4.5 unable to program 27C256

By: natuner - 2007-06-30 12:15:58

Hey guys,

I am very new to the willem programmer and only used it till now to read 27C256 chips that were used in car ECUs. I tried for the very first time yesterday to program a 27C256 and it failed. I heard that you could not use the USB power to do so, therefore I got a 16V DC adapter, but nothing. Error was :- Error 0X000000 chip 0xFF buffer 0x00

using software version 0.98D5 hope someone can help me out. It seems that reading any chip O.K i just cannot program a chip. Vpp voltage is around 4.5V even when I program or read or blank check.

Running windows xp SP2

I have no idea what is wrong. Can someone help? Thanks in advance.

natuner

Re: 4.5 unable to program 27C256

By: DL - 2007-06-30 14:56:44

Hi,
natuner wrote on 06/30/07 at 12:15:58:
I am very new to the willem programmer and only used it till now to read 27C256 chips that were used in car ECUs. I tried for the very first time yesterday to program a 27C256 and it failed. I heard that you could not use the USB power to do so, therefore I got a 16V DC adapter, but nothing. Error was :- Error 0X000000 chip 0xFF buffer 0x00
Enter the Test H/W tab and set switches as indicated.  With no chip in the ZIF socket, click on the VPP pin and measure the voltage on pin1 of the ZIF socket.  It should be ~12.5V.  The VPP supply on PCB3 designs (and derivatives) is known to have problems when the wrong type of coil is used.
Quote:
using software version 0.98D5 hope someone can help me out. It seems that reading any chip O.K i just cannot program a chip. Vpp voltage is around 4.5V even when I program or read or blank check.
Is your programmer set to PCB4.5 mode?  Be careful about assuming that a read with no errors is okay.  Reading using the wrong program won't give a read error.  Reading an empty socket won't give an error either...

VPP should not be enabled when you read or blank check.  If it is, it would cause your device to be programmed.
Quote:
Running windows xp SP2
Make sure you have the registry mod that stops printer port polling.  It stopped working for me around XP Pro SP2 and I had to disable legacy plug/play support on the port.

Your application requires programming EPROMs reliably (correctly).  You need to blank verify devices at a voltage lower than they would normally operate.  You need to program at a sufficiently high VCC to insure that they will read realiably over the full voltage and temperage range they'll be subjected to.  This means you can't program with USB power.

Personally, I wouldn't program EPROMs with any Willem based programmer that hasn't been modified to give more VCC options.  I use 4.45V for blank check and 6-6.5V for programming; no stock programmer will give you these voltages.  Some programmers have a 6.2 or 6.4V option, but it's not "good" enough for me.

Even some "professional" programmers don't program/blank check correctly.

Voltages to the sockets are switched via PNP transistors.  This is "okay" when current is small (less than 20-30mA); this includes most CMOS devices.  I choose to design more "robust" circuits and would have taken the non-linear, current dependent voltage drop of the transistor switch into consideration.

Power is the Willem 4.1 is the "best" so far.  It has an optional relay to take care of the voltage drop at high currents.

I haven't seen the power circuits for any of the USB data and power programmers that are becoming available.  It would be poor design if they didn't take the voltage drop of the transistor/FET switch into consideration.

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-02 03:35:09

DL wrote on 06/30/07 at 14:56:44:
Hi,
natuner wrote on 06/30/07 at 12:15:58:
I am very new to the willem programmer and only used it till now to read 27C256 chips that were used in car ECUs. I tried for the very first time yesterday to program a 27C256 and it failed. I heard that you could not use the USB power to do so, therefore I got a 16V DC adapter, but nothing. Error was :- Error 0X000000 chip 0xFF buffer 0x00
Enter the Test H/W tab and set switches as indicated.   With no chip in the ZIF socket, click on the VPP pin and measure the voltage on pin1 of the ZIF socket.   It should be ~12.5V.   The VPP supply on PCB3 designs (and derivatives) is known to have problems when the wrong type of coil is used.
Quote:
using software version 0.98D5 hope someone can help me out. It seems that reading any chip O.K i just cannot program a chip. Vpp voltage is around 4.5V even when I program or read or blank check.
Is your programmer set to PCB4.5 mode?   Be careful about assuming that a read with no errors is okay.   Reading using the wrong program won't give a read error.   Reading an empty socket won't give an error either...

