Electrical Engineering

« first day (1754 days earlier) ← previous day next day → last day (3478 days later) »

jippie

07:10

@W5VO Yeah that looks better. I wasn't sure what you were aiming at, but then I noticed the hinges ... and the radio was looking weird all along, but hey tastes in design differ I figured ;o) The original cabinet looks like a car stereo :-/

2 hours later…

PlasmaHH

08:48

high quality question of the month electronics.stackexchange.com/questions/186021

Asmyldof

09:03

Should have made a screenshot

No longer there

PlasmaHH

used the first piece of red heatshrink yesterday. it worked.

jippie

@Asmyldof does TVO ring a bell in ASML context?

Asmyldof

@PlasmaHH yay! :-)

@jippie Not in about 30 seconds no

jippie

09:19

@Asmyldof there you go.

I think he's describing a relay.

PlasmaHH

@jippie thats not what it looked like when I posted the link, see the first revision oof the question

jippie

@PlasmaHH just the image.

Asmyldof

I think anyone who knows me can figure out my feelings about the intellect behind the question: electronics.stackexchange.com/a/186033/53769

jippie

@Asmyldof you outsrouced tests?

Asmyldof

@jippie No, Plasma gave me his address when I asked him, weeks ago, so now he already gets to play with his thank-you-present

jippie

09:28

thank-you-present? I didn't realise you were that thankful

Asmyldof

... I literally asked the two of you if you wanted a kit

jippie

09:55

@Asmyldof threat & vulnerability assessment

@Asmyldof didn't realise you meant it. You are obviously not well known for being serious:-p

/me is out. Wish me luck for the interview

Asmyldof

@jippie Enjoy, though I'm late

Don't know anyone in that department, but knowing some other depts, you'll have plenty work if they hire you

2 hours later…

jippie

11:57

it was some high level conceptual architecture infrastrucure thing at corporate IT, the job description was suboptimal :-/

PlasmaHH

ouch

jippie

I know that business pretty well, just not too comfortable with conceptuals and architecture.

@PlasmaHH I rather create things that actually work, than think up things that are supposed to work according to vendor claims.

PlasmaHH

12:54

@jippie I am at a point where I wish I could outline most of the way and have other people deal with the dirty details.

PlasmaHH

13:45

anyone has experiences with stuff that switches 300V and a few thousands of amps really fast (on and off)? what would be realistic speeds ?

1 hour later…

W5VO

14:50

@PlasmaHH Would info on a 600V, 1000A buck converter be useful?

PlasmaHH

@W5VO unless it is done by parallelizing 1000 600V 1A buck converters, it could certainly be

W5VO

@jippie I couldn't quite put my finger on what was wrong until I saw all the other pictures. The ad "Philco 39-31 Radio case&Zenith am/fm radio w/phono input" doesn't read very clearly

@PlasmaHH It was a 750hp motor inverter converted into a DC power supply - large tall cabinet. Used massive IGBT arrays. What do you want to know? Switching frequencies, rise/fall times?

(I'll have to go ask the guy)

PlasmaHH

@W5VO rise/fall time is the most interesting, but frequency is nice too, yeah

W5VO

@PlasmaHH I'll let you know

PlasmaHH

cool, thanks

jippie

15:45

@W5VO Let me enjoy the brief moment that I feel smarter than someone else <:o)

2 hours later…

Asmyldof

17:32

@jippie You can start hanging around MBO Metselen if that's a feeling you cherish

Finished a couple of boxes!

happy dance

2 hours later…

Asmyldof

19:36

And built another one

jippie

20:35

I hate it when people paste an image screen shot from a pretty much text only display.

PlasmaHH

20:53

har har har

hm, the second meta question I answered this week, this is not going into a good direction

crasic

21:18

From a wikipedia article, kind of stood out to me as incorrect

> In many cases, the most expensive part of an integrated circuit is the pins, because they make the package larger, and each pin has to be connected to the integrated circuit's silicon.

PlasmaHH

@crasic sounds reasonable

crasic

21:34

Is there a term for PCB through hole single pins?

