Solid Body Electric Ukulele - the UkuLiLi

[Chad Schmidt]

I just need to get over the fear of using one of my routers.

And I need to get over my fear of using hand tools! Seriously, getting sharp chisels to do the volute carving was a real challenge for me. I can't imaging whittling out a neck pocket.

That is a pretty cool little uke!

Thanks Gordon! It's been really interesting and informative to follow your through neck build thread and see where we had similarities as well as differences in our processes. I hope my end result looks as good as yours...

[Chad Schmidt]

To finish off the pickup, I potted the wound assembly in a wax. The wax was heated in a water bath in a glue pot. I let the pickup soak for about 15 minutes; I couldn't see any more bubbles coming out after that period of time. The wax is a mixture of about 20% beeswax and 80% paraffin wax.

Here's a photo of the finished pickup. All it needs now is a cover- more on that to come. I like to use locking connectors on pickups and other interconnects that may need to be removed occasionally. Since I really wasn't sure if this pickup was going to work, I wanted to be able to swap it in/out with minimal effort. To magnetize the poles I passed the pickup between a couple rare earth magnets stuck to vise jaws. I made five passes. The magnetization is oriented South up.

[Chad Schmidt]

The next item up is the onboard headphone amplifier. I wanted the onboard headphone amplifier so the UkuLiLi could be played quietly but still retain its electric character. My requirements for the amplifier were:

1. Must use the same jack plug as the main passive external output.
2. Volume control through the same potentiometer as the passive output.
3. Must be able to be easily switched between active and passive mode.
4. Must have a visual indicator of when the circuit switch is on; I don't want to be replacing batteries too regularly.
5. Must be able to power both Left/Right speakers in the headphones.

All of these requirements except #5 are really independent of the amplifier circuit. Requirement #5 can be accomplished with either a true stereo circuit or with a mono circuit with output split to the L/R channels.

After some web searching, I found a schematic from the senior project of Michael Jenkins, "Analog Electric Guitar Distortion Effects and Headphone Amplifier", 2010. The Jenkins circuit is a mono amplifier so I'll need to split the output.

Here's the schematic from the paper. The only change I made from the Jenkins design is to add an LED to indicate when the circuit has power. To draw up the circuit I used EAGLE Light Edition. The Light Edition of EAGLE is freely available with the only constraint being a limit on the size board allowed. There are also nice demonstration projects/tutorials available to get you up to speed.

[Dan Smith]

Cool Chad,
I made an LM386 amp like that and used it as a tube pre-amp back in the old days.
Nothing like an old Fender amp overdriven and turned up full blast.
No wonder I'm half deaf.
Can't wait to see the finished instrument!
Dan

[Chad Schmidt]

I made an LM386 amp like that and used it as a tube pre-amp back in the old days.

Nice! I'm hoping that the circuit will also provide some boost if switched on while plugged into an amp...

Before starting the PCB layout, I wanted to make sure the circuit works and sounds decent. Here's a photo of the circuit prototype on a breadboard. It worked and sounded pretty good to me.

To simplify connecting the guitar into the circuit I made a simple panel that houses input/output jacks, a switch to put the circuit inline or bypass and some commonly used potentiometers.

Although I was happy with the circuit, I still wanted to try a stereo amplifier to see if it offered any advantages. I found a circuit for the Velleman Headphone amplifier and gave it a try. Here's my schematic.

After listening to both circuits, I decided that I liked the sound of the Jenkins amplifier better. The Jenkins amplifier also takes up quite a bit less space so I proceeded with laying out a PCB for the Jenkins circuit.

[Chad Schmidt]

Again, I used Eagle Light to lay out the PCB. My initial plan was to cut a foam insert for the electronics cavity and let the circuit sit in a cutout in the foam. Because I chose this "mounting" scheme, I did not include any mounting features on the board. Here's the board layout. Its a two layer board with the red traces on the top and the blue traces on the bottom. The gray lines show the silkscreen and the green circles are the solder pads.

I had the board fabricated by OshPark. I've been really happy with their quality, prices, minimum quantities and turn around time. Here's the board as received from fabrication and after being populated with the components.

