Kirk Humphries

Oscillator pb with mc9s12a256

Discussion created by Kirk Humphries Employee on Jan 29, 2006

This message contains an entire topic ported from a separate forum. The original message and all replies are in this single message. We have seeded this new forum with selected information that we expect will be of value to you as you search for answers to your questions.

 

Posted: Sat Sep 24, 2005  7:29 pm

 

Hello,

 

I'm the friend of ________ who posted the former message about the oscillator pb with a mc9s12a256.

 

Actually, all signals were tested and are correct except of XTAL and EXTAL which are 0V. So the oscillator doesn't work. What could cause that? We use the Pierce oscillator with pe7=0 and capacitors c1=c2=22pf and R=4.7Mohm and crystal=16Mhz.

 

Is there a position to respect with the two capacitors around the crystal? I mean, should we place physically the capacitors parallel with the crystal on the circuit? Because it's not like that on our circuit... Is that a pb?

 

Another idea : for a first use of this microcontroller, we think that the oscillator should run before we download a program. Is it a wrong way to think that? Maybe, the oscillator doesn't run before...?

 

And last idea : there is a WAIT mode for this microcontroller. Is it possible that our pb is related to that mode?

 

Actually we're searching for what could cause that pb in regards with all correct signals we can test and we don't know if it doesn't work because of the microcontroller or because of the oscillator...

 


 

Posted: Sat Sep 24, 2005  10:07 pm

 

> crystal? I mean, should we place physically the capacitors parallel

> with the crystal on the circuit? Because it's not like that on our

> circuit... Is that a pb?

 

It is fairly critical, keep distances short and do follow the recommendations about earth wires are the main things. Also some scopes put too much load on for it to work, try picking it up by holding the probe close first of.

 

> Another idea : for a first use of this microcontroller, we think

> that the oscillator should run before we download a program. Is it a

> wrong way to think that? Maybe, the oscillator doesn't run before...?

 

Nope, nothing happens when there is no clock.

 

> And last idea : there is a WAIT mode for this microcontroller. Is it

> possible that our pb is related to that mode?

 

On first startup without a program loaded Osc should just run.

 

> Actually we're searching for what could cause that pb in regards

> with all correct signals we can test and we don't know if it doesn't

> work because of the microcontroller or because of the oscillator...

 

Many things can go wrong, one common mistake when you lay your own board is to connect the power supplies incorrectly, especially watch that of these are powered from inside the chip, do not connect these to external power.

 

Cheers,

 


 

Posted: Sat Sep 24, 2005  11:35 pm

 

> We use the Pierce oscillator with pe7=0 and capacitors c1=c2=22pf

> and R=4.7Mohm and crystal=16Mhz.

> Is there a position to respect with the two capacitors around the

> crystal? I mean, should we place physically the capacitors parallel

> with the crystal on the circuit? Because it's not like that on our

> circuit... Is that a pb?

 

What's the distance from crystal/capacitors to XTAL/EXTAL pins? Try with some other crystal, especially different (lower) frequency.

 

> Another idea : for a first use of this microcontroller, we think

> that the oscillator should run before we download a program. Is it a

> wrong way to think that? Maybe, the oscillator doesn't run before...?

> And last idea : there is a WAIT mode for this microcontroller. Is it

> possible that our pb is related to that mode?

 

Oscillator does run "before". It runs when the -RESET line is active, so you can pull -RESET low and be sure the clock must be there.

 

> Actually we're searching for what could cause that pb in regards

> with all correct signals we can test and we don't know if it doesn't

> work because of the microcontroller or because of the oscillator...

 

As an ultimate test for problem location, apply clock signal from external generator, and check if microcontroller works properly then.

 

Regards,

 


 

Posted: Sun Sep 25, 2005  12:05 am

 

Hi,

Don't forget the XCLKS signal.

 

Check the documentation for the correct polartiy, but on the C32 it should be low at reset to select the Pierce. It has a pullup on the pin. Also the documentation often talks about the XCLKS signal, which may be a different polarity to the pin voltage level !

 

I have just solved a problem on a bunch of production PCBs. Whereas the prototypes had the previous mask set, the newer chips had a weaker pullup on the XCLKS pin, and so it was trying to start my colpits in Pierce mode. I had to put an external pullup.

 

Just a small "Gotcha".

Good luck.

 


 

Posted: Sun Sep 25, 2005  8:56 am

 

Hello,

 

I have read datasheet of A64 A128 and A256. This is not the same things on XCLKS!!!

On the A64 and A128 if XCLKS=0 i have pierce oscillator. But on a A256 the datasheet don't speak about pierce oscillator????

 

In fact we should use a A128 but we had problem of provisioning and we put A256 on the board. According to you is it possible to run the A256 with pierce oscillator? if yes what should be the state of XCLKS?

 

Thank you for help

Outcomes