I am trying to make a virtual rotating generator again with some modifications. I am opening a new thread to keep the builds separated and not bogged down with pages of other ideas, discussion and attempts..
Here are the highlights I am aiming for:
72 contact stationary commutator- (divisible by 36 for common 3 phase cores). Also easier to manage than 120 contacts.
Brushes rotate on the outside. My last one (rotating inside) the force required was immense! And accessing the brushes to make changes was impossible with completely dissembling.
Centrifugal force management- I will first try cantilever brush arms- meaning the brush arms are on bearing pivots with adjustable weighted ends extending outward. The rotational force should pull the heave ends out, thus applying pressure to the brush / drum inward.
Commutator drum body printed in ABS (more heat tolerant) and closer contacts. The less contacts and tighter spacing make the size more manageable- the drum is roughly the diameter of a coffee coaster.
Here is the start of the commutator..
I will be chucking it in a drill press to sand the contacts flat and smooth.
Next post I will describe the stator / rotor idea..
04-15-2025, 10:47 AM (This post was last modified: 04-15-2025, 10:48 AM by Jim Mac.)
The core is from an old truck alternator.. 36 slots.. I will not be overlapping like usual builds, but winding my tapped coils for the rotating primary every other slot like this:
The above pic was from a previous experiment but the coils were shorted against the unfinished core. So the windings have been removed and I grinded the rough parts off. Now I am enameling the whole core with barbecue high temperature enamel.
The core is interesting, as I may have the possibility of adding output coils in 2 ways.. First around the slots between the primaries in the same manner, but the core also has the inside facing stubs, which may also allow me to place an additional output coil assembly inside. We will see.
The primary will have 18 coils- that means I need to wind 4 taps per coil, all coils in series. The remaining 18 output slots is dividable by 3, thus allowing me to tap 3 phase output.
The coil and generation part may change, as my main focus is getting this commutator / switch assembly right and reliable. From that point, I can try different ideas and arrangements for the coils. But this is what I am thinking at this moment.. However it goes, I am convinced it needs multiple output phases so after rectification, we have a DC bias to feed the input.
Here's the update. I had to remake the commutator for 2 reasons.
1. The filament and crazy glue was not enough to withstand the force sanding and bending the contacts, so those bars need to be epoxied.
2. The first one did not have bearings embedded for a spinning shaft. I figured I could put the bearings on the brush carriage, but attaching the motor is a challenge like that.. So since I needed to re-do it anyway, might as well get it right..
The inside ring near the brush is ABS, and the bearing holders / epoxy tray is PLA. Should be no problem.
First side epoxy resin has been poured. Tomorrow I will pour the other side..
As far as the core / coils, I started the winding of the output 3 phase.. I got 12 of the 18 coils done.. So tomorrow is a day of winding, chilling, and start printing the other parts..
Other side epoxy has been poured. It's looking darn good so far.
Now as I said before, centrifugal force pulls the brushes out. Even they are spring loaded, the faster RPM, the less pressure the brushes will have against the contacts. With enough speed, we start losing contact, which needs to be avoided.
So the plan is this:
The drum stays stationary and the shaft spins via bearings. The brush carriage is fastened to the shaft and spins the brushes. The brush arms are on cantilever hinges. The horizontal rods are threaded- which weights can be added or subtracted via washers and nuts. The centrifugal force will pull the weights out, thus pressing the brushes inward to maintain pressure. And the pressure can be adjusted to ensure constant contact by tweaking the weights (adding / removing / repositioning the nuts) .
Instead of using split collars, you can see my shaft compression design to secure the carriage to the shaft. A simple hose clamp slips over the compression fitting and secures it to the shaft.
Now to winding the core.. Since this is a "Build Log" I am documenting step by step, so I can remember the exact steps in order.
The 3 phase output coils are in place. I figured I should wire the output first, because it's going to get real messy once I wind the primary 72 taps, and I want the taps to be totally accessible.
So I have 18 coils using every other slot of a 36 slot stator. I used 24 AWG to make 75 turns per coil, 6 coils in series per phase. So start- 1-2-3-4-5-6 is phase ONE.. 7-8-9-10-11-12 is Phase TWO.. 13,14,15,16,17,18 is Phase THREE.
I checked for shorts, and checked Ohms in each series. Each output phase is 2.6 Ohms.
Now to think about the primary. I plan on using 26AWG wire and doing 100 turns per slot X 18 coils. These 18 coils will have 72 taps equally spaced. I chose a smaller wire for the primary to ensure the output coil can handle more than the input current just in case this works.
I need to now think of a way to keep this organized and labeled as I wind the primary coils.
After some minor tweaks to the plan, it's starting to take shape..
I haven't sanded the drum much yet, and it already rotates and glides pretty smooth.. I will be printing out the motor mount overnight. Once that's done, I need to complete the slip rings and input brush assembly, do some sanding, then I should be ready to fire her up..
On the coil progress, the 3 phase output is all wound (green) and I completed 3 of 18 primaries.
I slapped together a simple jig in the bench vise, and it's doable..
May not get much done this weekend (holiday) but I'll be chipping away at it.
Excellent work man, I like the design... hopefully it proves out the brushes and slip rings - although it is gonna look hectic at a few hundred rpm Guests cannot see images in the messages. Please register at the forum by clicking here to see images.
(04-19-2025, 05:44 AM)unimmortal Wrote: Excellent work man, I like the design... hopefully it proves out the brushes and slip rings - although it is gonna look hectic at a few hundred rpm Guests cannot see images in the messages. Please register at the forum by clicking here to see images.
A youtuber commented the same thing, worried about high speed failure. I think it's going to depend if I can get the contacts nice and smooth and level. If it starts bumping on the contacts, it will be a vibrational sloppy mess.
I sanded the bars with a drill press till each bar was flat to ensure they match the curvature. I periodically checked and marked the low bars, then returned to sanding till all bars were right. Then I hit it with 320 grit..
Then I put her together and spun her up.. It takes roughly 1/5th the power to spin compared to the last build with the inside brushes.. I am aiming for around the 10 watt mark when completed, and it looks like I am on par.
The input sliprings and brushes have been installed. This should be an accurate test of input power vs RPM, as all the friction is added. I was aiming for roughly 10 watts input, and I came out roughly 12 watts input required rotate this 800 RPM.
Lasco and Escumo- since you both have mechanical commutators, what is your input wattage like on just the commutator unloaded? I would like to do some comparisons.
I don't see how I can tweak the brushes or contacts much further, the only possible improvement would be the tube shaft connector. I may be able to lower input a bit still, but it may not be too important.
Up to 800 RPM I can hear the brushes contacting and all looks working. I raised the RPM another volt and it spins true, barely any vibration but the brushes start to lose contact between 800 and 1000 RPM. Somewhere in that range, the centrifugal force pulls the brushes totally off the commutator and it air spins..
This is what I expected, and I did not yet add weights to activate the cantilever. I can still tighten the spring compression some, + the weights, so higher RPM may be possible. I am just being cautious at the moment because I am not sure what it can safely handle.
All in all- pretty good.. Next step is to do a test to see if it is dropping contact at any point, and what RPM is reliable.
