oneTesla kit review and build comments.

Posted on July 23, 2013 by Jay

Big Picture:
I was one of the first 100 backers of the oneTelsa Kickstarter (120 volt) and received one of their earlier kits (manual revision 1.3.0). Overall I was very happy with the kit. The overall build quality of the resulting Tesla coil (and especially the case) is much nicer than I could have achieved on my own without a kit, and the laser cut acrylic parts fit together very closely.

Because this is the first product the oneTesla team has shipped, and they are just starting out, some mistakes are expected. My kit did have one major problem: They accidentally shipped a 220 volt circuit board (for Europe or other countries that don’t use the 120 volt standard) and omitted the interrupter board entirely. Everything else in my bag of parts was set up for 120 volts. After a quick email to the oneTesla team they shipped me the correct circuit boards and I was able to solder them up. There were a few other minor hiccups with the kit (a missing screw, an extra fourth wire in the gate drive transformer pre-twisted wire set, weird reversed green LED’s, and a missing 2 pin jumper header that I replaced from my stock) but nothing else that was a real show stopper. The 1.3.0 manual was missing a few minor assembly steps, but nothing you couldn’t figure out from looking at the pictures and the parts lists, and most of those have been fixed in the 1.3.3 manual online.

Issues/suggestions for somebody else building a kit to think about:

Reversed Green LED’s
At least in the kit I received, the green LED’s were “backwards” in that the flat edge of the LED should be mounted OPPOSITE the flat edge in the circuit silkscreen. I believe they bought budget surplus gray market LED’s and got stuck with a batch with this weird manufacturing defect that a large company had rejected. This is documented in the errata online, and I didn’t make the mistake on the main board, but for some reason I didn’t remember to double check what color the LED was for the interrupter board. [Everything works with the LED reversed except you don’t get the power indication light ….but I decided it was worth fixing so that I didn’t accidentally run my battery down because I forgot to turn the interrupter off.]

Alternate Glue
They recommend using hot glue to affix all of the acrylic pieces. This would work, is somewhat reversible, and many people will already have a hot glue gun or can easily buy one locally. However, for the best visual appearance, I would recommend using acrylic welding solvent instead. Of course, acrylic welding solvent is scary stuff, so if you use it, read the warnings and wear nitrile gloves!

I used IPS Weld On 3 (McMaster Carr part number 7528A13) for both cases and the primary core base/clips. It will probably cost you as much as a hot glue gun to buy, but a one pint bottle will weld a LOT of acrylic. First, make sure you have the case assembled correctly, with all of the holes/slots lining up with the circuit board. Remember to try reversing the bottom piece if things are not lining up. Pay special attention to the slot where the wires from the primary core go into the screw terminals on the back. The cases are very nicely designed, but I wouldn’t have minded a few matching numbers etched in the corners to make it very easy to determine the exact orientation of each piece relative to one another.

After you are sure you have the cases assembled correctly (I held them together with rubber bands while testing), make sure that you can get your circuit boards into them from the top (they are a very tight fit, my case walls have to bend just slightly when inserting the main circuit board at an angle with a rolling motion).

To glue, lay the case down on it’s side with the base hanging off the edge of the table. (The base keeps the side pieces square.) Then use a q-tip to place a small drop of weld-on on the top of each tab that sticks up through the horizontal acrylic piece. You will see the water thin solvent wicking between the tabs and coating the bases of the joints between the tabs. I did one side (two sets of tabs) at a time with a ten minute wait between sides. Finally I turned the case upside down and affixed the base. Don’t glue on the tops!

For the primary core clips, I used clothespins to hold all six clips on. Make sure they are all straight and their spacing is evenly distributed. I used a coffee stir stick to place a small drop of acrylic welder on the top of the clip where it rests against the outside of the round acrylic primary form. After those set up (20 minutes) I took one clothespin off at a time and placed a drop of solvent on the top of the clip where it met the inside of the primary form, putting the clothespin back on to hold it in place for another 20 minutes. Finally, I turned the form upside down and placed a drop between the “U” bottom of the clip and the top of the form while rotating it slightly in my hand to use gravity to get the solvent to wick between the top of the form and the bottom of the clip’s “U”.

I did use hot glue to affix the top and bottom circles for the secondary core as that was an acrylic to ABS joint, which IPS Weld-On #3 isn’t good for, and it was a hidden joint anyways. I glued the bolt with magnet wire soldered to the terminal ring onto the acrylic circle while holding the bolt head in a pair of pliers, and used the wing nut to pull the bolt in tight. After the bolt was tight, I put extra hot glue all around the bolt head. Be sure to test your resistance between the bolts is less than 300 ohms (I got 228) before gluing the ends onto the secondary PVC pipe!

