Errrr, that means that there is a small chance that bad chips do not show as bad while programming. This could happen if the bootloader (cog0?) can read/write with no problems from/to RAM but a different cog couldn't. Bad chips showing up while programming are annoying but this can be easily corrected. Bad chips failing at the customer can get really expensive, though. This is very unlikely, I know, as we run an extra test that executes the actual software under real-world conditions (with multiple cogs).
Yes, that's right.
I thought I had written a stand-alone Main RAM tester a while back (extracted from our internal test jig code) that I could share, but I had only done so for the Cog RAM portion of the test. This thread makes me want to dive in and do the same for the Main RAM tester too.
Jeff,
I posted code to test the hub ram earlier in this thread. It checks all hub ram locations and is totally cog resident, including the serial routine.
A followup. I sent the problematic boards back to Parallax and they ran the tests and verified the error, both on my boards and on the chips after they had been pulled off the board. They put on new chips and returned my boards and now they work fine.
One thing I learned is Parallax tests 100% of the chips that come in from the factory and some small percentage do fail their exhaustive test procedure that involves everything from the memory and pin tests to power consumption and leakage. Maybe they learned something from these particular chips and ideas for the test procedure--but the failure mode remains sort of a mystery. The procedure involves moving chips manually into the tester, so there is chance of human error, and that is how we will chalk this up. Thanks Parallax.
Jeff told me the ZIF test socket that they use is the FPQ-44-0.8-19. Those usually cost around $40, but I found an offer on Ebay for $17 in case anyone else is interested in picking one up. I'm making a test carrier for it.
Some weeks ago we again had some boards not functioning at all. We soon found out that these boards had the propeller chip mounted in the wrong orientation. I have a very old pick&place machine and my wife blamed it on that. But I had some doubts. It surely is not very reliable but it does not randomly rotate 5 out of 100 propellers. The machine would either do 100% right or 100% wrong. I'd rather thought somebody has dropped a few propellers and put them back into the tray in the wrong orientation.
But today we changed the tray and put a new one in. I swear we took it directly from the sealed bag and surely didn't drop or even touch anything. We bought it from Mouser. Date code is 1727.
Spotted 4 in opposite orientation. Might have missed a few. If you have more trays it might be a good idea to see if the problem chips are in the same locations or a random scattering.
Comments
Yes, that's right.
I thought I had written a stand-alone Main RAM tester a while back (extracted from our internal test jig code) that I could share, but I had only done so for the Cog RAM portion of the test. This thread makes me want to dive in and do the same for the Main RAM tester too.
I posted code to test the hub ram earlier in this thread. It checks all hub ram locations and is totally cog resident, including the serial routine.
One thing I learned is Parallax tests 100% of the chips that come in from the factory and some small percentage do fail their exhaustive test procedure that involves everything from the memory and pin tests to power consumption and leakage. Maybe they learned something from these particular chips and ideas for the test procedure--but the failure mode remains sort of a mystery. The procedure involves moving chips manually into the tester, so there is chance of human error, and that is how we will chalk this up. Thanks Parallax.
Jeff told me the ZIF test socket that they use is the FPQ-44-0.8-19. Those usually cost around $40, but I found an offer on Ebay for $17 in case anyone else is interested in picking one up. I'm making a test carrier for it.
have you received the broken propellers? I've sent them the beginning of january.
BTW, find the error in this picture:
But today we changed the tray and put a new one in. I swear we took it directly from the sealed bag and surely didn't drop or even touch anything. We bought it from Mouser. Date code is 1727.