To give a sense of scale, the board dimensions are 1.25" x 1.35".

[Chad Schmidt]

So after giving it some more thought, I decided that the foam insert in the electronics cavity wasn't going to be a great way to hold the circuit and battery in place. I looked at some other mounting options and concluded that the circuit could be held on some standoffs that double as the nuts on the switch screws. The only drawback to this mounting scheme was I needed to have a new PCB made that had a shape to fit into the electronics cavity at the switch location and also included mounting holes with the same spacing as the switch mounting holes. Here's the layout from Eagle Light and the PCB as received from fabrication. I also took the opportunity to learn how to make curved surfaces and add graphics in Eagle Light.

Here's a photo of the populated PCB including the entire wiring harness. I used locking connectors for connections that may be need to be removed occasionally (pickup, battery, output jack). The wiring diagram is shown in the subsequent image.

[Chad Schmidt]

One more post on the circuit...

Two things I didn't mention in the previous post:
1. The battery ground is disconnected when the circuit is bypassed so the battery doesn't drain.
2. The signal at the jack tip is mono, regardless of whether the circuit is switched on or not. To get sound out of both R/L headphone channels I am planning to use a 1/4" to 1/8" converter with the necessary signal splitting.

Also, since I mentioned it, here's a photo showing the circuit mounted to some standoffs underneath the switch, followed by a photo of all the piece parts of the circuit.

Lastly, here's a photo of the battery mount. I used a small 90 degree drill adapter to drill pilot holes for the screws. The battery fits nicely under the pickup.

[Chad Schmidt]

My plan for the UkuLiLi is to use a bone saddle acoustic bridge with a trapeze style tailpiece. Since this project is also an excuse for me to learn more about CNC programming, EDM and milling I made my best attempt at fabricating all the necessary tooling to make the parts I need. Let's start with the tailpiece...

The tailpiece consists of three parts: the string retainer/anchor, the wire and jack/end plate. All parts are made of brass. To make the wire, I machined a plate with holes for 1/8" pins at the bend locations and a groove to indicate the final contour.

I used C-clamps to hold the wire in place during bending.

After bending, the wire was trimmed to length with a cut-off wheel.

I was pretty pleased with the way the wire turned out. It is flat and looks pretty symmetric to my eye.

[Chad Schmidt]

Next up is the string retainer/anchor. Here's the drawing for the part.

And here's a (poor quality) photo of the finished part. No in process photos; the part was made entirely on the mill.

To complete the trapeze portion of the tailpiece the wire needs to be attached to the anchor. I made this bond by brazing the two parts together in my oven (I did this when my wife wasn't home).

The braze material/solder I used had a melting temperature of 430 F. My oven goes up to 550 F, at least on the display, so there was no trouble heating the parts above the melting temperature of the solder. I assembled the parts, squirted a little of the included flux in the holes with the wire, put the assembly on a baking sheet and heated to 500 F. When the oven reached temperature I let the parts soak for about 30 minutes. Next I pulled the baking sheet out of the oven and touched the solder to the wire on each side. Thanks to the flux, the solder melted and flowed down into the hole. I tried to limit the amount of solder so I didn't have a lot of overflow. Once the solder was in place I put the parts back in the oven and let them soak for another 30 minutes before removing to cool.

Here's a photo of the final part after the braze, seems plenty strong! The stains from the flux cleaned up easily with some 0000 steel wool. The part looks a little skewed in the photo but that's an illusion due to camera angle.

[Chad Schmidt]

And finally on to the jack/end plate. The plate needs to wrap over the end of the UkuLili and provide a means of holding the wire on the trapeze. Here's a drawing of the side profile. The open loop allows me to slip the trapeze in and out as necessary.

To make the part I started with a piece of 0.5" x 3" brass. I cut the side profile on the wire EDM so when finished I had this profile with a width of 3". Next I moved to the mill to drill the holes and cut the outer profile. Here's a photo of the finished part; the second small hole is for the LED.

