Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Wave Soldering Ceramic PCB's

JohnW

#17147

Wave Soldering Ceramic PCB's | 8 August, 2001

Hi folk's,

I'm looking for some hint's here. Got a board from the customner that is ceramic, and it's 2.2mm thick. I've got a couple of RF connectors that I need to solder, problem is I can't get any flow up the PTH's. The pin's are gold on the connectors and after a little bit of digging I found a nice big copper signal layer in the middle of the PCB which is obviously acting as a great heatsink, as if the ceramic wasn't enough! Normally I'd go with the IPC of only needing 50% but the customer want's at least 75 preferably 100%. I'm throughing a hell of a lot of heat that this beast and can't get it above 95 deg C on the top side with I would have thought would be enough (although the flux spec is asking for at least 105...hey they always exagerate!) anyhoo I've tried running it slow over the wave to increase the dwell time which is now about 4.5 sec's heading for 5 so that should be well enough but no luck. I've asked for solderability check's on the components and the PCB but so far haven't managed to get somewher to do them as the plant I'm helping out doesn't have the facilities... any one any hint's on working with ceramics then?

John W

reply »

#17150

Wave Soldering Ceramic PCB's | 8 August, 2001

Get-out the razor blades, bud. You could run that sukka for 6 sec dwell, cook breakfast on it, and you still will not get fill. [Hole file that is, breakfast will go along ways if peckish, though.] You could fire-up the depleted ozone rework gun [the black one with the red trigger] and yud still be ...

The signal plane is connected to the hole and draining the heat. It needs thermal relief. Traditionally, connections from plated hole to BIG ol� planes like this should be with �wagon wheels�. These spokes between the plating on the hole annular ring and the plane allow heat to move up the hole, not get drawn down by the plane, but still retain signal integrity.

Look at IPC-2222, 9.1.2. It states that the relationship between the hole size, land and the web areas is critical. And goes on to describe that relationship.

[If you don't buy all that, a circumferential barrel crack of the through hole plating will do the same things to barrel fill, as you describe here.]

Next, it's unlikely that both the barrel and the leads of components have solderability problems on this very today. One or the other is gunna flow. [And you didn't even tell us about the breadth and extent of this mess you've got us into. Which would make it less likely there's a solderability problem.] Good try tho.

Two more things: 1 We have not wave soldered ceramic in years. We reflow ceramic exclusively. Could you give us more detail on your process, materials, etc.? How are you cleaning this thing? 2 We had a similar situation a couple of months ago. Went through the drill. And asked for relief. The QA type said �I don�t care what you do, just get fill.� Well, he signed-up and now there�s solder on the top and bottom of the holes. Not a real pretty thing to do.

reply »

Sean D

#17157

Wave Soldering Ceramic PCB's | 9 August, 2001

Hello John,

Just a thought, try contacting your equipment manufacturer. They usually see a bit of everything when it comes to applications. Maybe they can recommend a solution. Sometimes this involves purchasing something, but at least you can tap that resource for data and then examine your options. It couldn't hurt. With the current market, I'm sure alot of Apps guys are just looking for something to do.

Good luck

reply »

#17161

Wave Soldering Ceramic PCB's | 9 August, 2001

Thank you.

IPC-2222 applies to rigid organic boards. It does not apply to ceramic boards. Since there is no 2222 analog for ceramic boards, I used 2222.

reply »

JohnW

#17162

Wave Soldering Ceramic PCB's | 9 August, 2001

Dave ,

you know it's your cheery disposoition I love....

Ok so the process:

Process is Double sided reflow for SMT & PTH via the wave. Like I said or did I, I forget and can't check back...it's a spray fluxer with noclean mutlicore NR3s-06 water based flux...I've tasted better. Anyhoo, using an Electrovert Electra, you know one of those dross making devices, preheat is covnvection, convection and IR. Finish of the PCB is HASL, paste is std Tin/Lead.

Previosly the guy's at this plant used a selective solder machine to get the solder joint, I think they used bruit force anmd about 6 sec's a joint....not a good combination. It's already been suggested that they redesign the board to have the connectors changed to SMT, but that's gonna take a fiar time to be done, in teh mean time I need to help them get some production.

In term's pin to hjole ratio it actually look's peachy, well I'd be happy with it if the PCB was FR4 and ony 1.6mm or 2 mm thick, not sure if it changes for a 2.8mm thick ceramic!.

still I hear McDonalds if hiring folks ....

reply »

#17163

Wave Soldering Ceramic PCB's | 9 August, 2001

Cheery disposition??? Would a few ;-) ;-) ;-) help? [Oh and look who is talking, the one planning the Mickie D gig after the traces on the this board start vaporizing with the fuseable link schtik.]

Are you sure you're getting flux all the way up the barrel?

You essentially need to get the top half of the barrel blisterin' hot and then fire the solder up the barrel so that it doesn't loose too thermal energy when it blasts past the signal plane. Then you end-up with an hour glass shaped deposit with a good slug of material at the top and the bottom.

May be the selective soldering machine wasn't such a bad idea.

reply »

JohnW

#17208

Wave Soldering Ceramic PCB's - a result | 15 August, 2001

Ok so an update on the ceramic board. I've had some success but did have to cheet a little.

So Dave, on the fluxing, your right we're not getting it all the way up the barrel and in the end as a short term fix we're dipping the lead's in flux not the best I know but it's getting us the flux. Really had to heat the board but still kept our BGA's under the magic 150 deg mark, 1.3 meters per min conveypor speed and a really good 50mm contact width on the old contour / smooth wave and pop here's the joint! :0) Longer term I've asked the guy's to get the fluxer serviced big time and we're trying some other flux.

will keep you all posted on this.

John

reply »

#17215

Wave Soldering Ceramic PCB's - a result | 15 August, 2001

Coo

What do you mean ehen you say "kept our BGA's under the magic 150 deg mark"?

reply »

JohnW

#17236

Wave Soldering Ceramic PCB's - a result | 16 August, 2001

Dave,

we try and keep the temperture that a BGA is exposed to at wave solder below 150 deg C. At this temperature it's unlikely that you'll get near to starting secondary reflow of the device or that you'll grow any significant amount of intermetallic's. I've seen cases where you have SMT components on the topside getting to about 165, 170 deg C and they can pop from the PCB, sometimes leaving the solder in place with a nice impression of the lead in the remaining solder.

John

reply »

reflow oven profiler