Bedini Ferris Wheel Regauging Motor

[txaggie00]

Schematic and Observation

@All,

Just wanted to post the current schematic that we are using for our circuitry. Also wanted to note that the charging batteries have a huge impact on how well the machine does charge. We replaced one of the charging batteries with one that either has a bad cell or has some sulfation to see how it would act. The cap would not charge at all b/c the SCR would leak current continuously. I don't think we will see the true potential of this machine until we get 6 really good batteries. I'm not sure if the SIDAC circuit will respond the same way though, since it is either on or off and shouldn't leak current at all.

Thanks,
Scott

EDIT: This is the schematic we are using in the previous video. Bedini 2GT "Junior" Update video

EDIT 2: The earlier schematic showed two windings on the slave coil (copy and pasted from the main coil and forgot to delete all the extra windings), so I updated it to show the one winding so as to not confuse anyone!

[txaggie00]

Why we need a better SSR

Hey everyone!

Sorry for all the posts. Even though we aren't %100 done with our wheel, we learn a bit more about it everyday. Yesterday we put a bad battery (8.5V) in with the charging batteries to see how it would act and we noticed that the SCR would constantly leak and wouldn't shut completely off. Even with the good batteries, the SCR would still leak a little bit and I think that is why JB said you needed really good batteries with no sulfation. I was thinking about it last night, and our wheel is charging our cap up real good and I couldn't see why the cap dump circuit should have ANYTHING to do with this machine. The end result of this machine should be to charge that cap up as quick as possible with the spikes it's capturing. The cap dump and charging batteries should be independent of this machine and not another variable, IMHO. Someone correct me if I am way off base here. I think that is why JB mentioned finding a better Solid State Relay for the switching. Anyway, we shot a little video to show you guys what I am talking about. I don't know about ya'll, but I am a visual learner and seeing things vs. reading about it makes it much easy to understand.

Bedini 3GT "Junior" - The reason for a better SSR

Thanks,
Scott

[Les_K]

@All,

Just wanted to post the current schematic that we are using for our circuitry. Also wanted to note that the charging batteries have a huge impact on how well the machine does charge. We replaced one of the charging batteries with one that either has a bad cell or has some sulfation to see how it would act. The cap would not charge at all b/c the SCR would leak current continuously. I don't think we will see the true potential of this machine until we get 6 really good batteries. I'm not sure if the SIDAC circuit will respond the same way though, since it is either on or off and shouldn't leak current at all.

Thanks,
Scott

EDIT: This is the schematic we are using in the previous video. Bedini 2GT "Junior" Update video

EDIT 2: The earlier schematic showed two windings on the slave coil (copy and pasted from the main coil and forgot to delete all the extra windings), so I updated it to show the one winding so as to not confuse anyone!

Hi Scott,
Great stuff. I am still putting my circuit in a box. should be done today.
But just wanted to mention that I when I took the bridge out and used the STPSH100D that John K. showed in his schematic, my cap dump made significant improvements. Take a look and notice the polarity of the cap in relation to the circuit. Seems to me it should have been the other way around as the bridge is allowing charge during the pulse itself. But this way the power from the input battery is limited to the coils and only the calapse returns a charge to the battery. just a thought.

Just another note; I had put a 30uf 250 volt cap on the output and it charged to 250v in two pulses. So I know I am getting at least a 250 volt spike. Pretty certain they would have blown had I let it run and not stopped it then. If I had Higher voltage caps I am sure I would be able to see just how high my spike voltage is by the voltage the cap charges to.

Les

[txaggie00]

Hi Scott,
Great stuff. I am still putting my circuit in a box. should be done today.
But just wanted to mention that I when I took the bridge out and used the STPSH100D that John K. showed in his schematic, my cap dump made significant improvements. Take a look and notice the polarity of the cap in relation to the circuit. Seems to me it should have been the other way around as the bridge is allowing charge during the pulse itself. But this way the power from the input battery is limited to the coils and only the calapse returns a charge to the battery. just a thought.

Hey Les! You'll have to post a picture of your circuit after you put it in the box so we can all see it!

We had the Shottky diode in our circuit for a while, but didn't notice a difference with it. I think the biggest issue right now for us is the SCR and/or batteries. Until we get new batteries or we get the sidac, opto and FET circuit on there, I feel like we are going to be chasing our tail trying to get this thing timed right. We'll definitely try it again though. Maybe once we get the cap dump working great, we will notice an improvement with the other parts of the circuit like the Shottky.

Thanks for your suggestions and comments! We really appreciate everyone helping us out!

Thanks,
Scott

[txaggie00]

Sacrificing MJL's

After accumulating a fairly large graveyard of MJL's, I would strongly suggest that you have either a battery or large cap to absorb the spike.

