DashFest JJ's life and project dumping grounds…


Assembly and Review of the Imakr.com “Startt” $100 3D printer

Its been a LONG time since I've posted on here! Seriously, like 4 years! Whats changed? A lot. I have a ton of projects that have not been submitted. But for now, here's a review of Imakr.com's "Startt" 3D printer. This little kit has an irresistible $100 price tag. I figured it was worth a shot.

I ordered it on March 6th, and didn't receive it until April 17th. They kept me somewhat updated, claiming customs clearances and QA checks.

Here's whats in the box:

The checklist and packing was nice, and there were no missing parts. Good work here.
The assembly instructions arrived on the included 8GB MicroSD card, stuck in a cheap USB card reader. Although there are a few typos here and there, I found the instructions to be quite good, and concise.

Lets lay everything out:

They claim no additional tools required. The 3D printed wrenches were cute, but only somewhat useful. I still ended up using a pair of pliers, and a 10mm wrench to complete assembly.

Peeling the paper off the acrylic was tedious to say the least. Also, a few parts arrive pre-assembled. They left the paper attached, so these assemblies need to be taken apart to remove the paper cleanly.

One of the first things I noticed are the three 3D printed parts. They are extremely rough. They were hastily printed, with tons of ringing, and corner lift. More on this later, but here is how they looked.

Assembly coming along:

This has a Bowden style extruder, and a generic E3D V5 style hotend. The extruder itself is quite well built, mostly aluminum. (Sorry, I took these pics after assembly was complete)

Uh oh, an issue. The Y and Z rods were not cut to accurate lengths. Several mm difference between them. This can cause frame twisting, so I used my bench grinder to get them evened up.

Still coming along nicely:

Houston, we have a problem. Remember those poorly printed parts? Well, they have so much corner lift, that the Z axis nuts do not mount straight. They sit at an angle. This causes severe binding, and is unusable as-is.

Luckily the files for these parts are available in the product forums. I re-printed these using my workhorse printer, "The Beast".

The new parts resolved the issue. I have yet to receive a proper response from the manufacturer about why these parts are so bad. Really there is no excuse to put parts of that quality into production. I'll update if and when they respond.

One little gripe I have is the way the belts are attached and held tight. Zipties? Come on. It would be basically no additional cost to put a little acrylic clamp on the belts. Whatever, they seem to work.

Oh yea, here was a pleasant surprise I forgot to mention earlier! Real leadscrew on the Z axis! I expected threaded rod. It would have been nice if they included a little packet of lithium grease for these however.

Another gripe. The LCD assembly came preassembled, and did not have appropriately sized nylon spacers. Just nuts and bolts. This causes the PCB to bend and press against the acrylic. Just a few spacers would solve this.

I forgot to take a picture of the controller before assembly. Its a TronXY controller w/ Repetier firmware. All the steppers were labelled TronXY also. Probably because this printer is just a rebranded Tronxy XY-100.

Assembly complete!

A little calibration on the Z axis, and loaded up with Hatchbox Black PLA. Lets print!

Found this elephant on the SD card. Not great, but not bad for a first print. They didn't show well in the pictures, but it has some strange vertical and diagonal artifacts on the sides, that lead me to believe there may be some microstepping errors going on. You can see the vertical bar sticking out on the inside of his ear. I'll have to investigate further.

Another issue I noticed is the included 12v 5A power supply seems to be inadequate. It gets super hot, and you can hear the voltage drop in the fan RPM when the hotend kicks on. A 7A or greater should be used.

The Bowden tube gets close to being kinked at higher Z prints. The extruder is kind of in a weird spot, and should be moved.

Interestingly, the included bed is too small for the ability of the printer. It can print about 20mm further to the left of the bed. Weird.

That's about it for now! Overall the kit was quite good for the money. If you're a tinkerer, and want to try 3D printing for the first time, this is a great deal. The 3D printed parts were the only major issue. I hope the manufacturer gets that issue sorted on later batches.

Oh, and it fits inside my main printer:


I haven’t posted here in a while… This exists!

Thanks to Half Cut Tea for featuring me in this short doc!

Filed under: Life No Comments

Tabletop Tesla Coil

I built this little guy today.

It works pretty good considering it was built in less than 6 hrs, and being poorly tuned.  I tossed it together with junk I had around. Nothing was calculated or measured. Arcs are about 5-6" so far

The secondary diameter is smaller than a quarter.  I just wrapped ~33awg magnet wire around a piece of PVC pipe until I felt satisfied with its height.  The solid copper primary sits upon some ceramic insulators salvaged from an old runway strobe.  The box is some 3M marketing shwag.  The transformer is a 10kv that I found in a box of junk from an auction long ago.  Not sure on the current, probably about 10-20ma.  The tank cap is a Sprague 4nf doorknob.  An acrylic box from Ikea, with 2 carriage bolts makes up the sparkgap.  I'm running it though a variac, and found the cap begins to fail around 80v input, so it needs a better cap. Topload for now is a tennis ball wrapped in aluminum foil. Not much else to it.


