pulse width limiting max velocity? / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

pulse width limiting max velocity?

1 2 next

Move Thread

LAN_403

ramses

Thu Jan 03 2008, 05:52PM

Registered Member #1208 Joined: Thu Jan 03 2008, 05:30PM
Location: Chesterland, OH
Posts: 154

i have been simulating coils in Barry's LCR simulator for a while, an i was wondering if the half wave(when the current across the coil is 0) represents the the minimum time it would take for the projectile to get to the middle of the coil. so if it crossed 0 at 1 ms, and the coil was 10 cm long, would the max velocity be 5000 cm/sec, 50m/sec, or 162.5 fps. that's without the projectile. And as far as SCR's go, should i go by peak nominal current or the I^2t rating for fusing.

...

Fri Jan 04 2008, 04:28AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

As to setting the pulse length, it is really something best done by experimentation. There is a trade off, if it is too short the energy isn't transferred into the projectile as efficiently, if it is too long you start to get suckback. I usually aim to have the current back to 0 by the time the projectile is leaving the coil.

As to the SCR, I generally shoot to keep things around the peak current rating. It is rated to handle that for about 20ms, so you can probably run several times that but I quite like to keep my SCRs from exploding...

Barry

Fri Jan 04 2008, 04:38AM

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

That is the exactly the challenge to designing the discharge time ... we need to provide the energy over the duration in which the projectile is in the range of positions that it can absorb it.

If the discharge is too short, then it's not taking full advantage of the distance available.
If the discharge is too long, then it suffers from DSE (dreaded suckback effect)

My LCR simulator is great for visualizing the electrical aspect of the LCR circuit. But it doesn't include anything about the projectile's physical dynamics. To do so accurately will require treatment of several more variables such as mass, distance, magnetic coupling, magnetic saturation, initial velocity, and so forth. It's been on my wish list for awhile; I may get to it yet some day.

One such simulator has been built (and done extremely well I might add!) by Wm Harris, and is published on his www.gausspistol.com web site. It was last discussed on 4hv.org around April 29, 2007. Check it out! It is very nice work.

As for the specifics of your speed calculation, it might be fair (but still very crude) to assume constant acceleration of the projectile until it reaches the middle of the coil. In which case the position x(t) = 1/2 a t^2. And the velocity v(t) = a t. Use these to solve for an estimate of the maximum attainable velocity for a given pulse width.

Cheers, Barry
Barry's Coilgun Design Site

ramses

Fri Jan 04 2008, 02:05PM

Registered Member #1208 Joined: Thu Jan 03 2008, 05:30PM
Location: Chesterland, OH
Posts: 154

Barry, I'm honored :)
Now the only question is if that the the data sheet on some SCRs i found has the I^2S in amps^2/sec. can i push it to that limit for less than the 10ms, it specifies, or do i have to take the square root and multiply it by the time in seconds? thanks for confirming my theory! BTW that calculator is amazing!!! edit: the only bad thing is that it crashes sometimes (beta) and it assumes i can turn off my capacitors! any idea how to let it know that i can't? thanks

Barry

Fri Jan 04 2008, 03:41PM

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

For coilguns, I think the SCR's most relevant spec is I-TSM, the "nonrepetitive surge peak on-state current". This is the maximum surge current it can tolerate for a short period, typically specified for 8msec or so.

You can use I-TSM directly in your design work. Just compare your peak estimated current to this spec, and away you go. If your peak current will be a little higher then don't worry, SCRs are pretty tough, but keep some spare parts handy.

I've never used the I^2t specification, and heck I'm not even sure what it's for. I assume that it relates to some kind of heating effect but what? Anyone?

Cheers, Barry
PS - Can you send me an email and tell me a little more about how to make the LCR simulator crash, and explain the bit about turning off capacitors? I'm always looking for ways to improve it, and I have another version of it almost ready to publish.

ramses

Fri Jan 04 2008, 04:57PM

Registered Member #1208 Joined: Thu Jan 03 2008, 05:30PM
Location: Chesterland, OH
Posts: 154

Barry wrote ...

PS - Can you send me an email and tell me a little more about how to make the LCR simulator crash, and explain the bit about turning off capacitors? I'm always looking for ways to improve it, and I have another version of it almost ready to publish.

Your LCR simulator wasn't crashing (I'm not that good); it was the BETA version of the coilgun simulator (

)(which is fine because it is BETA). Your LCR simulator doesn't assume that i can turn the cap's off when the projectile reaches the middle. (I can't see IGBTs with that high of current + voltage. I'm thinking about using a solenoid spark gap at .9- 1.8kv, single stage, and I can't stop an arc...)

one last question: if you sweep the voltage on the gausspistol.com simulator, it also has to sweep either the energy or the capacitance - which one does it sweep? (hopefully energy, since charging a cap to a lower voltage doesn't increase the capacitance; it lowers the energy.)

PS: i can't wait for your new LCR simulator.

Barry

Sun Jan 06 2008, 07:19PM

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

Whew, glad to hear the LCR simulator wasn't crashing after all. I promise to try and publish my latest version this weekend. The changes are mostly eye candy for the presentation. The output area looks much more like an oscilloscope screen now, and the colors are much nicer. I started adding an option to simulate the anti-parallel protection diode, but it has some bugs so I'll defer that feature for now so people can start using what I've got so far.

Edit: The RLC Simulator page has been updated today. - bwh

As for your voltage sweep question, I don't have the experience to answer. I would expect it keep capacitance constant, as we would do in real life testing. But that's just my guess.

Cheers, Barry
When the going gets rough, the tough put a smoothing cap across V+ and ground.

ramses

Mon Jan 07 2008, 08:25PM

Registered Member #1208 Joined: Thu Jan 03 2008, 05:30PM
Location: Chesterland, OH
Posts: 154

(sigh) after using the gausspistol.com simulator, i have determined that i will put 20 teslas peak field strength through the projectile. since that is 9 times the saturation density, i have decided to move on to an inductance coilgun. if saturating the projectile has no negative effects (which i doubt), let me know. otherwise, mods, feel free to close this topic.

thanks for the help,
ramses

krenshala

Tue Jan 08 2008, 05:14AM

Registered Member #143 Joined: Sat Feb 11 2006, 04:25PM
Location: Austin TX, NorAm, Sol III
Posts: 28

I know this is old, but I had a question about your post, Barry.

Barry wrote ...

For coilguns, I think the SCR's most relevant spec is I-TSM, the "nonrepetitive surge peak on-state current". This is the maximum surge current it can tolerate for a short period, typically specified for 8msec or so.

You can use I-TSM directly in your design work. Just compare your peak estimated current to this spec, and away you go. If your peak current will be a little higher then don't worry, SCRs are pretty tough, but keep some spare parts handy.

If I'm reading what you say correctly (and I'm almost as far from being an EE as one can get ;)) the SCR can be in the "non-repetitive surge peak on-state" for approximately 8ms before it really runs the risk of damage/failure. Do the specs list the minimum time between one surge peak on-state and another that still allows the subsequent surge peak on-state to qualify as "non-repetitive"?

And, am I correct that this time would directly affect the theoretical maximum rate of fire for the coil gun. This of course is ignoring the loading mechanism bottleneck, assuming infinitely fast reloading and only one SCR in the circuit.

A curious ...

Barry

Tue Jan 08 2008, 04:21PM

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

krenshala wrote ...

Do the specs list the minimum time between one surge peak on-state and another that still allows the subsequent surge peak on-state to qualify as "non-repetitive"?

For any pulse that re-occurs, the specifications have a maximum RMS (root mean square) current limit. The RMS spec represents a limit on the temperature rise in the junction.

The exercise of computing an RMS value from your waveform is left to the reader.

From that, it is possible to determine the shortest interval in which you can fire again.

Cheers, Barry

1 2 next

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