PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2021-12-05, 20:58:02
News: A feature is available which provides a place all members can chat, either publicly or privately.
There is also a "Shout" feature on each page. Only available to members.

Pages: 1 2 [3] 4 5
Author Topic: AVEC Replication Attempt  (Read 73296 times)
Group: Moderator
Full Member
*****

Posts: 163
Coil placement is pretty critical here...

Instead of a toroid I'm using some nails connected to wire to measure any voltage where the output should be.
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
nails?
   
Group: Moderator
Full Member
*****

Posts: 163
nails?

Just testing for voltage in current in the field using some metal. I've switched to some thick copper bars (wires) for now.

Here is the comparator I'm using for clock generation, in case anyone is interested:

http://www.analog.com/static/imported-files/data_sheets/AD8561.pdf
   
Group: Moderator
Full Member
*****

Posts: 163
I'm seeing large spikes when connecting and disconnecting my SEP coil. I have some hash right now, nothing too interesting at the moment.

EDIT:
About 20 volts with a simple 50 turn air coil I wound around my fingers and stuck in the field. I'm out of time this week. Next week I'll finish the output coil and up my voltage (Using about 120 volts DC for the time being), my DC-DC converter voltage drops like a rock. I need to fix that quick...
« Last Edit: 2014-02-27, 08:14:26 by bte »
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
I'm seeing large spikes when connecting and disconnecting my SEP coil. I have some hash right now, nothing too interesting at the moment.

EDIT:
About 20 volts with a simple 50 turn air coil I wound around my fingers and stuck in the field. I'm out of time this week. Next week I'll finish the output coil and up my voltage (Using about 120 volts DC for the time being), my DC-DC converter voltage drops like a rock. I need to fix that quick...

Regarding the DC-DC converter, it is not designed to drive a dead short.  I use a few caps after it, and my pulse time is only a few nanoseconds (about 10 ns, set with a piece of open-ended coax as a delay-line).  I used a resistor to limit current before, when I had a longer pulse time, and this worked ok. 

What are you using as a switch?
   
Group: Moderator
Full Member
*****

Posts: 163
MOSFET at the moment. I don't know what I could replace the converter with, I even have a 1200uf 350V capacitor in the mix, it holds the volts up for a while, I need more current I suppose.
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
MOSFET at the moment. I don't know what I could replace the converter with, I even have a 1200uf 350V capacitor in the mix, it holds the volts up for a while, I need more current I suppose.

shorten the pulse length or use a resistor to limit the current to each coil
   
Group: Moderator
Full Member
*****

Posts: 163
How do you have the coax pulse width coil set up?

My SEP coil is rather poor at the moment, I'll need to rewind it, probably. Each input coil is around 680 ohms, 10,000 turns of 32 awg. Unsure about the inductance. Coil positioning may need adjusting (top coil). I've also ordered heatsinks so I can drive my DC-DC converter harder. The inductor I used may need to be bigger. Maybe an off the shelf 300VDC converter would fare better, I built this one myself, expecting low current draw. I'll see if I can whip up a pulse width control circuit, these MOSFETs and drivers are rather slow though.

I'll see if I can get a shot of the pulses next week.

Thanks for the help!
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
How do you have the coax pulse width coil set up?

My SEP coil is rather poor at the moment, I'll need to rewind it, probably. Each input coil is around 680 ohms, 10,000 turns of 32 awg. Unsure about the inductance. Coil positioning may need adjusting (top coil). I've also ordered heatsinks so I can drive my DC-DC converter harder. The inductor I used may need to be bigger. Maybe an off the shelf 300VDC converter would fare better, I built this one myself, expecting low current draw. I'll see if I can whip up a pulse width control circuit, these MOSFETs and drivers are rather slow though.

I'll see if I can get a shot of the pulses next week.

Thanks for the help!

The coax delay line discharges for a predictable amount of time based on it's physical length and impedance.  I guess it would work for any switch.

http://en.wikipedia.org/wiki/Avalanche_transistor

look at the image "Simplified transmission line avalanche transistor pulser."  You take your pulse out at Vout1 for a positive pulse.  Rc is to limit current.  You have to be able to fill the pot before our pulses drain it, and you need voltage not current so current can be very low.  I use 1 or 2 ma.

with 680 ohms, and 350vdc, the current is about .5 amps.

for .002 amps (2 milliamps) you need 175,000 ohms

This is just a ball-park idea, as you are using a pulse and the above is just Ohm's Law for DC.   The current is actually lagging and not instant, so if the pulse is very fast, the current doesn't have time to rise.  Spherics mentioned all of this regarding the use of MOSFETs.

You can avalanche mosfets too, by the way...

http://www.eng.tau.ac.il/~shmilo/20.pdf
   
Group: Moderator
Full Member
*****

Posts: 163
Just chiming in with some anecdotal things I noticed. When the bias is powered while the rotation is happening, I noticed a heat-like sensation. It was really weird, it might have been in my mind. The unit seemed cool to the touch but delivered some electrostatic shocks when touching certain parts that should not be charged or conductive. But these things happened when I wasn't using the right parameters. Once I boost the voltage and the bias field power, I wonder what I'll see?
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
low level effects
   
Group: Moderator
Full Member
*****

Posts: 163
So, a shorter pulse width, higher voltage, and a better bias coil. Do you think biasing each of the 4 coils would help? (Say, 50 volts?)
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
So, a shorter pulse width, higher voltage, and a better bias coil. Do you think biasing each of the 4 coils would help? (Say, 50 volts?)

