[amsat-bb] Re: Space Grade Soldering ?

[Bruce Bostwick]

Is 60 Sn / 40 Pb better than 63 Sn / 37 Pb or 62 Sn / 36 Pb / 2 Ag for  
space grade soldering?  I tend to prefer 63/37 because it's closer to  
eutectic and doesn't have as much of a "mushy" stage when it cools,  
and because it tends to contract onto the connection and seal gas out  
rather nicely.  I also don't seem to have had any problems with 63/37  
solder joints cracking like some of what I've seen from other solder  
types (which I suspect may be RoHS compliant stuff, and maybe wave- 
soldered), but I'm not sure how being in a spacecraft environment  
would affect that.  :)

The anti-wicking and looping/gluing treatment sounds well thought out,  
and I can see how it would make the components hold up a lot better  
under launch vibration.  I knew about the uncoated PCB's and the  
conformal coating already, as well as the clean room assembly, but  
hadn't heard of that particular wiring treatment.  Now I know.  :D

On Jan 29, 2008, at 5:10 PM, Howard Long wrote:

>> I've been reading about the construction of a few microsats
>> and I see "Space Grade Soldering" mentioned a few times. So
>> I'm curious as to what that entails ?
>
> This is a question where you'll ask three different engineers the same
> question and you'll likely get three different answers.
>
> I can't comment on the Microsats specifically, but in my experience  
> at ESA
> there are a whole bunch of things we have adhered to in the past.
>
> Firstly, 60/40 is good. Anything RoHS is bad. This is predominently  
> due to
> 'tin whiskers' that the lead in the solder significantly resists.  
> This leads
> to early failure in devices as the tin whiskers grow and short out
> previously insulated conductors.
>
> I am not sure of the specific flux they use in the solder.
>
> I am aware that reflow is often suggested as a way of soldering that
> prevents cracking of components by uniformly heating the whole  
> device. This
> isn't universal practice though. Again this is to reduce early  
> failure of
> devices due to weaknesses from stressing the part during conventional
> soldering.
>
> We also use special anti-wicking pliers when soldering  
> interconnecting wires
> (PTFE) to PCBs: that stops the solder running down into the wire  
> behind the
> PTFE insulation. Then a small loop is made and the wire is glued a  
> few mm
> (length dependend on the wire gauge) from the solder pad. This  
> procedure is
> to provide a uniformly strong bond without weaknesses when it goes  
> to the
> shake tests.
>
> Generally PCBs have no coatings at all prior to mounting the  
> components: no
> silk screen or resist. This is to reduce the chances of outgassing  
> from
> small bubbles that may have formed during fabrication.
>
> After the components have been mounted and the poulated PCB is tested,
> conformal coating is used, which is basically a guy with a tin of  
> goo and a
> brush. The goo is generously brushed over the all the components and  
> the
> entire pcb. Once hardened, it is designed to prevent outgassing and  
> provide
> some resistance to vibration shock.
>
> All this is done in a clean room, wearing disposable gloves and  
> protective
> clothing to try to stop any foreign object damage including grease  
> and other
> body gunge attaching itself to the parts.
>
> 73, Howard G6LVB

PHP is the P in PHP ... :)