To Thomas.
If you have an automated epoxy dispenser then you could dispense a 2 or 4 dots of chipbonder on either side.
If you can change the pad design later on the next batch then reference IPC 7525A and use the horseshoe pad design.
We have been successful in placing round SMT parts on square pads and not have them move out if spec only after I took several actions: 
1. Assured the SMT conveyor sections from the chip shooter to the IC placement machine and the last section before reflow were all perfectly aligned. Often the sections are manually cranked open on the conveyor systems and even if the fixed rail is perfect the rear or adjustable rail may not have been set up perfectly because it was good enough for other boards that didn't have the round parts. If the  board hits the beveled guides and shifts the board to let it pass through, the round parts will move. The alternative is to have the round parts placed on the IC placement machine instead of the chip shooter and possibly programmed to be placed last in the program to assure nothing moves. The board must have good bottom-side support in the placement machine.
2. The unload speed and transition to the last rail conveyor must be optimized. If the board goes to the reflow conveyor rail then it must be checked for a smooth transition.
3. If the circuit board is placed on the SMT conveyor's wire belt then the all the gear drives at the exit end must align perfectly with the wire belt. The master links must be well formed and the ends not long enough to come in contact with the group of drive gears that pull the wire belt at the exit end. This the case where some of the boards come out fine then a bunch come out with moved parts. Any bent wire in the belt or the wire master links can cause the shift. Check under the reflow oven to see if too much of the wire belt is hanging down and can get caught up on the long drive chain. On older machines the belt could have stretched and may need a 1-2 feet removed. Another rrot cause I've found were parts or stiffener bars that came off boards that  were trapped at and caught the wire belt as well as the sensor at the entrance that got caught in the belt as the belt stretched and sagged. 
The belt at the master link area must not be skewed or the belt will be smooth until the shewed section gets to the drive section.
4. With enough solder paste on a square pad and a good reflow profile with the above corrections you shouldn't have a problem unless the square pads were undersized or the pad spacing wasn't correct or the stencil apertures were reduced. 
I hope this helps.
John Goulet
MFG/Process Engineer
----- Original Message -----
From: Thomas L Gervascio
To: [log in to unmask]
Sent: Fri, 21 May 2010 13:53:12 +0000 (UTC)
Subject: [TN] Dispensable temporary SMT part staking material

I am looking for a material that I can keep round SMT parts in place during reflow. We use a peelable solder mask but the viscosity is too thin to dispense. Tried thickening with cabosil but no luck. The clay based water soluble mask swells up too much in the reflow profile. Any recommendations on any material that might be a good candidate?

Thanks
Tom


______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact KeachSasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------