UV SLA Printer Math - can someone check please?

[tom.k.cook]

I'm trying out the idea of a top-down SLA printer built with a 405nm laser (eg [1]) and a galvanometer (eg [2]). Can someone please check over the below math for me?

Assumptions:

• 0.1mm spot size - no idea if this is achievable with that laser. One issue is that I won't be varying the focus over the surface of the build plate, so the distance from the galvanometer to the build plate will have to be long enough that the deflection angle is small enough for the difference in distance to be fairly small across the plate.
• I'm assuming a square spot. It makes the math simpler.
• Resin critical irradiance of 45 mJ/cm^2. This is based on the spec for MakerJuice SF resin. There are other resins with much lower critical irradiance (eg MakerJuice G+ 10-15 mJ/cm^2). I've tried to contact other resin manufacturers for a detailed specification but no results yet.

Allowing an overhead of 50%, that requires an irradiance of 6.75e-6 J/m^2.

A spot size of 1e-4 m and a laser power of 0.5 W gives irradiance rate of 5e+7 W/m^2. (P / d^2)

That gives a dwell time on each point of 1.35e-13 s and therefore spot travel rate of 7.41e+8 m/s (approaching the speed of light...)

The galvo can do 30kpps (the headline says 20kpps but if it's only travelling 8 degrees then it can do 30kpps). Suppose each point means travelling 0.01m (10cm) then that's a spot travel rate of 300m/s.

That gives an actual spot dwell time of 3.33e-7 s and an actual irradiance of 16.7 J/m^2 operating at full power.

So to get my 6.75e-6 J/m^2 irradiance, I'd have to operate the laser at less than 1 millionth of its rated power. No doubt it could be operated at somewhat higher power than this with acceptable results, but I'm guessing that putting 1,000,000 times the recommended irradiance into the resin is going to cause problems.

Have I made a stupid math error here? Or is the idea of using a half-watt laser and an off-the-shelf galvo for SLA wildly unrealistic?

[1] http://www.ebay.co.uk/itm/405nm500mW-Sq ... Swl9BWMG6f
[2] http://www.ebay.co.uk/itm/20Kpps-HightS ... SwzgRWynrk

[iDig3Dprinting]

The quick answer is we don't know but we are trying to find out for you. If we get any info we will post back.

[tom.k.cook]

I've found one error: The conversion from mJ/cm^2 to J/m^2 is wrong.

45mJ/cm^2 is 450 J/m^2, or 675 J/m^2 with a 50% overhead.

A spot size of 1e-4 and a laser power of 0.5 W gives irradiance rate of 5e+7.

That gives a dwell time on each point of 1.35e-5 s and therefore a spot travel rate of 11.1 m/s.

The galvo can do 30kpps. Suppose each point means travelling 0.01m then that's a spot travel rate of 300m/s - much too fast. To get the irradiance I want, I have to operate the galvo at 1/27th of the galvo's capability, or just over 1kpps (but obviously that depends a lot on the distance between each point. Suppose each layer involves the spot travelling 10m, then it can do about 1 layer per second.

[tom.k.cook]

I've spotted the error in the math - the conversion from mJ/cm^2 to J/m^2 is off. With this corrected, the laser needs to travel about 7.5m/s - much more realistic. We're probably looking at something on the order of 1 layer per second, depending a lot on the print, of course.

[tom.k.cook]

More data points from manufacturers: Spot-A's Spot-HT has a critical irradiance of ~10-11 mJ/cm^2 (this test done with a broad-spectrum DLP light source) and MakerJuice's G+ resin ~10-15 mJ/m^2. So the value of 45 mJ/cm^2 used above is rather pessimistic and it would certainly be possible to obtain resins that would print 2-3 times faster than the math above suggests.

[iDig3Dprinting]

Well I we're glad you have been making progress. We tried to get some info from people we thought may be able to help but to no avail, nobody responded. We will have another try, but you seem to be getting there by yourself.