VPP should not be enabled when you read or blank check.   If it is, it would cause your device to be programmed.
Quote:
Running windows xp SP2
Make sure you have the registry mod that stops printer port polling.   It stopped working for me around XP Pro SP2 and I had to disable legacy plug/play support on the port.

Your application requires programming EPROMs reliably (correctly).   You need to blank verify devices at a voltage lower than they would normally operate.   You need to program at a sufficiently high VCC to insure that they will read realiably over the full voltage and temperage range they'll be subjected to.   This means you can't program with USB power.

Personally, I wouldn't program EPROMs with any Willem based programmer that hasn't been modified to give more VCC options.   I use 4.45V for blank check and 6-6.5V for programming; no stock programmer will give you these voltages.   Some programmers have a 6.2 or 6.4V option, but it's not "good" enough for me.

Even some "professional" programmers don't program/blank check correctly.

Voltages to the sockets are switched via PNP transistors.   This is "okay" when current is small (less than 20-30mA); this includes most CMOS devices.   I choose to design more "robust" circuits and would have taken the non-linear, current dependent voltage drop of the transistor switch into consideration.

Power is the Willem 4.1 is the "best" so far.   It has an optional relay to take care of the voltage drop at high currents.

I haven't seen the power circuits for any of the USB data and power programmers that are becoming available.   It would be poor design if they didn't take the voltage drop of the transistor/FET switch into consideration.

HTH,
Dennis


I have the same problem as described in this thread.  The Willem I bought is a Sivava 4.5, using the 0.98D5 software.  I have a 9v 1000mA power supply (actually puts out 13.5v).  I can read chips OK, but can't burn them.

I will admit to being out of my league here.  I just wanted a product that would work.  So maybe you can help explain how I should do the settings?

I have the vcc jumpers set to 5v.  I have the vpp jumpers set to 12.5v.  I can read chips and save the hex, and can load it to the program, but all I get is the first cell being changed from FF to 00.  If I push up the tWP and tWC, I can get a few lines written. and then the chip is NG until I can get it erased.  Are my fundamentals correct here?  Do I need a higher voltage power source, with vcc jumpers set to 15v?  You say that programming is better with 6.2v+.  Do I need to do anything other than set the vcc jumpers fpr 6.2v?

Is the Sivava product a piece of crap and I'm wasting my time?  I thought this was going to be a slam dunk, and it has just turned into a waste of money.  

Not sure if I should throw good money after bad and buy a new power source and eprom eraser.  Your thoughts?
Thanks,
Marc

Re: 4.5 unable to program 27C256

By: DL - 2007-11-02 04:51:48

Derbybrit1 wrote on 11/02/07 at 03:35:09:
I have the same problem as described in this thread.   The Willem I bought is a Sivava 4.5, using the 0.98D5 software.   I have a 9v 1000mA power supply (actually puts out 13.5v).   I can read chips OK, but can't burn them.
You need to qualify what you mean when you say you can read chips okay.  Are you verifying that the data read is correct or are you assuming that not getting an error during read means it was okay.  The former would be an indication of a good read; the latter isn't...
Quote:
I will admit to being out of my league here.   I just wanted a product that would work.   So maybe you can help explain how I should do the settings?
These programmers are intended for electronics hobbyists.  If this doesn't describe you, you might find a professional programmer more suitable.
Quote:
If I push up the tWP and tWC, I can get a few lines written. and then the chip is NG until I can get it erased.   Are my fundamentals correct here?
I have not found it necessary to change the default timing sliders to be able to program any EPROM.  Whether the EPROM has sufficient programming margin is a different matter.
Quote:
Do I need a higher voltage power source, with vcc jumpers set to 15v?   You say that programming is better with 6.2v+.   Do I need to do anything other than set the vcc jumpers fpr 6.2v?
First - no EPROM programmer should give you the ability to set VCC to 15V.  It would never be desireable to do that.

