Loading
makefast

TuneFast Harp

by makefast Dec 12, 2017
Download All Files

Thing Apps Enabled

Please Login to Comment

I'm trying to source alternative bearings (Amazon does not ship this item globally).

The 3124 pulleys I'm finding have a notch that is 2.8mm wide and 1.4mm deep. Will this work or mess with the tuning?

The diameter of the bearing where the string touches is the only critical element to maintain tuning. The 3124 bearings have a v-groove diameter of 9.7mm, so getting as close to that as possible will help (for example, 10mm diameter should still sound okay albeit slightly trickier to tune).

You can also use flat sided bearings, but then you may accidentally push the string off the side without the groove and need to completely restring/tune it. Once the string is under tension, it stays put pretty well, but as you're stringing it or playing it for longer amounts of time, the V groove really helps.

After a ton of searching I finally found something I think will fit.
This: https://www.aliexpress.com/item/10pcs-V623ZZ-V-groove-roller-wheel-ball-bearings-3-12-4-mm-embroidery-machine-pulley-bearing/32769540034.html

It seems to have an inner diameter of 9.8mm. Weridly, other v623zz berings have an inner diameter of 9.2. But this one shows a different diagram.

Do you have any recommendations on substitutions for the 3124 mm v groove bearings? I am going out of town this weekend and was hoping to make this as a gift for someone to take with me. I ordered the other materials on amazon but the bearings were out of stock until yesterday so they won't be arriving in time. The hardware stores around me don't seem to carry those specific bearings.

Woo Hoo! I was your 1,000 like! I'm printing this right now but I have to wait till the bearings come are delivered on Tuesday to play it. I can't wait to try it! Looks like an awesome design!

Comments deleted.

GREAT project and easy to assemble! If the holes are reamed after printing, older kids can put together with little supervision. Nylon strings seem to be stocked only in sets, but select "low tension for acoustic guitars" label on the package. The 4 thinnest ones work well. Appreciate that the base print is barely thick enough for the washer and nut on the tuner piece. If that part of the print lifts, you will have assembly issues. I get much better print on rafts, plus the back of the base is no longer slick from printing on your plate. Cut rounded triangles from old credit card plastic for picking the strings and loudest sound, if you have problems printing one that works.

Editing to add, great design! Ordering up some hardware now. Quick thought on reducing the parts list... did you try a version where the spacer/standoff is part of the printed part itself? I know the small diameter standoff would be tricky to print, but since there are 8 of them the layers should have time to cool as it prints. If you made them conical it would help with the flex under tension too.

DUDE! Way cool! When I was a kid we had a harp just like this but it was made out of wood, had a resonance chamber and about 15 strings or more. What was cool about it was that you had paper "music sheets" that slid in under the strings that were kind of a "connect the dots" song. So all you had to do was start on one side, pluck the first note and follow the lines/dots to play a song. All you need for this is a basic template and you could start printing on paper all sorts of little short songs. Might be cool start a google docks type thing that people could start making/sharing songs to print/cutout on paper. Definitely going to be printing/building some of these for my kids!

That's super cool, and you can even still buy those little multi-string harps with music that slide underneath if you're feeling nostalgic. But making your own (especially with the this harp's bearing configuration) would be even cooler!

One thing to consider is that the friction from the bearings (especially lower-cost version) will accumulate as you pass the string over more zig-zags back and forth. We found a practical limit of 8-10 notes per string before things get difficult to tune, so if you do want to make a 15-note version, consider using two strings/tuners.

if one were to CNC this out of wood, do you think it would still be effective? or does the infill or PLA material lend itself to acoustics?

See the other comment below about CNCing the harp out of wood (or manually just cutting/drilling it); in short, go for it!

A PDF for a paper template to trace is included in the files for download.

Wouldn't you want to make a cavity under the strings for the sound to be amplified ?

Ok, I'm pretty inspired. Really thinking about making one, then winding a thin pickup to slip underneath it, run it into my DAW and turn it into some kinda drone/ambient thing.

