USB Dual Power Universal Willem  Programmer

By: stuffyslicksam - 2007-10-30 21:20:41

Hi to all member!!!hope someone could answer my question...
I want to program sst39vf3201 chips and i am looking to buy a programmer,,Does the USB Dual Power Universal Willem Programmer with tsop 48 adapter would be a good buy for that kind of chip?????i have done alot of reading on it but i am still not sure wich programmer would be my best bet.. so if anyone could give me some direction on this would be greatly apreaciate..THANK-YOU.

Re: USB Dual Power Universal Willem  Programmer

By: DL - 2007-10-30 21:47:16

Hi,

EZo's database lists it as being supported.  EZo doesn't mention the device he selected or whether he used a 16Mb device and read/programmed in two or more passes (as is required for 64Mb devices)

A Sivava PCB4.5C with their 16 bit adapter will probably give you the fewest problems.  If you're going to program EPROMs, I'd go with Sivava's adapter and a Willem 4.1.

HTH,
Dennis

Re: USB Dual Power Universal Willem  Programm

By: stuffyslicksam - 2007-10-30 22:02:51

tHANK-YOU KINDLY for your quick response,,much apreaciated..
I will do some reading on the willem 4.1 version..

Re: USB Dual Power Universal Willem  Programmer

By: DL - 2007-10-31 00:58:34

Hi,

The "fewer issues" I referred to in using a PCB4.5C was with respect to the cable used to connect high order address lines to the adapter.   Sivava's cable will be compatible with their programmers and adapters.   Pin order on other programmers may be different...

The Willem 4.1 does low voltage interface to the computer better (I think because I don't have a PCB4.5C or a schematic to reference).   PCB4.5C have all of the power/voltage drop problems other PCB3 variants have.  I don't know how low voltage is implemented on the programmer.  Most didn't do this correctly.

PCB4.5C has the advantage of being faster when using the appropriate 0.98d* program versions.   Other Willem programmers can be modified to be compatible with those program versions.   Information has been posted, but I haven't tried it.

USB power shouldn't be used for programming any EPROMs unless VCC also uses a step-up regulator.   Mcumall's true USB programmer might be in this category; don't know because I don't have one or a schematic.   Willem's version will use step-up regulators for both power supplies.

HTH,
Dennis

*** The SECRET MODS *** REVIELED

By: Mr.Spock - 2007-10-31 08:58:41

DL wrote on 10/31/07 at 00:58:34:
[...]

PCB4.5C has the advantage of being faster when using the appropriate 0.98d* program versions.
  Other Willem programmers can be modified to be compatible with those program versions.   
Information has been posted, but I haven't tried it.


[...]

HTH,
Dennis


Was this the one you're referring to?
http://www.msxpro.com/py2bbs/willem_pcb45en.php

the willem v4.1 may just need the buffer MODs shown near the end of that page ...

.....




extra IC (74HC04) on bottom of board....   
pins 2 and 4 connected to pins of DB25
PAPER_END and SELECT




Interesting, I didn't fully read this article the first time, when I had a file from there, show as a worm virus (e-mail trojan) from his site ...
http://www.willem.org/cgi-bin/yabb2/YaBB.pl?num=1162099430/10#10
that he corrected later.

Sooo...this is the secret change on the PCB4.5c board made to the Willem 4.x

This makes it the PCB4.5cc as shown here ... http://www.darkwire.com.au/html/willem_programmer.html


....................................^ ^...................
with the special double x3 jumpers, making it able to switch between Version PCB3B/PCB3.5 and PCB4.5C

  

No-wonder why no one here was answering me when I asked what the mods were for the 4.1's
... it was there, all the time ...       ... i was trying to make the 4.1 like the willem 4.5

I may try this MOD after I make the EZoBUS-To-ZIF32 adapter.

specific v4.1 MODS- no trace cuts or IC needed...

By: Mr.Spock - 2007-10-31 17:11:57

wow, I just had a look at the 4.1 and it needs a minimum of modding.

NO extra IC is needed for the PAPER_END and SELECT lines at the printer port
because there are TWO(2) inverters not being used inside the IC9 chip at ...
pins 9 -input- & 10 -output- of inverter "D" ... and
pins 14 -input- & 15 -output- of inverter "F"

How sweet is that... LOL

No traces wouldn't have to be cut either, because the address IC's can be taken out of the socket
and piggy-backed onto another socket where the pin is bent out for a wire to be attached
to pin 15 of IC4 or pin 15 of IC5 , to attach the jumper wire, from the original socket pin,
to the jumper blocks. (2x3 pin)

then the IC's are put back into each of the piggy-backed sockets.

for example...

