Start up of microcontroller fails

Hello,

I use a Infineon C509 controller on a self-developed microcontroller board.

From time to time (quite rare) it happens that the microcontroller

doesn't start. Even resetting the microcontroller doesn't help !!! I have to remove power supply and reconnect it to get the microcontroller running.

Only a few components on the microcontroller board effect the startup process of the board:

- microcontroller C509-LM

- Address Latch (74HC573)

- Flash (AT29C256) (containing the program to run)

- GAL (Atmel 22F22V10) (doing some chipselect signals)

Every component has enough capacitors on the supply pins. When the microcontroller doesn't start, the oscillator works.

What I want to ask you is if you have any experience with the electrical limits of one of these components. Maybe you had some similar problems at power up, too. Right now I have no glue where to start....

BEST REGARDS

Christian

Reply to
Christian_ESP
Loading thread data ...

Christian_ESP skrev:

Remenber to pull the /HWPD and /reset up try 1kohm.

--
Hilsen Mikkel Lund
"Sund fornuft, har aldrig stoppet en tosse"
 Click to see the full signature
Reply to
Mikkel Lund

Does this processor have an /EA line to access external program? Is it tied down?

Reply to
martin.shoebridge

I'm not familiar with the C509.

If using a crystal or resonator, you might want to check the amplitude. Possibly adjusting the capacitance to ground. I recently had this exact same problem on an Atmel 89S8253. Datasheet said 33pF. Reality was it only needed 5pF. About a month later, I came across their revised datasheet....

Good luck.

Christian_ESP wrote:

Reply to
mpm

....

Lots of things could go wrong. Try to find something that makes things worse (!) You have checked all your timings during operation with a scope, I assume ...

Some devices/systems are sensitive to the Power On waveforms, rate of rise or non-monotonic behaviour.

Processors typically have a delay to allow a clock to stabilise - sometimes this process takes a suprisingly long time

Perhaps the timing on your flash chip/gal is marginal, so the uP gets fed junk

Other things to try - raising and lowering power supply rails, heating or cooling one chip at a time. Sometimes removing bypass capacitors reveals a sensitivity

Lots of fun !

Dave

Reply to
Dave

I don't use Infineon anymore but I'd check these things: Place reset on one scope trace, VCC on trace 2, oscillator on trace 3. Trigger on rising VCC, 25% of full nominal or so, long time base. You may not be able to touch the oscillator pins but taping the probe near it usually shows something on the scope. Now look whether reset holds longer than it takes for the oscillation and VCC to come up to speed. Pay close attention to how VCC comes up. Bite your lips and try not to cuss when you see that reset.

BTW the usual RC combination on the reset pin is one lousy way to do it. No matter what data sheets say I always provide an external reset of much better quality. An RC can fail and freeze the uC if VCC comes up but sags back a bit before coming fully up. For example because a hard disk is spooling up.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.