Second - you need to understand the difference between programming an EPROM and programming it reliably.  I assume you're programming 27C256.  That device should be programmed at VCC=6V for the "intelligent" algorithm (program in 1mS increments until the address verifies and then provide an overprogramming pulse that is 3 times the total of the pulses necessary for the address to program) or VCC=6.25V for the "quick pulse" algorithm.  Since no one (who's talking) knows what algorithm is being used, I assume it's "quick pulse" which would be the least conservative and therefore benefit from verifying at a higher VCC.

You can program any EPROM with VCC=5V.  You just don't know how much program margin you'll have.  For example, if margin were 5.25V (meaning the bit would read as unprogrammed at voltages above 5.25V); operating that device at high VCC (5.5V) would cause those marginally programmed bits to read as unprogrammed.  Clearly this is undesireable.

Another phenomenon you need to worry about is charge loss.  Electrons will "leak" from the floating gate over time.  If enough are lost, programmed bits will read as unprogrammed at progressively lower voltages.

Doing explicit program margin testing lets you avoid these problems for 10-20 years.
Quote:
Is the Sivava product a piece of crap and I'm wasting my time?
I wouldn't say that Sivava's programmers are junk though I do have issues with the way they're conducting themselves.  They started out working with the Willem open source group and later decided to branch off and built their product on the "shoulders" of work that others contributed.

There are some questionable practices with respect to some Willem variants.  Among them are using a step-up regulator that frequently has problems generating VPP=21 or 25V, excessive voltage drop in the VCC transistor switch, sloppy design in handling low voltage devices.

Sivava has made some good improvements to the base design, but they've chosen to keep them closed while building on the work of others.  Not something that you'd want anyone to do to you...
Quote:
I thought this was going to be a slam dunk, and it has just turned into a waste of money.
A PCB4.5C is definitely salvageable if you have the knowledge and ability to fix the problems.  All of the issues I mentioned can be fixed.
Quote:
Not sure if I should throw good money after bad and buy a new power source and eprom eraser.   Your thoughts?
I think you're giving up too soon unless you don't have the ability to fix the problems.  Granted you shouldn't have to fix anything; but if you had done some research before you purchased, you might have ended up with a Willem 4.1 which I feel is the best Willem programmer available (well, used to be available).

My recommendation is for you to obtain an EPROM that's electrically eraseable (actually a FLASH device) like 27SF512 to use until you're familiar and comfortable with the operation of your programmer.  Then start programming devices UV EPROMs.  Even if you had an eraser, EPROMs are only good for 100 program/erase cycles.

Read the troubleshooting information in the manuals section.  Check the voltages to make sure they're within tolerance.  Make sure you're setting the jumpers/switches/voltages correctly.  Are you using the PCB3 switch graphic?  Most Willem programmers use the PCB3 option (Willem 3.1 was Willem's last programmer that used the Willem option).  Give exact error messages.  Give your observations.  Do LEDs light as expected?  One problem with WinXP (yes, you should mention your OS too) has printer port polling enabled by default.  There's a registry mod that disables polling.  It stopped working for me around SP2 and I had many devices programmed while I was blank checking because of it.  I eventually found that disabling legacy plug/play for the printer port solved the problem once and for all...

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-03 13:58:40

DL wrote on 11/02/07 at 04:51:48:
[You need to qualify what you mean when you say you can read chips okay.   Are you verifying that the data read is correct or are you assuming that not getting an error during read means it was okay.   The former would be an indication of a good read; the latter isn't...


How am I supposed to verify a read?   The only files I have to compare against are the ones I created from reading the chip.   I can only say that the buffer is being filled with hex values other than FF, and appears to be complete.

DL wrote on 11/02/07 at 04:51:48:
These programmers are intended for electronics hobbyists.   If this doesn't describe you, you might find a professional programmer more suitable.
  No kidding.   I had no idea that I couldn't just read and write to a 27C256 without issue.

DL wrote on 11/02/07 at 04:51:48:
I have not found it necessary to change the default timing sliders to be able to program any EPROM.   Whether the EPROM has sufficient programming margin is a different matter.
  I didn't think changing those values was the answer.

