Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Vapor Phase Reflow

Vapor Phase Reflow

Subscribe to Thread

We are evaluating Vapor Phase right now, and there is someth... - May 07, 2010 by Hegemon

I assume you using a fixed mass vapour phase system? Yes ... - May 07, 2010 by josh

A huge help in the very first statement, Josh. And yes, fix... - May 10, 2010 by Hegemon

We are using VPS, but only for high mass assemblies. As t... - May 12, 2010 by Pete B

Thanks Pete B - great info. I am currently preparing a DO... - May 12, 2010 by Hegemon

Just a note, on fixed galden machines yes the way they compe... - May 12, 2010 by josh

Hegemon

#61782

Vapor Phase Reflow | 7 May, 2010

We are evaluating Vapor Phase right now, and there is something that bothers me a bit. The machine has a preheating stage where we can bring the board, components and paste up to temp to dry the paste out before descending into the Vapor. The heating in the preheat area is very gradual and smooth, less than 1degC/sec, and we hold the PCB at 150-170C for a minute before going into the vapor for about 20 seconds total. This is where I get a little itchy. From that 170C temperature, dropping into the vapor raises the temperature EVERYWHERE on the assembly almost instantly. Literally the surface of the PCB, and the rest of the thermocoupled locations go from 170 to 230 in about 10 seconds or less. Somehow that ramp rate does not jibe with what I have always considered a safe rate of heating, around a max of 2.5degC/sec for my sensitive components. Yes I know the heating is much more even across the PCB than with convection, and per 'the pitch' this is supposed to be my safety net. Is anyone aware of any articles that address this thinking, or might others have experienced this (for me)conundrum?

reply »

josh

#61785

Vapor Phase Reflow | 7 May, 2010

I assume you using a fixed mass vapour phase system?

Yes rates of rise are impacted by the mass of the PCB, larger the mass to mass of vapour phase the lower the rise generally. So if you load the machine with very little mass PCB rise times can be extreme. Fixed mass Vapor Phase needs to match the load to control the rise times.

Rehm has a system that solves this - galden on demand, can gauge the amount of galden required to transfer the heat to a given mass of PCB/s at any given time in the profile...you should check them out and have chat to them... ALso there systems offer lower usage of galden.

Vapor phase can have issues with electro caps - have to check with vendors as some state warranty void on vapor phase etc as rate of heat transfer is too agressive etc.

Hope this helps a bit !!!???

reply »

Hegemon

#61792

Vapor Phase Reflow | 10 May, 2010

A huge help in the very first statement, Josh. And yes, fixed vapor mass in the machine I am currently evaluating. OK, and a machine that can change the vapor volume based on a calculation of mass of the assembly sounds logical.

(whirrrr,click... and clunk! say the gears in my head)

But what about the components themselves? They are islands in the vapor until they attach to the the PCB and though they cannot reach more than the temp of the vapor, it eats at me that the last ramp to reflow pretty darn steep to the device itself. Kinda like a Thanksgiving dinner, many important parts, but the turkey (like the solderable device) is what it's really about about.

I'm doing some DOE with this in mind over the next week. We'll see how it goes.

I was hoping some of my esteemed bretheren here might know of some studies to which they could point that relate to this, or might just let me cheat on their answer sheets, OK?

'hege

reply »

Pete B

#61809

Vapor Phase Reflow | 12 May, 2010

We are using VPS, but only for high mass assemblies.

As the other respondent has suggested if the mass of the assemblies is very low then difficult to control temp. rise times to within the normally accepted parameters. The Rehm equipment does seem to offer a solution in these circumstances.

Our system has a means to control the rise time on the assembly by regulating the heater power to the fluid in the sump, thus controlling how fast the vapour is replaced after the initial condensation on to the assembly. The entry speed into the vapour zone is also slow.

We can achieve 1-2deg. C per second using this equipment and our assembly type.

For information we use Asscon equipment.

reply »

Hegemon

#61810

Vapor Phase Reflow | 12 May, 2010

Thanks Pete B - great info.

I am currently preparing a DOE to address concerns about reflow of componentry on the bottomside of the PCB while in the vapor. Specifically cosmetics and part dropping.

Bottom side has no flux, but will be fully reflowed. Inert environment protects against oxidation, but is there any other concern for the bottom side soldering? My thinking is that the practically zero delta across the PCB will mean complete and full reflow well into liquidus on the bottom side.

Yes I know there are formulas for surface tension and weight calulations etc. out there, and I will devote myself to the book end of this job shortly, but just for the sake of it, I figure reflow is reflow, and for all intents, I should experience increased surface tension of teh solder as I am reaching a lower peak temp (230C) than I would in my convections ovens, and shouldn't experience any increase in dropped parts.

Anyone with some Vapor experience want to check in on this?

Thanks all

'hege

reply »

josh

#61812

Vapor Phase Reflow | 12 May, 2010

Just a note, on fixed galden machines yes the way they compensate for rise time is as the previous post suggested, change power and also stage the lowering of the PCB into the vapor cushion ie not the main one but further up in the tank. The disadvantage some times with this is that cycle times can be impacted ie if you want a 4 minute cycle is sometimes hard to achieve, locked to reducing rise times vs mass of load. Might end up with 5 min or 6 min to get time you need. This is not gospel as all thermal responses of assemblies are different etc

reply »