Cold temperature boot up

[johnfielden]

Hi, on our board, the Tower, and the Phytec boards that we have in our possession, the Vybrid's part number start with a P.  We understand that this means the parts are pre-qualified parts, or pre-production parts.  Are the "P" parts temperature tested or binned?  Are they guaranteed to work over the Vybrid temp range of -40 to +85C?

I ask because we are seeing failures to boot at temperatures between -20C and 0C.  These failures occur on the Tower as well as on our own board design.

John

[xtrontec]

Something else to check when you are testing at -20C to 0C is that the board(s) are in a non-codensing environment - moisture causing shorting problems on your board etc. I am sure that you have this setup for your environmental test chamber and have it all sealed so that no moisture can get in etc. Just my 2 cents.

[xtrontec]

Maybe check the datasheet (specified operating temperature range) for the main 24MHz crystal used on the Tower System.

[johnfielden]

Says -40C.  So is the crystal on our target board.

[johnfielden]

Is it possible to get a response on the temperature question?  We're not sure if this is a real problem, or something likely to be seen with the pre-production parts.

Thanks,

John

[naoumgitnik]

Hello John,

The Prototype parts are not guaranteed, unfortunately. 

Based on our experience, what might help in this case is lowering value of the crystal oscillator loading capacitors to units of pF. Actually, due to the board stray and Vybrid pin parasitic capacitances, you may even try to simply remove them.

In this case, of course, the oscillation frequency will be higher by a few tens of PPM than nominal, but on this stage it might not be critical.

Regards, Naoum Gitnik.

[johnfielden · Accepted solution]

Thanks for the response.  We will wait till we get production parts to do further testing at cold temps. 

[johnfielden]

Production parts fail too.

[naoumgitnik]

Discussion continued in about Errata 7293 | Freescale Community.