Here's a vid:

Filed under: Life 6 Comments

Garmin Nuvi GPS Pin Brute Force Cracking Robot

I spent the entire day today working on this beast:

A little while ago I picked up a nice Garmi Nuvi 780 at a freight recovery auction. Its protected by a pin code, that cannot be reset unless either the correct code is entered, or it is taken to its "safe location".
I have seen a couple of other people do brute force cracks on these, and decided to give it a shot.

Entirely built from junk I had laying around, starting with an HP scanner. I don't have the hardware laying around to drive the stock stepper motor, so i decided to use servos instead. On top of the scanner's sled is a gutted DVD drive. 2 servo's for X and Y, and a 3rd servo for pushing a pencil eraser stylus onto the screen. I didn't have servo horns, so I used clothes pins instead. Most everything is hot glued together. Its all run by an Arduino Nano.

The code counts from 0000 to 9999, waiting for the appropriate delay between each attempt. At this rate, it will take about 36-40hrs to go through each pin.

Here it is in action:

That's about it. I'll report back when I find the pin!

That was quick, it was 0248. Only took a couple of hours!

Info update:
The robot does not currently have any check procedure to see if the device unlocked and will continue to pound the screen even after it unlocks. I figured this may not be a problem, and let it run. I had plans to add a camera to monitor its progress if this does not work or re-locks the device.
Coincidentally, I went downstairs to check on its progress about 2 hrs in, and I witnessed the unlock with my own eyes!

Here's the code (not cleaned up): http://www.dashfest.com/garmincrack.txt

Filed under: Life 16 Comments

Mac Classic Web Server

Last night I tossed this together:

Mac Classic Web Server (Down indefinately, I was sick of it running)

Its a Mac Classic, with 4MB RAM, 40MB HDD, and a Dayna Pocket SCSI/Link T Adapter. (Yep, this ethernet adapter runs on SCSI)

Running MacOS 7.0.1, MacTCP  V2.0.4, MacHTTP 2.2


We'll see how long I leave it running.

Filed under: Life 2 Comments

Professor Kill A. Volt

So I built a Tesla coil for Halloween. Me and Ash came up with the idea a week before, so I didn't have much time.

I already had most of the components for a coil ready. Already had neon transformers, a handful of doorknob capacitors, totalling 17.5nf. Since this is all I had available with capacitors, I was forced to keep the power down to 12kv @ 30ma (360 Watts). I had an old primary and secondary coil I built years ago for a static gap coil. But the secondary was only about 3" diameter.

I decided to build a synchronous rotary spark gap to get the most out of the little power I was putting in. I used a universal brushed food processor motor. I ran across a neat trick discovered by Clive Penfold that allows you to convert a brushed motor to synchronous with just a diode soldered between 2 opposed brush commutators. Synchronous Rotary spark gaps are a great way to get a efficient, and solidly timed spark discharge in the primary circuit. It allows the capacitor to discharge @ 120 beats per second, and also discharges at the peaks of the AC wave. The phase of the discharge time can be changed by rotating the motor in relation to the tungsten electrodes on either end of the gap. It worked quite well for me, as shown in these vids/pic:

The rotor is just a 1/4" barbed nylon "T" fitting for hose. I stuck a couple of set screws in it to lock the fitting on the motor shaft, and lock the tungsten electrode in position. I dremeled slots in the white plastic mount to allow for phase adjustment. (The slots were made bigger after that pic) Using my Variac I found it ran nice @ about 40 volts. I found a transformer @ Raelco that made it more permanent.

I'm mounting everything in a poly rolling cart that I found at the DI for $10. The cap is just a handful of 40kv doorknob caps, all in parallel for a total of 17.5nf. I used aluminum strip to mount them together, and a couple of IKEA plastic shelf brackets to mount it to the cart. A simple safety gap was built using a couple of carriage bolts (not shown).

Now the Transformer, Cap, and Spark Gap is taken care of, I start fiddling with the primary and secondary coils.

I already had old stuff available, and gave it a shot:

The sphere at the top is a pair of IKEA Bowls soldered together. Actually worked pretty well. The streamers were about 12"-14"

I decided to go ahead and wind a new primary/secondary. Due to only running 360 Watts, I know the performance wont be much better in terms of streamer length. But at least I'll have a better set up for next year after building a better cap, and increasing the current up to 120ma.

The new secondary was wound on a 5.25" poly post cover from NPS. Wound 24" of 23awg magnet wire. Here's some pics of the winding process.

It came out pretty well, although I had to tweak it a few times to work out the loose windings. Over the next few days I put approx 6-8 coats of Polyurethane from a rattle can on it.

The primary took a few attempts. I tried using clear acrylic boxes from IKEA, and threading copper tubing into holes I drilled.. It didn't work out so well. So I started over, and decided that zip ties and drilled PVC on a temporary plywood base should work.

I used 3x 20 foot spirals of 3/8" copper tubing soldered together. This was probably the single most expensive part, but turned out well!

I think the winding spacing is a bit too wide, but it works. I soldered together 2 more IKEA Bowls, this time for a 14" diameter sphere. And gave it a test!