... original IC inserted into top socket (piggy-backed) no pins bent out
... piggy-backed socket with bent out pin with wire soldered to it --- going to center of jumper block--- pin 15 of IC4a or IC5a
... original socket with wire inserted into opening where pin area above is bent out from--- to one end of 1 jumper block-   pin10 IC3b or IC4b
... then the other end of the jumper goes to the D4 or D5 location of MOD at IC8 Pin2 (D4) or pin4 (D5)

The only other change that most likely is need will be the voltage select DIP4 switch area, to a DIP6 to give more voltage selections for the programmable memories... I'll have to look at the willem v4.5 board for this.

the IC locations and numbers are different for the 4.1 and 3.1 ...
so here's the labeling for the v4.1 board...

Re: USB Dual Power Universal Willem  Programmer

By: DL - 2007-10-31 18:39:45

Hi,

While you're making your changes, you should put 10K pull-up resistors to the fixed 5V supply on the 3 inverters that drive the parallel port.  This will make the level shifting work better when the programmer isn't running at 5V; better rise time on the inputs which can affect the fall time on the outputs.

The inputs to the uncommited inverters appear to be floating.  Even if you don't do the mod, you should connect them to power or ground to avoid excessive power dissipation.

Did you buy a voltage mod from me with your programmer (I think I sold you your programmer)?  If you did, you can connect to that board to give more fixed voltage options.  Resistor calculations for my mod are simpler because there aren't any parallel resistances to calculate.  The stock Willem 4.1 and 4.5 use parallel resistances for all but the maximum VCC option.  But calculating the required resistors is still pretty simple.

HTH,
Dennis

Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: Mr.Spock - 2007-11-01 01:05:43

DL wrote on 10/31/07 at 18:39:45:
Hi,

While you're making your changes, you should put 10K pull-up resistors to the fixed 5V supply on the 3 inverters that drive the parallel port.   This will make the level shifting work better when the programmer isn't running at 5V; better rise time on the inputs which can affect the fall time on the outputs.

The inputs to the uncommited inverters appear to be floating.   Even if you don't do the mod, you should connect them to power or ground to avoid excessive power dissipation.


Hi there stranger, , are you referring to the Willem 3.1 board here, 'cause the 4.1 and 4.5 have the inverter inputs   connected to the 8x10k resistor pack (Pin1) to VDDS on the v4.1 board, and to VDD on the v4.5 board.   This was what was confusing me at first, when I originally looked at the two to see the differences between the Willem 4.1 / 4.5 ...
why the different location for the same resistor packs pin1 on the two boards?

Also, should the two extra inverters, in the IC9 chip, now used for the mod to IC6 pins 2 and 12, have the inputs with pull-ups to 5v, or VDDS/VDD power? (for the willem v4.1 board)

Quote:
Did you buy a voltage mod from me with your programmer (I think I sold you your programmer)?   If you did, you can connect to that board to give more fixed voltage options.   Resistor calculations for my mod are simpler because there aren't any parallel resistances to calculate.   The stock Willem 4.1 and 4.5 use parallel resistances for all but the maximum VCC option.   But calculating the required resistors is still pretty simple.

HTH,
Dennis

I purchased the v4.1 board kit from you, but you just told me about your voltage mod, and sent me the picture of it as a tease ..   
... you've since told me you've updated that Voltage MOD... yes?

BTW-do all your Willem products "live" in that small house, next to your computer, ...
shown in the lower left of the Mod-Collage?   

Re: Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: DL - 2007-11-01 18:30:48

Hi,
Mr.Spock wrote on 11/01/07 at 01:05:43:
... are you referring to the Willem 3.1 board here, 'cause the 4.1 and 4.5 have the inverter inputs   connected to the 8x10k resistor pack (Pin1) to VDDS on the v4.1 board, and to VDD on the v4.5 board.
No, I was referring to the Willem 4.1.  The pull-ups you're referring to are on the inputs of the tri-state buffers (which are non-inverting).  The inverter that sends the serialized data to the printer port should have a pull-up resistor on it's input to the fixed 5V supply if the programmer is running at low voltage (i.e. 3.6V).  This is because the that inverter is operating at 5V, but being driven by a signal that will only pull-up to 3.6V.  The pull-up would improve the rise time on the input which could affect when it's output goes LOW.  If you do the mod to send all three outputs from the CD4021 to the parallel port, all need pull-up resistors.

