Pulling my hair out problem. Solved.

Bean

I had designed a production product for my company using the SX20.

When the product was built I received 35 pieces initially.

When I tested them they kept failing.

After some investigation, I found that the SX would run for about 10 seconds then stop. Hmmm...

I spent a whole day troubleshooting the problem. Pouring over code, installing a new SX chip, etc...

Come to find out that the assemblers had switch the 10K /MCLR pull-up resistor and the 0.1uF by-pass capacitor. The cap would keep the SX running for a little while, then it would go into reset after the cap was completely charged up. After powering down (the cap would discharge) the unit would operate again for a little while.

I'm telling you this story in hopes that this might help someone else if you ever run into a similar problem.

Now to go raise some·*%$#* ·with the assemblers. Arrrggg. At least it is an easy fix...

Bean.
·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
·The next time you need a hero don't look up in the sky...Look in the mirror.




Post Edited (Bean (Hitt Consulting)) : 1/30/2009 3:38:30 PM GMT

Shawn Lowe

Wow, good story bean. I can imagine just how much frustration that must have caused!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Shawn Lowe


When all else fails.....procrastinate!

Guenther Daubach

Bean,

I completely feel with you...

A couple of days, I also was close to loosing a lot of my hair.

For a customer's project I designed a simple circuitry around an SX20 with a 4 MHz crystal, and a 1 MOhm resistor parallel to the crystal. I also added the typical "Vss-Vdd-OSC2-OSC1" 4-pin header for the SX-Key/Blitz. Next, my customer took care of designing the PCB.

When I tried to program the first assembled sample board for testing, I got a "Vpp generation failed" error.

Usually, such an error comes up when some excessive load exists on the OSC1 pin. I double-checked the board but could not find any cause for such a load. Just to be sure, I removed the 4MHz crystal, and the 1 MOhm resistor - no success at all.

Want to know the cause for this problem? Well, it took me a while to figure out: The PCB designer interchanged the OSC1 and OSC2 pins on the 4-pin header - Arghhhh !!!

The good news: The on-board SX20 survived!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

$WMc%

Bean

I get nervous when everything works rite the first time!

With todays tiny traces and multi layered boards, I can see this being hard to find.


____Look in the mirror Mr.Bean_________________$WMc%______

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Truth is out there············································ BoogerWoods, FL. USA

PJMonty

Bean and Guenther,

I did a project a few months ago where I used an SX-28 . It was the first time I had used a surface mount SX-28, so the first thing I did was check to see if the pinout was different. A quick look at the datasheet confirmed it was identical, so I could use my existing schematic part and just specify a different pad layout in the PCB program.

When I got the board back, everything was going great until I soldered in the SX-28. Suddenly, the LED on the PCB dimmed and the SX-28 got extremely hot almost instantly. I checked and re-checked, but the schematic and layout seemed good, and there were no solder bridges or shorts in the PCB. I removed that SX-28 and soldered a new one on, thinking I had a bad part.

Same problem.

After a lot of stumbling around, I finally went back to verify that the pinouts were the same. Yes, the DIP SX-28 and the SOIC SX-28 have indentical pinouts. Of course, just then I realized I was using the SSOP SX-28 and not the SOIC SX-28, and it has a completely different pinout. The SX-28 comes in two different surface mount packages, and I didn't pay attention to this critical bit of information.

A new PCB fixed the problem. Lesson learned.

Thanks,
PeterM