I don't think biasing each coil is applicable in the tetra.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1443
Frequency equals matter...


Buy me a drink
Please be careful. The electrostatic shocks and weird warmth are RF results. What is really really bad is the warmth is the field pulsations inside you. Then you send in electrical discharges through that internal bias field. You are disrupting internal processes that control nerve and muscle functions. Its been 6 years for me and I still have a numb spot on my thumb palm. So if you are going to stick your hand in there don't do like Keely did and test with your writing hand. That is why Elizabeth Clayton Moore wrote all Keely's notes.

Just chiming in with some anecdotal things I noticed. When the bias is powered while the rotation is happening, I noticed a heat-like sensation. It was really weird, it might have been in my mind. The unit seemed cool to the touch but delivered some electrostatic shocks when touching certain parts that should not be charged or conductive. But these things happened when I wasn't using the right parameters. Once I boost the voltage and the bias field power, I wonder what I'll see?


---------------------------
   
Group: Moderator
Full Member
*****

Posts: 163
Ah, that would explain it. Upon measuring the field, the oscope reported frequencies about 100 times higher than what I was inputting collectively. 3khz for coil A, 1khz for coils X,Y, and Z. The frequency read at around 250 - 300khz. It seemed to increase with time but I never had the unit on long enough to verify anything for sure. (MOSFET heating issues) Frequency mixing causing up-conversion perhaps?
   
Group: Moderator
Full Member
*****

Posts: 163
Got things rewound, now I don't have enough voltage at the coils. Ugh. I need a better PS.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1443
Frequency equals matter...


Buy me a drink
Build a flyback, MOT or ignition coil supply. Make a monster on your bench.


---------------------------
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
   
Group: Moderator
Full Member
*****

Posts: 163
The snap back time is around .01uS or 10 nanoseconds. I need a pulse rise time of less than 10 nanoseconds. I like how Spherics mentioned "10 units" of time for the rise time calculations.

This is for the AVEC. Not sure if the tetrahedral TPU has such constraints. Thanks Grumpy and GK, I have ordered one of those supplies.

My head hurts.
   
Group: Moderator
Full Member
*****

Posts: 163
I've gotten my pulse width down to 5uS. I'm generating a standard 50% duty cycle clock, splitting it with a shift register, and feeding each phase into a passive differentiator circuit with an op amp to boost the output back to 5 volts. I can't get it down much further than that. I might have to look into using microcontrollers eventually, seems like it's cheating though. I'm getting a clean 3 phase plus clock with identical pulse widths though. I'm ordering another EMCO DC-DC converter for the HV stage, the 500V version. My driver/MOSFET will only go up to 600V.

Everything is just laying out on a breadboard. Anyone have any tips on how to make this more "permenant"? I'd like to have it in a spiffy box and add some analog amp/volt meters and switches.

Out of time this week. Next week I'll reglue my support and start pulsing again.
   
Group: Moderator
Full Member
*****

Posts: 163
Sweet! I've got my pulse width down to 200-250ns. It varies (bounces) quite a bit at this time. I'm feeding a 50% duty cycle square wave in and feeding it through a passive differentiation circuit (Cap and Resistor) to get a pulse, which I feed into a comparator for the final square wave.

I have a tube transformer that I'm going to use to get the volts for the output.
   

Group: Administrator
Hero Member
*****

Posts: 3844


Buy me some coffee
bte
I've used Dual monostables before, some can get you a pulse less than a 100nS, what i did was use both mono's, the first is used to delay the pulse, the second varied the pulse width, so when you have more than one chan then you can fine adjust the phase position of each pulse as well as each width.
I can post the circuit if you are interested.

I went on to use delay chips which have a variety of flavours, 250ps, 1ns, 5ns, and you select using 8 bits, i first used dip switches but went on to build a controller.

I was able to get down to a 30nS pulse width and i could vary the phase in 1ns steps.
   
Group: Moderator
Full Member
*****

Posts: 163
Yeah, please do. The one I have now is crude, my MOSFETs wouldn't be fast enough for super short pulses, I do not think so anyway.

Which MOSFET did you use? I might even dabble with tubes, never used them before however.
   

Group: Tinkerer
Hero Member
*****

Posts: 3482
tExB=qr
pssst....


avalanche
   

Group: Administrator
Hero Member
*****

Posts: 3844


Buy me some coffee
Hi bte
Sorry for the delay, I've not been near my computer with the diagrams on.

There is some stuff published on OUR
http://www.overunityresearch.com/index.php?topic=217.msg2375#msg2375

You will see my digital monostable, that uses delay chips and dip switches, next diagram down shows my Fet driving Circuit.

Note i use MCP1406 or MCP1407 driver chips, i found these to be really good robust and pretty fast, the fet i use is IRF840

I went on to build a full digital controller
http://www.overunityresearch.com/index.php?topic=31.0

I tried a Avalanche transistor pulser
http://www.overunityresearch.com/index.php?topic=307.msg9600#msg9600

When i get some time tomorrow i will put up a diagram of my Analogue Monostable Phase shifter and adjustable width circuit.
   
Pages: 1 2 [3] 4 5
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2021-12-05, 20:58:02