It was marginally better, as expected. Not enough current, remember? Probably about 18" streamers. I put some hardboard covers over the spark gap area to cut down on noise and UV light hitting people. Also wired a foot operated switch from Harbor Freight to make operation easy. The Rotary Gap stays on continuously, and the transformer is kicked on with my foot on demand.

But wait, we need more! I also tossed together a few 12kv transformers for a 120ma Jacobs Ladder!

Also gathered some props.

I built a pair of grounded huge aluminum "tongs" with IKEA strainers at the end to use as a Faraday Cage. I let the kids put the candy in the tongs themselves, sometimes yelping really loud to make em jump. :D
I then zapped the candy with the coil, before giving it back to them. My wife Ash put together a spider web covered workbench with bubbling concoctions, sheets in the background, and other spooky bits. We both dressed in lab coats, her being my clipboard wielding "Lab Assistant". I'm also wearing a pair of welding goggles, and 40kv lineman gloves (for looks).

It all worked out well, and everyone enjoyed it!

A friend recorded this one.

This will be a new tradition each year. Each year will be bigger. I should be running a nice MMC Cap, and 1440 watts input next Halloween!

Filed under: Life No Comments

Game-tendo (Portable NES in a Gameboy)

I built and almost finished this project LOOONG ago, but abandoned it due to a fried LCD. I found a new suitable screen for $2 at the DI and decided to finish it.

Here's the guts (minus wiring):

Pretty straight forward.
The "NOAC" (NES On A Chip) was salvaged from a "Power Player" pirated NES system.
The LCD is one pulled from an Intec PS2 portable screen+controller combo.
The controls are the original game boy button board, trimmed to fit, and re-wired to a genuine NES controller shift register.
Wire it all up, and you got a portable NES!
Its still not quite finished. Need a battery and voltage regulator, as the NOAC takes 5vdc and the LCD needs 9vdc. Needs a simple audio amp for sound.
Other than that its good to go!

Filed under: Life 2 Comments

Not so quick n dirty iPhone controlled garage door.

Alright, so I lied in the last post. It didn't work well at all. The software had a memory leak and crashed often. The software was incredibly slow. The AMD Geode PC overheats trying to run XP, and is also UNBEARABLY slow. The wireless network signal was weak, and it didn't have any kind of feedback to tell me the garage door was open or closed. It just sucked.

So I started over:
The same Velleman K8055 USB Interface board.
Used a better mini PC this time. A P4 based box with 768MB of ram. (At least it has SOME grunt)
A high power wireless bridge for network connectivity (good signal now).
An Axis IP webcam to keep an eye on things.
It has magnetic reed switches on the door connected to the K8055 digital inputs to detect when it is opened or closed.
All sitting pretty on a little shelf near the ceiling.

I wrote all new software from scratch. A web based PHP front end running on my webserver. The PHP script communicates directly with the VB program running on the PC, and gives realtime status and control of the door, webcam view, and the VB software even sends me SMS messages when the door is open, closed, or jammed!
Its pretty slick now! Commands are nearly instant, and I can see the status of the door from anywhere.
Need to add security before I can show off the front end to you guys...


Quick n dirty iPhone controlled garage door

Soo uh... There is a PC connected to my garage door. Is that wrong?

I saw this piece on hackaday today, and thought I'd give it a shot with a different method. I actually managed to complete it in 3 hours!

I had one of these sitting around:

Its a Presidian PIC-1MM Personal Internet Communicator I picked up at the DI for $5. Its basically a tiny low power AMD Geode powered PC running Windows CE. I hacked the BIOS on it a while back and installed Windows XP (requires a BIOS hack to run a custom OS). It currently has a Cisco AE1000 wireless dongle for network connectivity. All running fine from a 12v 1.5amp brick.

Next part:

This is a Velleman K8055 USB Interface board. It allows you to control numerous inputs and outputs via USB digitally, or analog. I built it a while back for an abandoned project, and had no use for it. The relay connected to the side is what activates the garage door opener "button". The capacitor/resistor combo just above it allows the relay to click momentarily, even if the output is constant. We don't want to hold the garage door button down. :)

Now we just toss it all on top of the door opener:

The software:
This is a K8055 "Bridge" that runs in windows, and allows any iOS device to control the K8055 from anywhere. There is a $3 app in the iTunes store to install on the phone. Its secured with a "key" to prevent other phones from connecting.

...and thats it! It works great! I will probably eventually write my own software, as the K8055 Bridge software is clunky.


Retro Mouse

So how do you get a 25 year old Atari STM1 Mouse working on a modern PC?

Rip out these guts.

Took a Microsoft Optical USB Mouse, and hacked and glued the optical window into the "ball hole".

Trimmed the old PCB, so we can utilize the original button switches. Trimmed the new PCB so it'll fit inside.

Soldered wiring from the old PCB's buttons to the new USB PCB.

Jam it all back together... AND...

The original Atari mouse is NOT serial. its a proprietary connection utilizing the DB9 connector. So a USB/serial adapter does not work in this case.