Bedini/Brandt

[Ken]

Anyone have any experience with this type of controller? All comments welcome.

[Room3327]

Hi Ken,
 I have designed and built many circuits using this part (SG1524).  It is a pulse width modulator and is a very slick part with many uses.  In this circuit it is being used as  a Tesla switch to charge batteries.  I personally don't believe this circuit has any chance of gaining any extra energy as it just shuffels the available energy around.  This particular circuit can't work IMHO as there is no timing parts shown for the SG1524, these chips do not just oscillate on there own, frequency is set by a resistor and a capacitor which I do not see in the schematic.  There is also no pin out shown on the SG1524 so I'm not sure what their driving the transistors with as there is only 2 outputs on this chip, it seems to me that something is wrong with this schematic, as it stands, I don't think it is workable.  Hope this helps.
I guess it could be hooked up with all the offs to one output and the ons to the other (probably the way it was intended) but there is no support circuitry shown on the schematic that is important for the 1524 to opperate.  I could add it for you if you wanted to try it?

[Room3327]

Ken,
I don't believe any extra energy can be gained in a system like this but that doesnt mean there isn't a way to shuffel the available energy around a circuit in a way that can keep the load (motor) running for a very long time.  I would assume that to do so the load would of necessity have to be very very low impedance such as a powerful electric motor to keep the impedance very low for low losses.  A dead battery is a low impedance load also so a fully charged battery would transfer its charge through the load to the dead battery charging it up and etc. you know the rest through Tesla switching.  Repeatedly discharging one and charging the other could lead to longer running time for the load on the same or available energy.  Catching any BEMF and charging other batteries with it just adds to the running time. But eventually you will have to recharge the initiating batterys. Still it could be a great thing for electric cars which are way way past due for us.

[poynt99]

Ken,

Here are some things that may help:

EF post by PL

another EF post by PL

Mueller Report

.99

[poynt99]

Thaelin stated in an OU post that manually switching the batteries extended the time of usable power. Not quite scientific, and would need more rigorous testing imo.


Here's another pdf that might be interesting.

.99

[Peterae]

Hey Room
I've been riding a 1HP electric push bike to work during summer time for 3 years now, work is just over 10 miles each way and i go over some of the highest terrain down south England.

Good Fun.

Peter

[Thaelin]

   Thanks for the mention Point. I will do a quick reply as what I did and found. I used relays and a signal gen to do the switching with. No diodes at all. The output was put into a full wave bridge to keep the same polarity on the output. I achieved about 5X the run time but no charge effect. I didn't try for that anyhow. I wanted to do a bicycle setup with it. Sadly, 12v just is never going to make it work. Minimum would be 24v and better at 36v. At that, the weight of the batts would be prohibitive. They are still at it on EF with this if you have not been there yet.
   I am still looking into a 48v aircraft starter and golf batts with a small gas engine hooked to alternators for each 12 bat set. Only one alt grounds and the rest are isolated.

Good luck to you.

Thaelin

Thaelin stated in an OU post that manually switching the batteries extended the time of usable power. Not quite scientific, and would need more rigorous testing imo.


Here's another pdf that might be interesting.

.99

[Thaelin]

Hi Ken:
   If I had a way to make the amps, I would use 12v. On the average it takes
around a 1/2 hp to move the bike well. Any way you dice that it comes up to
being 350 watts there about. If you do the math for 12 volts you get about
29 amps draw from the batteries. Too much. Take that to 24 volts and you
wind up with 14.5 amps. Better but still too much for me to handle on a bike.
At 36 volts you draw 9.7 and that will fall into the area that size of batteries
can deliver.
   Now see how far that will take you. If you had the problems I have, it makes
it not doable because I can not peddle a bike any more. At 36v you would have
to have 8 batteries for the TS to generate the potential needed. Just not feasible
from where I sit. Just have to have a better way to make the amps.

thaelin

[Room3327]

I hear you thaelin, thats why I went to 72 volts on my e-bike, but you don't have to use car batteries to run them I use gell cells.
Peterae, did you put a video on You tube awile back, I think I watched it. It gave me incentive to build mine.

Anyway back to this Tesla switch, I have been studying this thing for the first time, as I didn't think it had much potential.  I have to say I think there may be something to this thing.  First I have to say it doesn't work quite like I thought, it shuffels energy around but not from a dead battery to a charged battery.  This thing is designed to work with all 4 batteries fully charged at the start.  Which is good because that gives us a lot more initial energy to play with.  This circuit connects 2 batteries in series and the other 2 in parellel, it then pumps 24 volts (2 in series) into 2 12 volt batteries (charging them).  In this process there is an extra 12 volts that is dropped through the load.  When it switches just the reverse occurs with the batteries and the drop through the load changes direction.  Being the energy is passed back and forth between the batteries through the load I believe some extension of run time should be able to be achieved.  This is not an easy circuit to deal with though even if it does look easy, the problem is there is no common Ground reference.  That is why all the transformers to drive the switching transistors. But every thing can be overcome and I am almost finished with a complete circuit schematic that can use almost any N channel power MOSFET or IGBT transistors you want.  I will be posting it soon. 

[szaxx]

Hi
take a look at OU stefans aluminum cell. Its worth a look and that bifilar coil switched circuit from ages ago they may be of no use but they came to mind.

[Room3327]

Here is a schematic of a PMDC motor controller I designed for my Electric Bike if anyone is interested.  It was designed for 72 volts but will run on anything from 24 to 92 volts and can be easily modified for 1 HP.  I will have to redesign for 12 volt operation.  It is small, cheap and works very well (very linear) using a ready made hall effect thumb throttle (bought on Ebay) and runs cool.

[Room3327]

Hi Ken,
 You're welcome, and yes please ask questions when needed.  I have not built this circuit so I don't know what problems may be encountered, but I don't have time to build it right now so I will let you go for it.  One tip I can give you is when picking a power mostfet I have found that they should have their current spec. be about 10X what you are planing running through them.  In other words if you plan an average current of 5 amps pick and use 50 amp mosfets, IGBT's are different as their output devices are regular junction transistors so a IGBT rated 300 amps will work for at least 250 amps.  Mosfet voltage rating should be picked to be about double what you plan on accross it, at least triple if there is any BEMF involved with the circuit.  I find this to be true only with mosfets if you want reliability in the finished product.
  I hope this isn't to complicated for you and like I said I have 1:1:1 pulse transformers if you need them, but they can be found at Mouser Electronics and other electronic suppliers for pretty cheap and their only the size of a sugar cube.