Introducing... The Adams Axial

[Jim Mac]

Question for ya,

Little off topic, but if you spin the rotor and use it like a generator, does the output produce a clean sinewave? Or is it more like an M shaped wave?

That's gonna require tools I don't have... yet

Even a cheap scope opens up a whole new world.  I am not the best with using one, but it's an invaluable tool.

[Shylo]

The spot where it switches is what should be looked at

[unimmortal]

Even a cheap scope opens up a whole new world.  I am not the best with using one, but it's an invaluable tool.

Yeah, on the list... my biggest headache is a balanced rotor. 3d prints on threaded rod ain't cutting it...

[unimmortal]

Ok, so we now have two coil sets, 1Kg each, both wound with 0.5mm (24AWG) wire in opposite directions. (Gotta love coiling...)

After trying to get it to run at 100 ohms with everything in series and failing miserably, I wired all of the coils in parallel and connected the coil sets in series to get that down to 3.2 ohms.

So this photo is from a shake out where I'm pulsing from the bottom coil set to the top coil set. I'm now getting EMF from all 12 coils and passing that to 3 x 12V / 0.1A globes to stop the NMOS from overheating - but this is all energy that I want going back into the coils via a supercap or similar.

The rotor now flies due to the coils handling it properly from both sides (push/pull), but I am still only riding this bike with one leg. With 12V, I'm seeing a current draw of 417mA which I expect to drop to sub 100mA or lower once I can pulse from the other side. The globes in the photo are adding ~60mA to the current draw.

So the next step is to switch the battery (-) with an NMOS, which will also drop a voltage divider to satisfy a PMOS gate, which will have a source of battery (+). The coils will be connected between the Drain of the NMOS and the Drain of the PMOS. Placing a pair of these switches at each end of the coil set loop, I 'should' be able to alternate current flow direction by switching the entire coil set from end to end. 

I'm not sure what the effect will be as I'm switching off flow from both sides of the coils at the same time, but slightly biased towards the NMOS side due to the voltage divider in front of the PMOS.

[Jim Mac]

After trying to get it to run at 100 ohms with everything in series and failing miserably, I wired all of the coils in parallel and connected the coil sets in series to get that down to 3.2 ohms.

What happened with the 100 ohms? What was the failure?

I never had much luck with think wire / High Ohm coils myself.  I understand the magnetic field is stronger with more turns, but my motors always seem to work better with low ohm coils.

You have built quite the contraption there!

[unimmortal]

After trying to get it to run at 100 ohms with everything in series and failing miserably, I wired all of the coils in parallel and connected the coil sets in series to get that down to 3.2 ohms.

What happened with the 100 ohms? What was the failure?

I never had much luck with think wire / High Ohm coils myself.  I understand the magnetic field is stronger with more turns, but my motors always seem to work better with low ohm coils.

You have built quite the contraption there!

The main problem I initially had was pulsing from the outside in - way to sluggish to begin with, then with ~880 metres of wire introduces way too much latency when I need a sharp pulse. I may have to revisit the high ohm wiring though - something Adams strongly recommended. Each coil is around 9-10 ohms so I've got plenty of options. I've always favoured 1mm wire for a bigger field with less resistance myself - and above all faster. Having said that the 0.5mm wire is running just as responsively now that it is in parallel.

In other news... I've been able to successfully switch one circuit in both directions (alternating) from the same source! Using 12V globes as the load and simultaneously connecting (+) and (-) to both sides from two different switches connected to a 12V battery. This is huge - as just like riding a bike, each side has a downstroke and upstroke - the rotor will be handled from both sides (N and S) and in both directions (push and pull) as the power is pulsed AND reversed across both coil sets, effectively removing all magnetic reluctance. 

You have built quite the contraption there!

Thanks, it needed to be strong and adjustable to use as a test bench, and I've thought about replicating it - hence the long threads. Funny enough, the Axial itself is barely 10" x 6" - so pretty small for what I am about to do to it  Guests cannot see images in the messages. Please register at the forum by clicking here to see images.

[unimmortal]

Quick update, we're all wired up and the bottom switch test is a success. 12V / 276mA in, ~900rpm, with a ~10mm airgap either side of the rotor. Reduction in amperage is encouraging, but I seemed to have lost a little EMF - possibly because of the (+) side being switched off at the same time - this is where a scope would be handy. So all things being equal I'll see the same results from the top switch.

Then comes the tricky part. I've decided to play it safe and wire in globes for each coil set and visually see if I've got crossover between the hall sensors before giving it a full blown run. Don't cross the streams as they say...

[unimmortal]

After crossing the streams and instantly killing 4 FETs, I'm back to pulsing each side individually with NMOS FETs for now...

Interesting results so far... each run was from a common (-) switched to a common (+), tuned to run at lowest amps possible without any surge or drag.... in a word 'humming'. The VAC column is just a multimeter set to AC across the coil. The 33V run was where this really woke up, although I'm gonna run out of road at 36V, so to maximise I'll look at dropping resistance by wiring 2 or 3 coils in series, in parallel - somewhere around 12-18ohms for each coil set and see if I can keep the current draw down with the resistance drop. 

Feeding all of this EMF back in is still the main goal, but at the moment I don't have any solid ideas on how to do that.

The other tihing I've realised with the 5mm airgap is the rotor is chasing the coil cores and causing a wobble because the threaded shaft (10mm) is flexing. This is all happening because DIY tolerances aren't going to cut it for much longer, so a trip to the engineers might be coming up.

Second pic is a 240V/50Hz 200W commercial floor fan motor that I'm mocking up to bolt on ... Guests cannot see images in the messages. Please register at the forum by clicking here to see images. ... oh yeah baby, we've still got shaft torque to play with too!

[unimmortal]

Another quick update: a little tuning and removing the power switches (yes I melted plastic 12v led switches) and input current is now down to 125mA @ 36V (4.5W), turning around 900rpm. Vac on each coilset is ~42.5V.

Edit - final entry for this testing phase.
36V in @ 57mA (1.8W), 400rpm, 27Vac @ each coilset.

[unimmortal]

Quick video of the Axial switching through individual 50 ohm coils sets. 10mm airgap.

https://youtube.com/shorts/MMF0cXhSq0o

...and switching from the bottom through both coils sets at 25 ohm

https://youtube.com/shorts/zXt1P9sF27U?s...ztPFIjkrtG