Idea For Laser Soldering

[Randy222]

Not a new topic in the commercial space, but was kicking around idea to DIY a handheld laser wand for pinning or soldering components. It would be a slow process compared to commercial robotic systems, but it's also DIY build. The main idea is using laser diode array fiber coupled, it runs very low power to create a visible strike dot on the target, then a button press to for a power pulse to do what's needed.

Thoughts?

[uer166]

Thoughts?

How good is your health insurance; and if you're selling kits, liability insurance?

[Randy222]

Thoughts?

How good is your health insurance; and if you're selling kits, liability insurance?

Selling kits? Health insurance?
No, and health insurance is up to you.

This project would discuss applications, diode wavelength, power needed, wand design, optics, holders-fixtures-jigs, control using an MC, safety, etc. It would be on you to source the parts, aka "DIY".

As example, pinning smd resistors to an sma connector when making precision Rf load module. Another could just be melting solder paste on items where soldering irons and ovens can't be used.

I don't know yet what the power requirement would be. 1w in 1mm² area is quite a bit different than 1w in 5mm², power density, etc. So actual power needed really depends on the target spot wanted, but I see many of the commercial systems will use 1mm² as the target area for delivering optic power.

[Gyro]

Thoughts?

My first thought is differences in reflectivity and thermal mass, not scorching the pads off the board. It must work of course for there to be commercial systems.

[unseenninja]

I fail to see the benefit of using an extremely dangerous laser for this application.

What benefit does this bring over all the currently existing ways of soldering which don't put the user's and bystander's eyesight at risk?

[Manul]

Emboss the buttons of the device with braille symbols, so people can use it for a long time.

[vivi-d]

A laser might be nice to have when soldering BGAs. Especially if you have BGAs on the opposite side. Hard to heat a board and solder one BGA without the other one on the bottom falling off. The laser would just heat up the package of the BGA which is enough to melt the solder (according to IPC). I haven't ever tried this though...

[Randy222]

I fail to see the benefit of using an extremely dangerous laser for this application.

What benefit does this bring over all the currently existing ways of soldering which don't put the user's and bystander's eyesight at risk?

Such device can be used in many types of applications, I just named one for soldering. Hermetically sealing is another but would likely need 50-100w of optical power in a tight spot to weld metal.
In some cases, soldering components in Rf modules leads to unwanted effects. If we can solder just a small fillet "weld" from smd edge to board tinned pad, you gain some performance (or take out unwanted affects).
Obviously the process is turtle slow compared to commercial robotic/CNC systems, but for the DIY'er doing one-off projects I think a tool like this could be beneficial.

I guess applications are broad, and would likely be limited to power and ability to aim the wand.

I am not sure why there's worry in high powered handheld lasers, they sell handheld laser welding apparatus today that you can buy, it can weld metals up to 0.080" thick and beyond, they even have wire feed units too.

[thm_w]

MiJing WLS-301 https://www.aliexpress.com/i/1005003800873309.html

I would not touch that thing because you'd want to be wearing protective glasses whenever it is in use.

[Randy222]

MiJing WLS-301 https://www.aliexpress.com/i/1005003800873309.html

I would not touch that thing because you'd want to be wearing protective glasses whenever it is in use.

So wear protective glasses, makes sense for safety.
But that's close to what I am talking about, thanks for the find. That one only makes a 6mm or 10mm spot, which is ok for some stuff, but I am thinking 1mm spot that you pulse to hit just a small spot.
Can it be DIY for less $?



https://youtu.be/ifu6k9FKjZ8

[thm_w]

So wear protective glasses, makes sense for safety.
But that's exactly what I am talking about, thanks for the find.
Can it be DIY for less $?

youtube.com/watch?v=Qvy50w-DeSs

The cost for two 15W lasers is in the $200 range. I don't know how you fiber couple those. I wouldn't bother.

If you want to DIY, you can do DIY IR stuff cheaply.