That would be awesome. You could probably even use an off-the-shelf pickup (perhaps a "rail" style so that the strings don't need to be placed exactly over particular points), just make sure it's wide enough to sense all of the notes. Or if it's primarily for drone sounds, just place it under the lowest 4 or 5 notes and don't worry about the higher ones.

Just ordered a bunch of hardware on Amazon. This is great. I am making a bunch of these for my wife and her new music class. This is absolutely perfect. I also have some color changing PLA, that will be fun for a handheld instrument...

Thanks for your work on this. I will post the pics when Amazon sends my stuff...

Happy to hear it! Let us know if you have any questions once you get the parts and are making/assembly it. It should (hopefully) be a pretty quick print-and-play instrument.

Quick design question. When using the formula to calculate proper string lengths. Do you use these exact values to space the centers of the holes for the screws/Vbearings OR do you adjust the distance between the holes based on the Vbearings radius?
I am guessing that the string stops touching the vBearing basically at the center, so you would used the exact length calculated for the center of the holes, but I could be wrong and you've done all the prototyping.

You're on the right track. In the end, all of the string segments can be placed such that the end of each is measured from the center of the bearing (off to the side by the radius of the bearing at the center of the v-groove) as you mentioned... plus a minor adjustment to makeup for two non-ideal factors:

  1. The slight tilt of each bearing when the string is under tension (i.e. the bearing/bolt each tip in a bit toward the center of the harp as the string is tightened).
  2. The overall bow of the instrument as the plastic bends depending on how tight you make the string.

For the bearing/bolt tilt, all of the bearings are the same except for the one closest to the tuner (since the tuner pulls the string to the side) and closest to the top of the high note (the final bushing has almost no tilt since the string is just guided along at a low angle). The overall bow is fairly uniform, but you may need to make some minor tweaks depending on how tightly you intend to string the harp.

Hope that helps!

thanks for the details. If I understand, both these factors shorten the actual length so you add a constant to make the holes farther part.
And for strings numbered 1-8.
B=bow factor
T=tilt factor
string1 and string8=lenght+B+T/2
string2 thought string7=lenght+B+T

Yes, that's correct. We ended up applying a constant to the middle string segments (on the order of 0.5-1.0mm was enough) to account for bow/tilt, and treated string segments 1 and 8 with slightly less (as you point out, half of the tilt is a good ballpark). In practice the first/longest segment is pretty comparable to the others and can probably use the same factor as the middle strings.

However, the shortest string segment (#8) was noticeably out of tune without compensating for the bushing (which doesn't tilt much at all). In the end we settled on moving it an additional 0.5mm closer than the theoretical placement (and could probably have gone even a bit more, maybe 1.0mm, since the final note only ever feels slightly flat depending on string type/tension).

3 hours to print? That seems a bit much. What layer heights did you use?

We wanted to give a realistic/conservative estimate on the print time since there's so much variation between printers and settings.

For our prints, we used 0.38mm layer height (0.5mm nozzle) at medium/high speed (but probably could have pushed it harder if pressed for time). If you do print one out, let us know how fast you're able to do it (and what settings you used)!

While they're super cheap I decided to just print some groove wheels for mine to just get started. Here's the thing model in case anyone else wants some or has a hard time getting them for whatever reason:
https://www.thingiverse.com/thing:2750787

VGroove Pulley

Do you think there would be any way to make semitones? Im trying to make a 3 octave dovetailed set and I think that would be amazing. The math is already beyond me, or I'd work on experimenting with it.

A dovetailed set would be super cool. Go for it! Just keep in mind that you'll likely only want 7 notes per string (instead of 8) or you'll have a repeated octave note (which could actually sound nice, so maybe that's not an issue at all).

As for using semitones, we tried a version (see the really wide harp in the upper right of the picture under the "Experiments Along the Way" section in the original write-up) and the complexity grew pretty quickly as the tension from tuning each string had a slight effect on the neighboring strings.

