my first coilgun, advice would be appreciated / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

my first coilgun, advice would be appreciated

1 2 next

Move Thread

LAN_403

Scanlight

Thu Jan 19 2017, 05:25PM

Registered Member #61389 Joined: Sun Dec 25 2016, 11:49AM
Location:
Posts: 4

Hi guys.

I'm new to this forum and would like to share my coilgun design.
Maybe you could give me some advice on how to improve the velocity.
Usually i build only small Arduino projects for a hobby, i know the very basics of electronics but i'm not a genius haha

I'm planning to build 6 stages. Optical triggered.
I have one coil up and running using a 4460uf capacitor bank (for first stage) Edit: wrong capacitor capacity calculation
The capacitor bank is built up from a lot of 470uf caps. (2x200v)
I use a kp50A scr as a switch. (Capable of 1200a)
My coils are 5cm long, 8 layer, 16 gauge wire. (1mm)
As a barrel I use a carbon fiber rod. 10mm outside, 9mm inside and will shoot 8x40mm bullets. (Tried multiple bullet dimentions, all have the same velocity so i stick with the most heavy one)
Only the first stage will result in a velocity of 20m/s.

some pictures of my project.

the full project will be build inside a modded Nerfgun (type: Longshot)
i paint it black to see how it will turn out, but its gonna get another paint job. this looks way to real.

i designed a basic pcb to control the system.
it will monitor the capacitor bank voltage and indicates when ready to fire.
some special FX are added like sounds and lights.
later it will be able to switch the charging system.
At this moment i charge the capacitors to 320v (bulb/rectifier)
I have a special capacitor charger so i can charge it to 380v but its slow. (takes minutes to charge)

at the moment i'm trying to figure out how i can trigger the second stage, i already added a second coil and capacitor bank (2430uf)

This project started with playing with coils and electromagnets just to learn more.
My most recent video:

...
Got my second stage working with an ldr light gate.
The second stage boost the velocity from 20m/s to 26m/s
Not a massive improvement
With kind regards
...

Robbert Loos

[Edit: Double post]

klugesmith

Fri Jan 20 2017, 02:07AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Welcome, Robbert!
>> The second stage boost the velocity from 20m/s to 26m/s

Why is that bad? Stage 2 added 70% as much energy as stage 1, and probably did it in less than half the time!

Bjørn

Fri Jan 20 2017, 02:07AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Typical LDRs are slow so possibly not the best choice as the light detector.

Remember that the energy rises by the square of the projectile velocity so your second stage added almost 70% of energy to the projectile, so not bad.

[Simultaneous posting]

DerAlbi

Fri Jan 20 2017, 08:46AM

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

I would even step up the answer: given your cap banks started at the same voltage, the second stage delivers 70% of the kinetic energy while consuming only 54% of the electrical energy. So everything is double fine.
However putting things in perspective your ~3J kinetic energy achieved with ~350J electrical energy leaves room for improvement.
...except if you got your cap bank wrong:
470uf caps. (2x200v) while suggesting you used 320V means you put those 470uF in Series so you got 235uF per series string. This implies you put 19 series capacitor connections in parallel using 38 capacitors in total.. is that the case?
(i just double check your math)

Scanlight

Sat Jan 21 2017, 12:23AM

Registered Member #61389 Joined: Sun Dec 25 2016, 11:49AM
Location:
Posts: 4

Thank you for the reply,

not all capacitors are equal in size. most of them are 470uf but mixed with 680uf and 560uf.
i salvaged them from old power supply's, most of them contain two pieces.
Rated 200v so i connected two of them in series (equal values).

Tomorrow i can draw a schematic of my total wiring.
The "charge" wire goes trough a diode to the capacitor banks, so all will be charged equal.

didnt know LDR's have a slow response. im gonna check if there is an option to use IR.
In this case i think the LDR is ok, but the next stages need faster response.

Reply to DerAlbi, the Bank UF is calculated in strings of 400v.
so in this case, 400v 4460UF total power.

a Little better picture

The capacitor bank has a strange shape but i needed them in this order to fit inside the casing.
See this picture for the capacitor arrangement.

Yes its not the best soldering, it was a hard time to solder the solid copper wire's.
in my opinion i could not use enough tin haha
and yes a lot of solder/desolder for testing purpose.

With kind regards
Robbert Loos

DerAlbi

Sat Jan 21 2017, 05:46AM

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

Your second picture worries me.

You got 4 caps with 470uF circled.
2 in series, 2 of those strings in parallel.
This is an overall capacity of only 470uF, not 940uF.

You somehow neglect that 2 caps in series half the capacity.

Scanlight

Sat Jan 21 2017, 12:32PM

Registered Member #61389 Joined: Sun Dec 25 2016, 11:49AM
Location:
Posts: 4

I draw a simple schematic with the same capacitor locations to make it easy.
yes your are right, i learned something today haha
didnt know 2 caps in sersie can hold only half the capacity
so the cap bank can hold 2410 uf

klugesmith

Sun Jan 22 2017, 03:56PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

>> i learned something today haha
>> didn't know 2 caps in series can hold only half the capacity

Good lesson. It works that way because capacitance is a measure of how much charge you need per volt.
In a series pair, the terminal current goes through both caps at the same time. So the terminal charge (current x time) accumulates in both caps at the same time.
The charge to bring series pair to 1 volt would bring an isolated capacitor to 1/2 volt.

Speaking of which, you were wise to choose equal-value capacitors for series pairs. That's not enough to balance their voltages.

In a charged bank, the partition of the bus voltage in each pair will depend on the relative leakage resistances. Please check that with a voltmeter.

DerAlbi

Sun Jan 22 2017, 09:43PM

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

The first step in that regard is to connect the center of each series connection to every other center. In that way statistics works for you. Maybe. Not sure if that matters in that build however

Scanlight

Wed Jan 25 2017, 07:26PM

Registered Member #61389 Joined: Sun Dec 25 2016, 11:49AM
Location:
Posts: 4

ok thank you, i'm gonna try this tonight.

have to glue my coils back in place. they moved out of place due to the recoil. this time i used an excessive amount of glue.
Yesterday i received my new capacitors Vishay 450v 680uf (20 pieces)
used them with 4 in parallel (2720uf) but achieved a very poor speed. 10m/s max with one coil (compared with 20m/s with the salvaged caps.)

Thanks you guys.
i learn a lot at this forum.

1 2 next

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