What happened to the buffered usb mod we were thinking of?
As I recall the VUSB was too close to +5 for 78L05 to work. He would need a LDO (Low Drop Out) regulator whic is most posssibly another part he can not get. If you want to use a 5V regulator I suggest you use it with an aprox 9V power source, be sure to use caps as suggested in the datasheet. If you can get a 5V LDO regulator that would work. I like the LM2931AZ5 which has a dropout of .2V at 10mA. It can be had in .1 and .6 amp versions.
The nice thing is that this part will deliver 4.8V with an input of 5V. A lot of (most) regulators will not do that.
I agree with mvs sarma that you need to have a VUSB close to 5V. Without that everythis else is suspect.
This is my understanding of how the charge pump works. If I am wrong I would
like to be corrected
The 18F2550 works between 2 and 6 volts, the extra half volt if it exits may not be a problem for the uC. But it may cause VDD to be too high.
The first thing to check for with a VPP short is the transistor connected to the inductor. It will not work if it is installed backward. I my case the datasheet supplied by the place that sold the transistor was wrong. You need to check it with a meter.
The PIC provides pulses on the VPP_PUMP pin. This causes the voltage in the inductor spike/jump above VDD each time a pulse happens. The diode D1 acts as a valve to charge the capacitor C1 with the + side of the pulse.
The PIC reads the voltage on pin VPP_FB (VPP_FeedBack( via the voltage divider formed by R6 and R3B. I do not recall what divider voltage maps to VPP 12V.
If the PIC see too little voltage on VPP_FB it causes VPP_PUMP to increase in frequency. When it see to little voltage it lowes the frequency. If it can not get the voltage to 12V it reports a short. The reporting of a short is misleading in that anything that prevents 12V will be called a short.
Check resistors R6 and R3B. If they are wrong VPP_FB will see the wrong voltage when 12V is right.
Check the direction of D1. If it is backward C1 can not charge.
I have been thinking about disabling the charge pump and placing 12V on the wire between R3B and the cathod of D1. If the voltage divider is working it should report a VPP of 12V. If VUSB is too high (6.5V) the internal voltage referance used by VPP_FB will be off. ....
Enough for now.
3v0