Evolution of a Blog

This blog has evolved as I have as a maker. It starts at the beginning of my journey where I began to re-tread my tires in the useful lore of micro electronics and the open-source software that can drive them. While building solutions around micro-electronics are still an occasional topic my more recent focus has been on the 3D Printing side of making.

Friday, March 2, 2018

My Powered Chair and 3D Printing

I don't think that there is a mention in this blog that identifies me as being disabled.   I am fighting MS and have progressed to needing a wheelchair to move around easily as muscle control and fatigue combine to make mobility a challenge.  Luckily I could afford one of the best powered chairs out there, the Quickie Jive Up, to meet my needs.  There is no doubt about it when I say that this is the Mercedes of powered chairs.  I say Mercedes as the manufacturer is Sunrise Medical and they are in Germany.  I say Mercedes and not BMW (my previous cars) as Sunrise Medical is near their headquarters and not in Bavaria.  If I could drive a Mercedes I would but since I can't at least I am lucky enough to be able to afford the Jive Up.

The chair is based on the Quickie Jive M which has a mid-wheel drive and an adjustable seat all controlled by a joystick and or some buttons.  The mid-wheel allows the chair to turn in its own length and the suspension that the chair sits on makes for a decent ride.  The seat is adjustable from sitting to laying flat.  To this the JIve Up adds the ability to stand with some useful positions in between.

Yes, it is a monster!  It looks even bigger than it actually is due to it being in a position call "transition" which helps me get out of it and into my office chair (or other chairs and seats).  It can continue to raise the seat until I am standing.  It is those extra servos and controls that make it so bloody expensive.  It is capable of 6mph and can run on streets legally.  A battery and motor upgrade allows for 8mph but at the expense of torque so I went with this one.

It has greatly improved my life already in that I can be much more useful to myself and Sara compared to being in a traditional wheelchair or on crutches.  I can actually carry coffee in one had while I drive with the other.  Believe me when I say that small things matter.  One of the positions the chair offers is between sitting and standing and is one that I use most frequently in the kitchen.  With it I can much more easily get things done that would otherwise be a struggle.

This does not even begin to touch on the ability that the chair will give me to get out and about this spring...combined with a new car to carry it.

This will not come as a surprise to anyone with a 3D Printer, or anyone that has read this blog for that matter, but within minutes of having it I was already thinking of how to modify and enhance it.  The first thing that I designed were two brackets that can hold my grabber for when I inevitably drop something.  This is a frequent occurrence as I have lost some coordination in my hands.

Soon after that it became clear that the arm holding the controller could be improved.  The one that came with the chair was very adjustable but came at the expense of a little added width.  So I designed one that was the right length and orientation for me, that included places for the two buttons that I had purchased for the chair, and was no wider than the chair.

Having done the controller arm I then did a mount for the other arm that I intend to be interchangeable.   You can see in the photo with the controller.  Right now the one inserted into the mount is for my iPhone.  As you can see in the picture there is a charging cable that leads back to the chairs battery compartment where I have installed a 24v-5v voltage converter that provides the 5v power to four USB ports.

The last two additions had nothing to do with printed plastic but add convenience.  First, a saddle bag that fits perfectly under the right armrest.  Second, a general purpose shoulder bag that fits perfectly across the back of the chair (not shown). Third, and finally, a retractable seat belt replacing the non-retractable version that came with the chair (also not shown).

Monday, February 19, 2018

Micro Studio - Photographing Tiny 3D Prints

The level of detail printed by my SLA technology is such that a macro lens is essential if you actually want to see the finest of those details.  I have been using a macro lens attached to my iPhone for this with pretty good results as described by a previous post.  I may have even gone a little further than I needed to set up a mini studio!

What the iPhone solution delivers in terms of convenience though is lost in quality.  Lack of depth of field makes focus across any distance problematic and the various forms of optical distortion caused by tiny lenses is not a help either.  Enter my SLR.  I have Canon SLR and a macro lens and macro flash that I had purchased a while back.  I have never really taken the time to set it up for easy capture of the small models that I print until now.

Here is the my setup:
Woeking Bits

  • 3D Printed stand for the camera and macro lens
  • The actual camera and macro lense
  • Rotating platform for the object to be photographed
  • Elevator to raise or lower the floor on which the subject will sit.
  • Backdrop holding paper that could be white or black
  • PC (or Mac in this case) running Canon's EOS Utility
    Close-up of the working bits
    EOS Utility Console
    Captured Image
The parts that I designed for the Micro Studio are a rough work in progress but they can be found here.  Rotating stage is based on a huge bearing that I had from another project.  It is really not needed for this one so I have one that can be printed but I have not tested it.  The concept is here though!

