Re: soldering phosphor bronze pins to PCB

Phosphor bronze is not the issue here -- billions of connectors with phosphor bronze pins have been successfully soldered. Whatever coats it is perhaps an issue, because the plating or other coating will determine how fast the solder wets the leads.

But you are worried about cumulative stress once soldered, because you allude to joints cracking. I would also worry about the effect of heat on your device if wetting the leads takes an appreciable amount of time.

And the best source of advice on how to solder a device is its manufacturer. Often manufacturers will have an assembly app note.

Brr, I wouldn't want to solder that--they cost a zillion dollars. How about a nice socket?

Cheers

Phil Hobbs

ze

ed

I got curious, so I googled Multi Pixel Photon Counter soldering

This first data sheet that popped up gives both the recommended temperature profile for reflow soldering, as well as hand soldering instructions. Such information should be available for your MPPC, as well.

On 9/11/2012 9:00 PM, spamtrap1888 wrote: [...]

The problem here is that Hamamatsu is designing a specific package for us and I was wondering whether particular care should be taken into account when mounting these components. A reference point could be the already available device for a 2x2 pixel:

which does contain reflow conditions and such, but also limits the amount of times to *one* which is kind of worrying for something that costs so much.

Understand, I need to investigate with the manufacturer to check the coating and any possible suggestions they have on this subject.

This is true, but I got kind of surprised when the amount of times for soldering is limited to *one*. Indeed they may be extremely conservative here but I'm not sure how hard is that limit.

Another point I'm worried about is how to unsolder these guys. Our pcb will host 12 of these guys with a pitch of ~2 cm. Considering that the pinout of our package is kind of off standard I can hardly imagine to have a tool designed to heat all the pins to allow unsoldering. The only thing that comes to my mind is with a hot gun with a large area, preheating the pcb first and then blowing hot hair only on the guy to be replaced. But I'm not such an expert in this area and any suggestion is appreciated.

p.s.: thanks to Rich Webb who re-posted this on a more appropriate NG.

On 9/11/2012 8:21 PM, Phil Hobbs wrote: [...]

because we have something like ~400000 of these guys and the failure rate of a connector may be something you do not want to consider.

On top of it the installation site of the detector is fairly in the middle of nowhere, therefore accessibility may be an issue, that's why soldering is the most robust solution that came to our minds.

I'd think very carefully about using flex. You can solder them to a flex circuit that attaches to the board via an FFC connector. That way, when the board needs reworking, you can just pop the MPPCs off for safekeeping.

Cheers

Phil Hobbs

On 9/12/2012 2:25 PM, Phil Hobbs wrote: [...]

using the flex can be indeed a viable solution. We want to avoid having the connector on the other end so I need to think about how to solder the flex onto the pcb and what is the bent radius with the flex since I cannot afford to have too much space between the pcb and those devices.

The device-flex soldering can be easily done I think, but from flex to pcb we may need to go across a slot in the pcb if we want to solder on the opposite side to facilitate the rework in case needed (assume you want to remove the device).

This is a very primitive sketch of the two situations: ________ |______| ========

Good flex will take a hard crease, once. You can accordion-pleat it if you need to.

(I've often thought about how badly Hamamatsu needs a racing division. They make great stuff, but their instruments could use some jazzing up.)

Cheers

Phil Hobbs