Also note that since a guitar string is a fixed length, you're limited by the number of zig-zags back and forth before you run out of string. And even if you do have really long string, the little bits of friction in each bearing start to become hard to manage (i.e. it's difficult to tune) when you get over 8 notes. So for our semitone experiment we had two strings per octave (6 semitones) and 3 string total (for 18 semitones total).

Probably if you rise string a bit it would be more comfortable to play. If you make it holow (acoustic guitar like) it will sound better and louder.

Just added a new version (AHarp22_v04.stl) that includes a parabolic sound reflector under the strings. This more than doubles the gap below the string and also projects the sound more effectively. You can see what it looks like in the picture of the harp printed in blue PLA above.

Thanks for the suggestions!

Making the strings higher does make it easier to play, but it also causes the instrument to bow/bend significantly more. That could be offset with an even thicker base, but we settled on a middle ground for the final design. Alternatively, there could be some structural reinforcements (like a truss rod or bolt the goes across the back side) to allow for the increased strain on the printed part from raised strings. If you want to try it yourself, you can easily raise the strings by just using a longer bolt and adding a spacer between each bearing and the base.

Increasing the volume would be a great next step, too. We considered making it hollow and adding a sound hole early on but settled on prioritizing the structural rigidity for the instrument, especially since it's designed with novices in mind with a lot of wiggle room for different printer capabilities.

edit: We're putting together a version now that includes a parabolic reflector to project the sound a bit. It adds a shallow dip under the center of the strings which hopefully addresses both of your suggestions (more comfortable to play and louder).

Hi there, do you think this would work in wood or MDF? I was thinking of making a paper template printed at scale to help drill holes in the right position.

Since the final design is rather flat, you could definitely make a version out of wood! The hole placement is critical, and the 3D printed part has a couple small details (like a raised guide to support the shorted string segment and recessed pockets to capture the nuts on the back side) that may be somewhat tedious to do by hand.

Printing a paper template sounds like a fantastic idea to line up all the holes. We'll take a look at adding a template file so that others can do the same. Note that you could even scale it up or down and the ratio of distances for each string should remain the same. However, the current design is about a large as you can make it where a standard guitar string is long enough to work.

In fact, for education/learning it'd be cool to make both 3D printed and wood versions to compare the building techniques, difficulty, strength, overall sound, etc..

edit: A PDF for a paper template is now included in the files for download. Let us know how it works out!

Thanks!
Good to know it could be scaled up/down, that hadn't occurred to me.
I had some other ideas for a wooden version:

  1. To avoid having the nuts standing proud on the back surface, you could perhaps have countersunk holes on the rear, and have the nuts on the front.
  2. For a raised guide to support the short string segment, you could perhaps have the string sit in a slotted screw, or the head of a hex bolt in that position.

1st to comment, this is actually really lit tho

congrats on the feature bro, you deserve it

I would make this if I had the supplies for it.

We've been thinking through what it would take to reduce the parts list. The washers can definitely be printed, and probably the smaller bushing that supports the final note (shortest string segment) could be printed too. If plastic bearings had low enough friction, that'd be super cool, but too much friction means that all the string segments aren't necessarily under uniform tension which is important for it to stay in tune.

IS that what you are going to try and tackle next?

Just included the files for printing out the M3 washers and M3 6x4 spacer to reduce the parts list.

Also tried printing out a version of the V-groove bearing but the significantly higher friction negates all of the tuning advantages (i.e. it takes a long time to tune, and if you pull on any note it shifts the tuning of neighboring notes until you carefully go back and re-tune it all). So at least for now, you'll probably need the metal bearings.

That said, if you want to give printed bearings a try (perhaps with a little oil/grease to let things slide under load), there's a nice version by geartechbrandon over here: https://www.thingiverse.com/thing:2750787

VGroove Pulley

Aw, you did that just for me?