[Chad Schmidt]

In addition to the tailpiece, I thought it would be pretty cool if I could make a pickup cover too. I tried several approaches to making the cover. In the end, the only method I could get to work was to mill the entire thing out of a single chunk of brass. Here's all the hardware ready for chrome plating. I had a hard time finding a shop that would take such a small job. If anyone has any recommendations on a chrome plating shop, I'd appreciate the input.

[Dan Smith]

Chad, you have some really cool custom parts!
I'm a PCB designer by trade, your circuit board looks great.
We have a 3-D printer at work, but I have yet to think of anything guitar related to print.
I put my little booster in a small aluminum box. The jack was screwed on one side, the plug on the other so the box plugged directly into the guitar jack.
You might get a kick out of stuart keith guitars. Stuart makes all of his hardware.
Dan

[Jason Rodgers]

The top flatwork is for the most part the same as the bottom flatwork but with an overall length of 1.791 inches and width of 0.492 inches. I made some MDF templates for both the top and bottom flatwork and used a small parts sled on the router table to trim the flatwork to shape. The top flatwork is 0.062 inches thick and the bottom is 0.094 inches thick.

Bobbin Flatwork Routing.jpg

I'm having a hard time seeing how your small parts router sled works. There's the toggle clamp to hold the part, but bottom of the sled is continuous: how does the router bit contact the template?

[Jason Rodgers]

Since this was my first home made pickup, and I don't anticipate making a lot of my own pickups, I opted to make a pickup winder rather than purchase one of the commercially available winders. The winder was cobbled together with parts from Mouser, Amazon and McMaster-Carr. It's a pretty standard design; the internals are shown in the following photo.

Pickup Winder Internal.jpg

The winder is powered by a 12VDC signal. The speed can be controlled with a potentiometer connected to the PWM controller. The polarity of the DC can reversed so the bobbin can be spun clockwise or counterclockwise. The LED display/counter uses a Reed switch to detect number of rotations. The motor spins at a maximum 500 RPM.

Next is a photo of the bobbin mount and Reed switch magnets. The screw at the top is for attaching to the motor shaft. The two holes at the bottom of the mount provide clearance for the bobbin brass eyelets. Double stick tape is used to attach the bobbin to the mount.

Counter Magnet.jpg

I like bright, shiny lights so I added a lot of them. The wire guide is removable for more compact storage.

Winder Front.jpg

Lastly, here's a photo of the overall assembly with a bobbin mounted and the wire guide in place. To control wire tension I used a piece of felt sandwiched in a C-clamp with the wire running through the felt. With this configuration it took about 20 minutes to wind the pickup; I need to switch out the motor for something faster now that I know what to expect.

Pickup Winder.jpg

Final specifications on the pickup: enamel coated wire 43 AWG, 7500 turns, resistance of 5.22 k-ohms, wound CCW when viewed from the top.

Overall, this is an excellent build! Great work!

I'm particularly impressed with your pickup winder. Over on the Music Electronics Forum, folks (myself, included) are constantly agonizing about winder construction and features. By the looks of this machine, as well as your circuit board design, you have a solid background in electronics. Very cool stuff going on in this little package!

[Chad Schmidt]

We have a 3-D printer at work, but I have yet to think of anything guitar related to print.

Yeah, I have some friends with access to 3-D printers too and the only thing I can think of to print are knobs and tuner buttons. Even with these parts though, I question the durability.

I put my little booster in a small aluminum box.

Ha! You mean something like this...

I thought this would be a great use for the three square'ish boards I made. I have a nephew that's learning to play guitar so I thought this would be a great gift for his parents! Thanks for the feedback on the PCBs by the way. I'm mostly winging it when it comes to the electronics; I just tried to lay out the traces so they look nice...

[Chad Schmidt]

I'm having a hard time seeing how your small parts router sled works.

Here's another photo of the small parts router sled, mocked up to show how it can be used.

In real use, the bobbin flatwork material (not shown) is attached to the bottom of the template with double-sided tape. The sled has sandpaper stuck to the bottom of the block on the toggle clamp as well as on the sled surface. The two handles can be loosened so that the side blocks can be pushed in to contact the part as well. Since the bobbin is so narrow I wasn't able to use the additional support of these side blocks. Routing is done with a pattern bit. I chose a bit with the bearing on top, but this would also work with a bit that has the bearing on the bottom. Half of the part is routed in the first pass, the template is rotated 180 degrees and the other half of the part can be routed.

you have a solid background in electronics

I wish that were the case... this project would probably have been done years ago!

