Hello all.

I've replicated Patrick Kelly's South African inventor's 'Tiny Generator', from his post last month.

I've put it all together, 3d printed parts and the rotor etc, but I'm struggling to get it to operate as advertised.
Likely my electronics and connections are at fault as I'm a newcomer to the practical side.

Contrary to the instructions, the magnets need to be placed in the rotor SOUTH facing out for the hall sensor to work, but I had a voyage of discovery of all the other mistakes I'd made along the way trying to get it to work, but got there in the end.

Problem is I attached the oscilloscope and I can't get the system to pulse the back emf as expected. In the beginning I left it running for a couple of hours, but found that the battery voltage dropped very slowly rather than increased, which suggested it wasn't charging. I had three 12v 7Ah attached, two running in parallel with the charging battery.

I attached a second coil out of curiosity and burned up the mosfet and the +i've line on the board.
Fixed that, then used a different single coil (close to spec but taken from a speaker and professionally wound) and the negative line fried, and mosfet again.

Now, I wound a new coil carefully to original spec, no flaws or overlapping, and I've replaced the mosfet but find it runs very hot within ~30 secs operating, so I switch it off.

I'm still researching and reading up to find the error of my ways.

If anyone has any suggestions I'd appreciate it.

I've attached a video of it's operation so it may help troubleshoot. Starting V ~12.6, I~0.212A.

Cheers,

Ian.

Here's the video (I think...)

Very interesting, many thanks for sharing :thumbsup:

Google pointed me to http://www.free-energy-info.com/ which offers quite a few interesting projects, each compiled into neat PDF's.

Don't know what's wrong with your design, but as with many, if not to say all, such 'free energy' devices it takes, unfortunately, a few trial and errors before you get them to work the way you'd like them to work.

Anyway, this design is so simple that it should be easy enough for most tinkerers to have a go at it :-)

MOSFETs might need a external cooling like heat sinks and fans.

I have burned coils and MOSFETs because I did not have a flyback diode. When you charge up a coil and then but the power, that magnetic field will collapse and it will result in a large charge that needs to go somewhere. This can very easily fry a mosfet.

So maybe a flyback diode or use that pulse to charge another battery? I think Rick F. had a circuit for that. Using the pulse

http://www.r-charge.net/Free-Energy_ep_56.html

Hello Belfior - many thanks for the input, and that was a very useful article on the battery charging/discharging. I'll also take a closer look at the Benitez patents again - I've seen them covered before (partially) in Bedini's work.

So I'm using 3 flyback diodes in parallel connected to a bank of 7x 7Ahr 12v batteries and then through to the driving battery (same type), as per the instructions from P. Kelly's South African designer. I've attached the circuit diagram.

This system is supposed to produce 600+ V of flyback impulses into the batteries, but I'm still learning how to properly measure the impulses, and how to adjust the inductor resonant frequency using a homemade switchable capacitor bank (thanks Master Ivo for the circuitry) and a resistor bank. I have a new laptop 4 channel oscilloscope and recently bought a x100 probe (because I was afraid to fry my computer!) so hopefully will make some headway soon.

I've put together the circuit board for the 150W solid state version of this and will be trying out some coils with that setup.

If you have any links you think would help my understanding I'd appreciate it - thanks!

well normally you would connect the flyback diode back over the coil, so the charge would keep going to though the coil. This will naturally kill the pulse :)

Now the charge coming from the coil is still going to be on the mosfet drain, if that gives in more easily than the diodes let the charge to the battery.

I am no expert of anything, but this would be my suggestion. Just keep tinkering. The solution will come if just work with it. most likely when you have gone to bed already... Try to go around the problem. Do you need the impulse from this coil? What if you drive another coil and use the impulse from that? Then protect this "driving" coil with a flyback diode? There is usually a simpler solution available, if you do not get into a pipe dream where you cannot think outside the box.

I have seen many people trying replicate Don Smith's circuits and they try to replicate them 100% to the schematics. Like Don would give everything way like that. You need to watch his videos 20 times before it hits you.

Hi Belfior - thanks for your reply.

 For me, the whole point of this experiment is to utilise the (rapid switching) back emf impulse to go through the diodes and charge the battery after the coil circuit (magnetic) collapse and use a high voltage mosfet to deal with the energy.
I'm just having a spot of bother to find the charging impulse.

put a spark gap

Hello all,

I am busy with a replication attempt of the Tiny Generator also. I have been told to use two hall effect sensors although that I think was more for the larger generator design that uses 10 coils instead of a single in the Tiny.

In the larger generator, there are essentially two "strings" of 5 coils each, one string being north and the other south in terms of winding. But these strings are connected together on both the positive 12V rail as well as the drain of the mosfet. The loving giving path diode or diode bank is used to bleed the back EMF off into the charge batteries or inverter when the drain is switched off from ground.

