Bedini Ferris Wheel Regauging Motor

[John_K]

Sometimes...

Sometimes I can't believe how dumb I am

On thinking of ways to slow the machine down I missed the easiest way to to it. It was literally staring me in the face.!!

You would think that all those years with working with Monopole Bedinis would help??

Ummm, how about if I adjust the coil to rotor magnet gap???

I increased the gap by about half an inch. Now it runs beautifully at 24RPM and about 6.7Hz - right in John B's ballpark.

Some times I just want to kick myself!


John K.

[John_K]

The slave coils

Hi All,

I'm starting to think some more about the slave coils and why John B has them the way he has.

The most obvious reason I can think of is to lower the total impedance of the coils, which should give a bigger radiant spike and therefore better capturing of the "waste" into the cap.

The next thing that comes to mind is that John B said the 3 coils are in interferometer mode. I did some Googling on interferometry and is is very interesting. This IMO is what creates the non-linear situation.

Thirdly, it's a neat way to slow the machine down. The outside coils can act as a "brake" as they will load the machine down if positioned correctly without degrading the performance.

Just my thoughts.....


John K.

[Les_K]

Hi Les,

I'm far from an expert in this area, especially with FETs. If it were me, I would spend some time with the "vanilla" Bedini/Cole circuit first and then modify it later once you know exactly what is going on.

I believe John B was talking about using a FET for the cap pulser section, not on the front end.

Good that you're thinking though!

I have learnt more in the last couple of days now I have a running machine in front of me than in the last month just reading all the posts in the thread, but that's the kind of guy I am.

The machine is not that hard to build. I have tried to keep as close as I can can to the instructions with the parts I had and also a limited budget. I still have a lot of unanswered questions in my head that I will figure out the more I play with it.

I am very impressed that I can turn a 20lb wheel and charge almost one to one with about 5W out of the primary! I'm not using the mechanical energy yet, but I may attach a fan to the shaft as this will be an easy way to load it down some and get the speed down.


John K.

I am the same way John, I have to get it in front of me for all the reading in the world...
And yes, as I stated I am staying with the vinilla....mmm, mmm, goood! for now.

I am saving for the upcoming conference so my budget for this got clipped a bit....

Thanks for the inspiration John K, It is greatly appreciated.
Les

[John_K]

John K,
I am still quite a ways from this part but have been thinking about it a bit.
I don't know if you ever had a chance to look at the file I posted on monopole_2 about using a fet in place of the MJL21194 on the SSG, but I noticed that with a Zener on the gate the fet would fire very quickly. (ie reduce the pulse width..A lot) and was wondering if that might be a good front end for driving the MJL's in the Bedin/Cole in this application?
I am sticking as closely to following the build as described but since JB mention a fet in one of his posts it just got me thinking. May not have had anything to do with this end of it.

Les

Hi Les,

I'm far from an expert in this area, especially with FETs. If it were me, I would spend some time with the "vanilla" Bedini/Cole circuit first and then modify it later once you know exactly what is going on.

I believe John B was talking about using a FET for the cap pulser section, not on the front end.

Good that you're thinking though!

I have learnt more in the last couple of days now I have a running machine in front of me than in the last month just reading all the posts in the thread, but that's the kind of guy I am.

The machine is not that hard to build. I have tried to keep as close as I can can to the instructions with the parts I had and also a limited budget. I still have a lot of unanswered questions in my head that I will figure out the more I play with it.

I am very impressed that I can turn a 20lb wheel and charge almost one to one with about 5W out of the primary! I'm not using the mechanical energy yet, but I may attach a fan to the shaft as this will be an easy way to load it down some and get the speed down.


John K.

[Les_K]

I've been studying my new toy and fine tuning it some.

I'm pretty sure I have the timing a lot better now. The hall switches when the rotor magnet is about 1/4 of the coil width after TDC. This gives the rotor magnet a good push when the coil is energized.
When it is timed right, as John B says, if you position the rotor so one of the rotor magnets is a couple of inches from the coil and you give the rotor a small push in one direction so the magnet approaches the coil, the coil will fire and push the rotor in the opposite direction and self start.

