notten has quit [Remote host closed the connection]
infobot has quit [Ping timeout: 258 seconds]
infobot has joined #neo900
herpderphurr has quit [Ping timeout: 240 seconds]
xmn has quit [Quit: Leaving.]
jonwil has quit [Quit: ChatZilla 0.9.93 [SeaMonkey 2.46/20161213183751]]
cc___ has quit [Ping timeout: 258 seconds]
gry has left #neo900 [#neo900]
jonwil has joined #neo900
knttl has quit [Ping timeout: 246 seconds]
knttl has joined #neo900
pagurus` has joined #neo900
pagurus has quit [Ping timeout: 240 seconds]
infobot has quit [Ping timeout: 260 seconds]
DocScrutinizer05 has quit [Disconnected by services]
DocScrutinizer05 has joined #neo900
neo900 has joined #neo900
neo900 is now known as Joerg-Neo900
Joerg-Neo900 is now known as Guest66482
_whitelogger has joined #neo900
herpderphurr has joined #neo900
chomwitt has quit [Ping timeout: 240 seconds]
galiven_ has joined #neo900
galiven has quit [Ping timeout: 240 seconds]
jonsger has joined #neo900
jonsger1 has joined #neo900
jonsger has quit [Ping timeout: 240 seconds]
jonsger1 is now known as jonsger
tsuggs has quit [Ping timeout: 264 seconds]
Kabouik has joined #neo900
jonsger1 has joined #neo900
jonsger has quit [Ping timeout: 260 seconds]
jonsger1 is now known as jonsger
cc___ has joined #neo900
ecloud has quit [Ping timeout: 246 seconds]
ecloud has joined #neo900
jonwil has quit [Quit: ChatZilla 0.9.93 [SeaMonkey 2.46/20161213183751]]
chomwitt has joined #neo900
chomwitt has quit [Ping timeout: 240 seconds]
chomwitt1 has joined #neo900
Tsutsukakushi has joined #neo900
cc___ has quit [Ping timeout: 256 seconds]
jonsger1 has joined #neo900
jonsger has quit [Ping timeout: 240 seconds]
jonsger1 is now known as jonsger
herpderphurr has quit [Ping timeout: 246 seconds]
drathir has quit [Ping timeout: 260 seconds]
cc___ has joined #neo900
drathir has joined #neo900
Kabouik has quit [Ping timeout: 268 seconds]
Pali has joined #neo900
drrz has joined #neo900
illwieckz has quit [Ping timeout: 240 seconds]
illwieckz has joined #neo900
illwieckz has joined #neo900
illwieckz has quit [Changing host]
chomwitt1 has quit [Ping timeout: 268 seconds]
_whitelogger has joined #neo900
infobot has joined #neo900
mzki has quit [Ping timeout: 240 seconds]
jonsger has quit [Quit: jonsger]
<ravelo>
good news
<ravelo>
I got reply from former Nokia Manager
<Joerg-Neo900>
WOW
<bencoh>
oO
<ravelo>
:)
<ravelo>
He could not help me directly
<Joerg-Neo900>
dang, you know how to scare Hitchcock with suspense ;-)
<ravelo>
but pointed us someone who "worked with the matter during Nokia years"
<ravelo>
hehe
jonsger has joined #neo900
jonsger1 has joined #neo900
jonsger has quit [Ping timeout: 264 seconds]
jonsger1 is now known as jonsger
<Joerg-Neo900>
an FAE! GREAT!
<Joerg-Neo900>
this is exactly our guy
<atk>
a what?
<Joerg-Neo900>
(for later on, when we have to solder the PoP stacks)
<Joerg-Neo900>
Field Application Engineer
<atk>
and he will help?
