Searching \ for '[AVR]: [EE]: Reset problems - 8535' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=reset+problems+8535
Search entire site for: '[EE]: Reset problems - 8535'.

Exact match. Not showing close matches.
PICList Thread
'[AVR]: [EE]: Reset problems - 8535'
2001\03\12@132958 by Attila Muhi

flavicon
face
Hi
I'm using an LS8535 qfp at 3.3V. I get reset problems from time to time, sometimes when I use the dongle fron STK200, sometimes when the application is running normally with the dongle disconnected.
I have a routine pointed to by the reset vector. It seems like the 8535 jumps to that routine on an external reset only when it wants to, not when I want it to :-<
When it goes wrong, it ends up with :
1) My reset routine not executed

2) PortC all pins high
3) PortB,2 high.
4) Rest of the ports tri-stated
5) clock osc running
If it does like this, I have to make about 15-20 reset pulses before it executes my reset routine. And, after that, the next 10-15 external reset pulses DO execute my vector. Then it is back to (ab)normal again. Of course the rest of the software doesn't work when the 8535 get stuck. However, after one of those successful resets, the software behaves excellent, rersponds to IRQ:s and whatever.
There's a .068 uF cap between /reset and gnd, and a 1k res between /reset and Vcc. Is this a bad idea ?
Any comments are useful !
Best regards
Attila Muhi - SM4RAN

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu


2001\03\12@161555 by David VanHorn

flavicon
face
>
>If it does like this, I have to make about 15-20 reset pulses before it
>executes my reset routine. And, after that, the next 10-15 external reset
>pulses DO execute my vector. Then it is back to (ab)normal again. Of
>course the rest of the software doesn't work when the 8535 get stuck.
>However, after one of those successful resets, the software behaves
>excellent, rersponds to IRQ:s and whatever.


This dosen't sound like a reset problem.

Do you have supply bypassing at the chip, and proper crystal caps?


--
Dave's Engineering Page: http://www.dvanhorn.org
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam@spam@mitvma.mit.edu


2001\03\13@115320 by Attila Muhi

flavicon
face
Hi !

I have ok crystal cap:s, around 20 pF. The leads from the crystal are as short as they can be, connected to the caps on their way.

DVcc is decoupled very near the chip, so is AVcc. AVcc is normally switched off, and is only switched on by a command from the LS8535 for power saving reasons.

Normally the circuit is in power down mode, with Xtal stopped. It is woken up from sleep by IRQ:s and then put back to sleep again after IRQ service. That works excellent - if the reset goes ok, and the cpu doesn't get stuck.
I realy have no idea of what the CPU is doing when it hangs after a failed reset. There's no entry at all in the code that could cause that behaviour.
Yes,the xtal osc IS running when the cpu is stuck. A nice, good-looking sine-wave, I checked it with the scope. 3.58 MHz by the way.

The dongle can always communicate with the CPU. The result of the reset from the dongle is similar to a manual reset.

I hope I haven't blown it with ESD !

REgards

Attila Muhi - SM4RAN


-----Ursprungligt meddelande-----
Från: David VanHorn <dvanhornspamKILLspamCEDAR.NET>
Till: .....PICLISTKILLspamspam.....MITVMA.MIT.EDU <EraseMEPICLISTspam_OUTspamTakeThisOuTMITVMA.MIT.EDU>
Datum: den 12 mars 2001 22:17
Ämne: Re: [AVR]: [EE]: Reset problems - 8535


{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\03\13@134737 by David VanHorn

flavicon
face
At 05:48 PM 3/13/01 +0100, Attila Muhi wrote:
>Hi !
>
>I have ok crystal cap:s, around 20 pF. The leads from the crystal are as
>short as they can be, connected to the caps on their way.

The values you list make me nervous..

www.dvanhorn.org/Micros/All/Crystals.php
Have a read through the above article, it may prove illuminating.
Check the crystal spec. With the values of cap you list, I'd expect a
loading cap spec of about 12-14pF.
This may not be causing your problem, but I'm a firm believer in getting
rid of anything questionable before spending hours of debug time.

>Normally the circuit is in power down mode, with Xtal stopped. It is woken
>up from sleep by IRQ:s and then put back to sleep again after IRQ service.
>That works excellent - if the reset goes ok, and the cpu doesn't get stuck.

There is no "reset" when you use wake on interrupt.
You just wake up, and start executing from the int vector.
The processor I/O is left in whatever state it was in when you went to sleep.
Are you forcing a reset with external hardware, WDT, or rjmp 0000?

Hint:  I always implement a full int vector table. It's a piece of library
code I make up for each CPU I work with.
Initially, all the ISRs do is turn their respective ints OFF.  As I build
the app, I replace those with actual ISRs for the ints I'm using.  That
way, if I ever accidentally enable an int, it's shut off after one hit, and
it's mostly harmless.


>I realy have no idea of what the CPU is doing when it hangs after a failed
>reset. There's no entry at all in the code that could cause that behaviour.

If it's hardware induced, the code may not even be relevant.



>I hope I haven't blown it with ESD !

A blown device could explain a lot of squirreley behaviour.
I try never to get down to one of any device. Three is my min. One in
proto, one spare, and one for emergency.
I don't touch #3 until I am as sure as possible that I have resolved
whatever killed #1 or #2.
--
Dave's Engineering Page: http://www.dvanhorn.org
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\03\14@040607 by Attila Muhi

flavicon
face
Hi!

>The values you list make me nervous..

Hope I don't cause any sleepless nights ... :-)

>
>www.dvanhorn.org/Micros/All/Crystals.php
>Have a read through the above article, it may prove illuminating.
>Check the crystal spec. With the values of cap you list, I'd expect a
>loading cap spec of about 12-14pF.
>This may not be causing your problem, but I'm a firm believer in getting
>rid of anything questionable before spending hours of debug time.

I will check the cap values. I printed your doc. But, when the circuit hangs, one of few things working seems to be the osc.


>There is no "reset" when you use wake on interrupt.

I know, I don't want it either.

>You just wake up, and start executing from the int vector.
>The processor I/O is left in whatever state it was in when you went to sleep.
>Are you forcing a reset with external hardware, WDT, or rjmp 0000?

I force reset with external hardware. I use the reset to initiate the circuits and set up certain values, and to go to sleep and wait for irq:s...
>
>Hint:  I always implement a full int vector table. It's a piece of library
>code I make up for each CPU I work with.
>Initially, all the ISRs do is turn their respective ints OFF.  As I build
>the app, I replace those with actual ISRs for the ints I'm using.  That
>way, if I ever accidentally enable an int, it's shut off after one hit, and
>it's mostly harmless.

That was a very good idea. Why didn't I think of that ?
{Quote hidden}

I will start off with that complete interrupt vector table. Then, if the problem still is there, I will put my second proto together. In fact, I have material enough for 10 proto's :-)

Thanks

Attila Muhi - SM4RAN

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...