The dwell I have on the timing disk magnets now makes the hall switch off when the coil de-energizes about 1/4 of the distance to the next magnet, or about 6 degrees past TDC of the coil. At this setting the draw current is reduced (to about 300mA on my setup) and the rotor speeds up. The cap charges up faster and dumps more often which makes the secondary battery charge faster.

The transistors now run cooler, because of the reduced duty cycle, although the MJL21194 is still slightly warmer than the MJL21193.
I feel I can reduce the draw current further by reducing the dwell on the hall even further, but I'll study it some more before I change it.

I found it is really important to have the rotor magnets spaced perfectly and the timing wheel magnets spaced and positioned perfectly as well. My rotor magnets are not perfect, nor are my timing wheel magnets so this throws the timing off a little bit on one or two of the 16 pulses per revolution. You can easily look at the timing by slowly rotating the rotor by hand and watching the amp meter on the primary battery whilst noticing the position of the rotor magnet in relation to the coil core.

There is something fascinating about watching the wheel turn and the super bright LEDs flashing, as well as the charge battery crawling up. I' going to let the charge battery charge up to 16V and then rotate batteries and see what happens.


John K.

John K,
I am still quite a ways from this part but have been thinking about it a bit.
I don't know if you ever had a chance to look at the file I posted on monopole_2 about using a fet in place of the MJL21194 on the SSG, but I noticed that with a Zener on the gate the fet would fire very quickly. (ie reduce the pulse width..A lot) and was wondering if that might be a good front end for driving the MJL's in the Bedin/Cole in this application?
I am sticking as closely to following the build as described but since JB mention a fet in one of his posts it just got me thinking. May not have had anything to do with this end of it.

Les

[Les_K]

As I reread the entire post AGAIN last night, I found a place where John said that the core of the center coil is THREE inches below the coil, not the two inches I had down in my notes. Sorry guys. I guess I should not offer information without checking the source first. THAT won't happen again. I am annotating all my notes with the post number so I can quick reference to prevent that happening.

Quoting John... Post # 111 "The shield is adjusted to right angles as that indicates the Bloch wall of the big coil in the north hemisphere the south pole is extending below the coil by 3 inches for a reason to gain that right angle. so the cores goes right through the shield."

My apologies as well, I noticed that and got lost in the shop and then forgot to reply.. Thanks for the correction.


Les

[mklimesh]

How slow will it go

If I have the hall switch on the outer edge of the timimg magnets I can get to about 8 magnets per second (MPS) aprox ( 30 rpm).
If I put it on the inner edge of the magnets I can get to about 4 MPS (15 rpm).
It ran it all night at 7 MPS.

The leds on the scr/zener/2leds switch are interesting to watch while changing cap size and rpm. I was able to get them to blink like JB showed on the BFW, 2 short blinks, a pause, then 7 or 8 blinks in a row run together.

I tried the same zener/2leds to switch a mosfet, leds didn't light but the scope looked like it was pulsing each magnet pass.

Mike Klimesh

[harz]

COP calculations

I would like to use the next quote from Mr. Bedini's post #367, to pick up some values he has given in that post concerning the big Bedini Ferris Wheel regauging Motor, to make some calculations. I have marked those values red and bold.

Turion,
Yes I did say that, what did I say, That your not going to do this with a roller-skate wheel. The machine is very big I would have liked to make it much bigger. The batteries I used were equivalent to 9v transistor radio battery on that machine.

No body is stopping anybody from building this machine. I let everybody take all the pictures they wanted to I answered what I could. I'm not going to do the work for everybody I encourage people to think for themselves because it's the only way to learn anything.

I have tried many different ways to state what the energy is here and what the machine is. Everybody can see the geometry of the machine. On a small scale I do not think you will get the same results. The monopole always worked like this but it was on a small scale. The coils are what the difference is. Big Coils big output. The machine does not need to go that fast to do the work 20 rpm's is fast enough here. I also encouraged everybody to look at the meters since that is what they want to see, but I guess everybody missed that.

So I will state it again, what it was, 36v 1.8 amps input. output batteries charged to the pin, limit on 2.8 amps float 36 volts. I could have swapped the batteries anytime I wanted to or just plugged an inverter in and sucked off all the power and sent it back to the primary but I ran out of time building the machine I almost did not finish it.

