Electronics > Other

Just trying to recreate the waveform

(1/6) > >>

oddball:
I have moved this to here as I realised it was in the wrong place. ;D

I was sitting watching the TV last night and this idea came to me as I said in the posting above to use duplication of some of the components so that you have more than one waveform hitting the cell out of phase with each other.
Also some people seem to think the choke in the negative is not so important.
Anyway here is a doodle I have put together this afternoon, I thought maybe try pulsing the positive instead of the negative all the time so maybe a P-channel FET.
I also had a look at DOG's posting about the inhibiting resistor.
"Electron Inhibiting resistor is simply any resistance that stops the flow to ground (As I understand it). I have used resistance values from 10 Mega Ohms to 1 Ohm. Simply stated, as you increase resistance on the negative side of the tube, the voltage across the tube increases. Of course, this only works when you have the blocking diode and choke on the positive side (driven by a transformer). "

(http://propaholics.wolfchasers.com/uploader/users/oddball/ODDBALL%20PWM%20system1.jpg)

You could have both of the 555 oscillators with the same frequency components and just set one at a harmonic of the other it could be worth an experiment just add a 6 position switch instead of the 3.
I have just guessed at the biasing resistors and used what Lawton had but adapted the set up for P-channel, the problem is to turn a P channel on may take a negative voltage I don't know if just 0V potential will turn it on. :-\
As you can change the mark space ratio on the Lawton 555 set up there should be no problem there.

As to the chokes maybe keep them separate and not bifilar at all. :-\


EDIT
0 Volts will turn on the P channel FET so have ordered some for a play with. http://www.tranzistoare.ro/datasheets/70/352213_DS.pdf

Also when I used to make H-Bridge drivers I always used a 10K resistor to stop a FET from switching to an indeterminate state so have changed the 2 x 820R resistors that Lawton had to 10K, and many of the DATA SHEETS say feed these FET’s direct
From an IC so I have also removed the 220R from the 555 pin 3 to the FET gate.

See H-Bridge motor driver with N & P channel FET's http://www.cadvision.com/blanchas/hexfet/images/h-bridge.gif

Maybe switching the supply ON and OFF on the positive side of the cell/capacitor and via 2 separate feeds and oscillators will give that voltage spike on the waveform you are after.

(http://ionizationx.com/index.php?action=dlattach;topic=97.0;attach=190;image)

Anyway this afternoon I got my scope out and an old NAND gate oscillator I made for an other project a few years ago.
So hooked it up to the cell to see what it gave me, and here is the waveform.
It is just pulsing the 12v as does the lawton but only 1 oscillator.
(http://propaholics.wolfchasers.com/uploader/users/oddball/waveform1.jpg)

It got me thinking the top part of your square wave has a ripple but it is only a slight ripple maybe a few volts compared to the 13.8 V or what ever you get from you rectified 3 phase.
So that means that this ripple frequency depends of the revs of the alternator and in turn the engine. The only constant thing is the square wave you are creating out of this rectified voltage, and that is different for each cell design am I right?

So this means that the lawton CCT fails because it still pulses the whole voltage during the top of the square wave.
So the idea I have above could be modified so the 2 waves are in sink but applied to each other at different voltages after being switched by separate FET's as I said.
So this means that both waves still need to be in sync to work.

Steve:
Hi oddball,

Nice out of the box thinking.
Are you gonna make a circuit and do some tests with it?
If so, please try to keep a bookkeeping of the resultsish them.

I can tell that Bob Boyce had positive results with multi frequency's.
The pulse that Hydro and I have is derived of the tests with the alternator setup.
This is where Stanley meyer started with. He had positive results and nothing indicates that your path is the right one. Do not understand me wrong here. I really like people thinking out of the box.
Why? Because all normal road are already walkd on.

So, build your setup and let us know. If we can help you, just ask.

Br
steve

oddball:
Yes I know what you mean Stevie thanks.

I have been looking over your stuff and can see were you are going with some of it now.
I can see the alternator is a hard thing to beat in terms of it's output but I thought I saw you are trying to make a solid-state version but are not yet showing your ideas.
I think I saw a decade counter but if you are just trying to re-create that wave there must be simpler ways.
However I see your plan is low volts and higher current as well so if you have an AC source with a lot of amps already in the car it is stupid not to use it. :D

However I have a few ideas that I would like to use the cell for besides to run a car so a way to do it from DC but without an alternator and get good gas for low power would be good. ::) 

Anyway both my cars are Suzuki one based on an old SJ413 so I believe the alternators are a favourite for conversion.
MY JEEP HERE

oddball:
Here is a variation on the same theme using your 3 phase idea, I think I will give this one a go for fun ;D

I have ordered 4 x 11 amp  P-channel FETS on eBay and have a bag full of high power N-channels so the rest of it is just a few £'s worth, apart from the transformer I suppose you guys have found a source for that or are custom making them.
At a push I could just mess about with 3 standard transformers.
My thoughts were still try something different and see what happens, the idea is to set up harmonics of the 3 frequencies, then just PWM it at the end with an N-channel to make a square wave with a load of harmonics on the top. :-X

If the harmonic on the top of the square wave is from the RPM of the alternator and that is good at 3000 rpm then 3Khz for osc A, 3.5kHz for B and say 4khz for C but there is plenty of room to adjust each one.
I am opting to try with the NOR gate after comparing the Lawton 555 system with it on the scope, in my opinion the Lawton has too big a gap between pulses but it could have the components changed to fix the mark space (well that is how it looked to me).
Also there are 4 gates in each IC so only 2 chips are needed.
Much of the work is probably being done just by the higher current and the ripple may just be a bonus as may be the switching spike at the start of each pulse.
(http://propaholics.wolfchasers.com/uploader/users/oddball/OU%20waveform.jpg)

EDIT This CCT is being modified day to day and this diagram will update at this location!(http://propaholics.wolfchasers.com/uploader/users/oddball/3%20PHASE%20FREQENCY%20SHIFT%20PWM.jpg)

Steve:
Hi oddball,

There are many different theorys about making a 3 phase system. Yours is a good attempt. But not perfect , yet.
The trafo will be very important. Also the way you drive it.....(hint)

Also the volts is stil an issue ( for me).
In theory is 2 volts the best way to go, BUT....you will not get enough amps in with 2 volts!
Unless you start to use chemicals.

Have you ever run tests with a WFC with different volt levels?
If not, please go and try this. Make notes of the volts and amps...
You will see that the resistance will be different, if you step up voltage.

So, what will do that to your theory?
You want amps? then you need volts as well......

Have fun!
Go run the test!

br
Steve

Navigation

[0] Message Index

[#] Next page

Go to full version