they correct? (please ignore the position of the dip switches as they were not positioned correctly for the 27c256 / 512 eproms when I took the photo).
Settings look correct.   You can verify them by entering the Test H/W tab and (with no device in the programming sockets), click on the VCC and VPP pins and measure the voltage on the corresponding pins on the ZIF socket.
Quote:
In my reading, I learnt that to blank check correctly, you should set the VCC to 3.6v... I think the PCB45c can do this, and I think this function is associated with the area I have highlighted in the green box... Is it as simple as just removing the jumper where it reads "open=3v6"??
I think 3.6V is too low to try to blank verify a 5V EPROM.   I've had some that wouldn't blank verify below 4.4V no matter how long they were erased, but this wouldn't prevent them from functioning properly at 4.5-5.5V.   It might indicate that the device is nearing the end of its life due to charge trapping in the gate oxide.   I don't have any information on how Sivava implemented their 3.6V option so don't know if they simply inserted a couple diodes in VCC (which would make it relative to what VCC was set to), or whether they took care of lowering VCC to the logic so the device being read doesn't have it's inputs overdriven by 2 diode drops, or whether they took care of low voltage interface to the programmer/computer.   Take a look at the Willem 4.1 schematic to get an idea of how to do this right.   All the 4.1 needs are a couple pull-up resistors on the inputs of the two inverters that drive the parallel port.
Quote:
Also, I need to buy a plug pack to use this burner as I dont want to use the USB power supply.... is a 9VDC, 300-500mA plug back suitable??
You need something closer to 12VDC at 500mA+.   The 5V regulator will typically have a drop out voltage of 3V and you have about a volt of drop in the rectifier.   So you need at least 10.2V at the input.   The VPP regulator can draw around 300mA on the higher VPP settings, so you need a good quality 500mA adapter to be safe.
Quote:
Finally, can you get UV erasable 27c256 or 512 eproms?? and if so, are they named differently?
EPROMs with quartz windows are UV erase.   That information can be obtained from the complete part number.   M27C512-12F is UV erase, M27C512-12B isn't.
Quote:
If I read correctly, the willem burner will not erase these chips so I need to use a UV eraser?
UV EPROMs need to be erased with UV light.   There are some FLASH being market as EPROMs that can be erased by the programmer (27SF512; 27SF256 are end-of-life and may be difficult to find).

HTH,
Dennis

Re: jumper settings for 27c512

By: JEDI-77 - 2008-01-11 22:44:43

Quote:
Quote:
In my reading, I learnt that to blank check correctly, you should set the VCC to 3.6v... I think the PCB45c can do this, and I think this function is associated with the area I have highlighted in the green box... Is it as simple as just removing the jumper where it reads "open=3v6"??

I think 3.6V is too low to try to blank verify a 5V EPROM.   I've had some that wouldn't blank verify below 4.4V no matter how long they were erased, but this wouldn't prevent them from functioning properly at 4.5-5.5V.   It might indicate that the device is nearing the end of its life due to charge trapping in the gate oxide.   I don't have any information on how Sivava implemented their 3.6V option so don't know if they simply inserted a couple diodes in VCC (which would make it relative to what VCC was set to), or whether they took care of lowering VCC to the logic so the device being read doesn't have it's inputs overdriven by 2 diode drops, or whether they took care of low voltage interface to the programmer/computer.   Take a look at the Willem 4.1 schematic to get an idea of how to do this right.   All the 4.1 needs are a couple pull-up resistors on the inputs of the two inverters that drive the parallel port.
Quote:


Hi Dennis thanks for your help..... The part about blank verifying still confuses me a little... in my early post I was reffering to something I read posted by EZo about removing the jumper at open=3v6, and shorting out the nearby diode to lower the VCC. Do you have any comments on this, and in the pic I have attached, which is the nearest diode?..

Also, what is the correct order to do things? I mean obviously I would blank verify before trying to programme.... but in regards to removing jumpers and changing settings, logic tells me that its not wise to play with jumpers while the unit is powered.... So I would disconnect power from the burner when I am changing settings? Sorry for the silly questions, just want to be sure...

Thanks

Mark

Re: jumper settings for 27c512

By: DL - 2008-01-12 01:57:45

Hi Mark,
Quote:
The part about blank verifying still confuses me a little... in my early post I was reffering to something I read posted by EZo about removing the jumper at open=3v6, and shorting out the nearby diode to lower the VCC. Do you have any comments on this, and in the pic I have attached, which is the nearest diode?..
You never mentioned anything about shorting one of the diodes in your previous posts (unless I'm really out of it).

Based on the proximity to pin32 of the ZIF and the two diodes next to the header, it appears that Sivava is simply inserting 2 diodes to reduce VCC to 3.6V.   The problem with doing this is that the device will be operating at 3.6V, while the programmer will still be operating at 5V.   This will cause the programmer to overdrive the device you're trying to blank check by 2 diode drops and that's enough to forward biase the clamp diodes on it's inputs.   I wouldn't do this.   You can either destroy the clamp diode or cause electromigration in the wires.   It's not a good design practice.

Doing this, however, will avoid the other issue I alluded to.   That is, underdriving the parallel port.   You'll still be under driving the inputs to the CD4014/21.   This is more benign, but still sloppy design.   If you put pull-up resistors on all of the shift register inputs, it would be acceptable.

The short answer to your question is that it will probably work and it doesn't matter which diode you short.
Quote:
Also, what is the correct order to do things? I mean obviously I would blank verify before trying to programme.... but in regards to removing jumpers and changing settings, logic tells me that its not wise to play with jumpers while the unit is powered.... So I would disconnect power from the burner when I am changing settings? Sorry for the silly questions, just want to be sure...
You can change the jumpers/headers with the programmer powered as long as the programmer is in synch with the program (i.e. VCC and VPP disabled).   Some switch/jumper settings can create shorts.   If you close switches 6 and 7 on the DIP switch and VPP is enabled, you'll probably damage one of the CD4503 and an inverter and maybe the transistor that switches VPP.

Just make sure you don't insert/remove devices from any socket that's powered.

I sometimes tell people to do tests with a chip in the programming sockets.   If you do this, you need to sequence supplies in the correct order.   VCC always on before VPP and VPP always off before VCC.

HTH,
Dennis