Double Half Bridge - Worth a shot? / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Double Half Bridge - Worth a shot?

1 2 next

Move Thread

LAN_403

DerAlbi

Tue Jan 01 2013, 06:49PM

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

Hi guys... i need to discuss an idea, that doesnt let me go: having 2 Halfbridges!

One wouldnt need to wait until the first coil's SCRs shuts off, instead, when you turn off the half brige you turn on the second one. In this way the second coil can magnetized to pull the projectile much shorter after it leaves the first coil. This also utilizes the barrel better because there are now more stages per length.

Usually the Halfbridge is timed, that the current reaches zero, when the projectile is in the center of the coil. Now, the second Half bridge can pull the projectile (or at least build up higher current) while the current in the first coil is still decaying. This also removes losses of cap-ESR since the current is passed from one coil to the other directly - quite a big shortcut.

I also expect the average current pulling the projectile is higher - and in this way reducing copper losses, since the current can be smaller -> more stages possible

Is this all stupid talking?

Yandersen

Wed Jan 02 2013, 04:12AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Why two stages only? Utilize the whole length of the barrel by many stages each of them driven from own halfbridge controller and optogate while sharing the common cap's bank. Try to design compact independent optic feedback-controlled stage module easily mounted on a barrel and then duplicate it to create an efficient multistage coilgun.

DerAlbi

Wed Jan 02 2013, 02:51PM

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

Thats why: (its just my head, not a prooven theory - that would be the reson of this thread)

1) You only need two halfbridges, because the individual coil can be selected via SCR. So if you have only two coils at a time that are active, you need only 2 half bridges.
2) Making the coils shorter, to build a true moving magnetic field results in some problems:
2a) you will need some Inductance. The shorter the coil the thicker it gets. As of my knowledge this ist not good for the coils poerformance. (?)
2b) you dont need to have more than 2 coils active at a time: The coil current needs to be zero at the time the projectile reaches the middle of the coil - or suckback occurs. A third coil would be 2 times the length of the coil away an so it has "no" magnetic influence to the projectile. (?)
2c) Is the coil shorter than the projectile, the projectile will be in the middle of the coil an - in a magnetic way - it will stay there. So the inductance of the coil is verry high, and the current will not decay, so suckback occurs. You need to precicely time the current (ok, that needs to be done anyway) but the reduced inductance (shorter timing) builds up less maganetic energy. So less force with the same current (??????? verry unsure)

If i missunderstood you, i am sorry, if you do not want to shorten the coils, then i verry much doubt, that a third coil would have any influence.
This is should only be a way of pushing the average pulling force to a higher level.
Also, making more Halfbridges, reduces the current in each coil. I can parallel my IGBts to a 5kA-Haldbridges or two 2.5kA Halfbridge, or three 1.8kA and so on...
i know, i reduce the losses by reducing the current, but switching on coils without any influence due to the distance issnt much less wastefull.

If you can explain how shorter (or far way) coils can be of benefit, maybe i would agree to the idea. however you only need as many halfbridges as simutaniusly active coils.

My projectile will be 4cm long and weight 20g.

DerAlbi

Thu Jan 03 2013, 05:35AM

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

Hmmh. The whole thing seems to become a discussion about what coil length is good... if shorter ones are ok, there will be a practical minimum length.

Also the number of active inductors at once divides the maximum current by this value. It needs to be dicussed if the "missing" current in the coils is (hopefully over)compensated by the parallel action..

I really do have the parts for this (at least two half bridges @ 2.5kA) for a third one i will need additional diode pucks. But more than (or even) three active coils at once does not seem usefull, with a 4cm projectile.

Does anyone have experiences with coil length equal to the half projectile length? How fast does the current decay in a half-bridge, if the proejctile is still in it Is this thinking relevant??

Yandersen

Thu Jan 03 2013, 11:43AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Oh, Holly Sousage...
I really got tired of repeating that the ideal coil shape which is the one creating the strongest magnetic field while dissipating the minimum amount of heat is the short one which has length equal to inner diameter and outer diameter equal to 3 inner diameters. And the best length for projectile, acceleration efficiency-wise, is the same short as coil - length equal to diameter. And you don't need kA for single stage, especially for halfbridge - few stages operating at few hundreds A will push pretty much J into the bullet.
So the basic operation of halbridge driven stage is follows:
1) when optogate, located just before the coil, got shaded by a front tip of the projectile, open both IGBTs for x microseconds to energize the coil;
2) as soon as current reaches required value, close one of the IGBTs - the current will stay inside the coil cycling through the other transistor while projectile moves inside;
3) when optogate got lighted again (bullet approaching the middle of the coil) close second IGBT to let current leftovers back into the source.

In practice you need thick wire for your stages to minimize inductance to make sure energizing and deenergizing times are as short as possible.