PlasmaHH

single pins? like half a cap?

crasic

nah, like a single pin header

with no header

found something "Pin Insert"

PlasmaHH

must be too late, not getting it...

crasic

I mean they are just called "pins"

mill-max.com/assets/pdfs/206.pdf

but on an eng drawing, refering to "pins" is ambiguous

PlasmaHH

what do you use them for?

crasic

21:44

PCB to PCB connector

PlasmaHH

a single one? hm...

crasic

well two of them, but yep :)

second pcb is a bulkhead integrated into a hunk of metal

PlasmaHH

hm, blood everywhere on that case... someone is not going to be happy to clean that up...

crasic

what case?

Asmyldof

@crasic If it has a plastic stand-off like any other header it's a header, regardless of pincount. You could call it a pogo but that would be a bad misnomer.

crasic

21:55

@Asmyldof just a pin, looks like a nail, no header

the pcb is the header

Asmyldof

Otherwise, it can be treated as a pin. When you make a module that has 3 pins before its potted, they're called pins, plain and simple

it's

crasic

kinda like the business end of a bed'o-nails jig

Asmyldof

I have microscope eyes again

crasic

Yes, they are pins, but on an assembly drawing "manual inspect pins for solder bridging and whiskers" is a little ambgious

Asmyldof

I wish it wasn't covered under a strict NDA, because the 5 hours of HD video I just shot could very easily be cut into two full feature length wire-bonding-porn vids

crasic

21:57

wire bonding? like chip-scale?

Asmyldof

Specially if there were an extra camera from the side shooting my hands/tools

Not 22nm process, but yes

Well, not the bonding itself

Batch of 200 items went... wrong and we kinda need them, so since I'm the only one with the patience and no-personal-life enough, I volunteered to put in extra time to fix and/or bridge some mistakes by hand, since there's no time for back and forth

10 of each type, for initial verifications, and then the rest will be shelved while we wait on the next batch

crasic

like leads to chip wire bonding?

sounds tiny

Asmyldof

It is

Hence porn potential

crasic

how do you solder to silicon?

Asmyldof

You don't

crasic

22:02

or is there a metal layer that is accesible?

Asmyldof

The wire-bonding is done wrong, so the metalisation layer on the chips is fine

crasic

ah, gotcha

Asmyldof

Need to keep original bonds everywhere, so it includes super-fine-grained induim-based solder paste that's extra sticky, tiny little hairs of gold and a teensy little heater

And many different pincers and pliers and whatnots

And a X-many stereo microscope

PlasmaHH

and someone stoopid enuff to do it

Asmyldof

snip one bond, attach wire extension, pull over, attach other end quickly to prevent first joint from coming loose (took 2 extra tries first time)

then move original over to its neighbour after you snipped that, etc

crasic

22:06

wow, sounds painful

Asmyldof

Normally you'd say "oy, idiots: do it again, but right this time!" but, we're alreayd behind :-)

crasic

would this pass conformance or is it r&d only?

Asmyldof

Ah, by the time I have kids (if ever) the NDA will be gone and I can show them how awesome dad is ;-)

R&D

Need to go through a couple of cycles of D before we can conclude the R phase, so all needs to be quick as possible

crasic

neat cv line though

Asmyldof

2nd stage and we expect to have marketable design after the next phase, but that needs to be in production before mid 2017, or the startup will fall in defunded

crasic

22:09

"Proficient at chip-scale rework"

Asmyldof

:-D

Fair dues

Just wish I could show the video too ;-)

...

Should make a scar on my right index finger before I do the last batch, so I can prove it was me

Everytime I move out a tray-thingy-gel-pack-like-object you shortly see my finger in the vid :-)

(watched back a tiny bit at 4x)

crasic

awesome procedure video to have for the company though

Asmyldof

Anyway, on to home. 0:11 here now. ZzZzZ time

That's one reason I opted to record

crasic

15:11 here, good day

Asmyldof

The other is to be able to see if any aborations in measurements (optical chips) can be caused by a mistake I made

Very important that my mistakes don't influence the dev cycle, or we'll be hunting ghosts

crasic

22:13

I'm the only one here that does any video or photography of my rework, so what you did is awesome

Asmyldof

:-)

nitynite. Sees yas later

PlasmaHH

hm, bed, yeah, or reverse engineer this lcd ...

crasic

22:28

@PlasmaHH unless its a particularly special or rare lcd, why?

PlasmaHH

22:42

@crasic I dont like throwing away stuff

« first day (1754 days earlier) ← previous day next day → last day (3478 days later) »

all rooms

Transcript for

Aug16

Aug '1517

Aug18

Electrical Engineering

A place to talk with friends from the EE community about vacuu...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC