|
|
| Author |
Message |
everest
Joined: 31 May 2010
Posts: 96
|
|
Posted: 09 December 2010, 5:12 AM Post subject: Pulse Signals Jumping Pins |
 |
|
Don,
I'm trying to drive some servos with my ZX24r processor and I'm seeing some thing strange on several pins. Here's an example:
| Code: |
Call Pulseout (9, 1500.0 * 0.000001, 1)
'Call Pulseout (10, 1500.0 * 0.000001, 1)
|
This code produces nice 1.5Ms pulses on Pin 9 as I expect. What I did NOT expect was a "phantom" pulse on Pin 10 which is exactly what I'm seeing. I've attached a snapshot from my Oscilloscope showing the issue. Note that I'm running the code above, and I'm clipped into Pin10 on my ZX24r.
I can uncomment the Pin 10 line in the above snippet and then I get a double pulse on Pin10 (and on Pin9 as well). I've shown a shot of that below too.
Any ideas on this? I'm totally stumped. . .could I have a damaged chip somehow?
-Jeff
| Description: |
| This shot has pulses running on both pins 9 and 10. Note that you can see the generate pulse on Pin 10 here (where I'm clipped) but you can also see the Pin 9 pulse. |
|
Download |
| Filename: |
ZX24r Pin Bleed #2.jpg |
| Filesize: |
216.43 KB |
| Downloaded: |
1326 Time(s) |
| Description: |
| Here's a O'scope shot with Pin10 supposedly doing nothing. |
|
Download |
| Filename: |
ZX24r Pin Bleed.jpg |
| Filesize: |
214.22 KB |
| Downloaded: |
1326 Time(s) |
|
|
| Back to top |
|
 |
spamiam
Joined: 13 Nov 2005
Posts: 689
|
|
Posted: 09 December 2010, 14:15 PM Post subject: |
|
|
Well, the unconnected pin's (10) signal is not the same frequency as the one getting the pulseout (9). The unconnected pin appears to have 60 hz power line hum.
Hook up a lead to the power supply and see if it matches the signal on pin 10.
Put a big electrolytic cap near the ZX (across the VCC and GND) and also add a couple 0.1uF ceramics in parallel with the electrolytc. Does that reduce the noise on pin 10?
-Tony
|
|
| Back to top |
|
|
twesthoff
Joined: 17 Mar 2006
Posts: 199
Location: Fredericksburg, VA
|
|
Posted: 09 December 2010, 15:33 PM Post subject: RE: Pulse signals jumping pins |
|
|
| You did not say if the pin you were measuring was connected to anything. If it was floating free I would expect to see what you are getting. Temporarily connect a 10K resistor (or any other similar value) from the pin to ground and see if the phantom signals go away.
|
|
| Back to top |
|
|
everest
Joined: 31 May 2010
Posts: 96
|
|
Posted: 09 December 2010, 16:31 PM Post subject: |
|
|
Sorry, I should have mentioned that I did hook up a pull-down resistor to the floating pins and still saw the same behavior. The issue is not with the power supply, it's directly related to the other pin, I can change the timing on the "phantom pulse" but adjusting the other pin. VERY strange.
UPDATE: As of this morning the entire chip has died. I left it running a series of tasks all night last night, and this morning it is totally DOA. IT gets extremely hot and pulls a lot of current so I suspect something is shorted out internally???
Don,
Is there any way to repair one of these? I hate to drop $50 for another one, particularly when I'm not really sure why this one decided to go belly up. I've tested/checked my schematics, used ZX24p processors with no issues. . .and last night the chip actually wasn't even connected to anything at all, I dropped it right into a development board with nothing hooked up, just letting it run several tasks. Quite a mystery from my perspective. . .maybe static zapped it somehow???
|
|
| Back to top |
|
|
spamiam
Joined: 13 Nov 2005
Posts: 689
|
|
Posted: 09 December 2010, 17:23 PM Post subject: |
|
|
| everest wrote: | it's directly related to the other pin, I can change the timing on the "phantom pulse" but adjusting the other pin. VERY strange.
|
But I don't think it was directly related! The frequency of the switching on pin 9 was between 72 and 85Hz and the "phantom" signal was 59.7 and 59.8 Hz. That is quite a bit different to me!
It looked to me that they were completely unrelated....
I have only killed AVR CPUs by miswiring them and pulling a lot of curent through a pin, or by having nasty spikes, like inductive spikes hitting the uC.
-Tony
|
|
| Back to top |
|
|
dkinzer
Site Admin
Joined: 03 Sep 2005
Posts: 2593
Location: Portland, OR
|
|
Posted: 09 December 2010, 17:30 PM Post subject: |
|
|
| everest wrote: | | Is there any way to repair one of these? |
One could de-solder the 44-pin TQFP chip and replace it but that would require a special-order chip.
| everest wrote: | | I hate to drop $50 for another one [...] |
If you'll send back that one for analysis, we'll send a replacement. Send it to the address below (First Class is fine if well-packaged).
Elba Corp.
10915 SW Avocet Ct.
Beaverton, OR 97007
|
|
| Back to top |
|
|
Don_Kirby
Joined: 15 Oct 2006
Posts: 329
Location: Long Island, New York
|
|
Posted: 10 December 2010, 1:07 AM Post subject: |
|
|
| everest wrote: |
UPDATE: As of this morning the entire chip has died. I left it running a series of tasks all night last night, and this morning it is totally DOA. |
While the excessive heat issue is certainly something to be concerned about, have you tried an emergency update using the techniques specified in section 9.1.1 of the Language Reference?
|
|
| Back to top |
|
|
everest
Joined: 31 May 2010
Posts: 96
|
|
Posted: 10 December 2010, 4:46 AM Post subject: |
|
|
Thanks for all the replies, and Don I'll ship out the chip tomorrow. I'd be very interested if you could let me know what the issue was/is, and particularly if it could have possibly been anything I did to the processor.
I have a hard time with my static electricity theory given that my ZX24P took a moderately close proximity lightning strike without batting an eyelash.
So I'm just stumped. Fortunately I've got a few 328n chips with interface boards lying around, plus a spare ZX24p. . .so the fun goes on!
-Jeff
|
|
| Back to top |
|
|
|
ZBasic Home
Forum Rules
Memberlist
Usergroups
RSS Feed
Site Search
Links
Download
Digests and Subscriptions
Profile
Log in to check your private messages
Log in
Register