Bonetti Machine - 2.5 Joule sparks like gunshots.
Вставка
- Опубліковано 12 бер 2022
- The Bonetti machine is an influence type electrostatic generator, similar to the Wimshurst machine, but without the metallic sectors arranged around the spinning discs.
I made this one with a bit of an unusual arrangement; more like a windmill design using an old DC motor instead of a hand crank. Rather than using the traditional Leyden jars, it uses a more conventional (by today's standards) flat capacitor made of a big sheet of ABS plastic, and some aluminium foil.
To get the required counterrotation of the two plastic (acrylic) discs, I made a belt plus gear drive using some old printer parts, but made the actual gears (14 tooth cycloidal profile) using a CNC mill. Since the motor shaft wasn't very long, this was the most compact design that was reasonable to make.
Machine is still under development. There's quite a bit of corona leakage that limits the voltage build-up. Currently it maxes out at 120kV based on jumping a 120mm air gap.
Молодец👍 , много хороших идей тут есть
Aaah, that's so awesome! I love that gear, gonna figure it out, I want to create a 3d print model with a brushless motor. PETG turns out to be a great dielectric
Very beautiful construction! Thank you for the quality video, can’t wait to build one!
Cheers
To resolve your corona leakage ... You see these spiky things? There is a spacing between them and the outer edge of the circle. 'This length * 2 + plate thickness + plate spacing' is your max voltage aka corona leakage aka unwanted sparks limiting factor.
Thank you very much, it’s very good Machine😉🤝
Some things you did are wonderful, but others are just asking for trouble. You can not use that wire and banana plugs to carry the high voltage. I don't know your disc diameter, but the maximum spark length (if the polarity and electrode size per that polarity are correct) is 2/3 the disc diameter on a Bonetti. Multi-tooth charge collectors are a BAD idea, you only need a single point closest to the outer edge of the disc, otherwise, the charge collector will discharge to the nearest neutralizer in the direction of rotation when trying to draw long sparks. The thin wire / Banana plug arrangement cant handle more than a few hundred volts, in a 12" (about 30 cm) machine you produce close to 340,000 volts, so each electrode to the capacitor(s) needs to handle 170,000 volts without leaking it to the air, that is impossible with your wire. Use tubing or rod at least 1/4" in diameter (1/2" or 3/4" copper tubing works fine, solid conductors are unnecessary because the high frequency pulsed dc develops a massive skin effect just like AC and will not travel anywhere but on the outer surface of the conductor, larger diameter conductors will work better (1/4" is enough, but larger works even better). See my videos on my Quadruple Bonetti machine. Join the Facebook Electrostatic Machines group as well, and we will help you make this machine perform the way it should! Look at the spark lengths on my Bonetti verses yours. You can do just as well with some advice.
Thanks for the advice. As I mentioned, this was a quick setup, hence the use of too thin wire and banana plugs etc. The machine was finished up a lot better, with the feeds actually running coaxially through the plastic pipe supports (using 19mm metal tubing). But I didn't know about the better charge collector design, so I'll have to give that a try. The discs are 400mm diameter, or so, so yes I should be able to do a lot better. As it is, the sparks max out at about 170mm. I also improved the spark gap by machining some ball ends - still not as good as the hollow brass balls, but an improvement from what was in the video.
Re skin effect: you're right, but with the enormous charging voltage and low current it doesn't much care whether the conductor is 1 ohm or 1k. The discharge path is more critical, but although you can't see it in the video, it is pretty much direct across the capacitor plates, which are actually hidden under the flat ABS base (the vertical tubes are only standoffs, not Leyden jars. The capacitors in the final design are formed from flat foil and ABS sheets. The video shows a temporary arrangement with much bigger sheets, since the edges were not sealed so needed to be way oversize. I think the capacitance ended up at about 600pF each side. The 3mm ABS sheet is a very strong dielectric, but I've had trouble sealing the edges; I think because when ABS cement dries out it leaves voids from the acetone evaporation. Next time I'll try polyester resin degassed in a vacuum.
What capacitor are you using?
On that machine, the capacitors are formed by a sandwich of 1/8" ABS plastic sheet with aluminium or stainless steel foil. The ABS is the big black area. Also used some Delrin sheet (it's what I had at the time - the white sheets on top), but the final machine just used ABS. It has a very high dielectric strength. Note that you can solvent glue it, but the joint has a very low dielectric strength - almost as bad as air. Better to use epoxy or polyester resin i.e. something that does not leave air pockets from solvent evaporation.