@John K, That looks better than the tranny that I blew up the other day. Not sure if you can see it in the more recent videos, but if you take a look where the hall circuit is you can see the burn marks left from the tranny catching fire. When I added the MJE after the MPS, I forgot to put the 220R and 470R resistor in and just went straight to the base of the 93's from the MJE emitter. It was a brilliant blue flame. Wish we had it on video! You couldn't tell it was a transistor when it was done!

Did I read this right? I have always tested my circuits with the charge
batteries connected. I have tested without the cap, but only for a short
time as I dont know where all that energy will go if not thru the cap.
I try not to take un-necessary risks when testing, or I would be replacing
trannies all the time.

@Mark: Yeah, you read it right. I was thinking about sacrificing some extra 93's and 94's just to be able to see the spike. After what you and John K are saying, we probably won't do that, even though I am really curious to see the spike coming off those coils! We'd have to pick up a 100:1 scope probe I'm sure. Of course right now, we are just trying to get this thing working right.

If either of you have a good scope shot of your machine's waveform, we'd love to see it.

Thanks,
Scott

[John_K]

Big Bedini Machine (3GT)

Hi Scott,

Yeah, that tranny I posted a picture of caught on fire too. It was on my SS SG at the time. The board was blackened pretty bad too.

It sounds like you are making progress.

Les_K is right, note the polarity of the cap on the output. This is pretty much SSG 101 theory. If you look at the circuit, it also places the cap in series with the primary battery, and yes some of the energy finds its way back to the primary battery in this arrangement.

Another trick you can try from SSG 101 theory is to place a large cap over the primary battery, as close to the trannys as possible. I would put it over the two emitters of the '93 and '94. This will fix the impedance problem between the primary and the coils.

Remember that the batteries are half of the system. Your tests have shown this as well. IMO, you can use bad batteries but it will take a while to re-condition them - as long as they are not too bad that the SCR won't fire. SCRs get tricky when working with large capacitors.

The STPS8H100D may not be the ideal part there as it is only rated at 100V, it was what I had on hand. I think a 1N4007 would work better, rated at 1A @ 1000V. You could even parallel a few of them that are matched to improve the switching speed. (Branch currents - Kron)

I'm just finishing up another project. As soon as it's done I'll get those coils wound and get mine up and running again. I think one of my YT videos shows a good scope shot (koornj channel)

Another thing is that you will want to start trying to get some long runs in. John B says to "walk away". You should start to notice that some strange things will start to happen after a while. For example, run the machine for around 24 hours or so and then switch it off and empty the cap with a 120V bulb or something. Let it sit for a while and then measure the voltage on the cap again. Notice anything strange?


John K.

[John_K]

Good test, better than blowing transistors!

Hi Scott,

Just another note; I had put a 30uf 250 volt cap on the output and it charged to 250v in two pulses. So I know I am getting at least a 250 volt spike. Pretty certain they would have blown had I let it run and not stopped it then. If I had Higher voltage caps I am sure I would be able to see just how high my spike voltage is by the voltage the cap charges to.

Les

Hi Les,

That's a good test, better than the sacrificial transistor test

Photo flash caps would work well, I have a 80uF 350V one I pulled out of a disposable film camera (remember them LOL). You can pick up the whole camera for under $10. To buy just the cap by itself is about $30! These will charge up really fast so if testing just turn the wheel by hand and then catch just one spike in the cap and measure the voltage over the cap.

Hope it helps...


John K.

[Web000x]

Hi Les,

That's a good test, better than the sacrificial transistor test

Photo flash caps would work well, I have a 80uF 350V one I pulled out of a disposable film camera (remember them LOL). You can pick up the whole camera for under $10. To buy just the cap by itself is about $30! These will charge up really fast so if testing just turn the wheel by hand and then catch just one spike in the cap and measure the voltage over the cap.

Hope it helps...


John K.

I had though about suggesting this technique but started to think about the downfall.

You would not be able to pulse the capacitor with just one spike from the coil to see the true voltage unless it is a low capacitance/high voltage capacitor. This starts to present the problem as I perceive it.

If you are trying to look at the voltage in a cap, you must let the spike fill up the cap with as many pulses that it takes before the voltage stops rising. That voltage would then be equal to the magnitude of the spike.

When the cap is filled up to the level of the spike, there is no where else for the spike to go so it starts dropping across the transistor again. This is the same thing as having no capacitor connected.

If you will notice the charging waveforms of most of the replications posted, you will see the "coil on" voltage in the positive direction. Immediately after the coil shuts off, you see the dip in voltage where the capacitor is filling up. As the capacitor increases in voltage, notice that negative voltage dip (Spike trying to manifest) increases in magnitude until the cap dumps and this process is repeated. If there was no SCR/battery to dump this potential into, it would continue filling the the voltage level of the spike.

If you try to view the voltage level from the spike by filling a cap, you will ultimately be letting the full potential of the spike become developed across the collector/emitters of the transistors thus possibly killing transistors.

