Coil gun project / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Coil gun project

1 2 next

Move Thread

LAN_403

Jack

Sat Apr 30 2011, 10:58AM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Hi everyone,
I was assigned a project about "mouvement and mobility" and my team and I decided to work on coil guns for the reason that they are kind of new in concept as a propulsion system (or projectile accelaration system) and not so hard to actually build!
Since in the theoretical field, we can talk very little about "mouvement and mobility", because the magnetic field is nowhere near constant and an actual equation to calculate the force created by the magnetic field isn't available (correct me if I'm wrong), we decided to actually build one and mesure speeds, calculate kinetic energy and efficiency...etc.
So, we will basically be talking about the different components in the circuit and their functions, plus all the calculations we made to choose the different components. We will finally expose all the results we got from testing: efficiency from: voltage sweep, positionning, projectile length.

So our system is composed of:
- 2 x 6800uF 200V electrolytic capacitors wired in parallel which will be charged to a maximum of ~ 150V using a 220-110V transformer and rectifier bridge (and a series resitor of cours). (already tested)
My first question is: Is it normal for capacitors to bleed on their own? ( In ~ 4 hours I've gone down from 130V to 100V)

- A 200A-800V SCR (No other information about it is available)
Question 2: I understand that if I'm not critically damping my circuit, I need to protect the SCR with a diode wired in parallel to the coil in reverse bias. How should I choose this diode? Should I check in Barry's sim for the maximum absolute value of negative current and choose it like the SCR (i.e.: would a 25A diode be able to handle a reverse current of 250A?)

Question 3: For my triggering circuit, would a 4.7V battery pack be good enough if I'm using a 270 ohm resistor? (the battery pack I have is rated for 28A so no worries about drawing too much power) or should I be using a capacitor instead? I already have one (22uF and 50V) and a 7V AC-DC adaptor so I can easily install either one! I simply prefer the battery so I don't have to worry about creating an extention for my circuit to charge to secondary cap!

Finally, assuming we have 2 coils of the same induction, is it better to use a long coil with less turns or a short coil with more turns!

Thank you very much for your help!
I will be posting pictures soon!

Barry

Sat Apr 30 2011, 02:00PM

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

Welcome to the forum, I look forward to seeing what you come up with. With 136 joules in each of your two capacitors, you'll get some exciting results.

Jack wrote ...

My first question is: Is it normal for capacitors to bleed on their own? ( In ~ 4 hours I've gone down from 130V to 100V)

In my experience it's not unusual to have capacitors bleed like that. Slow voltage leakage is not dangerous by itself. They might get better over time, as the insulating layer can re-form, after a long period of unuse. Keep a close eye on it; if it gets much worse then it can be very dangerous.

Jack wrote ...

Question 2: I understand that if I'm not critically damping my circuit, I need to protect the SCR with a diode wired in parallel to the coil in reverse bias. How should I choose this diode? Should I check in Barry's sim for the maximum absolute value of negative current and choose it like the SCR (i.e.: would a 25A diode be able to handle a reverse current of 250A?)

In theory, that approach would work perfectly. In practice, you want a considerable safety margin. If it were me I'd get diodes with about the same total current-capacity as the SCR. Fortunately diodes are relatively inexpensive and you can add several in parallel if needed for the total current. By the way, it's better to mount them on the capacitor bank than on the coil.

Jack wrote ...

Question 3: For my triggering circuit, would a 4.7V battery pack be good enough if I'm using a 270 ohm resistor? (the battery pack I have is rated for 28A so no worries about drawing too much power) or should I be using a capacitor instead? I already have one (22uF and 50V) and a 7V AC-DC adaptor so I can easily install either one! I simply prefer the battery so I don't have to worry about creating an extention for my circuit to charge to secondary cap!

The SCR gate is triggered by current (not exactly voltage) so almost any combination of battery + resistor that provides sufficient current will work fine.
In your case, ohm's law says that 4.7v / 270 ohms will strictly limit current to 17 mA or less. This might be a bit low for your big SCR, but it's a good start and you can gradually reduce the resistance until it works. Just don't go too high; most SCRs risk gate burnout with about 50 mA or more.

Build an SCR test jig to experiment with your triggering circuit. That is, let your SCR drive a simple power source and a load of a few amps. I recommend a car battery and 12v light bulb load. This is much easier to test with than a coilgun load. Besides, I can almost guarantee that halfway through the project you'll want to know again if your SCR is still working. Start with your 270-ohm resistor, and swap in successively lower resistance until the SCR turns on.

