Page 1 of 1

Anyone have XMOS L1 assembly issues?

Posted: Thu Jun 24, 2010 12:26 am
by kster59
I assembled 3 prototypes for the L1-128 and it was very difficult. I got numerous bridges which I attribute to the .40mm pitch. I got it to work eventually by clearing all the bridges with solder wick but boy it was a pain.

I assembled several L1-64 prototypes with no problems. .5mm seems industry standard.

My pain isn't a big deal but have people who have gone to production with .4mm parts experienced any problems from assembly houses?

The L1-128 seems like a great chip on paper but it's way more assembly work than a L1-64 (many more caps, much smaller pitch) that I will avoid it in the future.

I'd rather use 2x L1-64 than go through assembling another L1-128...

Re: Anyone have XMOS L1 assembly issues?

Posted: Thu Jun 24, 2010 12:41 am
by leon_heller
I'd get a stencil made and ship the board, chip and stencil off to ASK Electronics. They don't charge much for assembling that sort of thing, including X-ray (I paid £30 each to get a couple of Telit BGA modules assembled). It's a sort of loss-leader, they hope to get their costs back when they do the actual manufacturing.

Re: Anyone have XMOS L1 assembly issues?

Posted: Thu Jun 24, 2010 12:46 am
by kster59
That's my question if even professional assembly might have problems at .4mm pitch for qfps.

I have a low cost professional assembly in the USA but it still adds quite a bit to the cost of a prototype.

They also will install only specific chips and I can do the rest of the discretes, etc.

Re: Anyone have XMOS L1 assembly issues?

Posted: Thu Jun 24, 2010 12:48 am
by kster59
Also below .65mm pitch the assembly house I use won't accept plastic stencils and you need to go with a steel one.

Re: Anyone have XMOS L1 assembly issues?

Posted: Thu Jun 24, 2010 1:01 am
by kster59
Also I looked at other small fine pitched QFPs such as a Xilinx Spartan PQ208 (the finest pitch qfp they make I think) and that's also .5mm

.5mm is no problem but seems like XMOS went ultrasmall at .4mm.

I would be interested in hearing assembly stories from others.

Again it works and its possible but if it is something to look out for in case you go to production and the assembly house wants double now because of the .4mm part.