Problems with Hardware / Firmware (using v0.7.6) ?

[TobiZi]

Hello,

First of all - this is a great Project.

I built up a hardware circuit based on the "Open Programmer v0.7.6 Hard- and Firmware" with a PIC18F2550 Programmer and PIC16F887 as Target Controller.
The Firmware was left unmodified.

I run into some trouble.
The USB Voltage was temporary dropping down to <4,5V while initial generation of the 13V VPP.
It seems that this caused an abort of the generation process @ VPP=~9V.
Buffering the USB Voltage with a 100uF seems to help me with that problem, but now I have a big new one.

While I am debugging the Target PIC Device via the Programmer/Debugger PIC "Pin 13" of the Programmer ends up constantly HIGH what causes a short from +5V to GND via the 100uH inductor.
Using the OpenProg Software works fine, but if I press "Reconnect" the same happens. Pin 13 is HIGH constantly.

In which cases could PIN 13 stay high?
Has somebody any idea how to solve or prevent this?
Do I have to modify the Firmware ?

It would be nice if someone can help me out.

[Alberto Maccioni]

Hello,
one possible explanation is an abnormal load on VPP; you should carefully check the circuit for unwanted loads/shorts.
Also first try without target chip using the hardware test.
Pin 13 cannot stay high indefinitely, even if a load prevents VPP from rising to the proper level; there is a maximum duty cycle but you can only see it with an oscilloscope.

[TobiZi]

The hardware test is working fine except that Vpp is not correct.

I measured constant 5V DC with an oscilloscope @ PIN13 after pressing the Reconnect Button in OpenProg.

Sometimes The Condition is constant low @PIN13 but in most cases constant high. It seems that the last state/condition of the Capture Compare Output is "stored" at PIN13 after a "Reconnect". Does the firmware (initialisation) set a defined start condition for PIN13?

[Alberto Maccioni]

PIN 13 is low when PWM is off, so it could be that the state is forced from outside; try to remove the series resistance and see if it stays high (VPP will not work).

[TobiZi]

I measured again with an oscilloscope (see attachment).
The picture shows PIN 13 Output with series resistor not distributed.
5V VCC is supplyed via USB connection.

I connected to the programmer (pressing Reconnect), loaded some hex and clicked on write in OpenProg.
In the left picture the result: normal PWM at max dutycycle because VPP is not reached.
After that I pressed Reconnect again: The right picture shows constant 5V.
Nothing is connected to Pin13 right now and there is no shortcut.

Any ideas ?

Edit: I am using OpenProg v0.7.7 and unmodified programmer Firmware v0.7.6

ReEdit: I changed the series Resistor from 10k to 4k7 and now the error discribed above is not occuring. But I don't realy know why.

[Alberto Maccioni]

When you press reconnect the programmer executes a reset command, which disables the pwm block; it's not obvious from the datasheet, but it may be that the output stays in the same condition, so high if it was high. I will check.
Anyways I think that the components used in your DCDC section require more base current for the transistor to work correctly, so with 4.7k everything works correctly.
I was myself going to lower the resistor value because the circuit is more robust and has better performance.