Busy couple of weeks

For those interested in the DLP projector project I will be posting additional information in a week or so. Between a daughter graduating from college, the robomagellan competition coming up this weekend and work I have had a busy couple of weeks.

Here is where I’m at since I last posted.
I am re-using optics that are left over from the light path optics to focus the light from the uv led into the light tunnel. It looks like I will only need to use two of the lenses to focus nearly all of the light into the light tunnel. For close focusing I am going to move the entire projection/focusing/zoom lens “module” forward to obtain close focus using a spacer. The projector is very modular and the lens module is mechanically independent from the dlp “module”, it is attached with only 4 M2 screws. I should have details and photos some time next week.

HL

Posted in Uncategorized | Leave a comment

I’m going to go over this ground once and then I’m going to leave it alone.

This sort of thing pisses me off.

Yesterday I read this:
16-year-old-girl-arrested-and-charged-felony-science-project-mistake

TWO FELONIES AND EXPULSION FROM SCHOOL.

Really

When I was in high school for one of my science fair projects I strung 4 12 volt automobile storage batteries together in parallel and ran all that power through a couple of carbon electrodes and shot a thin stream of air through the resulting arc (They call those plasma cutters these days) it wasn’t a perfect experiment (as a very good physics teacher pointed out) but it was a damn good show. It was a project that I now realize was quite dangerous but there were a lot of science fair experiments back then that I’m sure would be seen as hazardous these days, but you know what, people who did those sort of things built the future you’re living in now. The curious sort of people that do things like build robots and 3d printers (and potato canons) did it and whether you think so or not they did a pretty good job.

However, I’ve seen this movie before. Some times it’s over the top as in the case of this teenage girl, some times it’s more (or less) subtle:

In my senior year of high school I took standardized tests like everyone (act/sat) and had very high scores, very high. When I sat down with my “official” school counselor I was told that with scores like that I would do just fine working on the line at Allied Tube (basically factory manual labor). I did not take his advice.

These attempts to shut down the future for some kids happens more than you realize and it needs to stop. I’ve seen very bright kids lose that light because of this sort of thing and it needs to stop. I don’t expect the people reading this post to all of a sudden rush out and change the world but I have seen what people of good will doing what they can when they can are able to do over time.

My daughter is graduating from college next week with a PharmD (Doctor of Pharmacy), her best friend from grade school is a marine biologist, her mother is an environmental engineer, my young niece wants to be a doctor. All are/were intelligent and curious teenage girls just like the young lady that was arrested. I wonder how they would have fared going to school in Polk county Florida.

Am I optimistic or pessimistic in view of the appalling arrest of this teenager you decide:
When I was 4 years old my mother took me to visit my great, great grandmother who was 104 years old, she was born a slave in this country prior to the end of the Civil War. My daughter is graduating as a Doctor of Pharmacy next week. That’s what I’ve seen in one lifetime. Am I optimistic, you bet I am but I also know that I have to do what I can when I can to make sure things continue to improve. I also know that killing our kids initiative or throwing them in jail at the drop of a hat is not a road to the future I would choose.

Charged with two felonies and expelled from school.

Un effin believable.

Me And Conestoga

Me And My Robomagellan Robot Conestoga

“All those who almost burned down their parents house when they were kids messing around with stuff they shouldn’t have been messing around with raise your hand. Ok put your hand down. Now all those who almost burned down their parents house twice raise your hand…”

Here is a link to one of a number of petitions asking that charges be dropped against Kiera Wilmot.

Henry Locke

Posted in Un effin believable | Leave a comment

Aligning Conestoga’s digital compass to the base (Conestoga)

The first robomagellan competition is in a few weeks so I’m working on conestoga quite a bit these days.

One of the issues I needed to resolve for conestoga was aligning the robot base and the digital compass on the stalk so that they both “point” in the same direction. Doing this has some difficulties, the robot drive and compass are generally not close to each other, you may have stray magnetic fields affecting the compass etc. I think I have come up with a way to align the compass and base that is very accurate, easy to use and very “old school”.