DL wrote on 11/02/07 at 04:51:48:
First - no EPROM programmer should give you the ability to set VCC to 15V.   It would never be desireable to do that.


So what do the J6 and J7 jumpers on a PCB45 board actually set? options are 12.5v, 15v, 21v, and 25v.   What do I need for a power supply and jumpers to burn a 27C256 successfully?

DL wrote on 11/02/07 at 04:51:48:
Second - you need to understand the difference between programming an EPROM and programming it reliably.   I assume you're programming 27C256.   That device should be programmed at VCC=6V for the "intelligent" algorithm (program in 1mS increments until the address verifies and then provide an overprogramming pulse that is 3 times the total of the pulses necessary for the address to program) or VCC=6.25V for the "quick pulse" algorithm.   Since no one (who's talking) knows what algorithm is being used, I assume it's "quick pulse" which would be the least conservative and therefore benefit from verifying at a higher VCC.

You can program any EPROM with VCC=5V.   You just don't know how much program margin you'll have.   For example, if margin were 5.25V (meaning the bit would read as unprogrammed at voltages above 5.25V); operating that device at high VCC (5.5V) would cause those marginally programmed bits to read as unprogrammed.   Clearly this is undesireable.


So 5.0v setting on the Willem is what I need for the 27C256?

[quote author=DL link=1183202158/0#3 date=1193975508]I wouldn't say that Sivava's programmers are junk though I do have issues with the way they're conducting themselves.   They started out working with the Willem open source group and later decided to branch off and built their product on the "shoulders" of work that others contributed.

Quote:
Not sure if I should throw good money after bad and buy a new power source and eprom eraser.   Your thoughts?


DL wrote on 11/02/07 at 04:51:48:
I think you're giving up too soon unless you don't have the ability to fix the problems.   Granted you shouldn't have to fix anything; but if you had done some research before you purchased, you might have ended up with a Willem 4.1 which I feel is the best Willem programmer available (well, used to be available).


Searched for that.   Can't find any assembled units available for sale.

DL wrote on 11/02/07 at 04:51:48:
Read the troubleshooting information in the manuals section.   Check the voltages to make sure they're within tolerance.   Make sure you're setting the jumpers/switches/voltages correctly.   Are you using the PCB3 switch graphic?   Most Willem programmers use the PCB3 option (Willem 3.1 was Willem's last programmer that used the Willem option).   Give exact error messages.   Give your observations.   Do LEDs light as expected?   One problem with WinXP (yes, you should mention your OS too) has printer port polling enabled by default.   There's a registry mod that disables polling.   It stopped working for me around SP2 and I had many devices programmed while I was blank checking because of it.   I eventually found that disabling legacy plug/play for the printer port solved the problem


My multi-meter took a dump.   Will get a new one and check this stuff.   Thanks very much.

Re: 4.5 unable to program 27C256

By: DL - 2007-11-03 15:54:44

Hi,
Derbybrit1 wrote on 11/03/07 at 13:58:40:
How am I supposed to verify a read?   The only files I have to compare against are the ones I created from reading the chip.   I can only say that the buffer is being filled with hex values other than FF, and appears to be complete.
You either need data to verify against or know what data should be read and check the buffer visually to see that data was read correctly.   Subtle problems (i.e. one or more address lines not working) would still fill the buffer with seemingly good data, but be incorrect.
Quote:
No kidding.   I had no idea that I couldn't just read and write to a 27C256 without issue.
These programmers usually work as advertised.   But, being hobbyist programmers, the User is expected to have enough knowledge to use them.   For EPROMs, many thought that programming at VCC=5V (and the appropriate VPP, more on this later...) and having it verify was the same as programming the device reliably.   This couldn't be further from the truth.   For example 27C256 specifies a couple of VCC's for programming and neither is 5V.   That doesn't mean that you can't program at VCC=5V and have the device verify.   It just might not be programmed reliably (i.e. read correctly over the full temperature and voltage range for 20 years).   If you're programming EPROMs for ECU's, unreliably programmed devices could be very bad.