When I checked a 4.1, the unused inverters seem to have floating inputs.  This is likely to cause unnecessary power dissipation.
Quote:
... you've since told me you've updated that Voltage MOD... yes?
No functional changes.  I changed the way I attach to the programmer.  I used to use a machined pin socket; now I hardware to avoid problems with it getting disconnected.  My board still uses a socket and it's a nuisance to have to keep checking to make sure it's still connected.  I left mine that way so I could move the mod to other programmers.
Quote:
BTW-do all your Willem products "live" in that small house, next to your computer, ...
No...

HTH,
Dennis

Re: Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: Mr.Spock - 2007-11-01 21:36:41

DL wrote on 11/01/07 at 18:30:48:
Hi,
No, I was referring to the Willem 4.1.   The pull-ups you're referring to are on the inputs of the tri-state buffers (which are non-inverting).   The inverter that sends the serialized data to the printer port should have a pull-up resistor on it's input to the fixed 5V supply if the programmer is running at low voltage (i.e. 3.6V).   This is because the that inverter is operating at 5V, but being driven by a signal that will only pull-up to 3.6V.   The pull-up would improve the rise time on the input which could affect when it's output goes LOW.   If you do the mod to send all three outputs from the CD4021 to the parallel port, all need pull-up resistors.

When I checked a 4.1, the unused inverters seem to have floating inputs.   This is likely to cause unnecessary power dissipation.

Hi again to you toooo, I was actually, originally, referring to the CD4049 IC9-C, & B sections, in addition to the unused ones at D & F that would now be used ... but yeah, I wasn't thinking of the CD4021    ... thanks for the heads up there, masked stranger, ...
One more question, before you ride into the sunset, where's the stable 5v source point?
The LM317's are adjustable, so I can't tap the output of those, and the v4.1 schematic doesn't have very clear printing at most of the spots showing these areas.

Quote:
BTW-do all your Willem products "live" in that small house, next to your computer, ...
DL wrote on 11/01/07 at 18:30:48:
No...
I hope you know I was joking....   
... with no emotion icon and just one word in your reply, I couldn't be sure...   
... I gathered it was your kids, but just couldn't resist   

Re: Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: DL - 2007-11-02 01:55:03

Hi,
Mr.Spock wrote on 11/01/07 at 21:36:41:
...I wasn't thinking of the CD4021
One other point about the CD4021.  Willem programmers typically have CD4014, but I and others have found that the CD4021 is interchangeable.  The only difference is in how the parallel data is loaded.  '14 is synchronous and the '21 is asynchronous.  It seems that Sivava may have changed the timing for reading the data because, as EZo pointed out, they're using CD4021 now.
Quote:
... where's the stable 5v source point?  The LM317's are adjustable, so I can't tap the output of those, and the v4.1 schematic doesn't have very clear printing at most of the spots showing these areas.
There are 3 LM317's on the 4.1.  The one in TO-92 is set to 5V.  This is used to power the CD4049 and the relay.
Quote:
... I gathered it was your kids, but just couldn't resist
It's just the base of the lamp I use to get natural light when taking pictures.  I didn't have a snappy rejoinder so the one word was about all I had to say about it...

HTH,
Dennis

Re: Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: Mr.Spock - 2007-11-02 04:12:27

DL wrote on 11/02/07 at 01:55:03:
Hi,
One other point about the CD4021.   Willem programmers typically have CD4014, but I and others have found that the CD4021 is interchangeable.   The only difference is in how the parallel data is loaded.   '14 is synchronous and the '21 is asynchronous.   It seems that Sivava may have changed the timing for reading the data because, as EZo pointed out, they're using CD4021 now.

Interesting, ...you included the CD4021 with the KIT you sent me, so I'm good to go , thanks.
Quote:
... where's the stable 5v source point?    ...   There are 3 LM317's on the 4.1.   The one in TO-92 is set to 5V.   This is used to power the CD4049 and the relay.

Ahhhh ... , it's a fixed regulator.   I didn't know that.   It goes to the 8x10k resistor pack pin1 too, but the Willem v4.5 has that pin going to the second LM317's output, which is being changed by the DIP4 power switch, so is that pin1 of the 8x10k Resistor pack (on willem 4.5) "floating" up or down depending on the setting of that regulator?   If so, why?

