Need Help on 3 coil Bedini

[Sephiroth]

I think you are right... 450 is very low and is most likely the problem... though remember having more windings on the trigger than the primary will have the effect of a step up transformer so ideally you would need to increase the power windings as well...

can't wait to see it going!

[Peter Lindemann]

Last Few things

Bits,

Here are a few final pointers. First, look at the pictures of the build we did. We used 1/16th inch diameter, copper coated welding rod and did NOT insulate them from each other. Eddy currents in the core are NOT a problem in these machines. Second, when the rotor is finished, we ALWAYS wrapped it in strapping tape as a final insurance policy against glue failure and magnet launch. Do not fail to do this. Third, 120 feet of #18 wire is more than sufficient for your coils and trigger. Take all of the extra windings off and balance the system with the resistors. Fourth, as you can see from the photos, the trigger winding comes out and goes directly to the ON/OFF switch, then to the light bulb, then to the silver colored loop, then to 6 identical resistors, one each to the transistor blocks, and then one more resistor to each transistor base. The trick is to get all of the transistors to SHARE the trigger current equally, so that each transistor draws about .25 amps. Balancing the system to run on 12 volts will require a different set of resistors than other run voltages.

Good luck,

Peter

[Bit's-n-Bytes]

Sephiroth

I do have a video of this working but can't upload that here. Can I email to you?

[Sephiroth]

We used 1/16th inch diameter, copper coated welding rod and did NOT insulate them from each other. Eddy currents in the core are NOT a problem in these machines.

Wow... I didn't expect that!

But did you intentionally choose to allow eddy currents in the core? You say it isn't a problem but does it have some benefit?

[Bit's-n-Bytes]

Peter, thanks for pointing out the trigger circuit. In this 24v model, is the resistors connecting to the transistor blocks 22 Ohm and the individual base resitors 470 Ohm? I will get some strapping tape on the magnets, thanks for the advice.

[uusedman]

Hi Peter,

Is there a reason why the extremely short circuit paths? How or maybe why the use of 12 transistors when there is only 6 coils?
What kind of tests did you run? If yes, can you please share them with us?

Many Thanks



@Bit's-n-Bytes
beautiful design, I guess may question is at you too, you have 3 coils and 14 transistors, what for?

Dear Bits,

Nice work. Here are two photos of the closest machine John and I built in 2004 to the one you are building now. Please notice the extremely short circuit paths made of large diameter wire or copper tube. This machine was a 24 volt machine designed to run on 6 amps, which is 1 amp per coil (4 wires per coil) and .25 amps per transistor. Only one coil had the trigger winding that produced sufficient trigger to fire all 24 transistors simultaneously.

For more pictures of this machine being built, go to: New Page 2

Good luck,

Peter

[Bit's-n-Bytes]

Uusedman, 3 coils (spools) 5 total wires per spool, in top spool, 1 wire used for trigger circuit. 15 in all coils, minus 1 for trigger, leaves 14 coils to be activated ea by their own transistor.

Hope this helps

[Peter Lindemann]

Not 470

Peter, thanks for pointing out the trigger circuit. In this 24v model, is the resistors connecting to the transistor blocks 22 Ohm and the individual base resitors 470 Ohm? I will get some strapping tape on the magnets, thanks for the advice.

Bits,

470 Ohms is always the best place to start for a single coil unit, but in these multiple coil systems, with shared triggers, lower values were used. 22 Ohms to the blocks is a good place to start, with 220 Ohms to each transistor base. We had bins of every value resistor available, so tuning was relatively easy. Also, we used 2 watt resistors in the bigger machines.

Start with one block and get it running right with the transistors drawing about .25 amps each. Then, just duplicate that system to the other blocks and it should work great.

Peter

[Bit's-n-Bytes]

Great information Peter, I can't thank you enough.

[tjnlsn255]

Bit's,

Are you using 18 g wire for your trigger? If so I have been told that is way to large.... they normally are 26 or 24 .... I use 22 but only because I have it on hand.... as I understand it you do not want the trigger circuit to consume extra current as that makes the motor less efficient..... also it makes a large current available to the base of the transistor and that can make it very hot and that is not a good thing.....LOL

I hope this is correct.... someone if not please let me know.....

Have a most excellent adventure....

Tj

[ren]

Not necessarily so TJ.

