new mega88pa low voltage startup issue ?

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

New Mega88AP with entirely default fuses (no BOD, 1Mhz etc). 10k pullup to VCC on the reset line.

+ If I run it at 5V, all is well
+ if won't run if powered from a stable 2.0V/120mA buck.

+ If I reset the AVR at 2.0V, even repeatedly, it doesn't start.
+ If I ground the reset line, apply 2.0V power, and release the 2.0V reset line, the AVR doesn't start either.

+ If, with the VCC power on, stable at 2V (and the AVR not started as mentioned) I prode the reset line with 5V, the AVR starts (without even puling the pin down/up!)
+ Any further reset (with the normal 2V line) works fine once the AVR is running.

As mentioned, I don't have a BOD, and I tried with two different AVR. same behaviour..

I just don't understand, it's like that AVR is "hung" in power-on-reset, despite being given .2V over it's spec.

The code on the AVR is tiny. It disables most peripherals in the PRR, first thing.

Anyone see that before ? tried a LOT of things already (caps on reset etc) with no differences...

Author of simavr - Follow me on twitter : @buserror

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
I don't have a BOD
But you have tried everything else why not that? :)
What do you have connected to the I/O pins? ...and yes I'm only trying to get the conversation started..no experience with anything lower that 3.3V.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Err well, I bricked a 88pa by trying to reprogram the hfuse. no explanation as to why, but I suspect the ISPEN fuse went too accidentaly :>
So this is a second, reworked in 88pa here, I'd rather mention the problem here before I risk frying another one :>

I have 2 PWM pins running on timer 1, very low duty (6%) to run a couple of red/orange leds trhu a 100ohm resistors each. Thats all that is plugged in.

AVR had a 32k crystal, and the timer 2 is running of it, at very low rate (20hz) for doing precisely nothing but wake the main() between sleep()

Author of simavr - Follow me on twitter : @buserror

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There have been some reports of code not starting with slow rising VCC.

Can you try it with say 3.3V? Would anything get upset by doing that?

That could rule out any code problems I guess.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

When you say "won't run", can you still get contact with ISP?

Are >>all<< Vcc, AVcc, Gnd, AGnd pins hooked up?

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Err well, I bricked a 88pa by trying to reprogram the hfuse. no explanation as to why, but I suspect the ISPEN fuse went too accidentaly :>

Tell more about this attempt. With ISP? Then it is impossible to accidentally or purposefully change that fuse. debugWire? Then use it to change it back. HVPP? Then use it to change it back.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think I found the problem: I didn't have AGND/REF/VCC hooked up. I assumed that since I was disabling that block and not using the ADC nor BOD I was "fine" -- turns out it's "optional" in higher voltages, but mandatory getting lower...

The ISP problem is probably due to my ISP "custom" programmer. I use a ftdi ft245 to do the programming, and the could be a problem with the fuse programming code...

I'll continue testing...

thanks theusch! found it all by myself in the end, but you got it right none the less :>

Author of simavr - Follow me on twitter : @buserror

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

I didn't have AGND/REF/VCC hooked up. I assumed that since I was disabling that block and not using the ADC nor BOD I was "fine" -- turns out it's "optional" in higher voltages, but mandatory getting lower...

Not really. There is a minimum voltage to get out of POR. There will be some leakage running around in the chip, plus a "floating" pin can have any value. Applying higher than Vcc to the /RESET pin (which has no protection diodes) can cause who-knows-what levels.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.