Let's Make Robots!

A PCB printer for photoresist which uses a PHR-803T Blue-Ray™ pickup

Print PCB's on photoresist materials with a blue-ray pickup


About a year ago my brother and I engaged on a quixotic project to build from scratch a low-cost PCB printer for photoresist. 

Finally we did it using a Blue Ray™ optical pickup (PHR-803T) and his UV laser/electronics to sensitize Dry-film.

The optical pickup is used through his own connector without any modification. We deduced the pinout using reverse engineering and designed a driver for an Arduino UNO to control it. Basically, we can adjust the laser power, turn the laser on/off, move the focus lens servo and read the photodiode array signal.

The control of the pickup allowed us implementing a laser auto-focus algorithm based on the astigmatic method. Is the same method used to focus the laser beam on the Blue Ray™ disc.

The mechanics is a X/Y Cartesian robot which moves the pickup with two Nema 16 stepper motors and T2.5 timing-belts.  Most of the printer framework has been printed with a 3D Reprap printer, so it’s replicable.

We also developed the Arduino UNO firmware, based on 3D printer firmware Marlin, and the server side software which send the HPGL commands to the printer. Eagle Cadsoft complete the “tool-chain”. It supports HPGL in his CAM processor, so is easy to design the circuit with Eagle, generate the HPGL file and send it to the printer.

DiyouPCB is a prototype and it's still not perfect: we have some resonance and vibration issues that affect the quality of printing, but it's a first step to have an alternative method to indirect methods as the famous toner transfer.

We shared all the information about how to build it and use it in our web site www.diyouware.com.

Our next project will be to build an UV resin 3D printer using also the PHR-803T.

Charlie & Victor

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I was very impressed with how much effort you guys put into reverse engineering that laser assembly. Ocassionally when I have had a dead CD or DVD player I have looked at the laser assembly and hated to throw out such a lovely piece of engineering.

As for your vibration problem. I have 1 suggestion, replace the stepper motors with coreless motor / gearbox assemblies and use encoders. The reason I suggest this is that most motors, but in particular stepper motors suffer from "cogging" where the magnetic circuit causes the rotor to "click" into set positions. Stepper motors do this on purpose for accuracy but it causes a lot of vibration.

A coreless motor has no cogging. This allows it to run very smoothly without vibration (assuming a good quality gearbox or belt drive).

I know that "play" in the gearbox can be a problem for accuracy but here we can learn a lesson from the inkjet printers. As the printhead runs left or right, the encoder tells the inkjets (or in your case the laser) when to fire.

To compensate for the "play" in the gearbox an offset value is used. That is why inkjet printer do those test pages where you need to tell it which setting causes the lines to line up. Your project can auto align itself in a similar way that you do the auto focus. Have a small piece of reflective material outside of the normal print area and get the head to scan across it left to right and then right to left.

When you scan left and right, your code looks at the encoder values as the laser hits the edge of the reflector. From this an accurate offset value can be achieved to compensate for "play" in the gears or belt drive.

You never cease to amaze me at the mountain of knowledge in your head. That is a very interesting description of how the printers work and how you could adapt it here.

Spent too much time playing with electronics when I should have been chasing girls :D



On the other hand, I spent too much time playing with girls until one caught me and now I have to play with electronics...:-)

Thanks very much for the tip. I also think servo motors will be the solution. I found a cheap one at ebay (see link) and it seems has a good resolution. We should only add a gearbox to reduce speed and increse torque. I 3D print some stepper motor gearboxes prototypes with the RepRap with good results (see picture) so I think will be possible to print one for a servo motor. About the backlash I already write some code for the firmware to compensate it, but t's a static solution: always compesates with the same amount, but your idea is really interesting.



Stepper motor gearbox



I saw in some UV resin 3D printers that they use an Iris Diaphragm to reduce the laser spot diameter.


I unknow the cost. May be you can use or create something similar.

Anyway, if you go for a solution with a focusable laser diode with alu case, lens plus iris diaphragm, then I will also add two galvos to move the laser beam because the laser feedrate can be really fast using galvos. Galvos are simple: a stepper motor with a piece of mirror.


I'm think you can make them with a stepper motor like this https://www.sparkfun.com/products/10551 and a piece of thin mirror or even better use two RC servos.




Looks great.  Phase II should have a drill head to drill the through holes after etching!  Same chassis, I don't see why not.



Yes, drilling is still a nightmare.

Yesterday I received an email from a guy who wanted to add our blu-ray pickup head to his CNC machine like any other tool.

Some expensive CNC machines has this feature and can automatically exchange the carving or drilling tool. Why not print in 3D a tool holder and install there the optical pickup?. The CNC machine could cut the board, drill, them paint the circuit with the laser, all with the same machine.



I used more simple solution from two old DVD drivers.

I named this BluePCB project :-)


(sorry on Russian only)

Photos at bottom of page

Good idea to use the CD/DVD stepper motors. Really, we should have done our
machine smaller because we never we are going to print so large PCB's so we
think small and precise will be better. I saw that you are using the laser diode
with a laser case. We decided to use the original pickup without
modifications through his own connector and move the lens to do auto-focus.
In our approach the lens are very close to de PCB.

I have tryed to resolve problem with autofocus. Second version of my machine has Z axis small stepper motor (used FDD parts).

But while without success :-)

Maybe I will try to use your solution with stadard optical pickup module. But is is not easy.

About precise. My solution allows move XY with 0.04 mm per step of motor. But my laser beam is too big for such resolution.