[Randy222]

So wear protective glasses, makes sense for safety.
But that's exactly what I am talking about, thanks for the find.
Can it be DIY for less $?

youtube.com/watch?v=Qvy50w-DeSs

The cost for two 15W lasers is in the $200 range. I don't know how you fiber couple those. I wouldn't bother.

If you want to DIY, you can do DIY IR stuff cheaply.

The fiber coupled is a module that is already built. Wavelength and power are the unknowns. Perhaps just a 30w module would suffice.
I was thinking more on the individual pin level vs the desoldering process, but it is possible to have both in one system.

With pulsed power you can get a weldment without creating any HAZ, we know power delivered is related to watts x time. A slower heat gives time for the surrounding area to heat up (HAZ), which I think we can avoid with more power in short pulse.

I think my 1st test is to buy a laser diode fiber module and test pulsing it using PSU and an MC for control.

[CatalinaWOW]

You seem to want a hand aimed device.  Human hands aren't notable for their pointing accuracy and stability, particularly when pulling a trigger.  If you are willing to accept machine control of the laser safety becomes much easier and packaged solutions are already available at moderate cost.  They are called laser cutters.  Developing the fixturing to hold the board and program solder locations becomes the task and isn't that hard.

Perhaps a more fundamental issue is how the heat is delivered.  The commercial products deliver a tiny spot, which may be appropriate for heating a single lead on a small component.  But optics delivers that heat very inefficiently to reflective surfaces.  What doesn't get delivered to the solder or lead bounces to unknown locations.  PWB material happens to absorb the light pretty well so you could easily burn the board without making the solder connection.  This isn't a purely hypothetical issue.  Both my son and I own laser cutters and have had wiring in the cabinet severed by random reflections.  It certainly motivates me to keep the cabinet closed whenever the laser is on.

If the beam is defocused to provide heat over a BGA package the problem of heat transfer is magnified.  Low power density and coupling which ranges from OK to terrible.

[thm_w]

The fiber coupled is a module that is already built. Wavelength and power are the unknowns. Perhaps just a 30w module would suffice.
I was thinking more on the individual pin level vs the desoldering process, but it is possible to have both in one system.

If you know they exist, can you link the sources you found.

I see a few on ebay, again in the $200 range, without any drivers or accessories.

[jbb]

…
Both my son and I own laser cutters and have had wiring in the cabinet severed by random reflections.  It certainly motivates me to keep the cabinet closed whenever the laser is on.
…

Well that’s disconcerting. Does it just burn through the insulation and lead to a short circuit, or can it sever the copper conductor too?

[CatalinaWOW]

…
Both my son and I own laser cutters and have had wiring in the cabinet severed by random reflections.  It certainly motivates me to keep the cabinet closed whenever the laser is on.
…

Well that’s disconcerting. Does it just burn through the insulation and lead to a short circuit, or can it sever the copper conductor too?

In one case burned through insulation and caused a short, in the other case burned through and unclear whether damaged conductor or if arcing from five volt supply.

[Randy222]

You seem to want a hand aimed device.  Human hands aren't notable for their pointing accuracy and stability, particularly when pulling a trigger.  If you are willing to accept machine control of the laser safety becomes much easier and packaged solutions are already available at moderate cost.  They are called laser cutters.  Developing the fixturing to hold the board and program solder locations becomes the task and isn't that hard.

Perhaps a more fundamental issue is how the heat is delivered.  The commercial products deliver a tiny spot, which may be appropriate for heating a single lead on a small component.  But optics delivers that heat very inefficiently to reflective surfaces.  What doesn't get delivered to the solder or lead bounces to unknown locations.  PWB material happens to absorb the light pretty well so you could easily burn the board without making the solder connection.  This isn't a purely hypothetical issue.  Both my son and I own laser cutters and have had wiring in the cabinet severed by random reflections.  It certainly motivates me to keep the cabinet closed whenever the laser is on.