<atk>
wow
<Joerg-Neo900>
we'll ask him if he could help, yes
<atk>
Does he have a donation page? :P
<Joerg-Neo900>
he's exactly the job description role that had to deal with stuff like the warp issue
<Joerg-Neo900>
nah, he's employed by another major company now
<Joerg-Neo900>
I however hope for his memory about what they did in Nokia to solder those critters
<Joerg-Neo900>
and he been *the* dude who was responsible for exactly that... so I hope, deducing from his title FAE
<Joerg-Neo900>
let's see if we can contact him
<atk>
But isn't the reason of these problems because the samsung PoP is different to the old PoP memory that is no longer available?
<atk>
As in, that email said that the samsung PoP is more difficult to solder.
<Joerg-Neo900>
no, this is what the fab is using for their own excuse
<Joerg-Neo900>
the PoP RAMs are hardly dufferent, I can't see how they are, mechanically and thermally
<Joerg-Neo900>
but they are the only thing /except fab internal changed machines and personal) that differs between successfull assembly a long time ago, and the current failed one, so the fab blames the PoP RAM, of course they do
<Joerg-Neo900>
they could as well have claimed "you chosen the wrong season, last time we did it in autumn. In spring those chips are harder to solder"
<Joerg-Neo900>
It's about responsibility and ensuing from that: cost/expense for fixing
<Joerg-Neo900>
>> This is blocked by not having finished the negotiations with the production company. They claim that the problems come from using some aged components provided by us... I claim that it is the warping of the (newly bought) DM3730 (and they knew the problem from the OpenPandora where they finally got it soldered but were not able to repeat the success this time).<<
ashneo76 has joined #neo900
<Joerg-Neo900>
cimponents don't 'age' the way they claim. They may need baking to remove adhered humidity from the solder balls, but that's it. You can solder a 20 years old chip just like a new one, unless it's oxidized or 'wet', and for both there are proven cures
<Joerg-Neo900>
but the failure modes seen in GTA04A5 are evidently (by mere logic, note '*lower* OMAP interface'!) _not_ related to the PoP chip at all
<Joerg-Neo900>
when the toast breaks, you can't blame the jam you spread over the butter on the toast
disly has joined #neo900
<atk>
Toast can break? :P
<atk>
But I see what you mean
<Joerg-Neo900>
toast can't break... and BGA chips can't warp ;-)
<Joerg-Neo900>
at least *usually*
<Joerg-Neo900>
I'm EE (doing schematics etc), soldering usually is the domain of a FAE, I have to admit I didn't even know of warp issues for a long time, until recently
<atk>
I mentioned the situation to some guy from another channel and he's of the opinion that really it's all down to temperature curves of the reflow oven.
<Joerg-Neo900>
yes, finally it is
<Joerg-Neo900>
I already mentioned lack of, or too luke/short, preheat phase in vapor phase soldering as a possible culprit
<Joerg-Neo900>
anyway my suggested remedies are orthogonal to reflow temperature profiles
<MonkeyofDoom>
best of luck with the fragile toast problem! ;)
<Joerg-Neo900>
to put it plain straight just once more: there's no real 'problem' in all this, except of a woeful lack of funds to do a sufficient number of proper test runs with first 95% of the test devices being paperweights
<Joerg-Neo900>
the 'problem' definitely can get soved by tuning the parameters of soldering, the methods etc, it just needs the resources
<Joerg-Neo900>
when you're aiming at a final production batch volume of THIRTY devices, you have very limited resources for test runs
<atk>
I know desoldering and re-balling BGA is possible, but it needs a re-balling grid thing (mask?), how hard is it to separate, clean and re-ball PoP packages?