Using a “sun compass” that I can place level and flat against one of the wheels I use the suns azimuth angle for the day and time (from susdesign.com/sunangle) to point the robot exactly north then I rotate the stalk until the magnetic compass reads 0 degrees, at that point I lock the stalk so that it does not rotate. The robot base and magnetic compass at this point are perfectly aligned.

sun compass for aligning magnetic compass with robot base

sun compass for aligning magnetic compass with robot base

HL

Posted in Conestoga and Robomagellan | Leave a comment

A few quick notes on the casio projector uv mod (3d printer)

some notes for those who may be wanting to do this mod:

When disassembling the projector you have two types of flat cables, push in and connectors that have a latch. The latched connector have what looks like a black bar across the connector, that’s the latch, to remove the cable use your fingernail and lift the latch up, the cable will be released. MARK THE CONNECTORS, although the cables in general cannot be plugged in wrong mark them anyway with a permanent marker.

the screws that hold the top of the housing on are on the bottom and edge of the projector. The black plastic vent that goes around the edge of the projector also helps to hold the top on.

I cut the first row of large fins on the “red led” heat sink off and flattened it up with a sheet of coarse sandpaper on a flat surface. I also removed the several rows of the right side large fins as seen from the rear of the heat sink large fins down (see the photos in earlier posts). The heat sink would not fit in the position facing the light tunnel without those fins being removed.

The peltier cooler is only stuck to the heat sink by heat sink compound you can just slide it off.

Tape the top of the heat-sink so that it does not short any of the circuits above it when in position (I used electrical tape temporarily. I will probably purchase some kapton tape, use a square of mylar or something similar. For testing, tape or insulate the power terminals on the led to prevent shorts (I will probably use silicone after testing).

I epoxied the peltier cooler to the heat sink and the uv led to the peltier (take care that the bottom of the peltier cooler does not extend below the bottom of the heat-sink). If you are going to use epoxy (use the good stuff) Make sure you clean the parts very well. Acetone, tar remover alcohol etc would probably work fine. The center of the uv led should be 15.5mm up from the bottom of the heatsink and 14.6 mm from the left edge of the heat-sink viewed looking at the heat-sink face large fins down. Use heat-sink compound on both the peltier and the uv led, any decent heat-sink compound will probably work fine. Note: I used only enough epoxy on the uv led to hold it in place for testing, once I have finished testing I will add more epoxy or a fabricated faceplate.

In the projector I purchased only the blue laser block is missing the red led assembly is not missing, also the projector is NEW.

Bear with me here: Once when I was driving cross country many years ago I came to a sign when I was crossing from one state (region or province to my non US readers) to another, it said FAIR WARNING followed by a long list of road hazards you might encounter driving through the state (including rock slides!). I have always thought of that sign when dealing with possible hazards. FAIR WARNING: the apparent brightness of the uv led is quite a bit less than it’s real brightness, use uv blocking safety glasses they are cheap on ebay. SAFETY FIRST.

If you are thinking of doing this mod (no guarantees) contact me at the link in the image below and I will send you the sources of the parts I am using (projector, uv led, power supply, UV BLOCKING SAFETY GLASSES, etc).

email1

HL

Posted in 3d printing | Leave a comment

First Light, Casio projector uv mod. (3d printer)

First test with uv led in position.

Even without having optimized the light path optics a lot of uv is being projected. This photo was taken in a well lit room, when I took the photo my camera immediately “stopped down” to adjust for the amount of light coming from the projected image. The projected image is all 405nm uv.