If the beam is defocused to provide heat over a BGA package the problem of heat transfer is magnified.  Low power density and coupling which ranges from OK to terrible.

Let me try and put things back into the correct context.

1) Safety would be part of it, no different than buying anything else that you can buy now. Safety would be in the design, like glasses, clothing, perhaps a cabinet, etc.

2) Next is "handheld". I am not wanting to build lots of jigs with stepper motors and cnc programming, and I am not talking about just holding the wand and blasting things, although I guess you could. This is a wand style one shot idea, you steady the wand, aim it to a spot, make a weldment, very slow process, perhaps more in realm of "slow precision". The wand itself would be small like this pic of a dremel wand attachment, but in the process of using the laser there would be a holder for the wand, you aim the dot to location and the make a pulse to do the work. The laser wand would look something like this pic. As for commercial tiny spot, I have seen some vids where user is actually doing handheld work and the fiber is in a clear poly tube, the poly carries a shielding gas and the tip looks like a big syringe end (plastic lab type), user just puts it right up onto the spot to be welded, and a pulse comes.

[LaserSteve]

When you get into laser diode fiber bundles, such as FC-Bar etc, you learn very, very, quickly that back reflections from shiny metal and certain optics cause the output face of the laser  diode(s) to blow off from excess energy density and resonances.  So, even with high power optical isolators, polarization schemes, etc.. You run the risk of sudden and complete laser diode death, Since by definition your aiming at a potential molten  metal mirror, even a tiny amount of stray laser light returning to the laser diode(s) can/will result in damage, if not a full on laser casualty.

Hence why you do not see inexpensive laser diode based soldering tools.

Not to mention that coupling 808 nm near IR laser diode wavelength  (typical surplus diode for ND:Yag or ND:YVO4 laser pumping) into solder, which averages 60-90% reflectivity in the near IR, is very, very difficult to achieve. Once it starts to melt you'd need to ramp up the power very fast to continue the melting process.


Even with Optical Density Six, very professional, very expensive, very broadband, laser safety goggles the eye hazard to you and those around you, is huge.


Not to mention the NOHD (Nominal optical hazard distance) for a roughly collimated fiber output laser might easily exceed two hundred feet.

I once was installing a 7,000$ diode laser array in a system repair in the field.  I warned the graduate students to under NO circumstances ever arrange for a back reflection into the fiber.  One of them, trying to teach the old man "Whos Boss" did just this while I was on site.  Whoopsie..   I caught him on cell video aiming 50 watts at a reflective chrome trim piece.   I'm sure his professor enjoyed two weeks of downtime while another diode came over from Europe, 7,000$ for the array, 1000$ for my travel, and 2000$ in fees from my then employer.    The young gentleman had an Ego / paranoia / racism / Trust Issue problem and wanted his own measurement of device power.  I had told him not to touch the fiber...  But he and his friends wanted to show off how smart they were, and did what the American told them not to do.  Took less then then a millisecond for an expensive, deliberate mistake.  My employer sent another tech with me for round two, for obvious reasons. My boss told the customer that his warranty ended the moment the student unscrewed the SMA fiber when warned not to.   My point is, if you have no idea what your doing, high powered diode lasers are NOT for you.
 
NOT a great idea to try to solder with a laser,  systems that do use this method use high power, ultra short pulse,  lasers to partially melt an existing  solder joint, then watch the heat decay curve on a thermal scanner to verify the integrity of a solder joint.  That is not a candidate for home construction, needing a Q-Switched laser source. Key word here is partial melt, not soldering.

Older laser diodes that you might find on Ebay etc.. Often have  a severe  aging problem that might not be apparent when first purchased surplus.

Did I mention the sensitivity to current surges, turn on spikes,  and static discharge events? As well as  the need for a  very high amperage constant current driver working into what can be a near dead short?

Get yourself an ERSA iron and enjoy life.