DIY smd reflow

I need to quit watching damn youtube videos and listening to 100 different answers. All I want to do, once in a while, is solder some SMD parts.

I'm ordering a hot air gun but --

also playing around with a toaster oven today and need to know --

do you preheat the oven? (I mean, after all you don't preheat when you solder using an air gun)

is it that big a deal to go through all this shit of going through different stages or can I --

(1)just preheat, put the board in.

or (2)just put the board in (cold), set the temp and let it come up to temp, which takes only a few minutes.

no in depth science please, can I just do 1 or 2 and which is better?

Reply to
mkr5000
Loading thread data ...

mkr5000 wrote in news: snipped-for-privacy@googlegroups.com:

The idea here is that you do not want the temp difference to be so high that the thermal shock introduced causes a mechanical break inside a chip or the end of a cap. So you bring the entire board temp up to near reflow temp, and then you can perform single chip adds or removals with the heat gun. That makes the transistion between reflowing solder and the board itself so small that there is not as much mechanical stress placed on things. if you are doing a new board and are trying to reflow the whole board, you would use the oven, but must be a very good process monitor to catch when it gets to reflow temps. If you are doing a single repair or part add, the heat gun works, but you still want the whole board brought up so that you do not do the thermo-mechanical shock thing.

Reply to
DecadentLinuxUserNumeroUno

That's good to know. How does a thermal profile impact IR preheat? It seems to me that excessive exposure to preheat may damage components. Yet my limited, undisciplined use of preheat with an air gun to remove components has yet to damage a component. Practical experience indicates that hot air and preheat melts solder at a lower temperature than an soldering iron. Any insight into why that's true is appreciated.

Thank you,

--
Don Kuenz KB7RPU 
There was a young lady named Bright Whose speed was far faster than light; 
 Click to see the full signature
Reply to
Don Kuenz

The approach recommended by the solder folks (and a lot of the component folks) involves a ramp up to a preheat temperature, a soak at that temperature for a while, another ramp up to above the solder's liquidus temperature, a short wait, then a fairly rapid cool-down.

The ramp-and-soak preheating process helps get the solder "ready to melt" properly, in that it allows the solder to gradually evaporate (and out-gas) whatever paste-solvent is still present. It also allows the components to warm up properly... slowly enough to avoid physical stress from too-rapid expansion, and thoroughly enough that all of the pads and pins are warm.

Then, as the temperature rises further, the flux becomes active (typically melting, I believe, and actively reducing oxides and lifting off other contaminants). A bit higher temperature and the solder alloy itself melts, and its surface tension helps pull the parts into proper orientation on the pads. Once that's happened (and it doesn't take long) you cool down the board.

Too-rapid heating can have several bad effects. It can damage parts by thermal stress (some parts expand faster than others). It can result in the flux and solder boiling (residual alcohol coming out rapidly rather than evaporating slowly and gently) and this can push parts around or even off of their pads. The flux may not have enough time to "clean" the pads and pins. You may get "cold" solder joints if some parts of the chip (e.g. the under-chip thermal pads) haven't warmed up enough to melt and bond solder before you start cooling down the board again.

So, there are benefits to warming up the oven at something resembling the recommended rate, with a soak/preheat period partway up. You can do this with a commercial temperature controller, or do it manually by watching a stopwatch and thermometer, and turning the switch/thermostat manually.

A quickie version would be to preheat the oven to a temperature a bit above the "soak" temperature, pop in the board, wait a couple of minutes, then turn the temperature up to the recommended liquidus temperature and keep your eye on the board. Once you see the solder melt all the way across the board, turn the heat off and open the oven door.

Reply to
Dave Platt

Ok -- I'm just going to put it in and watch for silver, wait a few seconds and take it out. I'd do trial and error but don't have that many boards to play with today.

Reply to
mkr5000

Some ovens (I speak from bad experience) have a heater element in the top that blasts parts with IR and destroys them. A toaster oven could well do that. Mine sure scorches muffins.

In theory SMD parts need a carefully controlled temperature-vs-time profile, but you can probably get away with 2 or three steps up and down. Preheat, put the board in, soak a few minutes, raise temp to melt the solder, turn heat off.

You can't just hand solder?

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

fredag den 31. januar 2020 kl. 21.31.34 UTC+1 skrev John Larkin:

or for few dollars you buy a controller that does the temperature profile

or a real reflow oven, I think you can get one for ~$150

with lots of part is is a hassle

Reply to
Lasse Langwadt Christensen

Sure, but stencils and solder paste and toaster ovens are a hassle too.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Foil lining the top of the oven should fix that.

Reply to
Tom Del Rosso

This one may be useful:

formatting link

It's a hot plate instead of a toaster oven. But you still have to deal with paste and stencils.

Reply to
Peabody

"Don Kuenz" wrote in news: snipped-for-privacy@crcomp.net:

It isn't a lower temp. It is about thermal mass.

A solder iron can only sink so much heat into a single joint and on a cold board that offset means the tip either has to be a meaty one or the tip heater has to be feeding heat to it all the time you are sinking it away. So the thermal mass of a solder tip matters too. Fine pitch tip puts down a part real well, but reflowing to remove it usually calls for a bigger tip to make up for the heat the board soaks in while you are on the joint.