Ive used AWG 18 as a trigger before, and so has John I believe. I read somewhere that the huge 240watt machine (10 pole) used OTT number 6 (?) stranded wire, with 5 strands available on each coil.

I believe this wire is pre fabricated like this, in other words, it comes pre twisted, however I could be wrong.

Anyway, I read John talking about this machine saying it had 49 power strands, implying to me that the 1 final strand was left for triggering.

So saying, it may work better in some instances to have a thinner gauge for the trigger, perhaps more so in a smaller build.

Id start back on one coil and get it running like the specs shared above by Peter. Try 100 ohms and a 1k pot and fish around for a good resistance setting on one coil with only one power winding functioning. Seps suggestion of switching on/off power windings is a good one. I have done that and it helped me track down a non functioning transistor.



Regards

[Peter Lindemann]

Just for Clarification

Not necessarily so TJ.

Ive used AWG 18 as a trigger before, and so has John I believe. I read somewhere that the huge 240watt machine (10 pole) used OTT number 6 (?) stranded wire, with 5 strands available on each coil.

I believe this wire is pre fabricated like this, in other words, it comes pre twisted, however I could be wrong.

Anyway, I read John talking about this machine saying it had 49 power strands, implying to me that the 1 final strand was left for triggering.

So saying, it may work better in some instances to have a thinner gauge for the trigger, perhaps more so in a smaller build.

Id start back on one coil and get it running like the specs shared above by Peter. Try 100 ohms and a 1k pot and fish around for a good resistance setting on one coil with only one power winding functioning. Seps suggestion of switching on/off power windings is a good one. I have done that and it helped me track down a non functioning transistor.

Regards

Ren,

You are mostly correct. The 10 coil machine was made with 18 gauge wire. Each coil had 4 strands of #18 wire, except the coil that had the trigger in it. That one had 5 strands. John and I would measure out the wire from the back of the shop wall all the way out into the parking lot (about 120 feet). Then, the strands were bundled and tightened into a drill chuck out in the parking lot and tied off to the back wall of the shop. Then, using the electric drill, we created the "twisted wire sets" ourselves. We kept twisting until there was about ONE TWIST PER INCH of linear length. So that machine had 41 strands, not 49. That's 10 coils with 4 strands each, plus one trigger.

I hope this helps.

Peter

[ren]

many thanks Peter for setting the record straight.

Its great to see that even the masters stretch out their wire and twist it together just like us amateurs. The local sports field has my tracks worn into one of its lengths

I was trying to relocate the information I read regarding the monopole in question. I believe it was on a PES site? It was a series of emails between John, yourself and Sterling discussing this and that.

In it I am quite sure John said the wire was OTT stranded wire, which lead me to search for WHAT OTT stood for. Funnily enough I found some OTT wire on Ebay, and thus I drew the conclusion that the wire was pre twisted upon purchase.

I can see now that you guys built this from scratch, doing the twisting yourself. Johns reference to OTT could have simply meant "off the table" or something equally non important, or perhaps I misunderstood the sentence.

Of course there is the ever looming possibility that I got it all wrong, and there are no references to OTT at all

Im sure I will come across that page again in my travels. Thanks for your help.

Regards


EDIT: Page found here John Bedini and Peter Lindemann's Medium and Large Motor-Energizer Project

[Bit's-n-Bytes]

Success!

Success Guy's you are awesome. I changed the resistance as instructed by Peter, tuning one by one each transistor. Let me say that I am now down to about 3 amps "total draw" and have picked up RPM. I am going to implement further suggestions has time permits. I will be designing a method in which to "auto sense" two separate 24V battery banks that I use for my grid tie inverter and while being charged from this device, be able to "roll" the charging / load process. I built a data logger using a PICAXE 18x chip design for my wind gen's and will program this to monitor and log this process. Would it be safe to say that a cap should be installed on the outgoing charge terminals? If so, what should the uf and voltage rating be? I will provide more info on the finished product as I get to that point.

Thanks again

Bit's

[Iotayodi]

Of course there is the ever looming possibility that I got it all wrong, and there are no references to OTT at all

He was probably thinking "ought". Electricians use this term on larger sized feeder wires. They are designated by the zeros. One zero is called 1 ought. 2 zeros are 2 ought and so forth. The more zeros the higher the amperes that can be used.
American Wire Gauge table and AWG Electrical Current Load Limits