I also warned Rick to not grab the shaft as it would rip the skin off your hand. Forty six foot pounds on that shaft can run many things so you did not even consider that power it all adds together.

Take the big leap of faith and just build it with the knowledge you have it will work like all my toys do. See the way I see it is, it is just a hobby for most people and it does not mean anything, but this is a job to me building machines and chargers and switchers.

The time is coming when we may find everybody without power, that is why I said I'm running out of time. Since it used something like ED's L , Radus type arrangements you need to build small scale to understand how it works, that is why I wanted to test the output switching as Mathew and I-Ron did and they did make it work because of the youtube video's they posted, and I did watch them 5 or 6 times to make sure it was working.

John K also stated you can skin that cat many ways, he is right, what works for me may not work for you. Jerdee has done a wonderful job posting all those drawings, we are all lucky to have him here. He has almost everything right to the "T" in documentation on this machine, Great drawings. I will do my best to keep going on here.
John B

So we can calculate that the input power is P = 36V * 1.8 A = 65 W and the electrical output power is P = 36 V * 2.8 A = 100 W.

Mechanical output power would be P = torque(Nm) * speed of rotation(rad/s). Torque is 46 foot pounds, which is equal to 62 Newton meters. Speed of rotation is 20 rpm, which is equal to 2.1 radians per second. So we get mechanical output P = 62 Nm * 2.1 rad/s = 130 W, which is equal to little bit more than 1/6 hp.

Now we can calculate the electrical, mechanical and overall efficiencies of that machine according to those values Mr. Bedini has given to us.

Electrical COP = 100 W / 65 W = 1.5
Mechanical COP = 130 W / 65 W = 2
Overall COP = 230 W / 65 W = 3.5

I think maybe the machine can be tuned to even much better results including the use of the machine at much greater power levels, these values to be somekind of initial values proving it is a real overunity machine. Great work mr. Bedini!

Greetings to all!

Harri from Finland (sorry my imperfect english)

[txaggie00]

@ John K.

Thanks John! I appreciate your assistance and input!

We are planning to get everything cut with a CNC machine so that we know all the angles and rotor magnet spacings are perfect. I was hoping to do the same for the timing wheel by incorporating it directly into the big rotor (It also saves some money for not having to pay for extra machining and material for a separate timing wheel!), but from your last post, I am beginning to think that I should leave all the holes to test, with the hopes that one of them will work out nicely. Our timing wheel magnets are about 20mm and we have 5 options for spacing: 30mm, 40mm, 50mm, 60mm, and 70mm (center to center on the arc).

I am glad to see you got your timing better! Loved your video and hoping to see your S1GT in action some more. If you have any more tips on the timing wheel, I would love to hear them. Chances of us getting the machining done before the new year is slim, but we are excited the ball is rolling!

Thanks,
Scott

@ LutherG - I read your PM and have responded to you

[Turion]

Length of center core

As I reread the entire post AGAIN last night, I found a place where John said that the core of the center coil is THREE inches below the coil, not the two inches I had down in my notes. Sorry guys. I guess I should not offer information without checking the source first. THAT won't happen again. I am annotating all my notes with the post number so I can quick reference to prevent that happening.

Quoting John... Post # 111 "The shield is adjusted to right angles as that indicates the Bloch wall of the big coil in the north hemisphere the south pole is extending below the coil by 3 inches for a reason to gain that right angle. so the cores goes right through the shield."

[John_K]

Studying and fine tuning the S1GT

I've been studying my new toy and fine tuning it some.

I'm pretty sure I have the timing a lot better now. The hall switches when the rotor magnet is about 1/4 of the coil width after TDC. This gives the rotor magnet a good push when the coil is energized.
When it is timed right, as John B says, if you position the rotor so one of the rotor magnets is a couple of inches from the coil and you give the rotor a small push in one direction so the magnet approaches the coil, the coil will fire and push the rotor in the opposite direction and self start.

The dwell I have on the timing disk magnets now makes the hall switch off when the coil de-energizes about 1/4 of the distance to the next magnet, or about 6 degrees past TDC of the coil. At this setting the draw current is reduced (to about 300mA on my setup) and the rotor speeds up. The cap charges up faster and dumps more often which makes the secondary battery charge faster.

The transistors now run cooler, because of the reduced duty cycle, although the MJL21194 is still slightly warmer than the MJL21193.
I feel I can reduce the draw current further by reducing the dwell on the hall even further, but I'll study it some more before I change it.

I found it is really important to have the rotor magnets spaced perfectly and the timing wheel magnets spaced and positioned perfectly as well. My rotor magnets are not perfect, nor are my timing wheel magnets so this throws the timing off a little bit on one or two of the 16 pulses per revolution. You can easily look at the timing by slowly rotating the rotor by hand and watching the amp meter on the primary battery whilst noticing the position of the rotor magnet in relation to the coil core.

There is something fascinating about watching the wheel turn and the super bright LEDs flashing, as well as the charge battery crawling up. I' going to let the charge battery charge up to 16V and then rotate batteries and see what happens.


John K.

[John_K]

Hi Scott,

Nice drawings!

I haven't figured out the timing properly yet, but I would pay close attention of the pictures that have been posted of John B's timing wheel.

The way I understand it, John B said the the fields overlap on the timing wheel. I have used 15mm diameter timing wheel magnets that are spaced 12mm apart. I feel that these are probably too big for a 4 foot wheel as I have had to adjust the hall switch so the dwell is very small, meaning only the bottom part of the magnet passes the hall.


John K.

Hello All!

First time poster, long time lurker. Went to the conference in Nov and was blown away. I am anxious for July, but have a lot to do before then. I have thoroughly enjoyed this thread and am learning a lot. My friend, Brent, and I have started engineering our mini BFW and I just had a few questions. We are finishing up some CAD work on the wheel design and wanted to incorporate the magnet placement for the Halls timing onto the one of the main wheels. We are unsure about the dwell arc for the timing. It's a 6 foot wheel and have designed three different diameters for the timing magnets placement: 10", 12" and 14". The magnets we are going to use are about .78" in diameter

@ John Bedini
Is there a certain size ratio or is the dwell arc dependent on the rest of the system?

Just wanted to see if I could get any clarity before we send the acrylic to get cut. I guess if it isn't perfect, we can always add a separate timing wheel to the shaft. We are eager to "just build it"!

Thanks,
Scott

Edit: I have added a picture which includes two more diameters for the timing magnets. The picture shows the diameters for the timing magnets and their corresponding "dwell arc" distance between each magnet (all measurements are in inches). Obviously, I don't want to lose any structural integrity on the wheel, so I would prefer not to machine all of these. Based on the picture, is there any diameter which could be preferable to another?

[John_K]

Thank you John
Somehow I missed those
Would the MJL21194 &93 work for the circuit instead of the MJE's?
Brian

Hi Brian,

Yes, I am using those for the main transistors. I'm using an MJE2955 for the transistor that the hall triggers.


John K.

[John_K]

Hello John K

What is your trigger transistor, couldn't here you in the video sounded like 2N2055. I've been using the MPSA06, is there much difference?

Thanks, Mark

Hi Mark,

It is an MJE2955T, probably overkill but what I had on hand.


John K.

[ren]

Hi Folks,

I have my setup running on a hall triggered Bedini/Cole circuit now as well as cap pulsing.

It's running well but I didn't get time to buy some zeners for the SCR today, so I'm using 4 x 1N4004's in series with the 2 x 2.8V super bright LEDs.

It is going way too fast, about 50 RPM so I think I need to reduce the dwell time on the halls some more. (Thanks again to Ren for the halls!)

It's charging the heck out of the secondary battery and the run battery also appears to be gaining a bit. But I've been tricked by these crap gel-cells before so I'm not reading too much into that.

Here's a short (low quality) YouTube video YouTube - Bedini S1GT

Thanks John B for all your tips, it works just like you said it would


John K.

Nice one John. Glad that hall worked for you.

Ill give you a call sometime this weekend and you can fill me in. Lightning took out my phone line at home, so I have no net or phone till next week sometime, and there is no youtube at work, where I am writing this.

Good to see you back too Steve