No, almost all were SMT, and most were 1206 or smaller sized. Maybe the soldering tech applied too much heat for too long; there was some usage of leadfree solder at that time, which I've managed to get us away from more recently. [Most of my stuff is for in-house ATE, as was this board.] I do believe all soldering was done with an iron, no hot plate.
One thing I noticed aftwards was that I placed all the 1206 parts parallel to the long edge of the board, which I want to say I read (afterwards) to not do that, in the IPC document.
Shawn Upton, KB1CKT
Test Engineer
Allegro MicroSystems, Inc
[log in to unmask]
603.626.2429/fax: 603.641.5336
-----Original Message-----
From: Kenneth Wood [mailto:[log in to unmask]]
Sent: Thursday, July 30, 2009 8:14 AM
To: 'TechNet E-Mail Forum'; Upton, Shawn
Subject: RE: [TN] Hashing-Out Hatching
I'm not really sure what to say without seeing your boards and soldiering setup.
The only time I have seen boards warp significantly during hand soldering is when far too much heat for far too long was applied.
Is this a possibility? I mean, do you have large parts that require an enormous amount of heat?
Maybe since all the parts are on one side and hand soldering usually takes much longer than a wave or oven the PCB is just getting too hot on one side.
Ken
_____________________________________
Kenneth J. Wood
Saturn PCB Design, Inc.
[log in to unmask]
www.saturnpcb.com
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Upton, Shawn
Sent: Thursday, July 30, 2009 7:59 AM
To: [log in to unmask]
Subject: Re: [TN] Hashing-Out Hatching
How does cross-hatch recommendations change for single vs double sided assemblies (parts on one or both sides of the board)? Does it change with hand soldering? [Given that heat is applied unevenly to the board with hand soldering.] I ask as I had a design that warped noticeably after hand soldering--I'd call it reasonably dense, SMT with parts on only 1 side, 0.062" with 8 layers and 3 continous gnd planes (Mid1,Mid4,Mid6 as GND planes, out of Top/Mid1/Mid2/Mid3/Mid4/Mid5/Mid6/Bottom).
Shawn Upton, KB1CKT
Test Engineer
Allegro MicroSystems, Inc
[log in to unmask]
603.626.2429/fax: 603.641.5336
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Kenneth Wood
Sent: Thursday, July 30, 2009 7:45 AM
To: [log in to unmask]
Subject: Re: [TN] Hashing-Out Hatching
I would not use cross hatching in modern high-speed or RF designs, cross hatching is ok for designs that have no high-speed though.
We are to the point of taking the tightness of the prepreg weave into consideration
for impedance issues so a cross hatched reference plane is defiantly out of the question today.
To your question, the only issue I have ever seen, and continue to see, is warpage of unbalanced layers.
Use outer layer thieving pads and balance inner layers to avoid this and you will be ok.
Even the outer layer thieving is optional but I do it as often as I can. (some customers hate seeing it)
It's not very scientific, if you have a 6 layer PCB and layer 2 is a solid plane then make sure that layer 5 (at least) closely resembles.
Ken
_____________________________________
Kenneth J. Wood
Saturn PCB Design, Inc.
[log in to unmask]
www.saturnpcb.com
|