Regarding VPP.   A professional programmer will automatically set the voltages appropriately.   With Willem programmers, you need to know what to set it to and how.   For 27C256, it's simple; most are VPP=12.5V +/- 0.25V.   For devices like 2732, you need to know that 2732 are VPP=25V, 2732A are VPP=21V, and 2732B are VPP=12.5V.   Setting VPP too high is one of the easiest ways to damage an EPROM.

Similarly, you need to know what the appropriate programming pulse width is.   2716 and 2732 EPROMs are specified to be programmed at VCC=5V, providing you program with a 50mS pulse.   The program can do this for 2716, but not 2732 (longest pulse is 25mS).   But 50mS isn't the default pulse width for 2716; so you can't assume that the program is correct.   This is one of my complaints with the software; no one who's talking knows what it's doing and no one has source for the recent versions...
Quote:
So what do the J6 and J7 jumpers on a PCB45 board actually set? options are 12.5v, 15v, 21v, and 25v.   What do I need for a power supply and jumpers to burn a 27C256 successfully?
Those jumpers set VPP.   For 27C256, you need VPP=12.5V and VCC=6.25V.   The default programming pulse width for that device is 140uS.   This isn't correct, but it's close enough.   The program should program each address until it verifies (up to 25 attempts before failing).   I suspect that the program only tries once, but this is usually suficient.   If a device fails, you could try additional programming attempts; but you could run into another phenomenon called programming erase where programming a bit may cause some of it's neighbors to be slightly erased.
Quote:
Searched for that.   Can't find any assembled units available for sale.
Willem may order a small batch of 4.1 boards.   If he does, I plan to stock up.

When you get a new voltmeter, enter the Test H/W tab and set switches/jumpers as indicated in that screen.   If more than one of the LEDs are lit, do the simple hardware test in the Help menu.   This will get the program and programmer in synch.   Then, with no device in the ZIF socket, click on the VPP pin and measure the voltage on pin1.   It should be whatever you set VPP to.   Do the same for VCC.   If those voltages are correct. you should be able to program if the rest of the logic is working.

If you notice the VCC or VPP LEDs turning on when they shouldn't, it could be XP printer port polling.   You can try the registry edit in the more downloads link to the left.   If that doesn't work, use Device Manger to disable legacy plug/play support for your printer port.

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-03 16:30:10

DL wrote on 11/03/07 at 15:54:44:
    What do I need for a power supply and jumpers to burn a 27C256 successfully?
Those jumpers set VPP.   For 27C256, you need VPP=12.5V and VCC=6.25V.   [quote]

So what is the recomended DC power supply voltage/current to provide the 12.5v VPP?  I get the impression that these board have their own step-up transformers, but can't tell from the limited documentation that comes with it.

Thanks,
Marc

Re: 4.5 unable to program 27C256

By: DL - 2007-11-03 16:52:32

Hi Marc,

Your unregulated 9VDC supply is fine.   On PCB3 variants, the input voltage needs to be VCC+4V; you can get any needed VCC with 13.5V.

The step-up regulator usually works fine for VPP=12.5V.   They usually need a better inductor to get a stable VPP higher than that.   I don't think it's appropriate for this application and it usually causes more problems than it solves (low VPP, voltage drop, difficult to troubleshoot without a scope).   Not really a problem if you know how DC-DC regulators work.   But it has the potential to cause more problems than it solves...

I have a couple PCB3 variants and I replaced the step-up regulator on my PCB5 from Silvotronics -- it had other power problems that I fixed too.   I keep the other one unmodified for testing...

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-03 17:01:26

DL wrote on 11/03/07 at 16:52:32:
Hi Marc,

Your unregulated 9VDC supply is fine.   On PCB3 variants, the input voltage needs to be VCC+4V; you can get any needed VCC with 13.5V.

The step-up regulator usually works fine for VPP=12.5V.   They usually need a better inductor to get a stable VPP higher than that.   I don't think it's appropriate for this application and it usually causes more problems than it solves (low VPP, voltage drop, difficult to troubleshoot without a scope).   Not really a problem if you know how DC-DC regulators work.   But it has the potential to cause more problems than it solves...