On a preheated board the temp offset is only a few tens of degrees so reflow occurs right where/when it is supposed to... the melt temp for that solder.

A (spot) heat gun takes a spot region up to temp very fast and damage ocuurs on surfaces of nearby succeptible parts like capacitor shrink sleeves, which all (thru hole) EL caps have if the flow rate is too high. Connector headers soften and deform even if you don't spot it. Using a heat gun I use a very slow flow rate and very small spot for anything smaller than 1cm sq. The heat soaks into the part very well, so speedy air is not needed, just hot. But the preheat is needed as well to make the job happen faster. You only apply the heat for as long as you need to to pull the chip/part etc.

Reply to
DecadentLinuxUserNumeroUno

snipped-for-privacy@coop.radagast.org (Dave Platt) wrote in news: snipped-for-privacy@coop.radagast.org:

+1

Top info. Very good, accurate and very well stated.

Reply to
DecadentLinuxUserNumeroUno

I've been known to use: - saucepan with glass lid which keeps the heat in - gas hob so heat input and temperature is rapidly variable - thin layer of sand to act as a thermal buffer - non-contact IR temperature "gun" - wire through mounting holes to lift board out to cool down - watch progress through lid

No burning, have to brush sand off, no components on bottom side.

That's surprisingly successful, but I'd be somewhat reluctant to recommend it for professional use :)

Reply to
Tom Gardner

Neon John wrote in news: snipped-for-privacy@4ax.com:

You don't put it up next to the element. You put it down in the cavity, but above the work product. It homogenizes the temp of the cavity area below it at that point.

"burn out"? What kind of cheap chinese toaster oven brands are you cheaping out on? :-)

Using only the lower element works, IF you also place a foil shield OVER it. MUST be between it and the reflow target surface.

The goal is to stop directly radiated IR and homogenization of the cavity temp. Makes for a better, slower ramp up too.

Reply to
DecadentLinuxUserNumeroUno

The hotplate method is the bomb for things with, say, 100 parts. Ours is an old Corning ceramic-top hot plate, with a 10-inch square piece of half-inch aluminum jig plate on top. (We ordered the plate from McMaster-Carr for a few bucks.) We also got a nice Extech SD200 triple-thermocouple data logger to get the reflow temperature right.

You get a stainless or Kapton stencil for cheap (PCBway sends them epoxied to these Godzilla aluminum weldments, for amazing cheap), plunk it over the board, shim the edges so the stencil lies flat on the board, squeegee on some paste, remove the stencil, place the parts, and them reflow them on the hot plate.

It takes a couple of tries to get the temperature right, but for protos we've had good luck with the hotplate method.

Nowadays we usually pay to have PCBway or Screaming Circuits or Pentalogix do the assembly of first articles. It isn't that cheap, but overall it makes economic sense unless you have lots of technicians and/or assembly folks on board.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Empirical temperature data collected by me.

Hot air apparatus: T-8280 Preheater Aoyue 906 Rework station Chemical stand that vertically secures hot air nozzle BK 630 Thermometer Measured temperature to remove component: 220 - 230 C

Hot iron apparatus: HOKKA 472D Temperature setting to remove componet: 330 - 340 C

This reflow profile confirms my hot air data:

formatting link

Thank you,

--
Don Kuenz KB7RPU 
There was a young lady named Bright Whose speed was far faster than light; 
 Click to see the full signature
Reply to
Don Kuenz

Thanks.

Reply to
Robert Baer

It's my pleasure. Allow me to correct one glaring typo. It's a Hakko

472D desoldering station.

Many websites include the Kester reflow diagram above as an adjunct to their DIY oven controller, which the OP wants to avoid (and me too).

My Weller solder station must be set to a 315 C minimum to melt SN62PB36AG02 solder on its tip.

The solder composition of the boards in my bone pile, from which parts are salvaged, is unknown to me. Regardless, the Kester reflow diagram pertains to SN62PB36AG02 solder.

Thank you,

--
Don Kuenz KB7RPU 
There was a young lady named Bright Whose speed was far faster than light; 
 Click to see the full signature
Reply to
Don Kuenz

All I want to do, once in a while, is solder some SMD

Here's my system - I have done over 2000 boards this way. I got a thermocouple ramp and soak temperature controller off eBay. The ramp and soak option allows you to set temperature points and times to go from one to the next. So, I do 3 minutes from ambient to 180 C,

2 minutes from 180 to 235 C, hold at 235 for 1 minute, 2 minute ramp to ambient. I poke the thermocouple wire into a plated through hole in the board, as the air temp never rises ANYWHERE NEAR the board temp. Having the sensor measure the actual temp of the board gets you out of all the calibration and differences between different boards.

The oven is a large GE toaster oven with a solid state relay cut in series with the thermostat.

So, I don't preheat, the board goes in before heating starts, and I don't disturb the board until it cools to about 100 - 120 C. I usually open the door when in the cooldown cycle.

Jon

Reply to
Jon Elson

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.