Quote:
... I gathered it was your kids, but just couldn't resist ...
  It's just the base of the lamp I use to get natural light when taking pictures.
I didn't have a snappy rejoinder so the one word was about all I had to say about it...

HTH,
Dennis

I think you just did .. ...    ... so it's a simple lamp? ... must be a collectors item, being so fancy.

thanks for all this extra info regarding the additional pieces of the MOD puzzle ...   Smiley
This is going to be sweet to upgrade now.  I have to start on the EZoBUS to ZIF32 adapter before I do this mod...
I promised EZo, and myself I would

Re: Willem 4.1 to PCB4.5cc MODs Uncovered :-)

By: DL - 2007-11-02 05:33:27

Hi,
Mr.Spock wrote on 11/02/07 at 04:12:27:
... , it's a fixed regulator.   I didn't know that.   It goes to the 8x10k resistor pack pin1 too, but the Willem v4.5 has that pin going to the second LM317's output, which is being changed by the DIP4 power switch, so is that pin1 of the 8x10k Resistor pack (on willem 4.5) "floating" up or down depending on the setting of that regulator?   If so, why?
The Willem 4.5 pre-dates the 4.1.  It, and the 4.0, were the first I'm aware of to attempt to deal with low voltage correctly.  Previous attempts all dropped VCC to the device being read/programmed, but operated the programmer at 5V (or 5.7/6.4 in some cases).  This causes problems with overdriving the inputs to the device in the programming sockets and creates a situation where the device is underdriving the programmer.  The only thing Willem didn't take care of was the voltage difference between the parallel port and the programmer.  The parallel port operates at a nominal 5V, so it would be overdriving the inputs to the CD4503 tri-state buffers.  The parallel port signals would also be trying to raise VCC on the programmer -- through those 10K pullup resistors on the tri-state buffer inputs.

The 4.1 fixed that problem by keeping the pullup resistors on the CD4503 at 5V, so there's no contention between the programmer and computer.  When the programmer is running at low voltages (i.e. 3.6V, the 2.2K series resistors limit overdrive current to the CD4503 and prevent the input protection diodes from carrying enough current to cause electromigration.  The only thing the 4.1 didn't do (WRT level shifting) was put a pullup resistor on the input of the inverter (which is operating at 5V) to 5V.  This will improve fall time on the output of that buffer when the programmer is running at low voltage.  All of those resistors aren't needed when the programmer is running at 5V.

When other programmers run at high VCC, 6.4V for instance, they're overdriving the parallel port by 2 diode drops.  If the inputs on the parallel port didn't have series current limiting resistors, the programmer could cause electromigration in the chipset that controlls the parallel port.  If you damage that chip, no more parallel port (at best).

Any circuit designer worth employing would know to avoid those types of issues.

HTH,
Dennis

4.x Board MODs to use 0.98Dx software

By: Mr.Spock - 2007-11-18 17:07:37

Hi Dennis,
How'd your MODs go to your Willem 4.x board, allowing use of the newer software?

You mentioned, in the eMail to me, you had a speed gain of 50% for verifying a 4Meg chip, but, how 'bout the other functions... programming, erase, reading ... ?

Have you had any programming errors from the speed gain?

Please let us know everything ...  

Re: USB Dual Power Universal Willem  Programmer

By: DL - 2007-11-18 20:08:33

Hi,

I've only had time to test with two devices: AT27LV040 (pre-programmed) and AM29F040B.

Read, verify, program, and erase (for the FLASH) worked without problems.    With 0.98d2, operations are 1.5-1.7X faster; with 0.98d5, it's 1.8-2X faster.   I did all of my testing with the default timing settings.

For the benefit of other members, I modified my Willem 4.0 to add a level shifter in addition to the PCB4.5C mods.   I haven't tested the level shifter, but it was used for all of my tests.   The extra buffers required for 0.98d5 are on the level shifter board which is inserted between the parallel cable and the programmer.   I bypassed the two inverters that used to send data to the parallel port and put them on the level shifter board so it would be more convenient to add pull-ups to the fixed 5V supply on all of the inverters that drive the parallel port.   The pull-ups are only needed if the programmer is running at low voltage (below 4.5V).

My 4.0 had already been modified to give more VCC/VPP options, add a VPP disable, and a relay to fix the voltage drop in the VDDp switch.

Modifying a Willem 4.1 would only require using the unused inverters, adding pull-ups to the inputs of the inverters driving the parallel port, and modifying VPP/VCC to give more options.

HTH,
Dennis