I have a couple PCB3 variants and I replaced the step-up regulator on my PCB5 from Silvotronics -- it had other power problems that I fixed too.   I keep the other one unmodified for testing...

HTH,
Dennis


Last stupid question (until the next one).  When measuring the pin voltage, where should I pick up the ground?

Re: 4.5 unable to program 27C256

By: DL - 2007-11-03 17:18:58

Hi,

Use any convenient GROUND.  I use one of the screws on the parallel connector.

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-05 12:49:50

DL wrote on 11/03/07 at 17:18:58:
Hi,

Use any convenient GROUND.   I use one of the screws on the parallel connector.

HTH,
Dennis


Bought a new multimeter from (don't laugh) Radio Shack.  Actually pretty nice for under $30.  Anyway, Vpp = 12.89V abd Vcc = 6.36 volts.  Am I supposed to measure these voltagews with no chip in the socket?  

Can you please confirm the following:
Pin 1 and Pin 31 are closest to the 'lever' on the ZIF socket?
Orientation of the 27C256 chip in the ZIF socket?
Is the plain end of the chip (no notch) closest to the lever, or the opposite end of the socket?

Gosh, I may just be putting the chip in the wrong position.  I searched and cannot find a definitive answer.
Thanks,
Marc

Re: 4.5 unable to program 27C256

By: DL - 2007-11-05 15:54:17

Hi,
Derbybrit1 wrote on 11/05/07 at 12:49:50:
Bought a new multimeter from (don't laugh) Radio Shack.   Actually pretty nice for under $30.
I have several Radio Shack multimeters.   If it's anything like VCRs, they're getting the guts from someone who provides the same (with slightly different specs) to others...
Quote:
  Anyway, Vpp = 12.89V abd Vcc = 6.36 volts.
VPP is a little high, but not high enough to damage devices if it's well regulated.   Do you have VCC set to 6.2/6.4V?
Quote:
Am I supposed to measure these voltagews with no chip in the socket?
Normally you measure with no device in the programming sockets.   When you do measure with a device in the socket, it's important to make sure VCC is always enabled before VPP and VPP is disabled before VCC.
Quote:
Can you please confirm the following:
Pin 1 and Pin 31 are closest to the 'lever' on the ZIF socket?
There are two ways to orient the ZIF socket.   I think most put the lever near pin 1.   The other is to put it on the ground end of the chip (my Stag does this).

If you were able to measure VPP and VCC on the ZIF socket, you've identified pins 1 and 32.
Quote:
Orientation of the 27C256 chip in the ZIF socket?
You always insert devices so the VSS/Ground pin is in pin 16 of the socket.
Quote:
Is the plain end of the chip (no notch) closest to the lever, or the opposite end of the socket?
Pin 1 is closest to the notch.
Quote:
Gosh, I may just be putting the chip in the wrong position.   I searched and cannot find a definitive answer.
If you insert a chip backwards, you'll get a "check power and connections" error message because you're putting a short on VCC.   With a Willem 4.1, you might be able to fry the device because the relay and regulator will supply up to an amp (limited by your power supply).

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-28 01:07:18

Took your advice and purchased a couple of 27SF512 flash eproms.  I'm still not able to get the results I expect.  Can't write and can't erase.  To review:

1)  Verified pin 1 and pin 32 are closest to lever.  Notch of eprom closest to lever.
2)  Vpp = 12.89V abd Vcc = 6.36 volts.
3)  Switches set on DPSW12 for 27SF512
3)  Jumpers for JP1 and JP2 set for SST27SF512

Blank eprom should have all FF's?  At 000000 - 00000F I have all E5, for example.  Try to erase eprom:
1)  From version 0.98D5 software, go to Device tab.  Select check box  for 2/3 pin abnd 4/2 pin, and DIP DW Settings.  Click on Erase Chip.   Get error 'Device Erase False'
2)  Blank Check says Device is NOT Empty $000000

What else can I check?
Thanks,
Marc

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-28 01:19:52

This might be related to my previous post.  Bought a Datarase II to erase the eproms that had previously failed.  (2) stints of 15 minutes each should be enough to clear first time burns?

Put them in the Willem, and they still read as not blank.