Friday, February 9, 2018

Leveling the Moai

One of the two complaints that I have stated in regards to the Moai concerns the leveling process.  Standing a big and heavy printer on its head does not sound like a good idea! I play with a couple of different vats and I have found that they are different enough to demand re leveling (unless you only print at 100um).  This would become a pain...beyond pain even!  As I worked with the printer, however, I was able to come to terms with a decent process.

The first part of the solution was this article being published on the Moai Wiki.  That led to a couple of refinements.

If you are starting fresh go ahead and turn the printer on its head.  Remove the four leveling bolts completely.  Print four of these knobs from Thingiverse.  Note that you need to do some test prints to ensure the knobs fit the bolts securely.  As in you should need to pound them on.  Once each bolt has a knob put the bolts back on the printer and do the leveling with the paper sheet and all.  Turn the printer back over!

The second part of the solution is to print the test file from this article (here is a direct link).  Measure each of the cylinders and record the measurements in your copy of this Google Sheet.  Each knurl of the knobs that you now have on your printer represents 1/8th of a turn.  With practice you can turn these knobs pretty precisely just by feel.  You can also do this through the opening where the vat sits and without removing a side panel from the printer!

As always, remember to take the build plate off the printer before removing the vat!

The final part of the solution is to come up with a way to roughly level things without turning the printer over.  Here is my technique.  Set the Z Reset Position to something like 985 or so.  Maybe even lower (which means the plate will be further from the PDMS).  Let it move into position.  Note whether or not the platform moved when the build plate settled.  What you are looking for is the first point at which you detect movement.  The Z Reset Position that I use is the one that is one lower than the first movement one.

IMPORTANT NOTE:  When the current version of firmware starts the Moai it does so with a raise of the build platform and then a tilting of the vat.  Why this order is inexplicable to me!  It will harm your PDMS and should be the other way around.  Lacking that you can protect your vat by starting a print and then turning the printer off after the initial tilt starts and gets PDMS separation.  When your turn your printer back on it will be gentle to the PDMS.

Sunday, January 7, 2018

Printer #5 (Though Only Temporary) - Peopoly Moai

Not the printer but rather the number of them!  Have added a Peopoly Moai to my collection of printers.   That means that I now have three different technologies represented.  Fused Deposition Modeling (FDM) in the case of the Ultimaker 2+ and 3.  Direct UV Printing (DUP) in the case of the two (going to one) Wanhao D7s.  Now with "traditional" Stereo Lithography Apparatus (SLA) for the Moai.

The Moai was attractive to me for a couple of reasons.  First and foremost because while I have enjoyed the Wanhao D7s they are not as dependable as I would like.  Second, because of the positive reviews from its small but enthusiastic user community.  Third, for the price point.  Fourth and finally, because the laser spot size of 67um.

Following is a brief review of my experiences with this printer in three sections:  the kit experience, design and component quality, and finally operation.

The Kit Experience

I was not sure what to expect from this kit but was very pleased by the experience.  The parts were all clearly labeled and the instructions were relatively easy to follow though I strongly recommend augmenting them with this video guide.  A five hour assembly is promised but I did not expect that to be the case (for me anyway).  It probably ended up taking me ten hours which is more to my slowness than anything to do with the kit.  I have subsequently read a post where someone actually beat the five hour estimate on a first build!

I think the printer looks good against a background of Ultimakers!  I don't have much else to say about the Kit Experience.  It was a pretty straight forward process and proof was in the pudding after doing a test print almost straight out of the "box".

Design and Component Quality

It is not fair to compare my Wanhao D7 experience with the Moai given the price difference but I will anyway with that caveat.  As you would hope the Moai is a world apart from the D7.  I would put the quality of the printer closer to that of my Ultimakers than I would to any of the three "clone" printers that I have owned.  One major difference to the D7 is that people seem to be able to reliably share profiles for resins as there is not the wild variability between lasers as we see between UV LED arrays.  This suggests a tighter attention to quality control for the Moai components in general.