Battery in the interrupter
I was a little worried about the lack of clearance between the metallic case of the 9V battery and the bottom of the leads and solder connections on my interrupter board, so I wrapped the battery in black electrical tape to make sure nothing shorted out.

Terminal substitutions
I substituted a ring terminal for the fork terminal they provided in the kit for the ground wire that affixes to the bottom of the secondary bolt. I feel the ring terminal is less likely to accidentally come off. Their assembly pictures shows a ring terminal on their demonstration model while the manual notes that they provide a fork terminal.

Instead of soldering a wire to the ground terminal on the top of the IEC power inlet, I bent the terminal upright and then used a crimp-on female spade terminal on the end of the ground wire to connect the wire to the IEC ground clip. I don’t plan on removing this connection, but this gives me flexibility for alternate grounding solutions if I need it in the future. For example, I can remove the ground wire entirely and replace it with a ring terminal going to an exterior ground. (It was also easier than taking the circuit board out of my case to solder the wire on, or trying to navigate my soldering iron inside the case.)

oneTesla build

Posted on July 21, 2013 by Jay

Who has a 99% complete tesla coil? Yes, that’s right, due to my extensive blogroll of the technorati (i.e. MIT students) I found out about the oneTesla kickstarter before it got super popular and got in on the ground floor (first 100 backers). Over the next month they got more and more interest, and I kept watching the features of the kit I had pledged/ordered go up. [Better interrupter, stamped toroid, etc…] Of course, like most kickstarters, it took them about three times longer than they had anticipated to actually ship the rewards, but I got my kit and (after a few customer service emails) put it all together!

The only thing missing is the stamped toroid for the top, so I can’t quite spit out long sparks to the tune of the imperial march, but since I was expecting to have to make my own toroid out of an inner-tube and aluminum tape when I made the pledge, I can’t complain about a few delays in such a custom component.

So until the toroid’s ship, you (and I) will just have to watch other people’s oneTeslas making music:

dyndns.org (dyn) ddclient configuration fix

Posted on May 28, 2013 by Jay

I have been using ddclient to update my dynamic dns records for various servers and laptops for many years now. Around a month ago they all stopped updating the DNS server. I eventually tracked the error down to a line in the configuration file:
server=members.dyndns.org

It’s not like I had made up a random server name…this server had been working successfully for several years. And it is the server recommended by the ddclient automatic configuration script generator on the dyndns.org website even now. [https://account.dyn.com/tools/clientconfig.html] However, at least for me on Ubuntu 10.04 and ddclient 3.8.0 it had stopped working.

I changed it over to:
server=members.dyndns.com

And this fixed the problem.

Suction Cup Cell Phone Mount

Posted on May 6, 2013 by Jay

This is a windshield mount for a (landscape mounted) cell phone that I designed and built. It has a ball & socket mount for easy adjustment.

The rectangular holder on the end can be customized to any cell phone.

And the whole thing takes about eight hours to print.

If you want to print your own, I posted the design files on thingiverse.

Top speed 275 mm / sec

Posted on April 20, 2013 by Jay

Although my print head can move at 300 mm/sec, my extruder can not reliably keep the plastic flowing at that speed. (Perhaps if I turned up my extruder temperature above 195 C…)

I have decided that 275 mm / sec is a reliable top speed for my extruder after printing a relatively large part at that speed with the temperature turned up to 200 C.

This video shows layers being printed in about 15 seconds with 3 exterior perimeters and 25% infill.

Here is another video of the twisted koch snowflake vase (scaled up to 150%) being printed at 275 mm/sec top speed. Due to the fractal nature of the sides of the vase the platform rarely got up to the top speed, as it never had a long enough path to accelerate up to full speed.

Printing at 300mm/sec

Posted on April 19, 2013 by Jay

I decided that I needed to switch to a different (larger) 3D object so that my printer could accelerate up to full speed on some long straightaways. Here is what 300 mm/sec printing looks like on a larger square object:

However, my extruder just couldn’t keep the plastic flowing (at least, not at 185 C), and it jammed. So I have decided to try 275 mm/sec with the temp set to 200 C (lading to an actual extruder temp that is closer to 195 C).

Stepping up the speed on my Rostock-Mini

Posted on April 18, 2013 by Jay

Now that my Rostock-Mini is basically finished, I have been adjusting the parameters of Slic3r to increase the print speed. Why? Because this is my sports car 3D printer….It’s small, looks cool, and is fast! In contrast, my Prussa Mendel is the family mini-van: Nothing to look at, reliable, with a large print volume.

