Coilgun help / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Coilgun help

1 2 3 4 next

Move Thread

LAN_403

Cloud

Thu Aug 20 2009, 05:55PM

Registered Member #2311 Joined: Thu Aug 20 2009, 03:07PM
Location: Houston,Tx or Troy, NY
Posts: 19

Been browsing the forums here for a few months. I have built a few coilguns in the past with pretty good results. My next coilgun is going to be in the 2kj-3kj range of 1050V range. I am having a few problems locating parts and designing it.

Here is what I have - none of it is built but I have the parts
2-3kj range with 1050v (3x 350v 2200uf caps in series, 7-8 of these sub-banks in parallel)
1000v 900A (10000A surge) hockey puck SCR
1200v 300A (3000A surge) stud SCR
.125" copper Class 1 bus bar for the bank and connections
2000v rated wire 12AWG for bank connections
2000v rated wire 16AWG for volt meter/resistor connection/etc...
And... almost unlimited amounts of LN2.

What I am thinking of doing..
1/2" - 3/4" steel projectile a few inches long.
LN2 cooled 10AWG-12AWG coil
Lots of iron surrounding the coil, inside the LN2

I would like to get the highest efficiency while having a fast moving projectile.
My budget is becoming limited after all the purchases...

Problems
I am trying to find a way to protect my capacitor bank from being negatively charged. I have looked into the huge diode (which extends pulse time and not good for efficiency), the V-Switch (which looks the most promising), and the diagonal half bridge (from which I gathered I can't use due to MOSFETS). Solutions? This is my big problem..
My SCR's datasheets say both are rated -40C to +125C storage/operating temp range. Some have told me that submerging it in the LN2 should be okay. Big chance of it breaking from the extreme temperature?

Also any tips/tricks/info/ stuff to say is much appreciated and hope that my stay at 4hv will be a productive one.

Attached is a simulation in Barry's RLC circuit. It was calculated using the bank and a crude coil of a few layers and it does factor in the lower resistance of copper (of the crude estimate).
Also attached is some pictures I drew in CAD what it should look like taking what I thought were the best ideas of many coilguns and using them in mine. The LN2 will be around the coil. The iron is not in the picture right now.
Lastly attached is the datasheet for my 1000v 900A scr in case someone needed it.

]powerexc447p.pdf[/file]

klugesmith

Fri Aug 21 2009, 01:15AM

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

Hadn't heard of that one before.
If your LCR circuit damping ratio is low to begin with, then cooling the coil to 77 K won't make a huge difference.
On the other hand, with LN2 bath you can make a geometrically more efficient coil, that would be grossly overdamped at room temperature.

I see no benefit to chilling anything except the coil. For your interconnect wiring and semiconductors, it is not hard to get negligible resistance and voltage drop at room temperature.
I don't know about the low temperature magnetic properties of your iron core (and projectile?), but expect the low temperature will aggravate any eddy current losses.

Looking forward to hearing about your progress, and material properties research.
Rich
p.s. when I was in college, departments were billed for their LN2 at $0.07 per liter. What's it go for today?

PYLEMET-mk2

Mon Aug 24 2009, 04:44AM

Registered Member #1513 Joined: Sat May 31 2008, 12:08PM
Location: Russia Klin
Posts: 48

Cloud, why don't you made a multistage coilgun? It would be much more efficiency to use your 2-3kj

Cloud

Mon Aug 24 2009, 03:12PM

Registered Member #2311 Joined: Thu Aug 20 2009, 03:07PM
Location: Houston,Tx or Troy, NY
Posts: 19

I will be doing that as well. I will first build a powerful single stage coilgun because I can and to test my LN2/designs out on a much less complex system. After that, I will break up the sub banks and go for 3 stages or so of 6 capacitors a piece.
I thought 4hv was a much more active place then this.
@Klugesmith - I have no idea how much the LN2 will affect it. At the very least it should cause some problems and keep my coil from overheating. As for the LN2 costs, I think the professor said somewhere around 1$ a liter or something.. but I'm not paying for it.

