quad wound coil & SCR scheme / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

quad wound coil & SCR scheme

1 2 next

Move Thread

LAN_403

Turkey9

Wed May 27 2009, 04:47AM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

Here's my problem: I'm going to build a spiral coil that is really four different coils all put together in one... There will be four leads and four different caps; one for each section of the coil. I'm hoping that this will give a much higher rise time without sacrificing magnetic field strength. I've seen a similar system like this and each coil had its own switch to control the current. I'm wondering if I couldn't just tie all the ground leads of the coils together and feed them to one scr that is really connected to ground. The other side of the coils would be connected to the negative side of each of their own caps. Would the coils share the current evenly? Or should I use a separate scr for each coil section? Thanks for any info!

klugesmith

Wed May 27 2009, 06:21AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Do you want a higher (slower) risetime, or a higher slew rate?

Are the four sections of your quad-wound flat spiral identical,
each with its own identical cap, and all fired simultaneously?
Then by symmetry, you could connect the 4 coils and 4 caps
in parallel without affecting the behavior.

The speedup (with respect to single-wound coil with the same
spacing and total number of turns) comes from the sections
now being in parallel instead of in series. Assuming the caps
are paralleled in both cases.

Theoretically the single-wound coil with 4 caps in series, would have
the same timing as the quad wound coil with 4 caps in parallel.

Turkey9

Wed May 27 2009, 05:27PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

Everything is identical. So you're saying that I could connect them together and switch it with one scr?

big5824

Wed May 27 2009, 06:12PM

Registered Member #1687 Joined: Tue Sept 09 2008, 08:47PM
Location: UK, Darlington
Posts: 240

if your just taking a normal coil and splitting it into 4 pieces to divide the inductance and resistance by 4, then just connect them together in parallel and use one bank of caps and one scr, much simpler.

KLH

Wed May 27 2009, 06:49PM

Registered Member #1819 Joined: Thu Nov 20 2008, 04:05PM
Location:
Posts: 137

Putting the 4 individual coil pieces in parallel will divide the resistance by 4, but the inductance can vary widely from expectations.

It would be safe to assume that the inductance would divided by 4 if the coils had none to negligible coupling between them. However, from what I understand about this setup, the smaller coils will be part of a single coil, and thus be very close together. The conventional methods of finding inductance in parallel would no longer apply, and a simulation of the magnetics would need to be run to find the actual inductance.

Please correct this if I made a mistake.

Turkey9

Wed May 27 2009, 09:15PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

If I put all coils in parallel and use one cap bank, would that increase the rise time? I need to have as fast a rise time as possible with these coils, that was the reason for splitting them up. If they're all in parallel, would they act like one coil with much thicker wire, or will the magnetic fields from each coil still add together? Sorry about all the questions, I'm not that solid on my inductors.

big5824

Wed May 27 2009, 09:39PM

Registered Member #1687 Joined: Tue Sept 09 2008, 08:47PM
Location: UK, Darlington
Posts: 240

Yes they will act very similar to a coil with thicker wire. Just remember, you are not just changing the rise time, the pulse will also be around 4 times shorter. I dont see why you dont just use a coil with thicker wire, this would achieve basically the same effect.

Killah:
Yes the inductance would be variable for each individually coil, but if they have similar geometries etc then it can be assumed they should have similar inductances.

Turkey9

Thu May 28 2009, 05:16AM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

I know that the formula for the magnetic field strength is dependent on both current and number of turns. I'm hoping that with this setup I can have a high peak of current (lower inductance and lower resistance) with the same amount of turns. Maybe it won't work like that, but I'll see. I can also set up the wiring to act like one coil with 4 times the voltage and anything in between.

KLH

Mon Jun 01 2009, 06:41PM

Registered Member #1819 Joined: Thu Nov 20 2008, 04:05PM
Location:
Posts: 137

Turkey9 wrote ...

I know that the formula for the magnetic field strength is dependent on both current and number of turns. I'm hoping that with this setup I can have a high peak of current (lower inductance and lower resistance) with the same amount of turns. Maybe it won't work like that, but I'll see. I can also set up the wiring to act like one coil with 4 times the voltage and anything in between.

If the coils are coupled beyond anything negligible, then the effect will not be the same. If you want a higher peak current with the same pulse time, then you need to increase your system voltage. As an alternative, you can make a thicker-wire coil with less turns and make a bigger capacitor bank to achieve the same LC pulse time. If you are just looking to increase the peak current without regard to the pulse time, just make a thicker-wire coil and use your existing system voltage capacitor bank. Doing so will reduce the pulse time, which, from what I understand, is one of your goals.

klugesmith

Wed Jun 03 2009, 11:36AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

I drew this picture to support our discussion of intra-coil coupling and external magnetic fields. Was momentarily surprised by conclusion that parallel connection of quad, or just winding fewer turns in the same area, speeds things up but does not increase the peak magnetic field intensity.

Coils Q and S are quad and single-wound with the same size, wire spacing, and total number of turns. Let's start with the 4 sections of coil Q connected in series, and carrying the same current as coil S. It's obvious that the total length, resistance, and magnetic fields are practically identical to coil S. Also the inductance, because each little wire segment in Q has a corresponding segment in S at same distance from center, and effectively the same flux linkage with the rest of the coil.

Now connect the four sections of Q in parallel, but don't (yet) change the wire current. Magnetic field is unchanged, but terminal current is 4x and terminal voltage is 0.25x. So the coil R and L are reduced by a factor of 16. Time constant sqrt(LC) and characteristic impedance sqrt(L/C) both go down by a factor of 4.

So if capacitor bank is unchanged, and we neglect all parasitic R and L and PIT plasma motor effects, then the capacitor discharge will happen in 0.25x the time, with 4x the peak currrent -- at the capacitor! So the peak current in each of the 4 parallel coil sections, and hence the magnetic field intensity, will be just the same as the series-connected and single-wound cases. (All the stored energy from capacitor is reversibly shifted into energy of the magnetic field, no matter how the coil is wound).

Not all is lost. Because the parallel-connected case has a faster timescale, the absolute rate of change of I and B, and hence the induced electric field, will be 4x greater. So more plasma ionization (?), currrent, acceleration power, and specific impulse during the shorter discharge time, we hope. On the other hand, the R and L of capacitor, switch, and interconnect (and even the R of coil itself) will be proportionally more significant in the parallel case.

It sounds like Turkey9 has not overlooked the intermediate option of connecting quad coil sections in series-parallel.

I agree with big that a single winding with heavier wire and 1/4 as many turns would be similar to the parallel coil. But it would have a less uniform field, and not save any copper.
Sorry to be rambling; it's way too late.

-Rich

p.s. don't forget the voltage reversal problem with your electrolytic capacitors. I bet real PITs (designed for efficiency) have an underdamped, oscillatory impulse. If you add a clamp diode then di/dt and plasma motor action are truncated after first 1/4-cycle, wasting the remaining energy in the circuit.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob