thermal pad need constant load in order to maintain certain thermal  
conductivity.  if you are going to change material- washer, you might  
have to consider insulating properties (nylon), mechanical strength  
and spring force, dissimilar metal coupling, etc.
my 2 cents.
       joyce
On Jul 7, 2014, at 10:09 AM, Steve Gregory wrote:

> Hi Phil,
>
> It's to isolate the three of TO220 packages from the bracket/ 
> heatsink that they are screwed to, the fourth TO220 device that is  
> screwed to the bracket/heatsink  is the  full pack type where it  
> doesn't need a washer for isolation. All these devices sit on top  
> of a custom cut Berquist SIL pad, and the bracket/ heatsink is  
> designed to capture the board next to it when it is broken out of  
> the panel and folded over. The two boards are connected by a flex  
> circuit which is soldered in during wave solder. So it winds up  
> being a rigid flex assembly without really being a rigid flex PCB,  
> I guess to keep costs down. There are two sets of these boards in a  
> panel, and you can't get to the screws easily on one set of boards  
> without them being broken out of the panel. They don't get broken  
> out of the panel until they've been tested and conformal coated,  
> which we don't do, the customer does that.
>
> So you can see the quandary we're in. Probably going to recommend  
> going to a different material washer, along with some sort of  
> chemical threadlocker.
>
> Steve
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Nutting, Phil
> Sent: Monday, July 7, 2014 7:31 AM
> To: [log in to unmask]
> Subject: Re: [TN] Screws losing torque by themselves
>
> Steve,
>
> First of all, why do you need the Nylon washer?  Nylon/wave solder  
> heat... bad combination... expect the Nylon to change shape.  Why  
> it didn't come loose before might be explained by someone  
> substituting cheaper Nylon 6/6 for a glass filled Nylon.
>
> My choice to mount the devices, if isolation is required,  Would be  
> to use T-Gard by Laird Technologies and an Aavid 118300F00000G  
> metal screw down clip with a square cone stainless screw using 10  
> inLb of torque. Or to eliminate the screw parts and operation  
> simply use an AAVID CLP212G clip, set it and forget it.
>
> But this is just me opinion backed be 4 decades of experience.
>
> Phil
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Victor Hernandez
> Sent: Monday, July 07, 2014 7:20 AM
> To: [log in to unmask]
> Subject: Re: [TN] Screws losing torque by themselves
>
> Is it possible that the screw can be torqued post solder instead of  
> pre solder?
>
> Victor,
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Stadem, Richard D.
> Sent: Monday, July 07, 2014 6:01 AM
> To: [log in to unmask]
> Subject: Re: [TN] Screws losing torque by themselves
>
> It is a good idea to double check the torque after soldering as  
> part of the final inspection.
> I would suggest after the final inspection, that the nut be  
> captured with Glyptal. The nylon washer will continue to cold flow  
> over time, as my good friend Mr. Bogden explained, and the torque  
> could continue to relax somewhat. The Glyptal will at least prevent  
> the nut from loosening.
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of stephen gregory
> Sent: Sunday, July 06, 2014 5:24 PM
> To: [log in to unmask]
> Subject: Re: [TN] Screws losing torque by themselves
>
> Hi Wayne,
>
> Thanks, and thanks to everyone that responded. I guess I didn't  
> think about things clearly when I posted my question. The reason  
> was that that we've been building these assemblies for a while and  
> just lately the problem has become an issue with our customer,  
> which leads me to believe that for some reason they've been  
> recently checking torque values on this hardware after we've  
> shipped the assemblies to them.
>
> I can surmise a lot of things about why they might have been  
> checking them, but I don't know for sure. But I do know one thing  
> now, those screws and nuts will lose torque during our normal build  
> processes.
>
> Steve
>
>
> ________________________________
> From: Wayne Thayer
> To: [log in to unmask]
> Sent: Sunday, July 6, 2014 6:26 AM
> Subject: Re: [TN] Screws losing torque by themselves
>
>
> Hi Steve-
>
> I love the video. I've been trying to tell the mechanical engineers  
> I've worked with about the problems with the helical washers  
> (sometimes stupidly referred to as "lock" washers!). This video  
> assists with developing an intuitive feel for what's going on. The  
> helical washers do have their place, but not in a high vibration  
> environment, and virtually always in addition to a thread locking  
> system of some type (double nut or chemical adhesive).
>
> The root of your problem is in the understanding of what Glass  
> Transition Temperature means. It means that above that temperature,  
> the material will deform PERMANENTLY if placed under a load. Nylon  
> has an extremely low Tg, but even most PC board materials top out  
> at about 170C.
>
> So you're asking us why, when you put a clamp on some plastic  
> material and take it way above its Tg, it seems to deform, thus  
> reducing the clamp pressure?
>
> Wayne
> ________________________________________
> From: TechNet [[log in to unmask]] on behalf of Yuan-chia Joyce Koo  
> [[log in to unmask]]
> Sent: Thursday, July 03, 2014 5:28 PM
> To: [log in to unmask]
> Subject: Re: [TN] Screws losing torque by themselves
>
> Nylon washer might made of nylon 66, that would be lost of  
> mechanical strength during your wave temperature.  see attached.   
> IMO.  (double check your datasheet for the washer for thermal  
> properties.  fyi.
> http://www.plastic-products.com/spec1.htm
>     joyce
> On Jul 3, 2014, at 4:19 PM, Steve Gregory wrote:
>
>> Hi all,
>>
>>
>>
>> Thought I might try to catch you before you left for your 4th of July
>> holiday. We have a strange little issue that we're trying to figure
>> out that's been happening with some torqued hardware on a board that
>> we build.
>>
>>
>>
>> The assembly has this heat-sink/bracket deal with four TO-220 devices
>> put together before we stuff the board and wave solder it.
>> The bracket is aluminum and we first put down a SIL pad, then the
>> TO-220 devices, three of them get nylon shoulder washers, then a  
>> split
>> lock washer, then a nut. The three devices with the nylon shoulder
>> washers get torqued to 3.4 in/lb., and the one without the shoulder
>> washer gets torqued to 6 in/lb. We tighten and torque from the screw
>> side, and hold the 3/16ths nut stationary.
>>
>>
>>
>> Then the whole bracket assembly is stuffed into the board along with
>> the rest of the parts, then the board is wave-soldered and cleaned.
>> Any touch-up is done, then there are certain parts and points on the
>> board that is RTV'ed and cured, then the board is shipped.
>>
>>
>>
>> Our customer has been finding that when they get the boards and they
>> check that the hardware is torqued, they can get ¼ to sometimes a ½ a
>> turn on the screws with the shoulder washers before the torque driver
>> clicks, and of course we're getting zinged for not torqueing the
>> hardware. The ones without the shoulder washers are not near as bad,
>> but they tell us that they have been finding a few that they can turn
>> a little before the torque driver clicks.
>> We've got new torque drivers and they're set correctly. We even sent
>> our drivers to our customer so they could check them out and compare
>> them with the drivers that they have. They were pretty much  
>> identical.
>> We use Lindstrom torque drivers, so they're not some cheap knock-off
>> brand.
>>
>>
>>
>> Here's a picture of the bracket and screws we're having most of the
>> trouble with:
>>
>>
>>
>> http://stevezeva.homestead.com/Bracket-Heatsink.jpg
>>
>>
>>
>> I've checked them here, and I find the same thing. I know they were
>> torqued properly when the brackets were assembled, but when I check
>> the boards that have been wave soldered and cleaned, I find I can get
>> ¼ turn one the screws with the nylon shoulder washers before the
>> torque driver clicks.
>>
>>
>>
>> I've always thought that when you used a screw and nut with a split-
>> washer it would hold it's torque, but then I found this:
>>
>>
>>
>> http://www.boltscience.com/pages/helicalspringwashers.htm
>>
>>
>>
>> So now I'm not sure. Has anybody ever run into this issue? Is it
>> normal to have things "relax" like this after they've been torqued?
>>
>>
>>
>> Steve
>>
>>
>>
>>
>> _____________________________________________________________________ 
>> _
>> This email has been scanned by the Symantec Email Security.cloud
>> service.
>> For more information please contact helpdesk at x2960 or
>> [log in to unmask]
>> _____________________________________________________________________ 
>> _
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]
> ______________________________________________________________________

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