Thanks for the feedback on the winder. I'm pretty sure I consulted one of your threads extensively when putting it together. If I decide to build pickups more regularly I'd definitely consider a faster motor and spending more time to get the bobbin carrier properly aligned. It's also a bit of trial-and-error to get the limits adjusted.

[Chad Schmidt]

The last thing I needed to do before starting to apply the finish was to make the bridge. The bridge is made from a scrap of some sort of Rosewood and routed with a 1/8" groove for a bone saddle. I cut the groove at an angle on the router table using a 1/8" bit and a small metal wedge with appropriate angle. Here's a photo of the finished groove along with the wedge.

I rounded off the edges of the bridge and saddle using a sanding drum on the drill press.

Here's a photo of the test fit of all hardware. By some small miracle, all of the electronics worked without any debugging! On the headstock you can see the graphics that I'll have printed on waterslide decal paper.

[Chad Schmidt]

On to finishing! Since this is the first guitar that I'll be finishing (I've sent the previous instruments I made away to a pro), I elected to use spray cans of nitrocellulose lacquer from ReRanch. The color I chose is Copper Tone Metallic. Before applying any lacquer, I painted the control cavity with conductive paint.

In hindsight, painting the control cavity with the conductive paint prior to applying the lacquer was probably a mistake. I did a little touch up sanding and proceeded to grind the shielding paint particles into the wood. I'm hoping it won't be an issue with the final result since it will be buried under many layers of lacquer.

Here's a photo of how I'm holding the guitar for finishing. To spray, I made an internal frame for a chest freezer box I had on hand.

[Chad Schmidt]

There's been a long lapse on postings on this project... I finally got it finished and have some photos I'm pretty happy with so it is time to wrap up this thread!

I did not take many in-process pictures during the finishing. Here are a couple of shots during the primer coat phase.

This was the first guitar I finished by myself and it was a learning process (understatement)! Some of the troubles I had were:

1. Fingerboard Binding: I masked the fingerboard binding at the start of the process and didn't remove the masking until I was ready to start clear coating. When I removed the masking the interface between the binding and the neck chipped extensively. To repair, I sanded away the existing finish until I had a smooth transition again and started over. The second time around, I removed and replaced the masking over the binding every three coats.

2. Headstock/Nut Interface: This is pretty similar to the problem I had with the fingerboard binding. I masked the nut slot at the start of the process and experienced some chipping in the finish when I removed the masking tape at the end of the process. I tried to sand/file away the built up lacquer in that area but still ended up with a small amount of chipping on the edge of the headstock where it meets the nut. It's pretty minimal so I did not try to repair it. In the future, I think the prevention is again to replace the masking tape throughout the process.

3. Headstock Sand Through: After adding the waterslide decals and working through the clear coat build up to bury them, I had a tiny sand through to the primer on the corner of the headstock. To repair, I masked off a small area and resprayed with color again (same exact spray can). However, when I resprayed this corner, the color from the can was slightly darker than the rest of the headstock. I'm not sure why the color changed. At this point the original color had been fully clear coated and allowed to cure for three weeks. One of my theories is that the color was different since I was spraying over the clear coat base. Another theory is that the finish color changes slightly as it hardens. And a third theory was that the color in the cans actually changed for some reason between uses. I did a few tests and think I can rule out the first two theories. If anyone has any insight into why this happened, I'd certainly appreciate thoughts on this one! I also decided not to repair this flaw...

4. Finish Cracking under Tuner Bushings: I'm guessing I pressed too hard on the tuner bushings when I installed them. Three of the four bushings caused a short crack in the clear lacquer when installed. The cracks all ran parallel to the string direction (seems too coincidental). I did a lot of work to clear the lacquer out of the bushing countersinks as well as from the edge of the holes. I'd appreciate any thoughts on this one too as to why this happened. Again I decided not to repair.