On the hardware side, the frame is solid and further stiffened by the exterior plexiglass panels. Everything seems to be well designed and well engineered to fit together to provide a sturdy and rigid structure.  The enclosure does, however, seem to be a fair bit larger than it needs to be for the size of the components stuck inside of it!  This all adds weight which goes to my first concern about the Moai.

Leveling the printer is slightly complicated by the process that starts with "turn the printer over"!  This is a printer, btw, that weighs nearly 20 kilos.  I am not sure how the leveling process should be done but it feels, to me at least, that this is a distinct area of improvement for the Moai.


Operation of the printer is very straightforward.  You insert an SD card, click on print, select a gcode file generated by Cura, and you are off to the races.  Beautiful print on the way.

This is, of course, unless you are switching resins and want to change the laser setting.  You have to do this via the firmware menu?  Wtf?  Why does this not come to the printer as temperature on gcode is beyond my paygrade to understand at this time though I have asked on the forum.   This is not at all a show stopped it just adds a cumbersome step to your workflow that does not need to be there. 

The bottom line is that the firmware that drives the Maoi is not as advanced as the hardware.  Don't get me wrong, it works and seems to work well, it is primitive!  The above is the biggest example but there are other things that could be improved:

My first wish, as stated above, would be for the laser power to be adjustable via the gcode print stream.  Extruder temperature?  And while that is being done do the same for the other settings that are adjustable and have an impact on print quality

My second wish would be to have an estimated print time displayed

Third wish would be for the knob to be reversed so that turning it clockwise results in a change up...as I believe is standard with most things?

Fourth is to add a menu for printer action controls.  Home, Print Position, Peel (Up and Down), Peel (Leave Down), Jog  Platform Down, Jog Platform Up.

Finally, given the above, a whole new menu structure would be nice!  Maybe three main items, "Print", "Setup", and "Control"?

Support and Community

I will close with praise for the support that Peopoly provides and for the active, and helpful, community around the printer.  

Saturday, November 25, 2017

Photographing Calibration Cubes

If it is worth doing it is worth doing over the top!  So it is when it comes to me taking photographs of calibration object (3DSlaTest).  The goal is to get consistent images so one cube can be compared to another easily.  My set setup is shown below.  The components of this setup are as follows:
  1. iPhone6
  2. Mount for iPhone6 from Thingiverse
  3. Lighting - Strip LEDs in a 3D Printed Frame
  4. Macro Lens for iPhone
  5. 3D Printed Cube Stand 
  6. 3D Printed Square Mount for Calibration Cube
  7. 3D Printed Backdrop Holder from Thingiverse
  8. Craft Paper for Backdrop

After Cropping

Wednesday, November 22, 2017

What $1000 SLA/DUP/DLP Printer Would I Recommend?

This was the year that you could start to buy (as opposed to build) a 3D Printer based on curing liquid resin (rather than melting plastic) for $500.  Pretty amazing when you consider where the costs of printers in this technology sector have been priced.  Of course at that price point you are going to have to make some sacrifices but at that price it is worth it!

The printers we are talking about come in three flavours.

The first, and the one that has been around the longest, uses a laser to cure the resin.  They are commonly called SLA printers which stands for (S)tereo(l)ithography (A)pparatus.  In reality, however, all of the 3D Printers in this category fit this title given its definition includes the following: optical fabrication, photo-solidification, or resin printing!  We will, however, just call the branch of the family that uses lasers to be SLA printers.  Think Formlabs Form 1 and Form 2.  The x/y resolution on a laser based printer will be 100um ish to 200um ish.

The second flavour is called Direct Light Processing (DLP) and uses a projector.  Think Moonray or B9Creator.  Some of these types of printers are capable of x/y resolutions of 30um - though with a smallish build size.

The third is called Direct UV Printing (DUP) and it uses UV light passing through an LCD display.  This is what the Wanhao D7 uses.  These types of printers are typically getting resolution of 50um on the x/y axis.

Generally these printers project onto the bottom of a vat of resin with a build plate that is dipping onto and off the build surface.  The z-axis will be capable of resolutions from 10um to 100um.  The whole workflow process for resin based printing is very different that for FDM and is described here.