If you will be able to build such a halfbridge you may beat my 20% efficiency record. ;)

DerAlbi

Thu Jan 03 2013, 01:21PM

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

Sorry that you get tired^^ hope clicking the ReplyButton helped your recovery. Maybe its time to put that in the Wiki

So my outer Diameter of the barrel is 12mm.
Lets say coil:
- d_in = 13mm.
- d_out = 40mm
- length = 15mm
- Select wire gauge according to the reulting volume^^

If i switch in your way (seems reasonable, but hard to manage) it dictates a verry low inductance. I ask myself (and you) if this does not reduce the force to the projectile: in my experience an electromagnet gets stronger with every turn. (or increased current)

So multiple Halfbridges seem to be a good idea

The question is: how many coils should be active at the same time...
using your coil-shape -> how far is a "usable" influcene to the projectile possible. Your switching dictates that the coil is magnetised when the proejctile is right in front of the coil. Is is bad to fire the coil earlyer? (in terms of kinetic energy - efficiency is not all the fun)

What would a longer projectile do?

Lets say i use a 4cm projectile (just because it has a big mass) i need more accelerators, yes, but would it really influence the transfered energy per coil?

Maxwell

Thu Jan 03 2013, 01:31PM

Registered Member #8497 Joined: Tue Dec 04 2012, 06:24PM
Location:
Posts: 74

Maybe I'm missing something with this half bridge circuit. I've been doing some reading on inductors and their physics.

Yandersen - could you put together a schematic of this proposed half bridge?

Maxwell.

DerAlbi

Thu Jan 03 2013, 01:42PM

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

Here a topology with 3 active coils at a time.

This coils create a cool moving magnetic field if controlled properly. The question is: are 3 halfbridges needed.. all this depends on the coil length and such stuff i want to discuss. I also do not trust the proposed low inductance in terms of force to the projectile - but i can be completely wrong.

Yandersen

Fri Jan 04 2013, 09:05AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

DerAlbi, your SCRs are shown upside down.
Actually, you are not the first one to discover the matrix way to connect coils - I know two other guys showing me that trick. But as far as I know noone has tried to build it yet.
IMHO, each module has to be independent - just a simple plain halfbridge (2 IGBT, 2 diodes and a coil). Cost issue? Mind twist, from the other side, will lead you nowhere - trust me.

Maxwell, see the schematic DerAlbi shown - that is the one except odd SCRs.

DerAlbi, the longer the projectile, the longer the pulse - more energy is dissipated as heat. Pulse time must be minimized to achieve higher efficiency. It is better to increase bullet's mass diameter-way rather than in length - pull force to mass ratio will be higher that way.
Yes, your goal is to minimize inductance to minimize coil energy-fill time. No matter which diameter of the wire you choose to wind the coil, power dissipation rate will be the same for all coils of the same geometry containing the same amount of energy initially. So check your IGBT's current rating - that is the goal for current coil will operate at. Coil energy must be around 10 times higher than expected kinetic energy achieved. From current and energy determine inductance (L=2*E/(I*I)) and the wire diameter. Now you can predict how much time it will take for coil to energize assuming voltage at cap's bank will not drop (t=L*I/U).

Go ahead, build that thing - be the first! :)

DerAlbi

Fri Jan 04 2013, 10:17AM

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

Haha, sry for the SCRs

So, lets say i cut my projectile to 2 halfs - thats ok.
-> 20mm long, 10mm in diameter - close enough to your suggestion.

my IGBTs all togehter can handle 5kA in a halfbridge. I can either split them up in 2x 2.5kA bridges oder 3x 1.7kA and so on.

A big open question in my head is: how long are the coils for a given number of bridges?

Lets say i have 2 bridges: would i half the length of each coil?
Or ist ist better to use only the half length when i have 3 bridges? So one coil can be ahead of the projectile, and the other two form a kind of a longer coil, while the last of this coil shuts down...

is the "shorter coil"-idea misplaced here?

Edit1:
here is a picture of waht i have in mind:

The question is: what would be a better arrangement? I notice that there some moments where suckback occurs... the spacing between the coils is min. 5mm. (optical trigger fiber + some mechanics )

Edit2:
I have played around with some coilshapes:
- the more turns, the higher the force. (short coil: bad)
- the more current, the higher the force. (long coil bad, slow curret rise)

Yan mentioned that the current need to rise quickly - i feel that this approach relaxes the issue and so i can have more inducatance through a bigger outer diameter (and so achieve more pulling force). Its really sad, that i cannot place triggerpoints inside the coils

this would give much more space for optimization.
Currently i feel that i need (a lot of) short coils, that create a moving field. eg for a 4cm projectile, 4 active coils (each 1cm long) would be great. however there will be a "big" gap in between the coils due to optical triggering.. so in the end: this sucks

Any chance of "slim" optical triggering?

1 2 next

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