JDM hardware help

I have made a JDM progammer and now tryinig to test it by the following steps which I got from:
**broken link removed**

THESE ARE THE TESTING STEPS:

If you're using Windows and ic-prog (which I'm using!), then go to Settings | Hardware Check. Plug in the programmer and leave the IC socket empty. You should observe the following:

1. Using a multimeter, measure the voltage on the IC socket between pin 5 (-) and pin 14 (+). It should be just below 5V.
2. When you tick Enable Data Out, the Data In checkbox should also become ticked (and vice versa). You cannot directly tick the Data In checkbox.
3. Measure the voltage between pin 5 (-) and pin 13 (+). It should read around -2V to -2.5V. When you tick the Enable Data Out checkbox, it should swing to 4.5V to 5V.
4. Measure the voltage between pin 5 (-) and pin 12 (+). Tick the Enable Clock box and it should go to +5V (or a little more). Untick it and it should return to 0V (actually about -0.5V to -0.8V).
5. Measure the voltage between pin 5 (-) and pin 4 (+). Tick the Enable MCLR box and it should go to +12V. Untick it and it should return to 0V (actually about -0.5V).

The thired step, bolded letters is what I am getting as +3.15V which should be -2V to -2.5V, after that every thing is right, I mean every thing except this bolded letters only not the whole steps!!
PLEASE HELP ME!!
Schematic on the link.(Also attached), I used 100uF/63V insted of 40V and 22uF/25V insted of 16V but I don't think that can cause problem or can it? (Exact values of capacitor voltage is not available!)

Best help and advice - see all the previous posts in this forum about JDM programmers.

Change it to use an external supply.
**broken link removed**
You can omit the 4 diodes in the center.

Also, have a read of this page.

Mike.

What transistors did you use? Did you wire it correctly? What PIC did you program?

**broken link removed**

I used the same transistors (BC547 B), those are in the schematic and wired them properly, I reffered the link but don't understand what's wrong, I also tried the circuit without capacitors as some people say to do so to have sharp rise from 5 to 12 V (actually it is 5.48 to 13.11V )but that did't work too, I'm using 16f628, the IC prog detects andreads it too ( of course ti is blank!) and write it too but on verification shows failure at h0000. I used 100uF/63V insted of 40V and 22uF/25V insted of 16V but I don't think that can cause problem or can it? (Exact values of capacitor voltage is not available!) as allready said.