Good luck,

Dave

[txaggie00]

Sorry Les

Les_K is right, note the polarity of the cap on the output. This is pretty much SSG 101 theory. If you look at the circuit, it also places the cap in series with the primary battery, and yes some of the energy finds its way back to the primary battery in this arrangement.

Alright, I think I am just now comprehending what it was Les was saying. Sorry for my ignorance Les. Let's see if I actually do understand though. You guys are saying that some of the issues we are seeing with the cap dump circuit and, in essence the SCR, is because of the way the circuit is configured and because we are using the bridge. With this setup, like John K said, some of the energy is finding it's way back to the primary (through the negative leg of the bridge) and that small current could be the cause of the SCR not shutting off. If we were to use the Shottky or paralleled IN4007's, then that energy could not make it's way back and the SCR shouldn't have a problem shutting off because the current would have no where to go. Are we on the same page or even in the same book? Just so you guys know, I'm not an analog guy, I'm a software guy, pretending to understand stuff outside of 1's and 0's logic (voltage and current tend to be a wicked curve ball for me). But I'm trying ! The transistors seem to make sense to me because I think of them as logic gates, but the SCR and cap dump circuit, I'm still trying to wrap my head around.

As far as the waveform goes, I'll look at your channel John, and compare with what you have shown. From what you have seen of ours, do we look like we are on the right track?

As a side note, we got the rest of the rods in today and got them all cut, but we are still about a 1/2 lb short. Brent is going to pick some more up tomorrow and we should have all the coils on by Monday!

Thanks again everyone!
Scott

[John_K]

Alright, I think I am just now comprehending what it was Les was saying. Sorry for my ignorance Les. Let's see if I actually do understand though. You guys are saying that some of the issues we are seeing with the cap dump circuit and, in essence the SCR, is because of the way the circuit is configured and because we are using the bridge. With this setup, like John K said, some of the energy is finding it's way back to the primary (through the negative leg of the bridge) and that small current could be the cause of the SCR not shutting off. If we were to use the Shottky or paralleled IN4007's, then that energy could not make it's way back and the SCR shouldn't have a problem shutting off because the current would have no where to go. Are we on the same page or even in the same book? Just so you guys know, I'm not an analog guy, I'm a software guy, pretending to understand stuff outside of 1's and 0's logic (voltage and current tend to be a wicked curve ball for me). But I'm trying ! The transistors seem to make sense to me because I think of them as logic gates, but the SCR and cap dump circuit, I'm still trying to wrap my head around.

As far as the waveform goes, I'll look at your channel John, and compare with what you have shown. From what you have seen of ours, do we look like we are on the right track?

As a side note, we got the rest of the rods in today and got them all cut, but we are still about a 1/2 lb short. Brent is going to pick some more up tomorrow and we should have all the coils on by Monday!

Thanks again everyone!
Scott

Hi Scott,

Are you still using the 36V zener on the cap dump? John B said he was dumping at 45.5V if I remember correctly. He also recommended a 4.7V zener. I think the 36V zener is too much to ask the SCR to dump at with such a big capacitor, which would explain the problems you have.
Just think of the zener as another switch, where the switch closes when the gate threshold voltage is reached (the zener + the LED + the LED) and opens automatically when the cap and the battery are at the same voltage.

Looking at your last video, the SCR is not "opening the gate" again after it dumps - called "latching closed". I'm sure if you swap your 36V zener for a 4.7V (or thereabouts) you should have it working fine.

As for your waveform, it looks fine. I'd be pretty happy with that.

I have the same problem as you guys - not enough welding rods. So I'll be ordering some more soon.


John K.

[txaggie00]

That's a good test, better than the sacrificial transistor test

Now that I have removed my foot from my mouth, we'll have to pick up some caps to test this on. Thanks for the more appropriate suggestion! I think I remember seeing a disposable camera at a local antique shop, so we'll pick one up if we don't have a cap like that lying around our office.

We are eager to see your build progress John K. You are trying to match JB's coils, correct?

Thanks,
Scott

[Les_K]

Got it all running again but had some problems... I connected the Collector of the mj2955 to the collector of the MJL21194. Amazing I didn't fry something!
Works great now. Better on bigger batteries for sure. I drained my little walmart u7 tractor batteries with a 40 watt bulb. then charged them. The first one was charged in about 15 minutes and the other in about 40 minutes. These are not deep cycle so I have no idea of the C-20 on em' but wham!

Ok, Scott;
What I noticed about John K's setup is what I read in the book John bedini was signing at the conference(which I sadly missed). page 16 of FEG. If you have a copy, read the last paragraph on that page about what is not in the text books. Very interesting.

Now to some tougher questions for everyone. I was reviewing my notes (I compiled what JB said here and with the Video 22) and discovered something I had not noticed before. He said that the upper section was run off the Bedin/Cole circuit. That it was pulsed twice! Then he said the Axial hubs are in parallel with the bottom coils but individually they are in series.

I also Had this in my notes but not sure how it fits?
Two short pulses (outside coils) and one long one to drive it around(central coil)

So As I understand it, should we not be ringing these coils twice?
And do we somehow pulse the outside coils twice then one long pulse to the center coil?

Oh, Scott I will try to get a picture or youtube tomarrow. Busy day today.

Dave, The test is just a rough it in sort of thing, so as to not harm the transistors. Just a quick and simple test to show there is a whole lot more than we think going on.
Les

[txaggie00]

Hi Scott,

Are you still using the 36V zener on the cap dump? John B said he was dumping at 45.5V if I remember correctly. He also recommended a 4.7V zener. I think the 36V zener is too much to ask the SCR to dump at with such a big capacitor, which would explain the problems you have.
Just think of the zener as another switch, where the switch closes when the gate threshold voltage is reached (the zener + the LED + the LED) and opens automatically when the cap and the battery are at the same voltage.

Looking at your last video, the SCR is not "opening the gate" again after it dumps - called "latching closed". I'm sure if you swap your 36V zener for a 4.7V (or thereabouts) you should have it working fine.

As for your waveform, it looks fine. I'd be pretty happy with that.

I have the same problem as you guys - not enough welding rods. So I'll be ordering some more soon.


John K.

Hi John,

I think in the last video we were using a 33V zener. The smallest one we have is a 5.1V zener. It's still on the board, so we will plug that back in. We were having the same latching issue with the 5.1V zener though.

We ended up needing ~65 lbs of welding rods. We went through a 50 lb box, a 12 lb box, a 1 lb tube we had on hand and we are picking up another 2 lbs tomorrow. That's the kind of stuff that delays a build. It seems we shorted all of our materials (except the wire) and end up waiting twice as long to complete one aspect of the build at a time. We have kept really good documentation of our build including the wood, magnets, wire, reels, aluminum spokes, hardware and the pricing for all of it in case someone wants to replicate it. I have even wrote an instruction pdf of how we built our wheel if anyone was interested.

Good luck with the winding!!

Thanks,
Scott

[Web000x]

As far as the waveform goes, I'll look at your channel John, and compare with what you have shown. From what you have seen of ours, do we look like we are on the right track?

I have posted my waveform before, but this waveform is slightly different because I have adjusted my coils significantly.

If you look at your coil waveform right after you turn the power off to the device (scope lead across collectors of transistors), you will see this in the first picture.

This scope shot shows the workings of the coil with welding rods as their core. The magnet comes in to the coil's core and causes a magnetic flux that induces a negative voltage potential (in comparison with zero or "ground") over the coil leads. The magnet will begin to move further away from the coil's core and the polarity of the coil will begin to shift directions. It shifts so quickly that it causes an almost equal voltage potential in the opposite direction. This is why there is a positive voltage as well on the previous scope shot.

So it seems to me that your magnet comes into the coil and is passing the coil's core about the time that you are firing it. It is a little bit earlier than mine.

In a recent video, you spoke about your curiosity of why the waveform was curving upwards before the hall fires. It is just the magnetic flux inducing potentials manifesting on the scope.

You may want to experiment with an super-bright LED and a resistor hooked up to the output of the transistors to see your exact timing. Just point that LED at some white tape that denotes your exact magnet spacing on the perimeter of the wheel. I think it is good for 'perfect' coil positioning.

My apologies about the quality of the photos. Cameras don't really pick up on CRTs at slow frequencies. I can make a movie if this is confusing.

Dave

[Les_K]

Hey Les! You'll have to post a picture of your circuit after you put it in the box so we can all see it!

We had the Shottky diode in our circuit for a while, but didn't notice a difference with it. I think the biggest issue right now for us is the SCR and/or batteries. Until we get new batteries or we get the sidac, opto and FET circuit on there, I feel like we are going to be chasing our tail trying to get this thing timed right. We'll definitely try it again though. Maybe once we get the cap dump working great, we will notice an improvement with the other parts of the circuit like the Shottky.

Thanks for your suggestions and comments! We really appreciate everyone helping us out!

Thanks,
Scott

Here is a quick shot of the unit. It is now running about 40+ RPM.
No idea what changed. I don't think it has sound so what I am showing is the hall sensor mounting it has a screw that allows virticle adjustment, and slides on the frame beck and forth, and the hub can me moved in and out to set the gap.
On the floor are some batteries but one was sulfated so It is not in the circuit. I showed the amp meter on the input but you can't see it because of some reflection. but it is bouncing higher and lower as the thicker magnets rotate by the hall. And the box is an older burned out inverter box.
YouTube - HPIM2126


Les