I've set a scope on the hall effect sensors and with the rotation there is a point just between the two where both sensors are low. I think that is the narrow gate we may be searching for.

Oh, and heatsinking the mosfets is mandatory.

I have setup an online multiway videoconferencing collaboration platform for all here to collaborate on this. Let me know if you are interested and I will post details.

Quote from Bogirish on March 4th, 2020, 11:44 AM

Contrary to the instructions, the magnets need to be placed in the rotor SOUTH facing out for the hall sensor to work, but I had a voyage of discovery of all the other mistakes I'd made along the way trying to get it to work, but got there in the end.

I have my magnets with south facing in and north out and the A3144's I am using work fine. I am using each sensor with a 10K pullup into an Arduino Nano to be able to tell me the RPM.

Quote from MikeTheSpike on March 30th, 2020, 10:08 PM

Hello all,

I am busy with a replication attempt of the Tiny Generator also. I have been told to use two hall effect sensors although that I think was more for the larger generator design that uses 10 coils instead of a single in the Tiny.

In the larger generator, there are essentially two "strings" of 5 coils each, one string being north and the other south in terms of winding. But these strings are connected together on both the positive 12V rail as well as the drain of the mosfet. The loving giving path diode or diode bank is used to bleed the back EMF off into the charge batteries or inverter when the drain is switched off from ground.

I've set a scope on the hall effect sensors and with the rotation there is a point just between the two where both sensors are low. I think that is the narrow gate we may be searching for.

Oh, and heatsinking the mosfets is mandatory.

I have setup an online multiway videoconferencing collaboration platform for all here to collaborate on this. Let me know if you are interested and I will post details.

Sure this sounds helpful - did you get yours to work yet as advertised?

Quote from MikeTheSpike on March 31st, 2020, 05:36 AM

I have my magnets with south facing in and north out and the A3144's I am using work fine. I am using each sensor with a 10K pullup into an Arduino Nano to be able to tell me the RPM.

Yes the north/south thing - I read a comment under the A3144's saying that they worked great with the south face - I changed mine to south out and it sprang to life! Perhaps there are different configurations for these sensors...

It'd be good to know if anyone has got this to work yet

Quote from Bogirish on March 30th, 2020, 09:59 AM

Hello Belfior - many thanks for the input, and that was a very useful article on the battery charging/discharging. I'll also take a closer look at the Benitez patents again - I've seen them covered before (partially) in Bedini's work.

So I'm using 3 flyback diodes in parallel connected to a bank of 7x 7Ahr 12v batteries and then through to the driving battery (same type), as per the instructions from P. Kelly's South African designer. I've attached the circuit diagram.

This system is supposed to produce 600+ V of flyback impulses into the batteries, but I'm still learning how to properly measure the impulses, and how to adjust the inductor resonant frequency using a homemade switchable capacitor bank (thanks Master Ivo for the circuitry) and a resistor bank. I have a new laptop 4 channel oscilloscope and recently bought a x100 probe (because I was afraid to fry my computer!) so hopefully will make some headway soon.

I've put together the circuit board for the 150W solid state version of this and will be trying out some coils with that setup.

If you have any links you think would help my understanding I'd appreciate it - thanks!

Also IRF840 has Maximum Drain-Source Voltage |Vds|: 500 V so not very good for +600V impulses

For sure, I was just trying to stick to the description which was supposed to work. I now have both 1200v and 1700v mosfets to try which should hopefully run cooler/better.

Quote from securesupplies on March 30th, 2020, 09:07 PM

put a spark gap

Spark gaps are great, but I'm trying to produce the spark gap equivalent through rapid switching. I got my kick into gear through Master Ivo's videos - thanks Ivo.
The ultimate aim is to harness the back emf for extra output energy, make it all solid state and easy for anyone to make.
I'm new to all this, so to get my first practical attempt to work with a little help should allow me to understand it all better and continue.

Cheers

Quote from Bogirish on March 31st, 2020, 10:41 AM

Sure this sounds helpful - did you get yours to work yet as advertised?

Still busy completing the circuit boards (etched and awaiting drill holes)  for this, so no, no motion quite yet. I have both the Tiny as well as the larger generator 3d printed and assembled and just waiting for said circuit boards. I am working between the two designs and the information I got directly from the original SA designer as they share common circuitry and concepts. I have reached out to see if he has the time and if he would be willing to give us assistance.

I have also attempted the early solid state version but that was unsuccessful. I guess I just didn't know what I was doing back then and had nobody to collaborate with.  I'll try make a small clip of what I have done so far.

Like you sir, I am also new to this, but if we all work together, I am sure we can do this!

To awaken a dead thread -

I just replied to your original post about this - check it out it