RE: The MCF52230 issue

  Since one out of six boards works intermittently, as a 1st pass assumption
let's rule out wiring error.  There may be assembly errors on the other
five, but one out of six seems like a high assembly error rate.  I'm sure
you have this covered, but verify that each power pin really connects to
power.  Sometimes things work with a power pin left off as power is fed
through the various protection diodes for the pins.

  The only thing I can see that is different between your schematic and the
demo board is the bypass capacitors.  The PLL bypass as a .001 uF (1000 pF)
on it.  The analog supply has a 10uF tantalum on it.

  In general, my bypass regiment tends to be board level bulk (low ESR
100uF), .1uF and .001uF.  The two chip caps cover a broad chunk of
bandwidth.  I've even put 100pF caps on things like SDRAMS and one or two by
the processor (on a 5282), particularly if it's not a multi layer PCB.
Another advantage of the dual cap approach can be seen if you take a
spectrum analyzer and put it on some of the wire coming off your board.  The
emissions aren't as bad.

  And finally, because of the nature of surface mount bypass caps this is a
very quick and easy thing to try to do.  You can put a .001 uF cap on top of
your .1's and see what happens.  You may also look at the power supply noise
with a scope and see if you can see any difference at the highest speeds.
You can also look at the power supply noise with a spectrum analyzer and see
if it's cleaner.
  I know most people that do digital design don't have spectrum analyzers.
I've got on because I do a lot of RF work too.  With EMI concerns, I'd claim
it's a tool that pretty much a requirement anymore.  The win is that once
you have one, you can look at power supply noise from two different "angles"
using your scope and your spectrum analyzer.

  So take an hour and solder some .001 caps on your .1 caps.  If the one
flaky board runs better, you're on the right track.  If there's no change
you've only lost an hour.
  I'm had some serial port problems on a HCS08 that I haven't found a cure
for in two years!  So I'm changing all the HCS08s out to 52233s.   It will
be the "little" micro in the toolbox so that all things big and small can be
on the network. So when you find the problem let us know!


  As a comment/question to the group:

	I'm moving away from PCs (after a decade of windows, I've just had
enough).  I've got a iMAC and am playing with it.  It will at least take
over email and documents.  Is there a compiled version of gcc for the MAC
out there?  I've also got to get a linux machine going to test software for
the products I'm working on, and could just migrate everything to that
machine.  If this is a better, easer approach, than I'd just get a bigger
platform for linux.  I guess I'm hoping that the MACs unix underpinnings
would make it equivalent to a linux box.  I like the idea of paying a bit
more for the hardware, but it all comes from one source even though I find
benign dictatorships distasteful.  I fully understand that this question is
a mix of aesthetics, religion, and politics and that my mileage may vary.

 - pete

> -----Original Message-----
> From: [email protected] [mailto:[email protected]]
> On Behalf Of Sanjay Morab
> Sent: Tuesday, February 05, 2008 7:41 AM
> To: [email protected]
> Subject: [ColdFire] The MCF52230 issue
> 
> Dear Mark
> 
> This is the extension the earlier problems of MCF52230. We had developed
> the
> all software on MCF52235 EVB and is working fine. We had made our board on
> MCF52230 instead of MCF52235. The code is well with in limit. When we
> tried
> to download to SRM it gets down loaded but does not execute. Secondly we
> had
> tried to download to the flash, it it down loaded but does not execute. I
> had cheked all votages ON VDD they are proper, The crystal frequency is
> also25Mhz. Reset is OK. I do not know what is the matter.
> 
> Out of 6 board made only one board is working that is also intermitantly.
> 
> I am attaching the pdf of the schematic please go through the same and
> tell
> me is there any thing missing
> 
> regards
> Sanjay Morab


---
[email protected]              Send a post to the list.
[email protected]        Join the list.
[email protected]    Join the list in digest mode.
[email protected]     Leave the list.