Jack wrote ...

Finally, assuming we have 2 coils of the same induction, is it better to use a long coil with less turns or a short coil with more turns!

It is often said that magnetic field strength is proportional to "amp turns". This is the simple product of current and number of turns. There is one important detail, however. They usually just say "turns" but it is more accurately defined as "turns per inch". So yes, given two coils with the same inductance, that are driven with the same current, the coil with the largest turns-per-inch will work better. The tricky part is that inductance is proportional to the square of turns-per-inch, so you just have to be careful with your coil design. My experience has been that my best coils are built with thick wire and just one or two layers; but then I shoot at relatively low voltage, and your mileage may vary. Just get the inductance right and keep resistance low.

Good luck, Barry
ROCKET SCIENTISTS NOT AS SMART AS ORIGINALLY THOUGHT
New Findings in Study Commissioned by Brain Surgeons

Jack

Sat Apr 30 2011, 04:09PM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Thank you very much for your reply Barry! I'll keep you updated on the build and hopefully post some positive results later!!!
As for my final question, I didn't express myself corretcly! what I meant was layers and not turns! So I was asking about the difference between a short coil with more layers and a long one with less layers! It's because I already got some 2mm thick wire and I was wondering about the geometry of my coil!
Thanks again

Barry

Sun May 01 2011, 02:03AM

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

I'm totally lost regarding your last question (a slow brain day for me) so ask again more more details if something doesn't make sense.

Here's a random thought, probably completely unrelated to what you really wanted to know...

A frequent puzzle people have is "where do I start?" The whole coilgun design process is very circular because everything depends on everything else, it can be confusing to get started. Size, length, mass, voltage, capacitance, inductance, starting position -- you change one thing and then everything depends on everything else.

So my suggestion is to choose your projectile size and shape first. Figure out what physical object you want to launch and then start designing the coilgun around it. For example, once you know about your projectile, then automatically the coil length is known too (same length as the projectile). Also the coil's inside diameter is then known (big enough to fit around the projectile). Secondly, choose your capacitors. The capacitance (and total stored energy in joules) goes into most of the rest of your calculations without leaving too many unknowns or arbitrary choices to make.

Hope this helps, Barry
Technology makes it possible for people to gain control over everything, except over technology.

Jack

Sun May 01 2011, 04:48PM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Thank you for your reply, the answer I was looking for was:
"once you know about your projectile, then automatically the coil length is known too (same length as the projectile)"
I'm actually going to fabricate the coil with a random geometry but of course with the correct inductance and prove by experimentation that the best projectile would be almost the same length as the coil!!!

As for the general thought you described, that's actually what I have already done so I'm on the right track: I mesured at first the velocity of an air gun pellet using sound software and came up with a kinetic energy of 3.6J I then calculated the capacitance needed at 100V to achieve the same energy, assuming an efficiency of 5% and so I came up with 13500uF. what I found was 2 caps that add up to 13600uF which is perfect. having these two fixed values made it a lot easier to find the other values.
As for the protection diode, I was giving it a thought and then I stumbled upon a problem that I think is huge. Please correct me if what I'm going to say makes no sense:

During the discharge of a capacitor throught the coil, the current cycle goes like this
zero max zero min zero
as for the voltage stored inside the capacitors the cycle would be
max(+) zero min(-) zero max(+)

looking at both in the SECOND QUARTER-CYCLE we can see that, in the beginning of this interval, when the current is at maximum, the voltage inside the capacitors is zero, the coil tends to keep the current flowing and therefore charge the capacitors negatively!!! And this is not an option for electrolytic capacitors!
So a protection diode for the capacitor bank with the same rating as the SCR is actually neccessary and not a precautionary mesure!!! It will serve to short out the capacitors after they discharge (at the end of the first quarter cycle) and keep the current flowing in the coil until it dies out due to heating!!!
By doing so no further oscillation will occur because we will be in the case of an RL circuit, so no negative current, and the SCR will be serving just as a switch and won't block any reverse current.

Is what I said correct or not?

Thanks again!