I will have details of the projector modifications in upcoming posts.

first test of dlp printer uv modification

first test of dlp printer uv modification

Some photos of the heatsink mod and asssembly positioning:

bottom of modified heatsink

bottom of modified heatsink

modified heatsink detail

modified heatsink detail

closeup of uv led mounted on heatsink

closeup of uv led mounted on heatsink

uv led and heatsink test position

uv led and heatsink test position

Posted in 3d printing | Leave a comment

casio dlp projector mods

I think that the casio xj-a140 (purchased cheap without the blue leds and heatsink) may be one of the better dlp projectors for dlp based 3d printers IF you plan to modify it for use with a high output uv led or led’s. If you would like to use it unmodified move on there’s nothing to see here. There are two modifications that need to be made to this projector one is to the pc board so that the dlp chip responds to vga or hdmi inputs regardless of whether the original led/laser light sources are present the other needed modification is adding the uv led light source.

PC board mods:
The first modification (thanks to the folks over at www.photonlexicon.com) basically makes the projector ignore the fact that none of it’s own light sources are connected. This allows you to install your own led light source (this is a led projector and I do not think light sources other than led’s would work very well, I don’t think the projector was made to handle high temperatures). The modification is pretty simple: find connector CN905 and jumper traces 4 and 9 together (there are test point type pads behind the connector that you can use). That’s it. The up side is that that’s all you have to do, the down side is those pads you have to jump are small, really, really small. If you are comfortable soldering very small stuff have at it, if not maybe it’s time to start hanging out with people with really good soldering stations and steady hands.

Notes:
You only need to jumper CN905 traces 4 and 9 together. You do NOT need to cut the CN905 cable in any way.
I used a short length of solid wire from a piece of cat 5 cable to jump the traces.
If you take the access door (bottom of unit) to the led assemblies off it will release a limit switch that will make the modification seem to not work, either leave the door on or cut the limit switch wires and short them together (the switch is normally open).

pc board modifications

pc board modifications

possible uv led mods:
The second modification is to the light path. The light path of this projector is a study in modern engineering. Consider this, you need red, green and blue light bouncing off of the dlp chip to give you a full color image yet the projector only has red and blue led’s/laser’s. Green is produced by the “color wheel” which instead of filtering white light fluoresces green when a part of it is illuminated with blue light and sends the green light back into the projector optics to the dlp chip. It uses a complicated set of dichroic beam splitters, mirrors and lenses to accomplish this. The light path in this projector is complex to say the least (look up a diagram if you get a chance). However, I don’t need all of the sophisticated optics all I need is a light path directly into the “light tunnel” for one color: uv.

I initially considered mounting the 20w uv led on the existing “red” led heat-sink and mounting that heat-sink with a custom fabricated face-plate/lens-holder in the same position that it was originally in and using optics and mirrors to have the uv light make the 90 degree turn into the light tunnel. After having removed the entire lens/dlp assembly I think that doing it that way would be harder than need be and not as efficient as I would like. What I think I will do at this point is still re-use the the red led’s heatsink/peltier and fabricate a face plate but mount the assembly so that it points directly into the projectors light tunnel.

I may not even need to have a lens on the face-plate just a sort of short tunnel (to prevent light loss) and a way of sliding the assembly towards and away from the light tunnel for focusing. The light tunnel itself has a lens in front of it that looks like it will do the job of focusing the uv light source down to the size the tunnel needs depending on the distance of the uv light source (the tunnel is a mirrored box roughly 1/4″ in square and 2″ inches long). I will have to raise the pc board that covers that area by using standoffs so that I can fit the heat-sink assembly and fan where I want it I will also need to make a mount and face-plate for the uv led/heat-sink. This setup would also gives me flexibility if I have to go to plan B (a 100w uv led which is quite a bit larger than the 20w).

dlp assembly with optics removed (bottom of assembly)

dlp assembly with optics removed (bottom of assembly)

closeup of dlp assembly showing light tunnel and projection lens

closeup of dlp assembly showing light tunnel and projection lens

closeup of the light tunnel

closeup of the light tunnel

HL

Posted in 3d printing | Leave a comment

DLP projector red led assembly and uv led (3d printer)

I received the 20 watt uv led today and thought I would post some images of the red led assembly from the casio dlp projector and the uv led (only the blue led block was missing from the projector). I added a “quarter” (about 24mm diameter) for scale.

It looks like I will be able to use the cooler from the red led assembly (and yes that is a peltier cooler attached to the heat sink) and I will be able to mount the assembly back in it’s normal position. I will probably fabricate a new faceplate (the black assembly with the lens) and replace one or both of the lenses (both plano convex “led” lenses) only because of the larger area of the uv led array. The only question now is if 20 watts of 405nm uv is going to be enough. The resins I have looked at seem to like that wavelength and unlike DLP projectors that use bulbs the uv led is all uv. The light path of the casio is pretty efficient (after bypassing the stuff you don’t need) and I doubt if there are any uv filters in the light path (what would be the point in an all led projector),however, if 20 watts of uv is not enough I do have a plan B. Note: I also purchased a matched led driver for the uv led and a cheap pair of “red” uv blocking glasses (a uv led will be brighter than it’s apparent visual brightness and I believe in safety first).

Casio xj-a140 projector red led assembly.

Casio xj-a140 projector red led assembly.

Red led assembly and uv led.

Red led assembly and uv led.

Red led assembly in place with mirror modification. Shows light path.

Red led assembly in place with mirror modification. Shows light path.

HL

Posted in 3d printing | Leave a comment

DLP projector for 3d printer

The DLP projector I am using for the 3d printer I’m building is a casio xj-a140 LED based projector. I am using this projector for a number of reasons:

When purchased without the blue led array they can be purchased new “No laser Diodes and Heatsink” for less than 60.00.

With a minor modification (jumpering two circuit board traces) the projectors dlp chip works with hdmi or vga video inputs to the projector without needing any of the original led light sources connected allowing you to use your own light source. I’m using a 20w 405nm uv led to start with.

Even though the casio projectors light path seems complex the modification needed to the light path is almost laughably trivial (moving one mirror gives you a light path straight to the “light tunnel” and thus to the dlp chip bypassing multiple beam splitters and the color wheel).

I may even be able to re-use the “red” led’s housing, lens and heatsink for use with the uv led, if so this is a slam dunk of an easy projector mod, more on this when I get the uv led.

I will have further details of the projector modifications in an upcoming post.

HL

Posted in 3d printing | Leave a comment

dlp 3d printer main assembly photos

Here are a couple of photos of the 3d printer so far. It’s a “bottom up” design with a tilting vat.

dlp-3d-printer-main-assemblies-1s

dlp-3d-printer-main-assem-closeup-1s

Here is a photo of a test of the dlp projector.

dlp-projector-test-1

Posted in 3d printing, CNC | Leave a comment

Design to CNC, the directors cut.

Design: Draftsight
GCODE Genration: Heekscnc
CNC controller: LinuxCNC

Draftsight drawing (saved as an r2010ascii dxf file) which is loaded into Heekscnc (where machining operations are set up and gcode is generated), the generated gcode is loaded into LinuxCNC for machining on the CNC Router.

It could be simpler but not less expensive.

HL

Posted in CNC | Leave a comment

Building A DLP 3D Printer

This was just too cool a project to pass up.

A DLP 3D printer is a simpler build than my cnc router was by a long shot. It also seems simpler, at least to me, than the extrusion type 3d printers.

I actually already had most of the parts and materials I needed with the notable exception of a dlp video projector and a motorized linear slide both of which I “scored” at a very good price. I have already modified the dlp projector to accept a high power uv led (more on this in a later post) and am in the process of designing and cutting parts for the few assemblies that are needed (a build plate plate to be mounted on the motorized linear slide, a tilting glass bottomed vat for photo-polymer resin and misc. widgets and gizmos). I figure I will finish it around the end of summer as conestoga takes precedence for the next several months.

HL

Posted in 3d printing, CNC | Leave a comment

What’s up with the stalk (Conestoga)

At the first robomagellan I attended I noticed that many of the entrants had their digital compasses mounted on a stalk. This made a lot of sense to me because you want to as much as possible isolate the compass from large chunks of metal and magnetic fields. However along the way I realized that there was a downside to having a tilt compensated digital compass on a stalk. The problem is that the tilt compensation is done using “mems” accelerometers. If the compass is being whipped around while the robot is moving the accuracy of the compass can be affected. So along the way I designed a stalk for conestoga that is damped so that it will move but does not whip around. I did this by mounting the base of the stalk in a thick rubber sleeve and adding a about a pound or so of cement (yes cement) into pockets milled into the assembly at the top of the stalk where the compass and gps are located.

stalk-cement-1

HL

Posted in Conestoga and Robomagellan | Leave a comment

Serial, let’s keep it simple (Conestoga)

I initially controlled conestoga’s sabertooth motor driver through the arduino:

arduino to pc: here’s a velocity

pc: ok let me do the PID thing.

pc to arduino: here’s a throttle setting from the PID, don’t screw it up.

arduino to pc: ACK (maybe, depending on how I feel)

arduino to sabertooth: here’s the throttle setting from the pc, don’t screw it up.

sabertooth to arduino: (ok I will set the throttle but it’s beneath my dignity to reply)

I have since simplified this making it less problem prone by talking directly to the sabertooth from the pc (through a rs232 to ttl converter)

arduino to pc: here’s a velocity

pc: ok let me do the PID thing.

pc to sabertooth: here’s a throttle setting from the PID, don’t screw it up.

sabertooth to pc: (ok I will set the throttle but it’s still beneath my dignity to reply)

HL

Posted in Conestoga and Robomagellan | Leave a comment

PID Tuning (Conestoga)

I’ve been doing some PID tuning on conestoga this week. The only thing I can say is start with P at a low value. My wheel encoders give me both a count (90 pulses per wheel revolution which is about a yard) and a velocity (between 0 and 1000 for the speeds conestoga runs at). I have hard coded conestoga’s speed limit to a velocity of about 500 which is a little over 2 mph. Right now I have P at .03, I at .02 and D at 0 with the sabertooth drive “ramping” in the intermediate range (if you were crazy enough to go from full forward to full reverse the sabertooth would do it in about .6 seconds with my ramp setting). Acceleration is reasonably fast to the set speed and when it gets there it does not “hunt”. If P seems low remember that the output of the PID must match the units that the sabertooth uses which are 0 – 127 (I have hard coded a limit of 60 for both forward and reverse). I expect to be tweaking PID settings quite a bit from here on out as I have 4 PID routines which have to interact: right wheel, left wheel, navigation heading and cone tracker.

I mentioned starting PID tuning with P low (which I didn’t do). The first test run spun the wheels up fast (on the test stand, it wasn’t going anywhere), so fast in fact that a weakness in drive gearing showed up. The pinion on one of the motors was not meshing with the gear it was driving correctly (either the motor mount was loose or the gear was sitting too high off the driven gear). I ended up pulling the wheel and gear, adjusting the motor mount and tightening it down (I actually ended up drilling and tapping the motor mount for a larger bolt. I will probably do the same thing on the other wheel when I have a chance).

HL

Posted in Conestoga and Robomagellan | Leave a comment

New Blog For Projects, Interests, Etc.

The blog I had over at lockeconsulting.com (lockeconsulting.com/lockeblog) did not seem to be a good fit for talking about my personal interests and projects since it is a business site so I am relocating my personal interests blog here.  I have a number of interests including robotics, cnc, 3d printing, computer stuff,  photography and even some automotive all of which I will be (somewhat randomly) talking about here, I hope you find it interesting. I will also be moving all of the conestoga (robomagellan) info to this site.

 

Posted in Uncategorized | Leave a comment

Original Robomagellan Project Pages From Lockeblog

Click here to see the original project pages from lockeblog.

Conestoga Robomagellan Robot

Posted in Conestoga and Robomagellan | Leave a comment