Please use EE to reply to this topic.

---
James.
 
 

> -----Original Message-----
> From: piclist-bounces@mit.edu 
> [mailto:piclist-bounces@mit.edu] On Behalf Of David Van Horn
> Sent: 2005 Aug 16, Tue 11:42
> To: Microcontroller discussion list - Public.
> Subject: [OT] or maybe [EE]? Ionic contamination on PCBs
> 
> 
> 
> I have a problem with a system that measures current using an 
> AVR's ADC.
> It's doing a differential high side measurement of current 
> into a battery.
> 
> The initial symptom was that a high percentage (20%-ish?) had 
> very low charge current. This was verified, and after looking 
> over the system very carefully, I came to the realization 
> that the ADC was reporting roughly twice (not exactly twice) 
> the current that was actually flowing.
> The only anomaly I could see in the circuit, was that the 
> in-circuit values of two of the input resistors were about 
> 20% lower than those on a known good board. Out-of-circuit 
> measurements show the resistors are good, well within their 
> 1% tolerance.
> 
> After looking at it a while, it appears that the problem is ionic
> contamination of the board.   
> 
> When operating, heating the input resistors even slightly, 
> causes the current measurement to go completely insane, and 
> the board may drop to 0 output current, or up to 3X normal, 
> depending on how you heat. 
> 
> Hitting the board with degreaser while operating, produces 
> similar results. 
> 
> Measuring the affected resistors, and the reference, show no 
> significant change over a fairly wide temperature range. 
> 
> A thoroughly cleaned board performs normally.
> 
> Many of these have been built, over about 2-3 years, without 
> ever seeing this problem, or anything remotely like it. 
> 
> Visually, under 10X mag, I can seem some faint whitish 
> "stuff", and some particulate gunk, and the solder joints are 
> grey, not silver.  
> 
> Further, I know that since the last run, there has been a 
> major shakeup at the factory, and the fellow who was in 
> charge of process control is gone. I don't have a ton of 
> confidence in his replacement, but up to this point, I hadn't 
> seen anything directly wrong with him either.
> 
> So: How can I <PROVE> that these are, or are not, contaminated?
> This will be a big issue, the factory is not going to want to 
> hear this.
> 
>  
> 
> 
> --
> http://www.piclist.com PIC/SX FAQ & list archive View/change 
> your membership options at 
> http://mailman.mit.edu/mailman/listinfo/piclist
> 
> 


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist