Parallel coil-layers? / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Parallel coil-layers?

previous 1 2 3 next

Move Thread

LAN_403

ScotchTapeLord

Fri Jan 09 2009, 02:18AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I was planning to parallel 3 strands of my 22 gauge wire in order to get a lower resistance coil without buying thicker wire, but also to help distribute current evenly over my SCR "bank." Each wire will be in series with the anode of each of my SCRs, with the cathodes to the bank ground.

Is this a good idea?

rp181

Fri Jan 09 2009, 02:20AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Just put all the wires in parallel, and wind it like a flat ribbon cable. But the SCR's in parallel and connect it to the coil.

Marauder709

Fri Jan 09 2009, 03:19AM

Registered Member #1895 Joined: Thu Jan 01 2009, 03:12AM
Location:
Posts: 22

I've done it my self to lower inductance and increase magnetic field strength. It also cuts back on resistance significantly (R= r^-1 + r^-1.....)

That's what I meant, putting the layers in parallel will reduce the inductance, just as you have said.
You're only increasing the magnetic field strength because you are increasing the peak current. Buy an LCR meter and measure it yourself. Or if you have a good voltmeter and a function generator, you can use phasor analysis to find the change in inductance.

I quote from Halliday Resnick and Walker, a Physics book used to teach AP Physics C and introductory college level Physics courses: "Inductance depends on the geometry of the device (the inductor)"

badastronaut

Sat Jan 10 2009, 12:18AM

Registered Member #222 Joined: Mon Feb 20 2006, 05:49PM
Location:
Posts: 96

Yes, that is true, inductance depends on geometry.

It also depends on how it is wired, which is the topic of this thread.

The inductance formula stated earlier in the thread makes the assumptions that the coil is either a single layer or all the layers are wired in series with an average radius.

If you have a two layer solenoid and wire the two layers in parallel, you end up with something equivalent to using only one layer with thicker wire. If this is not true then everyone would be doing it. The inductance of this configuration would be lower than the one where the two layers are wired in series as in a normal inductor. People do in fact use parallel layers to handle high current where flexibility and ease of winding are desired. There's nothing wrong with using parallel layers in place of thicker wire.

Say you have a solenoid with 2 layers and each layer is composed of 100 turns for a total of 200 turns.

The magnetic flux density depends on the number of amp-turns you have in a given area. If you wire the two layers in series, then 1 amp through the solenoid will give you 200 amp-turns of MMF. If, however, you put the layers in parallel, you will get 100 amp-turns with 1 amp through the solenoid. This is because the current divides among the layers such that half an amp flows through one layer and the other half through the second layer. 0.5amps*100 turns + 0.5amps*100turns = 100amp-turns This is only true if you believe in conservation of charge.

Each solenoid configuration, parallel or series layers, has the same current through it, but the series configuration has the higher number of amp-turns. Since the geometry of each configuration is the same, you can say that the series configuration has the higher inductance because it gives the higher magnetic flux per current.
QED

Yes, you may be able to get the same magnetic flux by increasing the current which may be made possible by the lower resistance, but it is not the same as using higher inductance. You can even use a single chunk of copper, i.e. one turn, and have the same magnetic field as long as the current density is the same as the original multiturn solenoid. The reason to use multiple turns is to match impedance with the power supply and reduce the amount of current required by the power supply even though the losses are the same for the same magnetic field since the current densities have to be the same. Inductance and magnetic flux are related but not the same.

When you say you increased the magnetic field by doubling its inductance, that is assuming you were able to keep the current the same, which you may not be able to do because of the higher inductance, you would need a higher voltage to overcome the extra impedance assuming there are AC components to the current flowing through the inductance.

In a coilgun, reducing the inductance might give you higher magnetic flux density because the lower inductance allowed the peak current to be much higher.

Don't go around quoting random texts whether or not they are true unless you understand them enough to prove that they are true. Anyone can steal other peoples work. Yes of course inductance depends on geometry, but it depends on other things as well such as how the individual turns are wired.

In fact, if you wire one layer of the solenoid and then reverse the winding direction on the second layer, you will cancel out most of the inductance of the solenoid. One layer will have X number of amp-turns and the other layer will have -X number of amp-turns for 0 effective amp-turns. This is how they make low inductance wirewound resistors even though the absolute number of turns is the same.

You can't go around citing equations as gospel unless you understand them enough to derive them yourself and know what the assumptions were in deriving them.

Is there anyone else that can vouche for this post?

Barry

Sat Jan 10 2009, 01:02AM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

Well written badastronaut (aka goodengineer), I agree with everything you've said.

A good analysis includes consideration of all the operating conditions (e.g. is the power provided from a voltage source or a current source, etc) along with all relevant assumptions.

Also Uzzors correctly reminds us that the LC timing constant is independent of resistance. That is, the discharge time depends on inductance but not resistance. You can control the rate of damping with resistance but not the natural frequency. Thank you.

Barry
Barry tickles 2009 under the chin. She wiggles and giggles and begs him to stop.

SamS

Mon Feb 02 2009, 10:07PM

Registered Member #1953 Joined: Mon Feb 02 2009, 09:23PM
Location:
Posts: 3

Just saw this thread and decided to drop by.

Would it be any good if a multilayer coil is made where each layer is a separate electromagnetic coil with it's own bank and switching. When all layers are fired:

a) Magnetic fields add up nicely?
b) If so, would this setup have any advantages compared to "traditional" multilayer coil wound from one wire if switching losses are ignored?

j.azz

Tue Feb 03 2009, 12:59PM

Registered Member #888 Joined: Tue Jul 10 2007, 06:52PM
Location: Hannover, Germany
Posts: 40

I guess it's just more to do if every layer has it's own switching and bank. I see no improvements over single switching.
But seperating the layers has advantages as you can vary the number of layers of your coil and thus making prototyping more easy. IIrc Barry has done this with his MarkII.

Hope that helps.

Quantum Singularity

Sun Feb 15 2009, 05:59PM

Registered Member #158 Joined: Sun Feb 12 2006, 09:53PM
Location: Central Ohio
Posts: 282

Just thought I'd point out if your making flat coils like I use then paralleling layers I dont beleive changes the inductance within reason. I've built a few with parallel layers as well as others here too. The turns per inch stays the same as well as the diamter of coil so the inductance should remain the same right? Just less copper resistance. Basically just using wider wire, which now I have the rectangular wire to use instead... a lot easier than doing parallel layers.

Signification

Mon Feb 02 2015, 01:19AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

Suppose you had two identical large pancake coils placed against each other face-to-face wired in parallel. The parallel inductor's "rule" says that the total inductance of the two will be reduced to half of the inductance of one. If you try this, I think you will find that this is far from the case--the reason being that the mutual inductance or 'flux coupling' is significant--especially in this particular configuration and at coilgun currents. The rule assumes that there is NO mutual inductance. It may help to think of the definition of inductance as 'amount of flux per amp'--and a lot of flux from one goes through the other. However flux will largley cancel if the paralleled coils are wired out-of-phase. The ohmic resistance, on the other hand, is indeed halved.

BigBad

Tue Feb 03 2015, 03:25PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

Why don't you model it in LTSpice (easily done) and get back to us?

previous 1 2 3 next

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