<Joerg-Neo900>
not really hard, but terribly tedious
<Joerg-Neo900>
actually labor worth more than the chip
<atk>
I'll do it for free :P
<Joerg-Neo900>
so not economically feasible
<atk>
just give me the tools, I'll work it out :P
<atk>
How hard can it be
<atk>
:P
<Joerg-Neo900>
getting the right tools being part of "labor", sorry for the over-simplification
<Joerg-Neo900>
with tooling for a maybe few thousand bucks, every chimp can do it
<atk>
I'll just use a blowtorch, a magnifying glass and my hakko :P
disly has quit [Ping timeout: 260 seconds]
<atk>
What could go wrong :P
<Joerg-Neo900>
that would earn you a top rank in Guinness book of records
DocScrutinizer05 has quit [Disconnected by services]
DocScrutinizer05 has joined #neo900
DocAvalanche has joined #neo900
akaWolf1 has joined #neo900
cc__ has joined #neo900
cc___ has quit [*.net *.split]
ecloud has quit [*.net *.split]
DocScrutinizer51 has quit [*.net *.split]
Oksana has quit [*.net *.split]
misv has quit [*.net *.split]
akaWolf has quit [*.net *.split]
akaWolf1 is now known as akaWolf
DocAvalanche is now known as DocScrutinizer51
ecloud has joined #neo900
Oksana has joined #neo900
xmn has joined #neo900
<Joerg-Neo900>
ohmy that's a dirty way of doing 'reballing'
<atk>
It works :D
* Joerg-Neo900
idly wonders if chips are also available unballed (blank pads) for glue-bonding
<Joerg-Neo900>
using conductive gue instead of solder paste, and placing chips without solder balls on pads directly into the glue, ideally under vacuum to avoid voids... I guess I just 'invented' a whole new assembly method
<Joerg-Neo900>
s/gue/glue/
<atk>
how would that work?
<atk>
wouldn't the conductive glue just make the whole thing into one big conductor shorting everything out?
<Joerg-Neo900>
this is a technique commercially used already. E.G the N97 camera's flat cable is glued to the N900-type camera module bottom with conductive adhesive. I had a hard time removing that stuff, it was so strong
<atk>
I see
<Joerg-Neo900>
maybe it's so strong it doesn't work for chips on PCB, noth rigid
<Joerg-Neo900>
both
jonsger has quit [Quit: jonsger]
<Joerg-Neo900>
might crack on thermal expansion stress
jonsger has joined #neo900
<Joerg-Neo900>
apropos thermal: no idea if that stuff is thermally conductive like solder
<atk>
Well, generally conductors are also good thermal conductors
<atk>
which is why good thermal paste is conductive
<atk>
of course, not always the case
<Joerg-Neo900>
:nod: there's a close relation, yes
<atk>
Although you would think conductive adhesive might have a relatively high resistance
<Joerg-Neo900>
and actually conductive glue is made of filler balls made of silver or whatever, which are touching each other. The 'gaps' in between are then filled by cyanoacrylate or whatever
ravelo_ has joined #neo900
* Joerg-Neo900
idly muses about using carbon nanotubes instead of silver balls
<atk>
that would be even more expensive
<Joerg-Neo900>
like a dish of spaghetti with sauce bolognese ;-)
ravelo has quit [Ping timeout: 240 seconds]
Kabouik has quit [Ping timeout: 260 seconds]
ravelo__ has joined #neo900
ravelo_ has quit [Ping timeout: 246 seconds]
ravelo has joined #neo900
ravelo__ has quit [Ping timeout: 246 seconds]
ravelo_ has joined #neo900
ravelo__ has joined #neo900
ravelo has quit [Ping timeout: 240 seconds]
ravelo_ has quit [Ping timeout: 240 seconds]
Pali has quit [Remote host closed the connection]
zencognito has quit [Quit: Leaving]
freemangordon has quit [Read error: Connection reset by peer]
ravelo_ has joined #neo900
ravelo__ has quit [Ping timeout: 240 seconds]
xmn has quit [Quit: Leaving.]
ravelo__ has joined #neo900
ravelo_ has quit [Ping timeout: 260 seconds]
drrz has quit [Quit: Leaving]
ravelo has joined #neo900
drrz has joined #neo900
ravelo__ has quit [Ping timeout: 256 seconds]
ravelo_ has joined #neo900
ravelo has quit [Ping timeout: 240 seconds]
ravelo has joined #neo900
ravelo_ has quit [Ping timeout: 240 seconds]
pagurus` has quit [Remote host closed the connection]