Hi all,

I have an interesting situation..

I had prototyped a circuit for a rocket altimeter on perfboard using a very basic set up. Basically a bare bones NK. I was unable to get the chip to program a simple LED blink code.

Notable information: - Running everything at 3.7 V Lipo battery - Only communication between PC and MCU was with serial TXD and RXD lines. - I have posted and seen posts of people being able to program the chip to about 3.6 volts, anything lower than that, the crystal cuts out.

I tried and tried, and looked for wiring errors, could not find any - in my frustration I ended up destroying all my work trying to find errors.

Then, I figured, I would try a fresh MCU with a known good NK bootloader installed and wire up the SAME simple circuit on a breadboard and give it a go...same problem. I get a butterfly error when trying to program.

Then, I tried to see if I could program using USB power...Bang, works every time.

Then, I thought maybe it was because I was using lower than 5 volt supply. I hooked up a separate 5 volt supply, got everything wired up, but I cannot get the chip to program. If I switch everything to USB, it works.

Any ideas on this? What should I be checking. I have never seen this problem before. I even tried to see how my new homemade dev board would handle this issue, it seems to be working just fine, even if I have a 3.6V supply.

I have to be missing something...

Thanks for your help!

Dave

Dave -

You said the only wires connected between the PC and MCU were TXD and RXD. How about ground? I think xodin had a similar problem and ended up using the internal clock and not the 14 Mhz crystal with 3.3 volts.

pcbolt,

Thank you! I can't believe I overlooked such detail like that. I went back and checked the original NK documentation and it has you connect the ground line from the programmer. I bet that is my problem. I am at work now, I will have to give it a try this afternoon/evening. I will get back to you with the results.

In response to the 3.3 volts and 14 Mhz crystal, I think you are correct, unless you change the fuses. However, I know you can get away with the 14 Mhz crystal at 3.6 volts. The limiting factor is the MCU/crystal, not the programmer, from what I understand.

Dave

Actually I've used 3.3v and 14 Mhz with no trouble at all. Guess it's just hit or miss.

Thanks again pcbolt,

Wow do I feel silly. Plugged in the ground to the programmer, and I can program using a separate power supply...oh well...chalk it up to a learning experience! Thanks for the quick response.

I will have to try putting my supply to 3.3 volts and seeing what comes of it. I have tried before and I don't think I have ever gotten it to work.

Have you tried 3.3 volts with the 14 Mhz crystal with more than one 168?

Dave