How does the gear work? Is it a planetary gear? Would make sense, 1:1 ratio.
No, it's all 1:1 spur gears, plus a 1:1 belt drive to the countershaft, driven from the rear. It's actually not that easy to make a planetary that does a 1:1 reversal. I do have a design, but the planet gears are two concentric spur gears (different numbers of teeth), which makes the whole thing thicker than one might like. Still, a planetary would be far more elegant.
Some commenters have mentioned the bad design regarding dimensions of the corona brushes etc. One of these days I'll see if I can improve the design and post an update. (Actually the machine was a lot improved from the prototype shown in this video, but still needs work.)
@@hardywoodlandcypress Ah, I see it now. Thanks, this planetary gear idea turns out to be exceptionally difficult, at least when trying an exact 1:-1 ratio
@@adrasx6999 Yes that's true. I don't know whether an exact 1:-1 ratio is needed. It would be interesting to see if 3:-2 or similar would work. If so, then a simple planetary gear would work, like 40T sun, 10T planets, 60T ring.
@@hardywoodlandcypress I played around with 8, 10, 28, using a double helix. But ... right now, I'm working on a design for two dc motors. Why go all fancy when not needed.
The gear drive of the motor is very interesting. Can you give more details? The distance between the discs? Nice job
Hi Francesco; I spaced the discs about 3mm. Initially I had a bit closer, but the electrostatic attraction was making them rub so I put in an additional spacer. Fast spinning actually helps because the air layer between builds up some pressure to keep them out of contact. But it does have its limits. BTW the acrylic is 1/8" thick and 16" diameter (it was conveniently cut that way at the plastic shop). I've made some other improvements since I made that video, such as having the capacitance neatly under the metal base of the machine, which was the way it was intended to be, but I had some problems with arcing. It turns out to be difficult to make a corona-resistant joint in ABS. Anyway, it can now go up to 150kV but I'm scared to run it at that level because once I get a dielectric failure, I would have to rebuild it.
For the mechanical arrangement, I used what I had except I made the gears because I'm a CNC software developer. Anyway, it started with an old mainframe tape drive reel motor, which had shaft at front and rear. The rear end drives toothed belt and pulleys, so that rotation is routed back to the front by an offset shaft. At the front, the offset shaft drives a flange which freely rotates concentric with the motor shaft. This flange holds the rear disc. The offset shaft transmits via gears to the flange, so that the flange will counterrotate on the motor shaft. The motor shaft is fairly long, and the front disc flange mounts directly on it. The trick is, of course, to find gears (equal size) which have a centre distance the same as the belt drive pulleys. I was lucky to have a CNC mill to make the size I wanted, but otherwise you could use an idler in the belt drive to make it match the gears.
I'm actually working on an improved design, which will use a planetary gearset to perform the counterrotation. It will be a lot more compact.
I discovered a bit of a trick for bending 6061 T6 aluminium: normally it will crack easily, but a few minutes heating on a gas stove followed by a water quench softened it up so it was easier to work, like copper. The arms holding the brushes do not need to be super strong, so this was useful for getting a nice curve.
@@hardywoodlandcypressI think that's what most of us would be interested in is the gearing concept. The rest is fairly simple.
Your gearing and one sided drive is not conventional and peaks interest.
Maybe go more into that depending if you want to share.
Hi, what sort of gear drive did you use? It´s a planetary, but which made?
No, it's just spur gears, which I made out of acetal resin (Delrin) on a CNC mill. The counterrotating drive is made by having the front disc directly driven by the front shaft of the motor, and the rear disc rides freely on a bearing on the front motor shaft, but is driven by a 1:1 spur gear. The spur gear is driven by a countershaft on the side of the motor, which in turn is driven by a belt drive from the rear shaft of the motor. This forms a reversal of rotation direction, because the belt drive keeps the same rotation direction, but the spur gears cause a reversal. This was the simplest mechanism I could construct. A planetary would work, but would have been much harder to construct. The motor shaft was not very long, so I needed it to be compact.
A guy working on new designs 30 years ago I read his books had sectorless designs also. Seemed to be way better but his you always charged with a PVC pipe at start