update 24-09-2004 This page is about eprom programmer versions 4.0, 4.5, 3.1
PCB3B from Gitti Ieo is much the same but there are also a lot of differences.
The voltage regulator section is totally different:3.1 uses a 7805 and 7812
regulator,Vpp 21 and 25 volt can be selected with zener diodes in the 7812 gnd line.
Pcb3b uses a MC34063A to generate different Vpp voltages.
The jumper configuration is also different. Visual inspection
after you have completed your programmer board do a thorough visual inspection.
This is very important:lots of problems are caused by bad solderings,small shorts,
bended parts and so on.Use a strong light source and a magnifying-glass. Checking the voltage regulators and power lines
Connect the power supply,the green power led should light if not try to reverse polarity.
Check or 12,7 volt is present at the 7812 output pin,if one of more zener diodes are mounted
you can also check or Vpp=21 volt by switching the 2 way DIP switch.Power supply voltage must
be 24 Volt min. for this check.If Vpp is too low,20.5 volt for example you can mount one or
more(serial) 1n4148 standard diodes in pos. Z2 in the opposite direction.If you want to use
Z2 for the 3.9 Zener diode mount the 1n4148 serial with the 8.2 zener,you can also use a 9.1
zener instead of a 8.2 type
The 7805 output voltage must be 5 volt of course.
Check all IC's or they have Vcc 5 volt,the Vcc pin is 16.(4049 pin 1)
If the 7805 output is not 5 volt but zero and gets warm there must be a short somewhere,if the
voltage is lower than 5 volt and gets warm it's most likely a chip mounted in the wrong
direction,this chip will also get warm.
With 4.0 and 4.5 boards it's a bit different:
the 4.0 board use 2 LM317's that are set with several precision resistors and a 4 way DIPswitch.
The 4.5 board uses a 7812 with a fixed 12.6 volt output and a LM317 with precision resistors.
Testing your programmer
Connect your programmer board and start up the software as described in the programmer manual.
If Test Hardware fails check or the power supply and the connection cable are properly connected.
You can also try to set to lpt2 or lpt3,it's possible that your PC uses one of these printerport
addresses even if your system has only one printerport(you can also check this in the BIOS setting).
If the Test still fails you can try to follow the test signal through the circuit:
go to the Test H/W section and check "serial out(D0 pin 2)" ,normally "serial in (/Busy pin 11)"
should become '1' but if the test fails it stays '0'.Vcc will also be switched on.
Start at D-Sub pin 2 and check or the voltages changes between 0 and 5(3..5V) if you check and
uncheck "serial out(D0 pin 2)".If nothings happens there is a problem with the connection to your PC.
If the signal change is ok follow the signal in the same way through 4049(Pin 14->pin 15),
(the signal is inverted now!),transistor BC549 Q6 and Q5(the signal will be inverted again twice)
and through 4049(pin 9-> pin 10)untill you have found the problem.
4049 pin 10 is connected to D-SUB pin 11,if the signal change is ok here it must be a connection
problem to your PC.
If Test Hardware is ok you can continue with other tests.
It's important to understand that Test Hardware says nothing about the parallel eprom(DIP 32)
section of the programmer,Test Hardware uses the serial devices circuit.
At this point you can try to program an eprom,it's best to use an 28F010 or 020.The whole programmer
hardware and Vpp will used this way and you can easily erase the chip with the programmer.
But you can also use eproms like a 27C020 or 29F040 of course.The test eprom must be empty of course.
If the Blank check fails and you are sure that the eprom is empty there must be a problem in the
reading section;see below how to test it.
Load a file that fills the whole buffer to use the whole device address range.Use a file with lots
of contents,a big *.zip file is perfect.Check device ID and Program if that works and Verify is ok
too your programmer has passed the test.As an extra you can test the address lines that are above
your test eprom address range,but normally that will be ok.
If programming fails you'll have to do more tests to locate the problem.
If programming fails at the first byte there can be different causes.
A failure at the same address every time like "error at 0x040000 buffer=0xXX ,chip=0xXX"
is caused by an addressing failure in this case A18(pin 31 or pin 1)
A random error,can be caused by a poorly UV erased chip or tWP maybe too low,
there can also be other causes like instability in the voltage regulator.
First set the DIP switches as shown in the Test H/W section(the DIP switches are not shown in the
reversed "Willem" direction here).Using the Test H/W function is not difficult:
in 'ZIF 32PIN' check the boxes one by one and use a voltage meter to measure or you get logic 1
(normally 5 volt,A9 (pin 26) is lower because of the erase circuit,3..4 volt)
You can find shorts to measure signals next to the one that is logic '1'.
For example:if you check A5,A4 and A6 should stay low.
To test the 4014 read function check and uncheck OE so that D0...D7 are "greyed out", you can now
apply a logic 1 (VCC 5 volt) or 0 (ground) to the different data inputs on the progamming ZIF
socket (pin 13,14,15,17,18,19,20 and 21).Use the "Read Data" button to test, all inputs at VCC
should read 0XFF, all inputs grounded 0X00 and all values between.You only need to test the 8
different data inputs of course.