PIC programming in low-voltage board

[jackab]

I'm trying to arrange ICSP for PIC16F914.

My board is powered by 3.3V, while PIC's programming voltage is 4.5-5V. This means I need to isolate normal CPU VDD from VDD coming from the programmer (ICD2).

I can feed VDD via a schottky but this means I will drop 100-200 mV on VDD during normal operation. The voltage drop will also change all the time since it's current dependant.

It seems very stupid to me that I need 5V on VDD to program PIC which was designed to run on low voltages (2-3V).

Any ideas how can I isolate VDD on board from 5V_VDD coming from the flash programmer without using this diode?

 

[Jay.slovak]

I'm trying to arrange ICSP for PIC16F914.

My board is powered by 3.3V, while PIC's programming voltage is 4.5-5V. This means I need to isolate normal CPU VDD from VDD coming from the programmer (ICD2).

I can feed VDD via a schottky but this means I will drop 100-200 mV on VDD during normal operation. The voltage drop will also change all the time since it's current dependant.

It seems very stupid to me that I need 5V on VDD to program PIC which was designed to run on low voltages (2-3V).

Any ideas how can I isolate VDD on board from 5V_VDD coming from the flash programmer without using this diode?

You can't program PIC @ 3.3V with ICD2 due to different voltage levels. It's stated in ICD2 manual. PERIOD.

Maybe you can place a simple jumper to switch VDD to 5V and MCLR to ICD durring program time.

 

[jackab]

no problem, I undestand that neither PIC16F914 nor ICD2 support low-voltage programming.
I'm just trying to build a low-votage board which can be programmed kind of a plug-and-play way without messing with jumpers during dubug and maintanace.

 

[Jay.slovak]

But you can try it, maybe ICD2's inputs are 3.3V tolerant and the LOW voltage PIC's diodes will clamp the voltage from 5V to 3.3V and it will work. Ofcourse PIC will have to be board powered and something should also be done do MCLR circuit...

 

[Jay.slovak]

And read this http://forum.microchip.com/fb.asp?m=105818

 

[jackab]

i just checked programming specification where it is said:

1. High programming voltage is generated internally.
2. Read/write operations voltage 2.0-5.5V; bulk erase voltage 4.5-5.5V

It is interesting that flash erase voltage is much higher than the programming one.

I'll try to check it tomorrow with the micro tech support.

 

[Jay.slovak]

i just checked programming specification where it is said:

1. High programming voltage is generated internally.
2. Read/write operations voltage 2.0-5.5V; bulk erase voltage 4.5-5.5V

It is interesting that flash erase voltage is much higher than the programming one.

I'll try to check it tomorrow with the micro tech support.

OK, according to your info it is impossible to program PIC below 4.5V. Interesting :roll: ...

 

[Mike - K8LH]

I suspect one of the problems programming PIC devices at 3.3v is that software for most programmers wants to perform a bulk erase/chip erase operation (which requires 4.5-5.5v) before writing code, data, idloc, and config areas on the target...

My melabs Serial Programmer has a "low voltage erase" option which will use the "row erase" sequence instead of bulk erase/chip erase sequence, but this "row erase" capability is not available on all PICs...

Regards, Mike

 

[Exo]

Feed the pic trough a fet with a low RDSon.
if you use a fet with only a few milliohm resistance then almost no voltage will drop over it. For programming just design a cicruit that opens the fet when you plug your icsp plug in.