I have an original PICkit 1 and I measure the waveform of the boost 
converter output and the MCLR pin waveform. I think the design of
the original firmware is flawed or maybe it is the limitation
of the USB requirement.

If you measure the output of the boost converter. You will find that
It also has the dip you noticed on the Vpp/MCLR pin.
The dip is not because of the turn-on of Vpp (i.e. turn of Q3 and
Q4) or because of lousy regulation but because of the +13V itself.

The reason is that it is normally at 4.8V (Vusb minus some drop) since 
Q2 is always off. Then when the host software issues a Read/Write command, 
the boost converter starts to generate the 13V needed. Then you have
the settling time of the boost converter (overshoot and undershoot) and
thus the dip.

The correct way is to generate 13V constantly. The potential problem 
is the slightly higher current consumption when there is no read/write
command. Another possible solution is to have longer delay time after
starting the PWM.

Regards,
Xiaofan



-----Original Message-----
On Sat, 2005-07-09 at 08:06 +0800, WH Tan wrote:
And now back to a previous discussion...

It seems like the 'boost converter' design is almost the same as 
PICKit 1. Then I wonder the firmware must be done some tweaking to 
achive the 'acceptable' level of 'fluctuation' when a load is 
connected, as if it is to 'support' most of the flash parts. I believe 
my firmware output is the same as what the real PICKit 1 does. And 
here is what happen when the Vpp transistor was turning on.

Best regards,
WH Tan 
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist