Announcement

**thedude** · 04-03-2016, 01:43 AM

Useful schematics

The schematic that i found helpful with regard to setting up an opto-microcontroller and 2N3440 to my SCR cap dump...

And two very artistically done that illustrate a similar arrangement with different triggering mechanisms...

**BroMikey** · 04-03-2016, 08:28 AM

Dude

That is so beautiful. Your post on transistor and diode details
is exactly what I first used and they got hot in GENMODE.

When I went from MJL21194's up to the brother MJL4281
and threw my cheap azz diodes out for some fast ones
that I mounted on separate heat sinks I doubled my
output. Think about that statement. Same input and
because I did not waste energy in the form of heat
my output went way up.

The diodes act as resistors when they get hot. I know this
is going to sound ridiculous but I used 30amp sinkable fast
recovery diodes on a single 14 awg strand, needless to say
with zero heat at any amp draw.

You want a friction free system. The R-charge is a quick
and dirty entry level teaser that gets you to explore.

I learned this the hard way also. I smoked my diodes in
GENMODE at about 1.5 amps per strand using 10 amp diodes
but they were not high freq diodes. 60hz is what some
diodes are rated for and anything higher the power
handling drops off.

Even at 300-500hz a diode may only be good for 1/20th
of it's ratings. I don't know how hard it would be to add
in some sets of diodes and sinks.

But all in all your set up is stunning.

Even my MJL21194's got hot over nothing when I went
from 12volts to 24volts. I learned that a 30volt dc output
using a 28vdc source battery produced a much higher
peak voltage across the transistor that bordered it's
max capability.

The industrial standards handbook says an MJL21194
operating at 70-80 volts is at it's peak and this is why
it has a 3X or 200volt max. Remember when your transistor
is fed 28vdc it is toggling at a much higher peak say 3X That.

And those spikes are gonna eat those junctions for lunch
in a short ime at only slightly elevated power outputs. Then
you will be soldering again, like me.

But the brother to it is the MJL4281 and has a 350 volt
max, so you see that transistor is at it's max handling range.

The MJL4281 does cost more but not that much unless you
are in a business to make money and are cutting costs
every way possible.

**thedude** · 04-03-2016, 04:00 PM

Originally posted by BroMikey View Post

Dude

That is so beautiful. Your post on transistor and diode details
is exactly what I first used and they got hot in GENMODE.

When I went from MJL21194's up to the brother MJL4281
and threw my cheap azz diodes out for some fast ones
that I mounted on separate heat sinks I doubled my
output. Think about that statement. Same input and
because I did not waste energy in the form of heat
my output went way up.

The diodes act as resistors when they get hot. I know this
is going to sound ridiculous but I used 30amp sinkable fast
recovery diodes on a single 14 awg strand, needless to say
with zero heat at any amp draw.

You want a friction free system. The R-charge is a quick
and dirty entry level teaser that gets you to explore.

I learned this the hard way also. I smoked my diodes in
GENMODE at about 1.5 amps per strand using 10 amp diodes
but they were not high freq diodes. 60hz is what some
diodes are rated for and anything higher the power
handling drops off.

Even at 300-500hz a diode may only be good for 1/20th
of it's ratings. I don't know how hard it would be to add
in some sets of diodes and sinks.

But all in all your set up is stunning.

Even my MJL21194's got hot over nothing when I went
from 12volts to 24volts. I learned that a 30volt dc output
using a 28vdc source battery produced a much higher
peak voltage across the transistor that bordered it's
max capability.

The industrial standards handbook says an MJL21194
operating at 70-80 volts is at it's peak and this is why
it has a 3X or 200volt max. Remember when your transistor
is fed 28vdc it is toggling at a much higher peak say 3X That.

And those spikes are gonna eat those junctions for lunch
in a short ime at only slightly elevated power outputs. Then
you will be soldering again, like me.

But the brother to it is the MJL4281 and has a 350 volt
max, so you see that transistor is at it's max handling range.

The MJL4281 does cost more but not that much unless you
are in a business to make money and are cutting costs
every way possible.

BroMikey, thank you so much! Your posts have been exactly the sort of information that i needed where i'm at right now. I'm also reading your ReGenX coil thread carefully and I'm very interested in working to cancel the Lenz effect in that side of the motor/generator. I realize this will take some careful planning, but i'm up to the challenge. These are the important next steps that i need to make IMO. I'm beginning preparations for a full tear down of my system for generator coils that will be of the right composition and impedance for the output my motor is capable of. The switch to MJL4281 transistors and some diodes capable of higher frequency feels incredibly important to me, and is something i can do in short order. Yes, all that heat seemed illogical somehow. I have ideas for making the whole system much easier to maintain in the future as well. I'm learning a lot about what will make the system more functional on a practical level and there are a lot of changes that need to be addressed. But by far, i agree that a rework of the passive components is paramount. I don't need the heat, i have an oven on the other side of the room. :P I spend a lot of time trying to conceptualize the movement of current in the system and its refreshing to hear that my instincts were in the right place. The diodes were always first to heat up and the hottest component on the board, which felt like a bottleneck in the system to me. With these alterations I'm quite hopeful that i will see performance benefits. While money is not unlimited, I'm on a mission to build the best running motor/generator that i possibly can. Keeping every thing open source is important, but only if i'm doing every thing in the finest way possible. No use in sharing bad information. If there is better composition that will improve performance i'm game!

I'm actually excited to have a clear direction to proceed now. I felt a bit like i'd stalled out on the level of progress and i have you to thank Bro!
Today i removed and replaced my diode array that was acting as a bridge rectifier for my gen output with a proper 1000v bridge rectifier. Small change with no apparent benefit. Not much i can do until i can afford some passive components. I'm sorta laid off of work atm so i'll be a bit getting them in. This business of being an electronics enthusiast is a catch 22. When ever i have the time, i seem to lack the money. When i have the money, i seem to lack the time. lol But i'll get the parts.

I mentioned that i'd ran some tests with my tesla pancake coil acting as the capacitor bank on the radiant output. My output leaped over 31-34 volts on the secondary banks and i had to increase resistance on the base of my gate transistors to turn off my neon bulbs while the system was running! Suffice it to say, that was a test that i could not run for very long. I found a way of increasing amps dramatically, but heat threatened to destroy everything. I'll be ordering spares of everything and make sure to have camera running for future experimental setups.

I'm pumped! Thanks again BroMikey!

**Donald Haas** · 04-05-2016, 05:05 AM

opto trigger circuit

Hi The Dude and all,
I have an 8 coil DP running with 8 gen coils.
Been trying to have good understanding of Opto trigger cct.

I wired mine to fire when the blade is in the slot.
40% on time.
I was not getting enough amp draw and changed the individual tranny resistors to 1ohm and was stilll not getting plenty of amp draw at 12v or 24v.

After studying the schematics from Rick F, it seemed that he wired the sensor to the HD113 wrong. I redid mine to agree with his except this apparent error.
I gained about 30% in amp draw but not what it seemed that I should have.

I studied his some more and could see that his should function, so I bread boarded his. His causes the circuit to fire when the blade is out of the slot so I had 60% on time and abundant amp draw. I had a 35amp fuse in place so I stayed under that at 12v but with my 1ohm tranny R's and 60% on time I couldn't get a scope shot that showed a full H wave without going into the "oh no zone."

Now please allow me a few "silly questions",

Do I have Rick's trigger circuit correctly in place with it firing when the blade is not in the slot.

What is the duty cycle of the timing wheel (blade) that he ships.

Thank you all for being here,
My pics are at:

https://www.dropbox.com/sh/l64umdoyq...nL8R3tLva?dl=0

bro d

**Donald Haas** · 04-05-2016, 05:16 AM

DP?

Quote:
"That is so beautiful. Your post on transistor and diode details
is exactly what I first used and they got hot in GENMODE."

Hi Bro Mikey,
Are you talking about a Dual Pole build?
One with strong neo's on both ends of the run coils which are present in the two rotors?

Thanks,
Bro d (your brother from a different mother)

**thedude** · 04-05-2016, 07:42 AM

Originally posted by Donald Haas View Post

Hi The Dude and all,
I have an 8 coil DP running with 8 gen coils.
Been trying to have good understanding of Opto trigger cct.

I wired mine to fire when the blade is in the slot.
40% on time.
I was not getting enough amp draw and changed the individual tranny resistors to 1ohm and was stilll not getting plenty of amp draw at 12v or 24v.

After studying the schematics from Rick F, it seemed that he wired the sensor to the HD113 wrong. I redid mine to agree with his except this apparent error.
I gained about 30% in amp draw but not what it seemed that I should have.

I studied his some more and could see that his should function, so I bread boarded his. His causes the circuit to fire when the blade is out of the slot so I had 60% on time and abundant amp draw. I had a 35amp fuse in place so I stayed under that at 12v but with my 1ohm tranny R's and 60% on time I couldn't get a scope shot that showed a full H wave without going into the "oh no zone."

Now please allow me a few "silly questions",

Do I have Rick's trigger circuit correctly in place with it firing when the blade is not in the slot.

What is the duty cycle of the timing wheel (blade) that he ships.

Thank you all for being here,
My pics are at:

https://www.dropbox.com/sh/l64umdoyq...nL8R3tLva?dl=0

bro d

Hi Donald, your replication looks awesome! Good work on those boards. You'll be running top speed in no time i'm certain.
I took a picture of the timing wheel on my motor from r-charge, the best that i could. I never measured its gauge, but i would estimate that it has a 1/3rd on time gap on it. Off for 2/3rds. By estimation.

Here is close up of your previous schematic.

and one of the trigger from R-charge

I'm pretty sure you've discovered this already, but your (top) above schematic has positive current running to both the collector and emitter of the H11D3. I may be incorrect, but I don't see that the emitter pin 4 on the H11D3 should go any where but the base of the trigger NPN transistor in order to get it to fire to your gate transistor. In the r-charge schematic, the opto emitter will operate when the opening in the timing wheel is present. So the duty cycle of the circuit is determined by the ratio of opening to blocked portion of the timing wheel appropriately. On when open, sounds like you had that figured.

I use this same relationship to fire to the gate of my SCR for my cap dump circuit with only one gap on my other timing wheel. Effectively dumping my capacitors once per full rotor revolution or 8 motor cycles (8 magnets on my rotor).

In aligning the timing wheel opening with your magnets on the rotor, choose the direction of rotation that you wish to operate in, then have the opening in the timing wheel occur as close to 1/4" to 1/2" past top dead center (magnet to coil relationship). Set it up there and then test, adjust slightly in either direction and test again. I found that much trial and error was needed. When i thought i was getting a strong "H" wave profile and higher amperage and what i thought was good rpm, i still wasn't where i needed to be. Additional adjustments were needed to find that sweet spot. Its remarkable how much performance can be gained by only the slightest adjustment. Important to have a lot of patience in tuning the timing wheel.

Also it is good to consider that the running performance of the motor does not change as much with varying the the resistance on the base of the gate transistor, when using a timing opto circuit and not a coil trigger. By this i mean that efficiency does not appear to be affected in the same way as it does when using a trigger coil to fire your power transistors. While the running RPM will obviously go up and down (using opto) with the base resistance, the efficiency profile is much more of a linear relationship.

Great to have you posting here Donald. I'll do the best i can to help. Please keep in mind that i am a layman grade dude. :P Hopefully we'll continue to have others contribute.

**BroMikey** · 04-06-2016, 08:40 AM

Originally posted by Donald Haas View Post

Quote:
"That is so beautiful. Your post on transistor and diode details
is exactly what I first used and they got hot in GENMODE."

Hi Bro Mikey,
Are you talking about a Dual Pole build?
One with strong neo's on both ends of the run coils which are present in the two rotors?

Thanks,
Bro d (your brother from a different mother)

There you all are, the internet was off here til just now.

Thank you kindly. My comments were centered on devices. I have yet
to build a hall effects triggering setup but THE DUDE came thru
for us all.

Even tho I have not built these forced type triggering circuits
I have speculated this as compared to JOHN B. basic SG dia.

Forced switching like with a hall effects must not ignore a coils
happy spot, if that is done a much lower duty cycle should be
used to prevent heating.

In other ways of looking at this we could use the teeter todder
analogy where as long as the persons on either side work together
the process glides along smoothly without using excess energy.

This is the coils happy spot.

I don't care what the triggering method is that a person uses.

Many times a hall effects system is used to give the operator
more control say down at the lower RPM's range and on up
to max. However we can not ignore the fact that the coils
want to operate at a couple of bands of fundamental and
harmonic frequencies where conflicts are diminished.

There are several rpm ranges in any system that are
perfectly tuned to give excellent results and the other
frequency ranges just a few harmonics over will cause
elevated stress.

This is heat generated by harmonic conflict.

Remember those coils still want to be set at their happy
spot or you are going to get wasted energy build up in the
devices.

We can take the analogy off the playground and apply it to
common sense solutions as simple as they may seem.

**thedude** · 04-06-2016, 10:12 AM

Originally posted by BroMikey View Post

There are several rpm ranges in any system that are
perfectly tuned to give excellent results and the other
frequency ranges just a few harmonics over will cause
elevated stress.

This is heat generated by harmonic conflict.

Remember those coils still want to be set at their happy
spot or you are going to get wasted energy build up in the
devices.

We can take the analogy off the playground and apply it to
common sense solutions as simple as they may seem.

Hey Bro!
Yeah, the forum was not talking to the database properly there for a bit. Good to see it back.

Your absolutely correct about bad alignment (inharmonious) generating heat. I found many adjustments of the opto trigger were horrible and would run with massive overheating. Or even just spinning the rotor in the opposite direction in its current setup will do the same for me.

I ordered some MUR8100EG diodes. They were the fastest that i could afford (in stock). other specs were comparable or better than 1N5408. i'm not certain i'll be able to heat sink them them very easily with my arrangement. I'm considering leaving as much of the legs of the old diodes attached to my board and installing double pin binding terminals to eliminate soldering in the future. I couldn't afford the MJL4281 upgrade just yet. Wish i could though. I'm hoping the new diodes will land by Friday and ill have something to work on over the weekend.

I have a Flir Thermovision Short wave infra-red camera and i've been itching to break it out for a few shots. So i'll make a video of the before temps and then do a comparison in my upcoming video after the new diodes are in. I'm pretty confident that we'll see a temperature decrease, even if there isn't a major change in efficiency. The transistors will likely be a contributor on the efficiency factor, more so than the diodes. Even if its just temperatures that improve, i'll be very happy and it will open to door to future experimentation.

I'm considering taking out the existing 18 gauge gen coils ( of varying impedance) , which are hardly contributing to the 12 volt battery bank in their present state and rewinding a flat alternator like a windmill generator uses that i can mount to a frame on the outside of the back of the motor. In that fashion i can have easy access to them, and make use of all that 18 gauge wire that i paid good money for. :P I'm not happy with what they are producing in their present state and feels like unnecessary drag on the motor with the core material that they have in them. Doing more harm than good IMO. Even just a store bought windmill PMA would produce some major current off of all that kinetic energy that could loop back into the primary to improve the COP equation. I'm pretty confident that it can handle quite a bit of counter torque efficiently.

@ Donald. In looking, and trying to recall the timing wheel, i've remembered that the opto sensor light mount just barely fit between the opening in the fly wheel. Was something i realized when i was making fine adjustments. That might give you a better idea of the size of the gap albeit without the actual radius of the wheel itself, you'll still have to make an approximation.

I'm going on a bit long here so i'll post again when i have something more to report.

Thanks again!

**Donald Haas** · 04-06-2016, 02:40 PM

[QUOTE=thedude;287420]Hi Donald, your replication looks awesome! Good work on those boards. You'll be running top speed in no time i'm certain.
I took a picture of the timing wheel on my motor from r-charge, the best that i could. I never measured its gauge, but i would estimate that it has a 1/3rd on time gap on it. Off for 2/3rds. By estimation.

Here is close up of your previous schematic.

I'm pretty sure you've discovered this already, but your (top) above schematic has positive current running to both the collector and emitter of the H11D3. I may be incorrect, but I don't see that the emitter pin 4 on the H11D3 should go any where but the base of the trigger NPN transistor in order to get it to fire to your gate transistor. In the r-charge schematic, the opto emitter will operate when the opening in the timing wheel is present. So the duty cycle of the circuit is determined by the ratio of opening to blocked portion of the timing wheel appropriately. On when open, sounds like you had that figured.

Hi Darcy,
Glad we are back on line.
In the trigger circuit that I posted there is no H11D3.
I have an LED to indicate power present.
When I started with Opto switching, I used IR diodes and the schematic looks the same as an opto sensor cct. I learned to wire it the way my posted circuit is wired. On my DP this didn't get it. With Rick's method it appears that the LED in the H11D3 and the Transistor in the opto sensor are in series.
and this bumps it up substantially.
In my posted circuit when the blade interrupts the gap it fires which is opposite to Rick's. Don't understand that yet.
I bread boarded Rick's version and have to wait for resistors of the recommended size. His version allows abundant amps.
Thank you for the much needed info regarding Rick's circuit and timing wheel. Thank you also for the encouraging words.

On another note it seems that all roads lead to RF engineering.
I'm looking into it. I prefer not to post rambling speculative thinking.

Keepin' on,

bro d

**thedude** · 04-06-2016, 04:44 PM

Originally posted by Donald Haas View Post

Hi Darcy,
Glad we are back on line.
In the trigger circuit that I posted there is no H11D3.
I have an LED to indicate power present.
When I started with Opto switching, I used IR diodes and the schematic looks the same as an opto sensor cct. I learned to wire it the way my posted circuit is wired. On my DP this didn't get it. With Rick's method it appears that the LED in the H11D3 and the Transistor in the opto sensor are in series.
and this bumps it up substantially.
In my posted circuit when the blade interrupts the gap it fires which is opposite to Rick's. Don't understand that yet.
I bread boarded Rick's version and have to wait for resistors of the recommended size. His version allows abundant amps.
Thank you for the much needed info regarding Rick's circuit and timing wheel. Thank you also for the encouraging words.

On another note it seems that all roads lead to RF engineering.
I'm looking into it. I prefer not to post rambling speculative thinking.

Keepin' on,

bro d

@ Donald
Hi Donald. Yeah, was worried we were losing the forum there. :P You definitely welcome and forgive my confusion on your schematic (i did miss read the polarity at your X1 as well). Ah, i understand now. I thought you had H11D3 in operation. I do recommend the R-charge schematic Rick Friedrich provided if you can, which sounds like the direction your already heading. haven't varied from it at all in my set up and it really works great. Has not given me any trouble and is built exactly as described in schematic. However, i do completely understand working with what you have on hand, as has been my situation many times in the past. I just want to be sure you don't feel i'm having any trouble with RF's trigger arrangement at all. My temp issues are not related.

@ All
On a side note, the temperature issue i'm resolving in my build has to do with using "Generator" mode which bonds the Negatives of primary and secondary via large blocking diode in conjunction with cap dumping setup (not to be confused with generator coils) and pushes my amps up over 30amps @ 24+voltages on the input (over 720 watts). I'm actually able to push the amps up even higher with experimental Tesla pancake coil (1.5" copper ribbon) in place of my discharge caps. In my build, leaving the motor running in "Generator" mode raises the temps up to 60c+ on my 1N5408 diodes and up to 40c+ on my MJL21194. Which is why i'm switching to fast switching diodes MUR8100EG and MJL4281 transistors. In regular operation mode, my temps are less than nominal (45c diode and 35c transistors) and not much of an issue.

**GlenWV** · 04-06-2016, 05:21 PM

Transistors, FETs, Diodes, & such

Greetings:

I purchased both of the mentioned transistors and some FETs from AliExpress.com and saved a bunch on them. Look at buying 50 or 100.

So far, they seem to work so I'm hoping that I didn't buy knock-offs.

If someone will post some numbers for high speed, high amperage diodes, I'll check those out too.

Also, Taydaelectronics is a good source for some things.

glen

**BroMikey** · 04-07-2016, 04:38 AM

This is the way I run my coils. This one runs at 3000hz so the fast
diodes are very much needed.

I drew the dia wrong so now it is fixed. This is my 14 awg wind.
I wanted to run more than an amp or two.

**Donald Haas** · 04-07-2016, 05:07 AM

new timing wheel and cct board

Hi Guys,
I have a high speed CNC router (27" X 30" table) and I cut my motor parts with it. It does iso tracks for cct boarts and wood carving also.

The new timing wheel for firing when the sensor slot is open and a cct board for Rick's trgger circuit were machined tonight.

I can make the change when the 330ohm, 3W R's come in a day or 2.

Mean time I took my old circuit and added the H11D3 and a 2n3439 (350v version of the 2n3440) and we're sending 12v from the run batt into the collector of the MJL Gate driver. It does 17A at 24V, which is pretty close to what I wanted. She runs about 1400rpm. 1 ohm resistors on the trannys and not heat problems. Close to one to one charging with 2, 24v banks.

I've installed 8, 100W LED's off the Gen coils over a section of my bench and they do a great job. More pics in my drop box soon.
My 8 gen coils are 29ohms of #22 wire and there is 1, 100W LED wired to each coil. No series or parallel connections. There is switching over the bench for 2 at a time. DPDT's

We win,

bro d

**thedude** · 04-07-2016, 08:45 AM

Great stuff!

@ BroMikey - Nice to see your schematic, its great to see alternative arrangements. I can appreciate the parallel transistor set up given your input voltage. I'm guessing that you see some pretty high rpms. I checked your 30EPF12 diodes to the ones i ordered and i think i'm in the same ball park, at least in the nanosecond department. Thanks for sharing Bro.

@ Donald - Great to hear Don! Your not wasting any time are you.

I'd love to have a CNC for fabricating. I've had one on the back burner for about 5 years now. Your setup looks great and your gen coils are doing better than mine, although i know the potential is there for great improvement in my arrangement. I'm looking forward to seeing more of your build.

Donald, I hope you don't mind that i took the liberty of uploading your images to my host modvid.com to display them here.

Hope to see more from you soon.

Thanks to you both for contributing to the thread. Its great to have company while working on these projects!
Will be posting more on my build by Fri night i hope.
Cheers!

Announcement

4 Dual Pole Bedini Motor - by Darcy Klyne

4 Dual Pole Bedini Motor - by Darcy Klyne

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment