Coaxial LC design for pulse compression using a moving short / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Coaxial LC design for pulse compression using a moving short

first previous 4 5 6 7 next

Move Thread

LAN_403

Signification

Wed Mar 11 2015, 01:25AM

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

DerAlbi, you previously posted:
===================================
Really: i am offering my help by questioning your fears. First show me a basic calculation that shows the relevance and thats it.
For basic ball park figures its often enough to know
Resistance R = U/I;
Inductance I = U*dt/L
Capacitance: U = I*dt/C
=====================================
This is an improper format for differential notation--you cannot have a differential notation of the independent variable (dt) with NO corrosponding differential component of the dep. variable in the same equation. I only bring this up because you seem to be laughing at a notation, DI=U/Ldt, which I agree is improper, but you called it my notation. If I actually wrote this differential eq. verbatim: ""DI=U/Ldt"" I would like you to point it out--I don't usually make this error.

DerAlbi

Wed Mar 11 2015, 01:58AM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

...........really? duuuude go buy a bigger car.

Uspring

Wed Mar 11 2015, 10:02AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Shrad wrote:

if there is AC, wouldn't the field even go back and forth in direction? the projectile would then just vibrate, no?

No, it wouldn't, unless the projectile is a permanent magnet. See e.g. Significations remark:

F(b) = (k) * d/dx(B^2) joules; where k is a constant.

Force is dependent on B^2, thus doesn't change under sign change of B. The force depends on a gradient of B. The projectile is pulled into regions of larger B.

There are other issues with the design:
a) Energy: You have 2 Joules, which can accelerate an iron projectile 3mm diameter and 2.5cm length (mass = 1.4g) to at most about 50 m/s. The distance it will travel during a say 1us pulse time is just 50um.

b) Pulse time: The velocity increase is F * tp / m. To accelerate a projectile of 1.4g mass to 50m/s within a time tp of 1us you need a force of 7 tons. That looks unlikely.

In order to get by using lower forces you need much longer pulse times.

Shrad

Wed Mar 11 2015, 12:36PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

to get back to our critically damped oscillator

with formulas this time,

For L=10ÂµH and C=10nF

half period is 1/2*2pi*sqrt(LC) = 1Âµs

resistor for critically damped behavior = 2*sqrt(L/C)=63.25ohm (take 64 as nearest R value)

I'm desperately trying to specify the L value as a decreasing function of time but LTSpice is throwing up errors

do anyone here know how to correctly set a dynamic value component in LTSpice?

DerAlbi

Wed Mar 11 2015, 06:36PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Use arbitary behavioral source to implement the differential equation. Component parameters cant change during a simulation run
You could also use the flux-extension of the Spice-coils....

Signification

Wed Mar 11 2015, 07:51PM

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

Shrad wrote ...

I'm desperately trying to specify the L value as a decreasing function of time

I really think there is a way--it exists and is a simple concept...

Shrad

Wed Mar 11 2015, 07:52PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

I have been able to include a .func directive in a .subckt block with a value decreasing from 10ÂµH to 0ÂµH in a set period of time, but it lacked the rest of the models

I don't want to implement a differential equation at this time, just to approximate an inductance which decreases over time

component parameters can change during a simulation by specifying a custom model or package, but this is not that easy to find documentation on how to do it

Signification

Wed Mar 11 2015, 10:39PM

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

Shrad wrote ...

I don't want to implement a differential equation at this time, just to approximate an inductance which decreases over time

FWIW, I understand EXACTLY what you want to do and why.
Similarly, I just want to implement a specific current shape (through an inductor) which increases over time. I also realize this will be different than the standard differential equation approach.

Shrad

Thu Mar 12 2015, 08:47AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

I'm glad someone has seen where I want to go ;)

Do your current shape need to sharpen over time? If so you can achieve pulse compression with a succession of LC in a coaxial ladder design : http://www.slac.stanford.edu/cgi-wrap/getdoc/slac-pub-5432.pdf

Ideally I would like the decrease of L to be a direct function of the I and time values but I can't figure what would better approximate this

I'm first trying to simulate a constant speed (constant decrease of L) and once done I'll try to simulate a constant acceleration... would a square function of time suffice? I still need to read about it

DerAlbi

Thu Mar 12 2015, 10:00AM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

I gave the solution above. You dont wanted it. Now you keed stuggeling. WTF is wrong with you guys ?

first previous 4 5 6 7 next

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