Re: 4.5 unable to program 27C256

By: DL - 2007-11-28 18:14:49

Hi,

30 minutes should be sufficient to erase any EPROM.

We've established that your using 0.98d5 with a PCB4.5C, but you never answered my question regarding whether you had your programmer configured to be compatible with 0.98d5.  You have a set of headers that can set your programmer for PCB4.5C or PCB3.  If you don't have the programmer configured correctly, the data won't read correctly.

HTH,
Dennis

Re: 4.5 unable to program 27C256

By: Derbybrit1 - 2007-11-29 00:13:37


DL wrote:
Hi,

30 minutes should be sufficient to erase any EPROM.

We've established that your using 0.98d5 with a PCB4.5C, but you never answered my question regarding whether you had your programmer configured to be compatible with 0.98d5.   You have a set of headers that can set your programmer for PCB4.5C or PCB3.   If you don't have the programmer configured correctly, the data won't read correctly.
HTH,
Dennis

There are 2 jumpers used to set either PCB45C or PCB3B  Mine is set to PCB45C.  Is that what you are referring to?

More info- - ran the self checks.  all were OK.  Checked the specs on the SST27SF512 eprom.  Manufacturer states "In order to activate
Erase mode for SST27SF512, the 11.4-12V is applied
to OE#/VPP and A9 pins, VDD = 4.5-5.5V, and VIL on CE#
pin. All other address and data pins are “don’t care”.

I am only able to get vcc=10.3 and A9=9.6 volts respectively from my unit, which is below manufacturer specifications.

I have tested with both a 9v/1000mA power supply and a 24V/1000ma power supply, with a slight change in output voltages.

When I try to erase the chip, the vpp red light flashes, but the 'Erase and Check Device is Empty' status bar does not move.  It does move 0-100% with no eprom installed.

Back to your question about programmer configuration.  Can I set the unit jumpers to PCB3B and use a different version of software and possible get better results?  I thought changing to an electrically erasable eprom was going to make this process simpler.  So far no luck.



Re: 4.5 unable to program 27C256

By: DL - 2007-11-29 00:29:20

Hi,

Derbybrit1 wrote:
There are 2 jumpers used to set either PCB45C or PCB3B   Mine is set to PCB45C.   Is that what you are referring to?
Yes.   With the jumpers set to PCB45C, you can only use 0.98d5 or later.   You could use 0.98d2 or later, but that wouldn't make much sense.   In PCB3 mode, you can only use 0.97* versions (with 0.97ja and 0.97g being the most likely versions)

Quote:
More info- - ran the self checks.   all were OK.
What checks are you referring to?   The simple test in the Help menu only uses a fraction of the logic and won't check for tolerance problems.   The checks in the Test H/W tab can be used to check for tolerance, but you would have noticed that VPP is out of tolerance...

Quote:
Checked the specs on the SST27SF512 eprom.   Manufacturer states "In order to activate
Erase mode for SST27SF512, the 11.4-12V is applied
to OE#/VPP and A9 pins, VDD = 4.5-5.5V, and VIL on CE#
pin. All other address and data pins are “don’t care”.

I am only able to get vcc=10.3 and A9=9.6 volts respectively from my unit, which is below manufacturer specifications.
I assume you meant VPP=10.3V.   That should be low enough that program and erase will fail.

Quote:
I have tested with both a 9v/1000mA power supply and a 24V/1000ma power supply, with a slight change in output voltages.
Are these DC or AC voltages?   Shouldn't make a difference unless the 9V adapter is low quality.

Quote:
When I try to erase the chip, the vpp red light flashes, but the 'Erase and Check Device is Empty' status bar does not move.   It does move 0-100% with no eprom installed.
I suspect VPP is too low to erase.

You need to fix the low VPP problem.   Were your measurements with VPP unloaded?   It would be unusual for any PCB3 variant to not be able to provide VPP=12.5V with no load.   The step-up regulator usually has problems with VPP=21V or 25V, and only because an inappropriate inductor is used.

If the programmer is new/still under warranty, can you return it?   You'll be happier with a Willem 4.1.   It'll cost more, but it will work better...

HTH,
Dennis