$50 fried brains and servos coming up

[salixfire]

Hi to every one here. I'm trying to build the $50 robot and seem to be having a few difficulties. I have built the circuit and have checked all the voltages and they are all correct. I started modifying the servos (Hitec 311) and the first servo seemed to work correctly. Using the hold_servo the gears turned until I set the pot, glued it in place and put it all back together. The second servo I opened up and started the hold_servo program. The gears rotated once then came to  stop without me touching it and even if I rotated the pot it stayed still.
  I then programmed the Atmega8 to the photovore code that I edited so that the servo test code was used instead. The servo rotated a couple of times then stopped and made a slight buzzing sound. I thought that due to the i=250 part it would only rotate like that so I edited the code so that the servo should rotate continuously but again it stopped after a couple of rotations and then even turning the power off and back on it wouldn't rotate. I tried reprogramming the chip and suddenly the AVR studio said that the programming failed and the signature didn't match. I checked the voltages and everything seems fine. I don't have access to an oscilloscope so I can't check the signals.  So do you think I've fried my servos and microcontroller? Is there any way to salvage them?
Thanks to everyone in advance. There's no real hurry for answers as I'm in no rush. So far fate has been against me doing electronics these past few weeks as even the Sparkfun embedded tutorial blew up when I tried setting the fuses for an external clock.  Ah well it's fun anyway. Thanks.

[pomprocker]

What doesn't kill you, only makes you a better roboticist   

look around the forum for signature errors, plenty of em.

[salixfire]

Yes indeed. Though technically I have quite a few years of experience with robotics as I have a MEng in cybernetics with 3 out of 4 major projects being robotics based.  I was trying this one out as I have never used servos and atmel chips before. It's surprising how much you can't remember after not doing this for a few years of not touching it. 

I tried searching for signature does not match, signature match and signature to see if there was any helpful posts before I posted this. The most helpful one was the person had a faulty chip. I also searched for help with the servos but didn't seem to find anything (unless I missed something) . Thanks for the fast reply.

[salixfire]

Well I received my set of new atmel168 chips today. I was wondering if anyone has a clue on what I did wrong/what to do this time before I go and fry 10 more chips and another set of servos? I'm not sure how to check on the servos to see if they are working and if it was only the chip or if the servos are busted as well. Oh and while I'm here could someone tell me what capacitors I should use for this crystal or how to work it out please. Thanks in advance.

[Admin]

Hmmmm it really sounds like a power problem, like a low voltage. Did the regulator heat up by any chance?

Oh and while I'm here could someone tell me what capacitors I should use for this crystal or how to work it out please

That xtal will work. According to the datasheet you linked, it requires 32pF caps.

[salixfire]

Thanks for the reply. No the regulator was fine, I checked the voltages when I turned it on and they were all fine. When the microcontroller stopped working I checked them again and still they were fine. I'll have a go again tomorrow with a new chip (remembering to alter the code for the atmega168 rather than the atmega8)
Thanks for the advice about the xtal. I had looked it up and new it had something to do with the load capacitance.

[salixfire]

Well I came back from a trip and tried to test my circuit and servos again to see if it was something simple as a low voltage. I plugged my modified servo in (the one I thought I fried) and plugged in my programmer. Checked the device signature and that came back fine, but this chip had been giving me signature errors the last time I tried it. So I reprogrammed the chip with the modified code again and it worked. Nothing had changed, everything was how I left it a couple of weeks ago. So I tried the servo that seemed to be broken when I used the hold_servo program and magically it worked this time. No idea what went wrong before but everything is working fine. Been over the entire thing to check for loose connections or shorts anywhere.

[Admin]

lol

I once had a circuit go up in smoke on me literally . . . came back a day later and not only did it still work, but I couldn't find any burnt marks either.

Never did figure out why . . .

[salixfire]

It's power gremlins  They're always messing around when you least expect them.