Tips on hand-soldering USB type C receptacles?

[SiliconWizard]

Currently designing a new isolated JTAG/SWD probe, and I'm going to use an USB type C connector for a change.
I selected the Amphenol 10137061-00021LF which seems a good compromise (pricing, build quality, availability and one through-hole row).

I need to be able to hand-build a few prototypes.
This connector has a shielding which will prevent (I think) direct hand-soldering of the SMD pins (one of the two rows) which are covered by it.
Most receptacles I've found so far have a similar design (I guess for EMI reasons?). I've found only one receptacle the shielding of which doesn't cover the row of SMD pins, but it's unvailable. Don't want to get into supply issues.

Obviously, using a hot air ironl ooks like an option but I'm wondering how difficult it will be given that the whole connector is covered by its shielding, so heating it from above will be a pain due to the thermal mass.

Any tip/real experience with hand soldering those connectors is welcome.

[wraper]

You would need metal stencil (preferably 0.10-0.12mm thick) to apply solder paste and use hot air. Preheater is not necessarily required but will make things much easier. Just applying solder paste on the pads without separation will guarantee shorts between the pins. Soldering iron is completely useless for SMT pins.

[wraper]

If you don't want to bother with stencil, you could also apply as much solder as you can get on SMT pads, clean flux residues, apply tacky flux on them, place connector and heat with hot air.

[SiliconWizard]

If you don't want to bother with stencil, you could also apply as much solder as you can get on SMT pads, clean flux residues, apply tacky flux on them, place connector and heat with hot air.

Thanks, but didn't you say above that without separation shorts are almost guaranteed? (And I would tend to think so as well, unless maybe you find just the right amount of paste...)

[wraper]

If you don't want to bother with stencil, you could also apply as much solder as you can get on SMT pads, clean flux residues, apply tacky flux on them, place connector and heat with hot air.

Thanks, but didn't you say above that without separation shorts are almost guaranteed? (And I would tend to think so as well, unless maybe you find just the right amount of paste...)

You should apply as much solder as you can but pads should not be shorted of course. I thought it was obvious. If you do this, you should notice that solder prefers to stay on the solder iron tip rather than on small pads.

[Howardlong]

Eww... I see what you mean. I was wondering what you were talking about, and then I had a look at some USB-C 3.1 receptacles I have in stock.

The 3.1 receptacles I have give visible (and even solderable) access to half the connections, but the other half are going to be difficult if not impossible to visually inspect.

I do have some with a half through hole design that are a lot more accessible.

Glad you mentioned this because I was looking to do a repair on a laptop that has lost one of its pins on a receptacle (don't try to plug in a lightning jack into a USB C receptacle when you're half asleep boys and girls) so it only charges with the plug in one way round, not both.

[SiliconWizard]

The 3.1 receptacles I have give visible (and even solderable) access to half the connections, but the other half are going to be difficult if not impossible to visually inspect.

Yes, I can see this. The first row should be solderable with an iron and a very fine tip.
Unfortunately, the receptacle I chose has no such access to the first (smd) row, the shielding goes almost down to the PCB. (The second row being t-h is no problem.)
Very hard to find a receptacle with both characteristics.

[wraper]

Yes, I can see this. The first row should be solderable with an iron and a very fine tip.

If you try, you'll find very soon that you are wrong about this. At best you'll be barely able to touch end of the pin with a point of a very fine tip. Barely enough to melt solder. And due to fine pitch it's almost guaranteed you'll get solder in between of the pins. When that happens, you are screwed as you won't be able to do anything with that.

[TimNJ]

I've done it by hand with a good iron with a small chisel-tip (maybe like 3-4mm wide). *Used a lot of flux.* Looks like your pins are a more concealed beneath the shell than the ones I worked with.

I think I'd try to do it pin-by-pin, inspecting each joint after you make it. I wouldn't bother drag soldering; The pitch is too fine.

[SiliconWizard]

I found this one from Hirose: https://www.hirose.com/product/en/products/CX/CX70M-24P1/
specifically marketed for "easy reworking". Looks like the pins are a lot more accessible. But I would have liked to avoid mid-mount connectors...

[wraper]

You would need metal stencil (preferably 10-12 mil thick) to apply solder paste and use hot air. Preheater is not necessarily required but will make things much easier.

Whaaaaaaaat? Do you have an idea how thick that is?
Common SS stencils are 3~4 mils thick.

Edit: For those Europeans who didn't know, a mil is 25.4um, not 10um.

That was a brainfuck. I meant to write 0.10-0.12mm thick.