IR or Vapor Phase for Prototypes

Phil
Welcome to Board Talk. This is Jim Hall and Phil Zarrow, the Assembly Brothers who sometimes go as ITM Consulting. We're here to discuss process problems, situations and possible solutions to all your electronic assembly needs, or at least the ones we think we're qualified to answer, or even some we're not qualified to answer. That never stopped us before, did it?

Jim
Never. What is the question today, Phil?

Phil
Today's question is from a Mr. P. B. from somewhere: "Could you please compare the advantages of IR convection reflow and modern vapor phase reflow with polymer solvents for small-scale prototyping environments. There are many suppliers for batch ovens and it's hard to find spares of vapor phase equipment. That should normally mean that the supposed process control and environmental advantages, the vapor phase process has some important disadvantages about which the cognoscenti such as yourselves (flattery will get you everywhere) could expound. So please do."

Okay, so this is a vapor phase question, which means time for our vapor phase expert: Jim Hall, front and center.

Jim
Thank you, Dr. Phil. Yes, from a process standpoint vapor phase is superior to convection-based systems. Think about prototyping: you have maybe only one set of parts, you want to get a functional board to get it operational and you don't want to damage anything.

Vapor phase allows you to take any assembly, put it into the vapor phase system and completely reflow all of the joints without any risk of overheating anything and damaging it because of the inherent temperature control of the vapor phase process: 215 degrees for tin lead, 235 degrees for lead-free assembly.

Whereas with convention-based, first off we want to talk exception to the term "IR Convection." To us that means the old Vitronics panel IR-based systems, which are definitely inferior in the heating characteristics and are even worse for prototyping.

We hope that you are talking about modern convection-dominant systems that have powerful fans that transfer most of their heat by convention and give you a much better temperature uniformity and allow you to keep the temperatures lower and minimize the chance of overheating.

But still, with the prototype you may only get one shot. So you've got to set up a recipe in your oven and you either have to be conservative and use a short cycle, or fast conveyor speed, and low heater temperatures to minimize the risk of overheating something and damaging it. In that case you may not get good reflow on all your joints. If it's underneath a BGA or another area array that could be a real problem in a prototype.

If you go the other way, and to ensure good soldering you have to use longer cycles or slower conveyor speeds and/or higher heater temperatures, in which case you take the risk because you haven't had an opportunity to profile the board of overheating and damaging something that may, as we say, be a one-up in a prototype environment. So vapor phase is definitely better and safer and easier to get a good reflow first time, every time.

Why don't you find much vapor phase equipment? Quite simply because the equipment is expensive and the fluids are expensive. With vapor phase even for a simplest system you have to have a lot of expensive hardware. You have to have a stainless steel high temperature tank, high temperature heaters, cooling coils with an active water circulating loop, either once in or recirculating. You've got to have a filtration system which adds a lot of money to the base cost, even for a relatively simple system to do small prototypes.

Also the fluid is very expensive. Typically $500.00 to $700.00 a gallon for tin lead fluids and for the lead-free fluids, which you have to boil at 235 or 240 there's only one supplier in the world. I've heard numbers above $1,500.00 a gallon.

Even with the best systems you're always losing a little, particularly in a prototyping system where you can't optimize your cycles and so forth. So you have a high equipment cost, initially, and a high operating cost because of the expensive fluids. That is why you don't see a lot of small vapor phase systems, even thought it has the advantage.

Most people feel that a good convection-dominant system is good enough, it gets them close enough, they understand it, they can get a profile that they feel comfortable and can give them reasonable soldering without overheating things. But there is always that risk, which you completely eliminate with a vapor phase system.

Phil
So maybe you might luck out and you find one of those ancient original batch systems that Jim used to build in the old days. But whatever you do, as Jim said, stay away from IR convection. It's the antithesis of prototyping: there's a lot of trial and error, a lot of burning of potential for overheating,

And remember: whatever you do and however you're doing it, be it vapor phase, convection IR, convection, wave solder or selective --

Jim
Don't solder like my brother.

Phil
And don't solder like my brother.

Jim
Keep those kids away from the flux pot.

With regard to DS' comments, thank you and some of your points are well taken. However, I do consider my colleague, Jim Hall, an expert on vapor phase. While at HTC, in the early 1980's, he was involved in the development of vapor phase for electronic soldering and you'll find his name on some of the patents.

He was also involved in the development of IR and Convection Dominant while at Dynapert, Vitronics and Vitronics-Soltec. I would, thus consider him one of, if not the leading expert in reflow and I think he approaches the topic with objectivity as well as open-mindedness. On the other hand, he's not selling anything. Just sayin'...

You mentioned that vapor phase soldering uses temperatures of 235 to 240C. We utilize a .8mm RoHS BGA that states: It is recommended to apply a soldering temperature higher than 250 C. Could we use a vapor phase system for this part? Do you have any idea why this part recommends temperatures higher than 250C?

Guys I think you have been buried in the North East in convection Reflow Central for a few years too many. Fluids are only expensive if consumption is high, it typically is not.

Remember expense is only a cost in the absence of value. Vapor Phase brings significant value to prototyping shops as it does all electronics assembly houses dealing in exotic products or situations. And lets be reasonable and logical. Vapor phase machines have no more stainless steel in them than a convection machine, operate LOWER temperature heaters as they do not have to create peak temperatures significantly above the target temperature, are easier and faster to profile and operate at ~40% of the electrical power of an equivalent Convection machine. This makes them extremely cost effective to own and operate before we factor in the benefits of costs saved due to higher yields and first time pass rates.

And just for the benefit of all - PCBs in a convection system are NOT heated as a result of convection, they are heated as a result of conduction. Convection describes the movement of thermal currents, conduction describes the transference of heat from one media to another. The thermal conduction characteristics of gases, air or nitrogen are a fraction of those of vapor and it is this in efficiency that results in Convection reflow machines consuming considerably more electricity than the rest of the line in aggregate.

I am happy to chat more about the process and the current state of the technology and machines, either in batch form or in-line, with Vacuum or without.

In a vapor phase system, is there an issue with condensing fluids removing flux from the items to be soldered? Vapor degreasers are very similar, they work very well at removing anything that was soluble in the fluid. Are special flux types sold for vapor phase use?

There is no issues with the vapor removing flux. Vapor Degreasers are a marginally similar technology using a completely different family of fluids. With vapor phase, start with the flux you're currently using, create a proper profile, and go from there.

As much as I enjoy your "Click and Clackish" banter, I felt it necessary to deliver a few facts regarding your "Vapor Phase vs. IR conversation for Prototyping" conversation.

1) Vapor Phase Expensive? Most units run at a fluid operating cost of well under $5/hr. They don't use Nitrogen; they have lower power consumption rates and generally smaller physical footprints reducing floor costs.

2) Expensive equipment? Units for Lab and Prototype processes start around $18K

3) You say that people can understand convection; they can get their head around it? So that's why it's better? Do you mean ignorance is bliss?

4) If Vapor Phase is the process that "lets you do it in one shot" why would you ever use something that required two shots?

5) The Pb free Vapor Phase fluids you're groping for are the Galden LS230 and the Galden HS 240. Both products are manufactured by Solvay Solexis. The 230 and the 240 in the product names denote the boiling point (vapor temperature) of the fluid. And yes, Solvay Solexis is currently the only manufacturer of these boiling points. However, they are not the only manufacturer of vapor phase fluids which range in temp from 200 C* - 320 C* and with the current rise of vapor phase due to its superior reflow methodology, I feel it won't be long before the fluid market expands just as the vapor phase equipment market has.

Leave the radio shows and Vapor Phase to the experts. Or, become experts.