Because the Rostock-Mini has the cold end of my extruder mounted on top of the frame (not on the motion platform) it doesn’t have to move the weight of the extruder stepper, gears and associated hardware. The filament is pushed down to the platform via a bowden tube (think bicycle brake cable) and the only part that needs to accelerate and decelerate is the hot-end and associated fan / air duct. The lighter the platform is, the faster it can move and change direction while maintaining positional accuracy.

The Twisted Kochflake vase that I’ve been using for my test print has 7 layers at the bottom with “infill” but above that, it’s just made up of four perimeters of plastic traced around the volume of the interior of the vase. This means that some layers require a relatively short amount of motion/time, especially near the lower part of the vase. I have my Slic3r software set up to not allow any layer to take less than 15 seconds to give the plastic a bit of time to solidify before we put the next layer on top of it, so in some parts of the videos below the platform is not moving at it’s true top speed because of this software limitation. Also, due to acceleration constraints, the platform can’t get up to full speed on small bumpy surfaces. When the printer is printing the bottom seven layers (you’ll see it going back and forth to fill in the circle with plastic) or the wider part of the fractal pattern as the vase grows up you’ll see where layers take longer than 15 seconds (4 times around the vase is a single layer) and the platform will be moving at top speed.

Here is my printer set to 225 mm/sec, which is faster than most printers that have a moving single extruder will be able to do:

Here is the twisted Koch Vase at 150 mm/sec, which is approaching the top speed of most gantry style homebrew 3D printers that move the cold end of the extruder.

This is a relatively slow 75 mm/second video:

Rostock Mini Z-Axis accuracy

Posted on April 10, 2013 by Jay

When calibrating the bed of a standard 3D printer, you can slide a piece of paper under the extruder and adjust the bed until it’s touching the extruder (but still able to be pulled out) in several places to level the bed about right. However, with a delta bot, your X/Y/Z coordinate system must be converted mathematically into the coordinate system of the three carriages ridding the towers, and determining if your calibration parameters are correct is not as easy. If your calibration parameters are incorrect, your entire coordinate space may be warped!

I was able to eyeball things to get my calibration parameters set up “good enough” for standard use, but it still wasn’t perfect. I finally broke down and shelled out $15 for a cheap Chinese made machinist’s dial indicator so that I could get my coordinate space transforms square and flat down to a thousandths of an inch. (I changed one calibration parameter by 0.5 mm…so it wasn’t terribly far off from the “eyeball” approach, but I feel better about it now…)

Here is a video of the machinist’s dial indicator in action:

The nine small vertical “bumps” in the beginning of the video is from me pushing the 0.1mm down button on the control interface multiple times until I got the indicator close to the top of the dial. As you can see from the dial it takes nine 0.1mm bumps to travel around 3.5 hundredths of an inch. Google says that 0.9mm = 0.0354331 inches, so my units appear to line up right.

I also jumped the head up and down 10mm at a time to show that the head comes back to the same Z height.

When I scrape the probe back and forth in the Y axis the indicator jiggles around due to friction, but you can see that the measurements don’t move more than 0.01″ when the probe moved across the entire glass build plate (and it’s very close to 0.001″ accuracy when stopped at the end and middle points). Overall I’m very happy with the positional accuracy and calibration of the motion platform now. Although I only measured the Z axis with my dial indicator, because it’s a delta-bot the z-axis is a joint effort of all three towers, so I figure that my positioning accuracy in the Z coordinate axis is a good proxy for the X and Y coordinate axis as well.

Greg’s Wade Bowden Extruder for Rostock-Mini

Posted on April 4, 2013 by Jay

The original direct drive Airtripper V3 extruder that I had made for my Rostock-Mini was almost able to get the job done. However, my stepper just didn’t quite have enough torque to push the filament directly, and it would “skip” steps relatively continuously. I could still print large objects, but they would have a “foamy” appearance due to using less plastic than they really should. Also, my stepper motor and drivers were getting hot due to all the extra current flowing through them.

I finally decided it just wasn’t going to work well enough for production use, and printed an extruder (Gregs Accessible Wade extruder) that has a printed gear system for a large mechanical advantage. I adapted it to feed into my Bowden tube and mount onto the top of my Rostock-mini frame with two printed parts.

It made all the difference in the world. My geared extruder can now easily feed filament continuously through the hot end at a 300mm/min rate.

I also printed an adapter plate that holds the stepper motor and attached extruder in the appropriate location/angle. In the future I may integrate this with parts from Gregs Wade extruder design to build an integrated extruder.