DIY PCB Using a Fiber Laser. Video.

[KG7AMV]

I have Only had the fiber laser a few days. Lots to learn anyone have any settings that will not burn the substrate? Using easycad2.

Laser Source is JPT 60 Watt MOPA M7 (YDFLP-60-M7-L1-R)

[jeremy]

Very cool! How much was the laser?

Unfortunately the way to not burn the substrate is probably by decreasing the pulse width. But that won’t be something you can do with these sorts of lasers.

[SiliconWizard]

That's cool. Tuning the laser so that it will cut off the copper without burning the substrate underneath will be tricky, especially since the copper transmits heat pretty well.

Just a note of caution: you should probably use some vacuuming while it proceeds and/or wear decent protection. This vaporises copper and I guess tiny particles of copper will be floating everywhere nearby in the air, not something you wanna breathe in.

[KG7AMV]

Very cool! How much was the laser?

Unfortunately the way to not burn the substrate is probably by decreasing the pulse width. But that won’t be something you can do with these sorts of lasers.

This is a MOPA Laser and Can Change Pulse Width.

[jeremy]

Very cool! How much was the laser?

Unfortunately the way to not burn the substrate is probably by decreasing the pulse width. But that won’t be something you can do with these sorts of lasers.

This is a MOPA Laser and Can Change Pulse Width.

Sorry, what I meant was more that the MOPA lasers from Chinese engravers don't typically go lower than about the ns range I think, but I'm pretty sure that the LPKF lasers that do this are more like 100ps or less (in research or high tech manufacturing, they use femtosecond lasers!). As far as I know, the key to working with very thermally conductive materials like copper and aluminium is maximum energy in minimum time, because you want to basically vaporise or ablate the material as fast as possible to avoid heat damage to other parts of the substrate. So this is one of the areas where shorter and shorter pulses ie ps and fs are driving the research into these lasers.

I haven't actually used one of these, but I would suggest making the pulse width as narrow as possible, and if possible when testing examine the board with a thermal camera to see how much heat is being spread out by thet copper.

Also, I believe the LPKF machines do some hatching process where they slice up tiny regions of copper and then blast them off with an air knife at the same time (I guess maybe they have poor adhesion to the substrate when heated due to degradation of the epoxy).

Definitely try to contact this guy: - he seems to have a good idea of the process window. Also please report back!

[free_electron]

Definitely try to contact this guy: - he seems to have a good idea of the process window. Also please report back!

he also forgot the invert the image .. all he has now are missing traces surrounded by a massive copper plane ....

[PlainName]

That fibre one is taking several passes, which would have a lower heat transfer per pass. That could be going a long way to preventing laminate burn.

[free_electron]

That fibre one is taking several passes, which would have a lower heat transfer per pass. That could be going a long way to preventing laminate burn.

you still need to reach temperatures to vaporise the copper. melting it is not enough . you are turning it into gas ... not a good idea.

[PlainName]

Sure, but a quarter of a layer at a time means it's high heat but low exposure time, so conduction isn't as bad as doing it all in one pass. Kind of poor man's fast pulsing.

[KG7AMV]

I will be trying more settings and hashes as soon as I get another shipment of FR4. But the boards are functional just burnt substrate.

With multiple passes the copper peals off along with the traces.  So right now its going to be 1000 ways not to do it but the setting will be found.


My 60W JPT M7 Fiber Laser Specs are.
Frequency Range 1-4000kHz
I can Pulse Range 2-500NS

[Marco]

Isn't a little carbon heat insulation basically the only thing which can protect the PCB from the laser?

Without some kind of transfer step, gluing the copper to a PCB with a sacrificial carrier, I don't see how you could avoid charring entirely.

[jeremy]

Isn't a little carbon heat insulation basically the only thing which can protect the PCB from the laser?

Without some kind of transfer step, gluing the copper to a PCB with a sacrificial carrier, I don't see how you could avoid charring entirely.

It is all about the pulse width. The goal is to hit the copper with so much energy so quickly that there is no time for the heat to move anywhere else. In terms of thermodynamics, the thermal gradient is incredibly large, but the timescale is also incredibly short, so your time averaged thermal conduction is basically zero. Laser ablation has a similar advantage to machining where you (ideally) remove heat from the workpiece in the chips; the heat is literally moved away from substrate the by the superheated copper exploding.

See this video for a general explanation and many examples with no charring (they use a 1ps laser pulse):

https://youtu.be/278VPs4eT7Y

[KG7AMV]

See this video for a general explanation and many examples with no charring (they use a 1ps laser pulse):

https://youtu.be/278VPs4eT7Y

That laser in the video the link gives specs.
Laser Wavelength 515 nm
Laser pulse frequency 50 to 500 kHz
Laser pulse duration 1.5 ps
Laser power Max 8W

I was using a pulse of 10-25ns from maybe try it out at 50+  The wave length on my laser is 1064nm. 

[analityk]

So you have 10000 times longer pulse so power density is 10000 smaller. As long as 25 ns really short period for copper vaporizing to to long.
Another problem is wavelength. Copper color suggest it absorbs green or blue light. Mirrors for infrared lasers and in some type of infrared apparatus are made from copper. So probably you have to have 100000 times more power to vaporize copper with your laser.