Build updates

Added a mirror to the path moved the mirror from the path went back to top down with a slightly adjusted throw then rebuilt a light tube to allow for the led to be added and removed for testing
purposes. Time to start looking at the sweeper mechanism

straight z with adjustable height

In the picture above you can see the projector is finally mounted to the frame albeit temporarily I think this is becoming the final setup. The frame is likely to be moved around some more to make room for the z screw when it arrives and the sweeper.

collimator attached to light-pipe with 90 degree pipe

I plan on adding a mechanical shutter on the collimator side, and attaching the power led to the shutter.

weekend work

I received the misumi order Thursday of all that beautiful extruded beam, and promptly set about procrastinating while I was deciding how to build the box itself.


So I cut up 4 beams into a set of  various lengths and set about building a frame.

I realized once I had a basic setup that the linear path was going to drive any freshly printed model right into the lens.

I then decided to try the mirror method.

Following the guide on calculating mirror distance I came up with a number for my height.

Cut some mirror

and spent some time re-arranging my beams to get various configurations looking for the right height/width combos with my now cut beams.

Even more on lenses and optics....

After some discussions on irc with some knowledgeable people in #dlp3dprinting

There is no need to modify a working projector. Make sure you have a projector with 3k lumens or more and you can either use a mirror ala this instructable  or with closeup lenses from cameras. The b9 creator I think it was is using a +4 closeup lens for their unmodified projector.  77mm lens packs with 6 assorted lenses can be had for 12$

At this point if I want a better DLP I dont need to worry about destroying a projector I can just use a good one and when not printing I can watch tv :) although i seriously doubt I would waste valuable bulb life on a movie.

I know my eyes are particularly sensitive to uv, its both inherited and sensitized from years of contacts and a little laser surgery. Because of that I've been waiting on some glasses but I'm not the only pair of eyes wandering around and while I have been making sure kitty is out of the room when I fire up the UV I thought I should take a look at case options for containing the near UV I'm working with. A little looking on line and I see tap plastics acrylics block a decent % of UV so I wandered down with a 405 nm light source and something fluorescent I found the amber worked well.

bare 405nm 9 bulb flashlight pointed at paper with highlighter

camera behind filter light source in front of filter

This picture shows that fluorescence is a re-emitted light its visible through the near UV filter

Light through filter, camera unfilterd

camera and flashlight behind lens, the brighter spots are reflections from the shiny surface

Here you can see that light has made it through but unlike the pic 2 above there is no fluorescence thus the light penetrating the filter is no longer 405

More on the lens

I wasn't happy with the way the lens focused... that's not it... I didn't have a build are big enough to build the kind of things I was after. so I took the lens unit out and messed around a little bit. It seem that by moving the built in optics forward around 2 mm in my case I was able to focus the pixels almost exactly to .1mm. Actually between the throw and focus I can get a variety of resolutions. From about 44 mm X 33 mm (.044mm per pixel) all the way out to 110 mm x 82 mm (.11mm ppx).

The macro lens I had is unnecessary.
I will need to make a collar or something to set it in place properly there's a little bit of wiggle with the screws in place and it would be nice if I could adjust it. The depth of field? if that's the right term is very short it is only in focus within a very small distance, and the throw length? distance from projector to the focused image is now 16.5cm when zoomed out and 9cm or so when zommed in

in place with the 2mm offset
Much damage to the case from trying to fit unneeded optics

The optics unit

The dmd and light prism

The optics unit part way in place

A focusing lens

I decided it was time to work the optics out.

Upon opening my projector I see it has this nice sealed lens system leading into the light path The tiny lens and focal point made me not want dig inside. I happen to have a few old camera lenses laying around an after a little fiddling found the macro lens that worked.

random macro lens from a view camera

The iris mechanics unscrewed which was good
I needed to get the lens closer to the aperture

I notice the light does not exit my projector in a level manner its angled "up" at about 15 degrees this meant the lens needed to be fitted on at an angle.