Sulaiman

Mon Aug 24 2009, 11:03PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I have a couple of thoughts;
- LN2 may make the copper so brittle that it will shatter or at least fracture due to the magnetic shock, maybe a couple of quick tests ?
- Do a quick calculation or simulation of your intended setup to check two parameters for the Thyristor..dV/dT and most importantly dI/dT.
- my (often unreliable) gut feeling is that you will have to move to IGBTs for controlled ON times, especially when you move to multistage.
- I hope that the Iron will be laminated, strip or sintered or there will be MASSIVE eddy current losses.
- with kJoules of kinetic energy your FIRST thoughts should be where and how to test the setup, specifically how will you 'catch' the projectile and how will you measure and record all of the parameters involved? especially projectile velocity.
- EVERY part of the setup is a potential hazard, with respect to electrocution and/or explosion..make suitable arrangements!

This could be a very interesting experiment ... please keep us updated.

P.S. Regarding activity, it's your first posting here and commonly we have all kinds of 'pipe-dreams' from new members, hence often we do not get immediately involved. You will find that the expertise of members here is not only immense but free too!

Cloud

Tue Aug 25 2009, 01:11AM

Registered Member #2311 Joined: Thu Aug 20 2009, 03:07PM
Location: Houston,Tx or Troy, NY
Posts: 19

There will be many tests done with the LN2 as I have no experience. The copper breaking, eddy currents, cracking of plastic firing tube, freezing insulation off, etc.. will all be checked.

For the dI/dT, a ROUGH calculation (with rough being a linear calculation) from my posted RLC picture. If this is wrong, could someone do it for me. I am trying to learn CAD/FEMM and I gave up on spice after about 10 mins with it.)
Goes to 8500A in 1.58/2 ms.
8500(A) / ( 1.58/2 (ms) ) = 107594 (A/ms)
107594(A/ms) / 1000 (ms/us) = 107.584 (A/us)
(Math is right... Right?)

The Hockey puck SCR rated for surge of 9000A - 10000A (another source said 10000A was surge) with a dI/dT of 800A/us
The stud mount SCR rated for surge of 3000A (which will be used in the small tests) has a dI/dT of 100A/us.

If I do move to multistage, I will most likely dab in it with SCRs at first until I move into lower power/max efficiency and use IGBTs.
It will be iron from transformers.. so laminated E / I leaves.
I usually will shoot into a cardboard box with sheet metal backing. The cardboard will be used to prevent it flying somewhere after it hits the metal.
For measuring, I will try both sensing coils and photogates. These will be mounted outside of the LN2 box of course.

I know about the pipe-dream with new people... but be assure most of the stuff is already bought. My total budge spent so far is around 400$ and I am getting some of the stuff at really good prices. But since I already bought most of the stuff, I can't turn back now. Also, I don't want to fail my 4 credit class

For safety, I am thinking..
Hiding behind book shelf that is 5 ft away (there is book shelf middle of room, hiding behind it and having 10ft wire and handheld trigger.
Plexiglass the capacitor banks/scr so that in case of explosion, I will be (mostly) safe.
I will be having a discharge bank of resistors triggered by another small scr. This will be used to drain the bank in case of something happens. It should drain the bank in 15 seconds or so.

I have another question. Dealing with the huge electric fields, how far should I keep laptop away? Also, will the photogates/sensing coils be effected by it?

Keep the questions coming. It makes me think more about it and solves problems before they arise.

Cloud

Sun Aug 30 2009, 03:01AM

Registered Member #2311 Joined: Thu Aug 20 2009, 03:07PM
Location: Houston,Tx or Troy, NY
Posts: 19

bump
Any opinions/experiences would be a great help!

Steve Conner

Sun Aug 30 2009, 09:06AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Sorry, bumping is against the forum rules

Cloud

Sun Aug 30 2009, 01:17PM

Registered Member #2311 Joined: Thu Aug 20 2009, 03:07PM
Location: Houston,Tx or Troy, NY
Posts: 19

G. Avoid double-posting (FYI: double posting is posting twice in a row in a thread, without allowing any replies in between). Double posting just to get a topic marked as new is a no-no. Every topic deserves a chance. If yours slips into obscurity, its chance came and went. If 48 hours have elapsed, and you have some new information to add, you may double post. This is the only exception, there are no other exceptions! Posting an identical or nearly identical question in two individual threads is also considered double posting.

Sorry, thought it said bumping was okay after 48 hours. Won't do it again.

Steve Conner

Sun Aug 30 2009, 01:56PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

It's 48 hours AND new information to add. Bumping isn't information...

1 2 3 4 next

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