Get best of the week

DIY UV lamp 400-405 nm for polymerization of 3D photopolymer models

Good day!


It will be a question of solving such a problem as polymerization of finished 3D models after printing on LCD 3D printers, such as ours - Anycubic Photon S. This printer prints on the basis of photopolymer resin - after printing, the models are washed in isopropyl alcohol, but the models themselves remain sticky, soft.

A prerequisite for writing was the urgent and rapid need for a lamp for curing the printed models before painting. An analysis of the Internet showed the lack of ready-made budget lamps for sale - it’s strange, with such a prevalence of 3D printers now ...

Anyway, I hope the article will be useful to those who start to print on photopolymer and are faced with the problem of drying printed models.

Go


According to the instructions from the resin, the hardener for it is UV light with a wavelength of 400-405nm.
A proven solution would be to buy a proprietary dryer for the printer, but the price is not at all understandable - with a printer budget of 32 thousand rubles, pay 23 more for a box with LEDs some kind of flattery.



The solution is to do it yourself with the available components.

Were purchased:

  1. UV 3W LED Unit - 19 units for 68 rubles each (purchased offline in their hometown, apparently these with Ali )
  2. LED driver stabilized by current 600mA at 22-38 volts (spoiler: immediately died and was replaced by this one ) - 450r
  3. Table lamp ERA NL-202 black - the quality is lower than the plinth, but for cramp it will go - 1299r
  4. Radiator in Chip & Dip - 170 rub
  5. Hotmelt - 190 rub
  6. 2 meters of refractory cable - 60r





Since the lamp in our form has a boat shape, we fix the radiator under it: We





disassemble the lamp itself and throw out all the guts:





The first option is the arrangement of LEDs: 9 pieces in series - totaling 32.4 volts 600mA.



We prepare the jumpers:



Degrease the radiator:



Apply glue:



Drill the finished radiator with glued units - we will use the remaining regular holes for screws:







We tinker the contacts:



Solder everything in sequence:



Finish: try on a driver that fits right into the regular place of the lamp electronics



First inclusion






Turning on was successful, LEDs gave powerful diffused light.
2 minutes - the radiator began to warm up significantly for
5 minutes - the LEDs began to blink, the radiator glowed like an iron - the system was urgently turned off.
My mistake was that I did not check the supply voltage from the driver in time - but it was around 41 volts, that is, initially with the driver there was something wrong.





After the first power-up and test in 5 minutes - the driver constantly began to give a ripple of about 1 hertz - I still did not understand what had happened.

I poked the driver board a little from the sealant, but without possessing sufficiently deep knowledge in the ringing of the circuits, I simply wrote it off to the scrap:



Error handling


As a result, the tested LS-AA-2.1 - 12V at 2.1A was extracted from stocks. Under it, I had to recalculate the current-voltage circuit. Given that the PSU has voltage regulation, it was decided to make a 4-row matrix in series of 4 units in parallel. Thus, we get 525mA power supply per unit (at 600mA by datasheet), and adjustable in the range of 8 to 14 volts (about 2 - 3.5 volts per unit). Roughly speaking, the current mode is quite sparing, and the voltage - there is a field for selecting the optimal range.

So it looks:



Test inclusion at 13,43v: The



flight is normal, I waited 5 minutes - the radiator heats up again very much, but decided to leave this question to complete assembly and debugging.

Refinement compounds shrink:



The board of the new driver, although protruding from the aluminum case, does not climb into any regular place. I decided to return it to the case and put it on top of the lamp - a household steam-punk.



General view:



During debugging, one 4-string of LEDs began to fall off (turn off), I connect this with the previous crash - all the same, some units “sailed” from their factory characteristics. Sadness, but somehow we also have to live with this.

At 3.5 volts per unit, the system worked stably for 10 minutes, until it was warmed up to about 60 degrees (radiator) - then the same 4-chain started to fail.

He decided to drill radiator holes on top to remove heat:





The most stable result was obtained at 3.15v per unit - the system never blinked, it gives enough light. The temperature does not rise above 51C:



During testing:



Total


The project budget turned out to be about 3300r, of which the shit lamp came out at 1300r, instead of it you can simply fix the radiator through an aluminum corner or tripod. Thus, 2000 rubles is quite an affordable price for home DIY.

If you know the solution is cheaper and better than the proposed one, please write in the comments.

On Monday there will be testing on blanks - there will be UPD.

More articles:


All Articles