Hung loosely in place just to get a rough idea

6 cm x 4.5 cm the dark edge is the camera taking the pictures shadow notice the pushpin in the upper left.

A couple of good links that will make everything easier...

Software that slices objects for dlp and lcd type systems by PacManFan

and A group of people doing various versions of the same thing I am BuildyourownSLA

I got my connectors and installed them on to the steppers, which I then attached to the board, and wired it all to the power supply with a fan.

Duino power and steppers next to projector

I mocked up a light source and by that I mean I pointed a 3w white bike light into the light path. I had some trouble getting the projector to find an input signal. Eventually I decided to just unplug my primary monitor and leave the projector as the only display option. The computer forced a vga signal and poof I had a display

#reprap projected onto white sheet of paper
form a modified Acto dlp Projector
powered by a bike light

OMG yet another easy way to get where I'm headed, TY Diyouware!

http://www.diyouware.com/

These guys have taken a dvd player mechanism and hacked it enough to hook to an arduino.
The current incarnation is a PCB printer, something that will sensitize a uv photoresist. They only need to add the z motion of a keyhole type SLA and they've got an ok resin printer. .1mm seems to be the current resolution available, I think that's a mechanical function of the frame and steppers and better mechanics will allow higher resolution.

I've ordered a few of the 7$ laser modules and am looking into getting the pickup board made. since they provide the files.They are running it off an ardiuno uno, and I'm pretty sure I'm gonna want to upgrade the core to a 2560 just because the full version of marlin will install. so i might need to "design" a new shield thankfully that's barely within the scope of my abilities... gotta love software...

eagle cad

gonna see if I can't borrow a ramps schematic and merge it with their shield schematic.

DLP Projector Now ready for any light source

I have an Acto AT-X51 dlp projector...


 This is a perfect projector for my purposes, it has a 1024x768 dlp chip, and they no longer make lamps for this projector its super cheap. After rather a bit of disassembly and tracing wires around I found the right thing to short to prevent the MB from checking the lamp. its actually accessible right under the keypad.

notice the solder blob on the first two pins of the "ballast"

 Next I see this model knows up from down so I need to disable the switch, on the other side of the same board so you dont have to remove the board it to short it.

The odd little can with the silkscreen sw1 Disabled by shorting with a solder blob
Shorting it makes the projector think it is upside down. desoldering it would keep it right side up

as well as the color wheel because this is to be a UV setup and the color wheel gets in the way. well there's an encoder and I cant remove the setup so its just breaking the colors off the wheel.

Broken color wheel glued out of the way of the light tube.
Backside of a 3w led star base

The color wheel has an encoder attached that it cant be removed since the projector uses the encoder for color syncing.

UV leds, pwm and UV curing oven

3W Leds are bright little suckers, dont work around without eye protection. I brushed a 5v across a pair just to make sure they'd light and my eyes are still irritated an hour later.

I built a constant current LED driver source ala this instructable, I bread boarded circuit 1 just to test the led's and will connect it all together so I can do pwm of the power and vary the intensity.

I need 1 maybe 2 of these 3w LED's for the printer, and the rest will be turned into a UV curing oven for the finished prints. Memory suggests stippled aluminum sheet does a great job as a reflector shield it scatters light and has high reflectivity. A quick browse around online and I think I might go with a combination of thin wall material and cheap turkey roasting pans for a oven.

single motor printing and bootstrapping

I need to get z figured out first. Once I have a good z motion I can start trying to print small parts. If i can get either of my projectors to a size of 10cm x 7.5 cm I can get a .1 mm resolution and print my own connectors for xy motion and etc. if I have a long z motion already I can print larger parts so I am aiming for a specific bed size but will take whatever I end up finding at the second hand store to salvage. the z motion should be independent of the tank I use.