At this moment there seem to be four major choices for a cheap printer in the resin printing space:
  1. You can build one from a kit or even from scratch as the parts are pretty common and there are both kits and good DIY instructions available.
  2. You can get a Wanhao D7...or any one of the numerous clones that look just like it.  I am not sure who was actually first here.  Wanhao is a cloner but are they also a clonee?  Here is an earlier article about the D7 ... note that a lot has improved since then.
  3. You could buy into one of the already released, or very soon to be released, crowd funded printers.  Two top this list, the Moai SLA 3D Printer ($1250) and the Phrozen Make DUDP 3D Printer ($980).
  4. You could wait for one of the many other crowd funded projects to come to fruition.  There are at least a dozen on Kickstarter with many of them having characteristics shared by the Wanhao D7 and the Phrozen Make.
Sooooo...what would I recommend?  I honestly do not know!  The market will mature a lot in the next 12 months so waiting could be your best option,  If you absolutely can not wait, and must be under $1000, then I would get a Wanhao D7.  Quality has been improving, there is a great support community, and they are ahead of the rest of the pack in terms of release date and number of printers in the system.

If you can go a little above $1000 then the Moai at $1250 is probably worth considering.  I don't have any hands on experience but it has a very enthusiastic following.  One slight catch...it is a kit!

A printer to watch in the short term is the Phrozen Make.  It enhances the basic D7 design with what looks like a better z-axis, and what will certainly be a better UV illumination for more even curing of prints.

Sunday, November 12, 2017

Methods for Creating a Mask for a DUP/DLP 3D Printer (running NanoDLP)

As evidenced by recent posts on this blog I have been busy creating masks for my Wanhao D7 printers.  I actually have masks that work but what I don't have is an easy process that can be done by anyone without a lot of extra hardware or specific knowledge.

Some of the techniques that can be used to create a mask are as follows:

Mask Creation Wizard of NanoDLP
Tools needed:  Light meter
Process: Project a grid of cells for measurement and adjustment.  Find the lowest light intensity cell using the light meter.  It will be one of the corners.  Note the light meter reading.  Adjust the grey value in all the other cells to match that of the lowest cell just noted.  Press the button to generate a new mask image.
Advantages:  Minimal investment in tools and/or technology (can even use a multimeter connected to a battery and voltage divider photo resistor as discussed here).
Disadvantage: Cumbersome and extremely tedious process.
Comments: Remember that the default screen orientation in NanoDLP is not the orientation of the Wanhao D7!

Measure, Calculate, Generate Mask
Tools needed:  Light meter and software to generate a mask. Note that I have developed two light meters that I need to release to the public domain. Both based on Arduino Nano's.
Process: Measure the light intensity using either a physical grid or the grid of cells projected by NanoDLP.  Use that set of measurements to model the mask with output being either an actual image file or the inputs needed by the NanoDLP Mask Generation Wizard.  
Advantages:     Moderate need for additional tools and/or technology.  Relatively easy process.  Flexibility to use the data to generate masks for different uses of the build plate.
Disadvantage: Thus far there has been limited success in creating a really good mask using this technique.
Comments: Garage Science has a piece of software that does build a mask but it does not seem to work for the D7 resolution.  When I fed it data from a D7, but using a lower resolution in the same aspect ratio, the mask was no more effective than the Excel model that I have developed.  A number of people on the Wanhao D7 FB Group has said that a model will not work but I do not understand why.  I am assuming it is just a matter of my not having the right maths skills!

Measure, Automatically Adjust, Generate Mask
Tools needed:  Light meter integrated with  software running on the printer attached Raspberry Pi..
Process: Measure the light intensity using either a physical grid or the grid of cells projected by NanoDLP.  First find the dimmest corner.  Then project a grid of cells, one at a time, so the meter can be placed on each cell.  With the meter measuring the cell adjust its intensity using a mask value until it is equal to that of the lowest cell.  Output the matrix of mask values for input to NanoDLP (or generate a mask).
Advantages:  Should generate an excellent mask (though I have not had time to do so yet).
Easy to run once set up.
Disadvantage: Fairly complex requirement for integrating a meter and software solution.  

Photo Interpretation
Tools needed:  Camera (e.g Smartphone) and Image Processing Software (e.g. Gimp, Photoshop, etc)
Process: Take a photo of the light pattern projected by the UV LEDs.  Manipulate it into a mask using something like Gimp or Photoshop.
Advantages:  Can be done without any addition of hardware or software on the printer
Disadvantage: Requires a very fiddly setup for the camera to be positioned above the build plate precisely.  Fairly complex process of translating the image to a mask.
Comments: I think this approach is the most promising to meet my requirement of something that most people could do.