A thought on power issues over in the support forum

[Link]

Could it be the silver paint is so thick that it raised the bed of the battery case to a level the batteries aren't making contact.


I don't have a silver case or have any idea of testing this. But if this has been applied in an amount with some form of hardener, it could be a few nano meters thicker.


I'm looking forward to seeing the replies that come in over there.


I'm going to give my dad a call to and see if he knows more as he is PPG certified with auto painting.

 

[gruso]

I can say that the battery in my silver unit is a tighter fit than either of the black units I've had. I know this feel can vary even between black Pandoras, so that's not really enough data to draw any conclusions. You are barking up an interesting tree though.

 

[Link]

My dad says that a layer of flat black can be about a half mil, so on the old units that's not much.


Potentially a flex agent could have been added to help bind up the cracking areas, a coat could be 3 mil per coat and they likely did two coats. He said that might not be a big enough difference, but he's not aware of the precision of the case as it is "tight". I then remembered that this is on both sides of the case, so its potentially lifting the board side of the case too, meaning 12 mil of paint vs 1 mil of paint and the plastic on the black casing.


Is Global Components testing the units with batteries, or are they doing it off an AC test rig of sorts?

 

[huniper]

This would make sense if there isn't sufficient contact maybe the board isn't getting enough voltage.


I would think that applying a little bit of pressure on the battery above the contact area, with the battery door off could be a troubleshooting step for this.

 

[EvilDragon]

GC are testing and flashing the units with batteries and doing the first boot with AC Adaptor connected (though they might not always see whether it is charging or not).


The batteries fit and work fine though. Are there more units which don't charge?


So far I've only seen that one - and one out of 700 produced ones wouldn't really scare me

 

[Link]

I've seen four having power issues of some sort that I've asked what color the Pandora was in the support forums. Maybe 3, but it's just a hunch.


Between that and any battery inconsistency in thickness could lead to non-contact or intermittent brown outs.


Again, I'm just going off some recent reads and knowing the failure rate, something else could be at play with the recent change to thicker paint.


Maybe the last thing to add, is if Global is doing the testing with these batteries, is a batch they have sitting there for testing purposes, or is it the battery included with the unit that is shipping. Because if they test with battery "A" & ship untested battery "B" which is fresh, brand new, and uncompressed tape compared to a battery that's been in & out of 2 or 3 hundred units.


I think they need to take note on whether charging is happening or not, as this is obviously important. And testing with the battery that ships, not a battery for testing only.

 

[centus]

My dad says that a layer of flat black can be about a half mil, so on the old units that's not much.


Potentially a flex agent could have been added to help bind up the cracking areas, a coat could be 3 mil per coat and they likely did two coats. He said that might not be a big enough difference, but he's not aware of the precision of the case as it is "tight". I then remembered that this is on both sides of the case, so its potentially lifting the board side of the case too, meaning 12 mil of paint vs 1 mil of paint and the plastic on the black casing.


Is Global Components testing the units with batteries, or are they doing it off an AC test rig of sorts?

just a thought here. you ARE aware of the fact 12mm is 1.2cm right?


something sounds wrong in that math for a paint thickness...

 

[steveway]

My dad says that a layer of flat black can be about a half mil, so on the old units that's not much.


Potentially a flex agent could have been added to help bind up the cracking areas, a coat could be 3 mil per coat and they likely did two coats. He said that might not be a big enough difference, but he's not aware of the precision of the case as it is "tight". I then remembered that this is on both sides of the case, so its potentially lifting the board side of the case too, meaning 12 mil of paint vs 1 mil of paint and the plastic on the black casing.


Is Global Components testing the units with batteries, or are they doing it off an AC test rig of sorts?

just a thought here. you ARE aware of the fact 12mm is 1.2cm right?


something sounds wrong in that math for a paint thickness...

12 mil != 12 mm

 

[Neelix]

Erm... if mil doesn't refer to millimetres what does it refer to? :huh:


- Neelix

 

[Prometheus]

^ My only guess would be millilitres.

 

[TrashyMG]

http://www.thefreedictionary.com/mil

mil 1

​

(m

l)​

n.
http://www.thefreedictionary.com/mil

1. ​

http://www.thefreedictionary.com/milA unit of length equal to one thousandth (10

http://www.thefreedictionary.com/mil-3

http://www.thefreedictionary.com/mil) of an inch (0.0254 millimeter), used, for example, to specify the diameter of wire or the thickness of materials sold in sheets.

http://www.thefreedictionary.com/mil

http://www.thefreedictionary.com/mil

2.

http://www.thefreedictionary.com/milA milliliter; one cubic centimeter.

http://www.thefreedictionary.com/mil
http://www.thefreedictionary.com/mil

3.

http://www.thefreedictionary.com/milA unit of angular measurement used in artillery and equal to

http://www.thefreedictionary.com/mil1

http://www.thefreedictionary.com/mil/

http://www.thefreedictionary.com/mil6400

http://www.thefreedictionary.com/mil of a complete revolution.

http://www.thefreedictionary.com/mil

[short for Latin mllsimus, thousandth, from mlle, thousand; see gheslo- in Indo-European roots.]​

 

[PaulForey]

Holy shit really? "mil" means a thousandth of an inch? That's messed up, America. I think you should apologise.

 

[Link]

Sorry, been on the road all day. I should have clarified that a mil = about a human hair diameter, and perhaps that is a thousandth of an inch.


What's the tolerance on the battery connection?

 

[jasay]

I then remembered that this is on both sides of the case, so its potentially lifting the board side of the case too, meaning 12 mil of paint vs 1 mil of paint and the plastic on the black casing.

I just had my silver Pandora open to fix the shoulder button and there is no paint on the inside of the bottom of the case.


Also I did a quick measurement with calipers and found the spring loaded battery contacts were ~.050-.055" (hard to measure bendy bits) above that recessed area of the battery compartment and the recess itself is about .045". That means the battery contacts should be compressed by .005-.010" when the battery is inserted. I could do more accurate measurements on Monday in our incoming inspection area at work, but I only have one sample so not sure how helpful that is.

 

[Link]

^


WOW


Thank you for that type of measurement and documentation. From the few hundred batteries I've handled, I can say they aren't identical.


By reading this increased "failure" or "issue" in the support area, I can't but wonder what else it could be. Unles something changed with the PCB outside of tolerance. Glad to know that the inside of the case is not painted. So that should bring the gap down from 12 mil to 6 mil hypothetically if this could cause an issue. Knowing the history of the paper tuck battery work around, I can't see why this could not be an issue.


Is the battery encasement area painted silver or was that left unpainted too?

 

[jasay]

Is the battery encasement area painted silver or was that left unpainted too?

The battery compartment in the lower case is painted. The battery facing side of the battery cover is not (the outside obviously is).