Ramblings on build chamber thoughts

Thoughts on the resin chamber;
light proof or uv proof min
simple interior I don't want resin able to harden anywhere critical
Must be resistant to salt water and plastic resins, I suspect this leaves stainless steel, glass (crock pot?), aluminum will oxide passivate so it will eventually stabilize in the salt water.
a method of managing resin level... like maybe the way a teapot is built with a narrow tube from the bottom, or since ardiuno can be a laser scanner build a rangefinder in to check the arbitrary level of resin.

a build platform with holes so resin can can flow through/around as the platform indexes maybe a chicken wire mesh

I think knowing exactly where the surface is isn't that important as long as its focused for the first pass. if I build a raft or support structure of some arbitrary height and start with an over cure then every pass after the over cure will be of a known z motion and the actual height could be calculated from cure volume shrinkage - z strut support structure displacement.

Haven't decided how to place the z screw for the platform but it will be outside the tank. some stability considerations... I should think about that more

Stuff I have ordered/assembled

I will add pics of it as it arrives, i mentioned earlier I bought too much stuff... lets say building a second printer shouldn't be hard...

1 Converted ATX power supply 5v at 35a, 12v at 23a max, I have spares to do this to as well

Power Supply I use for electro-etching

5 nema 17 steppers 78 ozin

10 stepsticks with heatsinks

1 aurdino mega2560 v3 + ramps + 5 matched stepsticks (that's 6 extra so far)

1 raspberry pi with too many goodies and a camera(octoprint path or just a hdmi interface)

5 3watt 400 nm leds

The lightpath and dlp from a 44nhm84 toshiba dlp television

 An acto dlp Projector the other option for a "printhead"

2 gt2 5mm gears

2m of gt2 belts

4 nema 17 steppers 58 ozin

1 9bulb 380 nm led flashight

Parts and pieces I gathered from a 20$ tv

Back of the Dlp with a Dvi-d(duallink) cable, power cable, lamp detect and color wheel cables

Optics

The light path just fire a uv LED in the little hole

The power supply for the tv

3.5 years later I'm considering this again with today's technologies

my current plans are;
I'm trying to mix the concepts of a sla device with plotter technology

Build a dlp printer with a mobile print head that moves on 4 axis xyz+ focus
The print head is to be a modified dlp projector or tv projection system with a base detail of minimum .1mm

Build area of 20.5cm*23cm*23cm
if I can focus in I can get better resolution, absolute minimum spotsize for 400nm and crappy optics with a numerical aperture of .54 is 770nm .77micron .00077mm or damn small... I remember reading achieving minimum spot size is very difficult, but dropping a zero makes it achievable. .0077mm or a little under 1000'th of a mm at this level of detail the print head is 1mm across I dont think this is achievable with my optics setup.At 100th of a mm resolution the image is 1cm across, this is my goal, but I will consider this project a sucess if I can get a full sheet print area. at .1mm resolution thats 10 pixels per mm or a 2x3 grid of 1024+768 pixels like in the tv I am parting out.

Like a standard SLA machine I will be lowering a platform in resin. the resin I am planning on using is makerjuice which is supposedly VOC free and will float on salt water. I think this is brilliant since the proposed build volume is about 3 gallons or ~400 but I can float a liter on salt water and print for 25$

To maintain a constant surface I think I will try with an overflow pipe attached to the bottom of the build chamber. with the Z platform in the top position I fill the chamber to some mark with super salty water and then add at least enough resin it oveflows from the pipe, then lower the platform to its start position. The salt water overflow should keep the surface of the build tank the same the entire time as the platform supports continue to be lowered in. I'm doing it this way to keep from having to build anything super complex inside of a tank that's under assault from salt water and resin could harden in.

I've ordered too many parts I'm not sure what all I will need I have a Pi and camera on the way to run octoprint with... using the hdmi output of the pi I plan on driving the video of the projector.

It seems from reading in #reprap I might want to use an arduino, the support of users makes problem solving seem easyier the pi is just newer.

I am planning on a variable print resolution as defined by the z axis print head, change size. I see some real challenges to anything involving software more  complex than printing images serially like  an inkjet does.
        I would like (not planning on pursuing yet) for the printhead size to optimize printing. for example back out and print a large area, zoom in and detail, type of thing.