[Solved] Willem 5.0 suddenly stopps working

[Marxon]

Hi all!

After one year my Willem 5.0C (dual/USB powered, SMD ICs)
suddenly stopps working and isn´t recognized anymore "Check Power and connection".
I am using 0.98D10 software with the remapped io.dll because of my PCI-LPT card
and everything was fine until now.
I allready tried another PC without success.

Some notes to the behavoir:
-only LPT connected: power LED lights slightly
-only Power connected: power and Vpp LED light fully
-Power and LPT connected: power and Vpp LED light fully
-Software start after
Power and LPT connected: power and Vpp LED light fully
(normally only Power LED remains on after start)

I tired the Test Hardware functions but the Programmer
semms not to react if i change some settings.
Wouldt anybody give me some tipps how to check my Willem?

Thanks a lot!
Marxon

[Ezo]

Go Test H/W and activate Data signal, check changes on related paralel port pin. Like D00 on/off and measure DB25/2. If there is no changes, it indicates no programmer problem. Check selected LPT port adress, your PCI-LPT card and driver.

[Marxon]

Thank you for your quick help Ezo!

I´ve checked the data lines like you said:
3,4V no matter if activated or not! My LPT cable is ok.
So it shouldt be a software problem...
My LPT card driver is correctly installed, polling is deactivated,
port adress is BC00 and "IO.ini" is also edited.
But i haven´t done changes in my system since he was working
To be save i uninstalled the software (and io.dll), deleted any willem
related in the registry and installed everything again:
still not working.
What can i do now?
May you know some tool to test my LPT port?

thanks
Marxon

[Ezo]

I am really surprised you have activated programmer with PCI-LPT card before.
Usually no success. I do not have similar experiments.
I recomend you check your cable, programmer and power supply temporary on other PC with original LPT port.

[Marxon]

Hi again!

I am really surprised you have activated programmer with PCI-LPT card before.

The only thing i´ve done was to use the remapped io.dll from web.

Goods news:
I tried again to control data lines D0-D7 via the software with success.
Maybe last time i unintended selected a wrong port
Now they change from 3.4V / 0V if i enable/disable them.
Also pins 1(Vpp), 14(OE), 16(Vss), 17(CE) are working but here i measure 0V - 5V.

If i connect the programmer i can control Vpp, Vcc LED and A19-A23
but measure 0V on the zif socket all the time.
Another PSU doesn´t help. Output of 7805 regulator on willem board ok: 5V
I will check if there are also 5V on all ICs´s Vcc pins... after sleeping a while

Thanks again for your help.
Marxon

Marxon

[Marxon]

Back again!

I checked Vdd/Vss on ICs 4015, 4021, 4069 and measured 5V.
I figured out one 4015 ontrolls A19-A23 so i guesss another 4015 must make problems.
Is there a way to test it?

Marxon

[Ezo]

Seems progress.
Go Test H/W for all signals and measure voltage on ZIF32 pins.
What is missing (no changes from 0/~4..5V) ? Are you able set Vpp (0>12.7) ?
4015is working serially , if adress set stops due to failure, next adress will not change. Check latest acceptable level change adress , find on hardware related 4015 pin , check lines/shorts.

[Marxon]

Hi Ezo!

I allready did measuring

if i connect the programmer i can control Vpp, Vcc LED and A19-A23
but measure 0V on the zif socket all the time.

So i can set ZIF pins Vcc, A19. A20. A21, A22, A23 to 0/5V
and Vpp has correct voltage, too according to jumper settings.
All other ZIF pins = no reaction, 0V

4015is working serially , if adress set stops due to failure, next adress will not change. Check latest acceptable level change adress , find on hardware related 4015 pin , check lines/shorts.

COOL... we getting closer to the problem

The 4015 has eight outputs. Vcc + Vpp + A19-A23 = 7! so one must cause the problem, right?
Or maybe the connection from one 4051 to the next one?
What lines/pins shouldt i test further?

Marxon

[Ezo]

So i can set ZIF pins Vcc, A19. A20. A21, A22, A23 to 0/5V

If A19 is working, all prior adress should work.
Let measure on contacts of ZIF32 opposite side (soldering side).
Check connection and resolder all ZIF 32 contacts !

[Marxon]

So i can set ZIF pins Vcc, A19. A20. A21, A22, A23 to 0/5V

If A19 is working, all prior adress should work.
Let measure on contacts of ZIF32 opposite side (soldering side).
Check connection and resolder all ZIF 32 contacts !

Hi again!

You´re right! i checked the ZIF socket exactly and noticed some piece of tin between some pins!
Now i can control all pins exept VCC.
I know i said i can control VCC before but now VCC is dead

[Ezo]

Find related/connected Vcc switch transistor and replace.

[Marxon]

Q1 is directly connected to the VCC pin
But its SMD, i cant read the type.
May you know which i couldt take?

Thanks again.
Marxon

[Ezo]

Check voltages on transistor pins (Test HW 32-Vcc on and off)
What mark on the top ?
You can replace to BC807 (5C) or BC857 (3F). Last not recommended if programmer is applied for old eproms.

[Marxon]

Fortunately I found a BC807 right away and replaced the old one.
Hurray! VCC is back again! Now, i can control all Test-HW pins
Ezo, you are my hero of the day!
But i still get this damn "Check Power and Connection" error.
After doing some google I finally found a solution:
PortTalk
After downloading and extracting PortTalk,
i copied Willem´s Eprom.exe, the recoded IO.dll and the IO.ini into the PortTalk directory.
Now i start PortTalk´s AllowIO.exe with this syntax via windows command line:
".../PortTalk/AllowIO.exe eprom.exe 0xBC00"
(0xBC00 is the adress of my PCI-LPT card)
AlloIO automaticly starts the Eprom software...
Finally i get "Hardware present"!!!
No idea why and how it works, but it works

And again: thank you for spending time to guide me