Strange RESET pin behaviour on AT90S2313

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

Hello Fellow Freaks,
I've just assembled a very small board to be used as as generic controller "core" based on an AT90S2313. The circuit is very very simple as you can see in the picture below.

However, I have this strange behaviour: after I power it up, the AVR will only run for a few seconds and then stop. After the first run, I need to let it turned off for some time to be able to power it again and it will run for a few seconds only, again.

Doing some measurement I found that the reset line is having a "capacitive" behaviour. After I power it up, it will start with a few volts and then it will start to fall until it reaches a few tenths of millivolts. So the AVR enters reset and the program stops. It looks like there's a capacitor somewhere that charges slowly (the few seconds) and controls the RESET line. Odd, hun? In a first fixing attempt I removed the cap on the RESET line, but it obviously (now thinking) did no change.

Now, I would say that the chip probably got ruined during soldering, but I would like to know if someone else has any other theory...

Thanks!!

Attachment(s): 

Embedded Dreams
One day, knowledge will replace money.

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

Well, If the RESET pin were deffective, it would probably pull to VCC or GND rather rapidly - mili or micro seconds.

Resistor, R1, should pull the RESET pin to VCC very rapidly without the capacitor in the circuit. I'd be looking at the resistor, R1, the traces or wiring and, all soldering very closely to see if there is an open circuit in the wiring leading to the RESET pim.

I had an issue a few days ago where there was a crack in the PCB trace, right where the trace and the solder came together. I couldn't see the crack with the naked eye but, I could see it with an Ohm meter.

If the RESET input is floating to GND, I'd say that the RESET pin isn't really connected to anything - even though you might think it is.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

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

microcarl wrote:
I'd be looking at the resistor, R1, the traces or wiring and, all soldering very closely to see if there is an open circuit in the wiring leading to the RESET pim.

But if some problem led to RESET being left unconnected, wouldn't the internal 100K-500K pull-up resistor do the job (I'm at home)?

I usually put solder all over the traces, to avoid oxidation of the cooper. This also helps to avoid crask on traces. Of course I spend a lot of time in all of this, but it's all hobby use.

Well, I didn't check the pcb but I've found the problem now: power supply too weak. I was using 2 AA rechargeable batteries but I keep forgetting that these are only 1,2V each, and since they aren't all charged, they were giving only 2,2V. I guess that at this voltage and 6MHz ressonator the AVR simply doesn't work. I know it is out-of-spec, but I've used the same power supply with a Tiny26 @ 4MHz (crystal) and MMC without problems (hobby use, of course!).

Sigthly touching the ressonator with my fingers would stop the AVR, and it would not recover after I take my fingers off. Also, the AVR would stop when connecting the circuit ground to the oscilloscope ground, although my DMM said the difference is ~3mV. And sometimes it would run at a higher speed that 6MHz!
Nothing of this happens with the stronger (4V) power supply.

Thanks for your answer anyway Carl, its always enlightning to ear a guru's experience :). I was almost ready to pick the soldering iron and remove the chip.

Embedded Dreams
One day, knowledge will replace money.

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

Try a smaller resistor on the reset pullup, maybe 10k. May not solve the problem, but it narrows things down.

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

Thanks dksmall, but it's already solved as you can see above.
Low Vcc can cause strange behaviours...

Embedded Dreams
One day, knowledge will replace money.