Anyway I'm stuck until wednesday with no testing because my multimeter just got busted

I might have put it in a different mode while I was mesuring voltage and it started to give me false readings!
So until then I need to make everything clear so I can buy the right diode and resume work rapidly!
Cheers

[Edit]
[It seems I didn't fully understand the concept of putting a protective diode until I gave it a thought! Anyway I found the same explanation in another thread after I posted here (lol), well I guess I was too anxious to get an answer that I haven't looked elsewhere carefully!
10x anyway]

Jack

Fri Jun 03 2011, 07:01PM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Hey everybody!
It's been a while that i haven't posted anything!
Well here are some updates! I finished building the whole thing and it looks neat! however performance isn't how i expected!
The maximum speed mesured is about 25m/s with an efficiency of 0.7%
My guess is either the capacitors have high ESR or it's just because the projectile is made of normal iron and not silicon steel.
The specs of the gun are as follows:
- 13600uF, 200V Cap Bank (2x6800uF) (Maximum Energy is 272Joules)
- Coil is 2mm wire, inner diameter 8.5mm, 5 layers, 7cm long (100uH, ~60mOhms)
- 200A - 800V SCR for switching + protective diode with same specs
- Projectiles are 4mm in diameter and I have one supplementary projectile 5mm in diameter all 5cm in length
- The firing tube has an inner diameter of 5.5mm and outer diameter of 8.5mm and is made of cardboard or some kind of paper
+ all main circuit wires have been replaced with 1mm thick 1cm wide copper bars to reduce resistance!
The half-pulse length is ~4ms.
I need some advice on how to improve performance, i would be satisfied with about 2%, I'm trying to find a silicon steel rod to try and shoot it but can't find one, what common items contain such material?
My project is due in ~ 3 weeks after that i will be free to upload all my results with pics and maybe videos! Now I'm just a bit too busy with exams!
Cheers and thanks for the help!

[edit] by the way it's not portable, it's all mounted on a wooden plank.

Forty

Fri Jun 03 2011, 11:50PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

solenoid plungers

Barry

Mon Jun 13 2011, 12:06AM

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

Hi Jack, sounds like great progress, thanks for the update. I've been travelling on vacation and couldn't respond earlier, sorry.

Is your main goal "efficiency" or "velocity"? The approach to improving each of these will probably be quite different.

Jack wrote ...

The maximum speed measured is about 25m/s with an efficiency of 0.7%

Your coilgun is very good, but unfortunately, it appears that you're operating in a region where gains will be small. This is quite good efficiency for a single-stage medium-energy coilgun in air. You might be able to gain modest improvements by:

use thinner firing tube. Your 3mm wall thickness (6mm total) is wasted space, where a thin tube could bring the windings closer to the projectile.

a firing tube with thinner walls means either a larger projectile (recommended) or a smaller coil with closer fit

using lower power. Cut the energy storage (joules) in half, by lowering the charge on capacitors. This, in turn, requires tuning the coilgun again, by trying a variety of projectile lengths or different coils. Of course the velocity will be lower, but the efficiency will be higher.

For larger improvements in efficiency, I suggest changing from air to iron around the coil. I've had good results using external iron to improve efficiency up to 7% in a low-energy coilgun, see for example the Mark 2.

I know your time is short, so good luck.
Cheers, Barry
The difference between theory and practice in theory is less than the difference between theory and practice in practice.

Jack

Fri Jun 17 2011, 03:41PM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Thank you for the reply Barry!

Actually the tube has a wall thickness of 1.5mm not 3mm. (It's 5.5 - 8.5mm in diameter not radius)
I also found some 5mm-wide projectiles that will fit more tightly than the 4mm ones I was using, leaving a total of 0.5mm of free space.

As for introducing external iron, I don't think I have the time to do it before the presentation. However I have plans to improve the design during summer so I'll try to work on that.

I guess I have achieved what I wanted for the presentation, our supervisor was impressed that it even works but I thougth I'd ask if there was any quick fix to improve performance a bit in the 100V range! But it seems low efficiency is a common problem in amateur medium power designs!

The deadline for returning our research is the 28th of june and the presentation is the first of July! So I'll be back after that to post the results and get a few tips and advice on what to do next because I really liked working on coilguns:-D! Maybe I'll try external iron, different coil forms maybe a double stage coilgun!

Jack

Mon Jun 20 2011, 01:55PM

Registered Member #3841 Joined: Thu Apr 21 2011, 10:15AM
Location: Lebanon
Posts: 6

Just a quick question! I made some simulations for current and magnetic field and I got at 200V:

I_max = 1464 Amps and B_max = 4 Tesla for the air cored solenoid.

I just want to know if the magnetic field sounds reasonnable! I have already posted earlier the specs for my coil!

Also I read on wikipedia that even silicon steel saturates at a little over 1 Tesla so projectile saturation is basically the most limiting factor in coilguns since, based on my simulation, magnetic fields more powerful than 1 Tesla are very easily achievable!

Can anyone please confirm or correct what I said? And thanks!

1 2 next

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