5 coil machine running

[knagtegl]

Hi,

I made a 5 coil machine and so far it runs very well but I would have hoped for even better results
because I do not see a non linear improvement of the charging.

I will for sure not have a perfect tuning of all the parts but I am close.
I would like some advice on further improvements.
I mean I can add 5 more coils and more transistors but that is obvious.
I would like some tuning tips and any comment for improvement is welcome.
Will start to do some performance testing today.

The specs of the machine:

MDF rotor with 20 magnet slots for the superpole magnets the rotor weighs 1.6 kilo and is
covered with glassfiber and epoxy resign to prevent magnets flying off.

The bearings are ceramic bearings made for rolerskates.

the coils are made of r60 rods and 130 feet (40 meter) of 0.3 mm (#28 wire) 24 strand twisted together. every coil is around 1 kilo and resistance of around 0.6 ohms

It wants to run at about 1500 rpm using about 1.2 Amps input

Every coil is powered by two transistors (MJL21194) wich are matched. one diode 1n4007 is going from the two colectors to the high voltage bus and every transistor set has one neon.
the base of all transistors have a 22 ohm resistor and they go to a 500 ohm potmeter.

One coil has a trigger winding for all coils

I have a film of the machine here: new 5coil bedini monopole - YouTube

[Tom C]

knag,

so you basically have 10 circuits running...... your rotor is very very heavy, you are expending a lot of energy just turning it. your coils are small so there is not much of a pumping action. bearings sound like they are not lubed correctly. I do not know how to get good charging out of it... how big are your batteries? are you getting 1 to 1 charging? if not parameters need to change.

Tom C

[knagtegl]

Hi Tom,

The rotor itself is actually very light but the 40 magnets are responsible for the weight.
What is the size of a good coil in your opinion and what would the resistance of that be?
I will look at the bearings, I am not sure or the ceramic ones need to be lubed the sound is also
because the frame is wood and vibrates with the machine.

thanks for your reaction,

Karel

[Tom C]

1 kilo is over 2 lbs not too bad..... what is the core size of those spools? should be 3/4 of an inch 20 mm? you can do 8 of those circuits off of one coil at 20 awg wire.... you just have a lot of stuff hanging off everywhere. the multi coil machines are also hard to time.... coils have to be very precisely aligned. scalar norths are hard to fire that is why your base resistance is so low.

Tom C

[knagtegl]

Tom,

The core is 20 mm (3/4 nch) and 75 mm long the total diameter of the coil is 60 mm and completely filled.
The base resistor can be a lot higher but at the moment I don't have enough of the same resistors.
I will go to 100 ohm at every transistor I think and then I will tie the wires of the bases so it does not
hang around like they do now.
I found that that the precision of the position of the scalar magnets on the rotor is far more important
than the position of the cores. The rotor is precision made (0.1 mm precision). I can slightly move the coils
this does make a small difference in rpm and charge but not much

Karel

[John_Koorn]

Knag, overall that is a pretty good setup. I would not have the coils held on with zip ties. I would also change the coils out to 4 strands of #18 wire for the slave coils and the master I would make 4 strands of #18 and 1 strand of #23 for the trigger. If running at 12v you could have 1 transistor for 2 strands safely.

Also, use an analog amp meter for draw current measurements not a digital one. As for tuning, once you have the base resistance set so you get 1 pulse per magnet pass and the highest RPM adjust the coil to magnet gap on all coils for faster RPM. I can't see what's on you scope.

good job and keep spinning

John K.

[knagtegl]

About the wire size, I think that for example three times .3 mm is about the same as 1mm as long as the resistance is the same.
I think that it might even be better to have more smaller wires. Since I only have a big spool of .3mm wire I have not much choice.
The resistance of all my wires together is around 0.6 ohms. I can switch this with only one trannie as long as the timing and tuning is right
if your pulse is to long your current go's up and you burn your trannie. In all the experiments I did until now I found that it is low impedance
and timing that is the main thing.You can almost go to 0 ohms coil as long as your pulse is short enough so the current can not build up.
Somehow the counter force of the zero point field gives you the full power back as a voltage.

I am still playing with the config of the trannies, I think you should have a coil resistance of 1.2 ohms to get the best effect with one transistor.
How to best configure the trannies is still not clear to me. It seems to make no difference to seperate the 1.2 ohm circuits and have a diode on every collector or just join all the wires and all collectors together with one diode (as I have now).
At the moment I have one pulse per magnet on all 5 circuits at 1.3 Amps in and 1520 Rpm.

As for the analoge amp meter, I have one but not on this location but so far I have not seen much differance but that has to do with the
fact that I use a total 40000uF caps at the input just before my coils so the buffer my current input.

Karel

[Tom C]

Karel,

if you build it like JB designed you will have good results, there is a reason he does EVERYTHING...... otherwise you will play with variables all day long. resistance of the coil is not the only factor inductance is also which has to do with turns which has to do with wire size and spool core diameter. I dont know why the transistor configuration is not clear to you.... one strand one transistor, common trigger bus, output diodes to common power bus..... you are not getting good charging now, which is why you posted. first build it like he says to then work your way to different builds. 4x18 as John K said is a viable coil.... every build I have seen from JB is 1 wire 1 tranny.. now on the output you can split the outputs to branches, single diode single battery (see forced oscillator thread) all batteries share a common ground.

Tom C

[John_Koorn]

I would follow Tom's lead and go one tranny per strand. I did try 2 x #18 per tranny on the 10 coiler I had and I spoke to JB about it and we agreed that as long as you stay within the SOA of the tranny you'd be fine. But JB is keen for people to experiment once they know what they are doing.
Is it better or worse? Well that's for you to find out by doing the proper experiments.

Karel, I once ran a very similar setup to yours with #23 wire and the charging was absolute crap. BroMikey, #14 is probably going to smoke your trannys. Don't know why you'd want to use a different tranny than the tried and trusted MJL.

John K.

[knagtegl]

Thanks for all the advise, I was wrong about the wire sizes, did not look it up but just guest it would be the same with more smaller wire
but I looked it up:

#18 wire has 0.0210 ohm per meter 40 meter is 0,84 ohms per strand 4 strands parallel is 0.21 ohm
#28 wire has 0.213 ohm per meter 40 meter is 8.52 ohm per strand 24 strands parallel is 0,355 ohm

So #18 wire is best to use.

Did the first tests today I took 1.8 Amp out of the batt for 6.5 hours which makes 11.7 Amps/h
I charged it with 1,3A input for 22.5 hours and it was up to 15,15V so this makes 29,25 Amps/h

Don't know if this is very bad, don't have a good reference for what the charging should or could be.

I will make the coils connections more permanent and loose the zip ties in the near future and maybe
install them again like Patric suggested. And for the not linear improvements you are right I did not do
the runs yet so i can not tell exact but I expected to much I think.

Karel

[knagtegl]

First 17 runs of the 5 coil machine on an old 60 amp/h battery.

SSG COP 60ah.zip

Karel