6 Month Gauss Cannon Project / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

6 Month Gauss Cannon Project

first previous 5 6 7 8 next last

Move Thread

LAN_403

blackgrunge

Sat Jan 03 2009, 10:33PM

Registered Member #1889 Joined: Mon Dec 29 2008, 07:36AM
Location:
Posts: 55

One bank of capacitors works really well with a power supply that charges with 10 Amps and is left on constantly charging the capacitors, not only between firing but also between coils.

I'm using 6 coils and the bank will power them all, this is the reason behind needing a 5kJ power source so that at the end of the chain I won't out of power. With the power supply replenishing a little bit of the energy I will be able to maintain a fairly decent output over 6 coils. Having seperate capacitor banks is very wasteful and I'm trying to achieve high efficiency.

Slotting comprimises the intergity of the barrel. Why would you ever go through all that trouble when you could get as guitardude said, a polycarbonate tube.

Edit: Using osmium oxiode epoxy could be very useful. the reason why other heatsinks don't work is because they have poor surface area contact. if the coil is dipped in os oxide epoxy then left to harden in a mold it will work quiet well but just slapping a heat sink on won't do anything

I'm taking the pictures to give you guys a better idea around 7 or 8 pm EST today because i'm currently out of the house.

guitardude012

Sat Jan 03 2009, 10:36PM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

blackgrunge wrote ...

@El_Roberto

Standard 3 inch nails with the heads removed but I'm in the process of getting a Mu Metal projectile made. I looked at the v-switch and it not even close to what I'm using. The v-switch quenches and I have no reason to quench using an SCR because I have a quenching diode for that rated for 4000 amp pulses and has a 800v reverse voltage breakdown rating.

The capacitor doesn't detonate because it pulse through the SCR immediately shutting it off. It basically shorts out the SCR and by the time it is done discharging the SCR is off so it doesn't become negatively charged. I think you may be thinking that the capacitor is negatively charged but its not. It is charge normally and pulsed through the SCR in the opposite polarity as the current is flowing through it. If you copy and paste the link mentioned earlier it might help you understand it a little better. Warning though, using electrolytic capacitors may cause detonation if they aren't able to turn the SCR off. That is why a ceramic capacitor is used instead.

Sorry I wasn't around because of the holidays to elaborate on the v-switch after my initial post. It is not negatively charged, usually it is charged to approx. the same voltage as the main bank. If you time things right with a V-switch you can actually achieve higher peak current just before quenching.

edit: Did you mean you are using a ceramic capacitor or those building V-switches(like EV) are?

blackgrunge

Sat Jan 03 2009, 10:44PM

Registered Member #1889 Joined: Mon Dec 29 2008, 07:36AM
Location:
Posts: 55

I'm using an electrolyic but its charged to 330v so i'm positive it will turn the SCR off. What I was saying is if it doesn't because you didn't charge it to a high enough voltage, the line voltage of the firing SCR will cause it to detonate

guitardude012

Sat Jan 03 2009, 10:52PM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

blackgrunge wrote ...

I'm using an electrolyic but its charged to 330v so i'm positive it will turn the SCR off. What I was saying is if it doesn't because you didn't charge it to a high enough voltage, the line voltage of the firing SCR will cause it to detonate

Gotcha. Capacitance is also important... too much and you lose efficiency, too little and you fail to turn the SCR off. This is hard to deal with I have found. Several parameters like coil resistance, inductance, and SCR shutoff time may remain close enough to "constant" to ignore after initial setup. However, the voltage of the main capacitor bank when you decide to quench depends on both initial charge, as well as how long you wish the firing pulse to be, and so you get stuck with a small amount of wiggle room before swapping caps

Caps rated to something like 450v 47uF are slightly expensive to keep replacing.

rp181

Sat Jan 03 2009, 11:15PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

A 10 amp charging supply is too powerful. With such charge rates, you will exceed the dv/dt of the SCR. I have exceeded it with less then 5A input. exceeding the dv/dt of the SCR does not damage it, but it will cause the SCR to trigger with no gate charge. Worse yet, you could exceed the di/dt from this false triggering, which destroy's the SCR. I am speaking out of experience here ( i destroyed 3 3/4in stud SCR's with a surge rating of 6kA for 10ms, the dv/dt and di/dt where high on these too). Your method will be putting lots of stress on the SCR, so you should employ the gate drive recommended for pulse power (Current spikes at 4A/us to 10 times the gate voltage, stays long enough to open, falls down to rated gate voltage, and stays on for the pulse duration).

Did any of your capacitors come with cosmetic defects? a couple of mine came with dents.

guitardude012

Sat Jan 03 2009, 11:25PM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

How are you protecting the charging circuit that is on during firing?

Edit: It seems at 10A input(this is before voltage multiplication?) you aren't accomplishing much in terms of charging between/during each stage firing considering you are peaking at over 2kA per coil(guessing)?

rp181

Sun Jan 04 2009, 12:17AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Keep in mind that it is constant current, rather then 1 pulse of current. Even if 10A is before voltage multiplication, it is still too much current to be safe (assuming your using a bank voltage of 400v. 800v and above would probably be fine).

guitardude012

Sun Jan 04 2009, 12:29AM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

rp181 wrote ...

Keep in mind that it is constant current, rather then 1 pulse of current. Even if 10A is before voltage multiplication, it is still too much current to be safe (assuming your using a bank voltage of 400v. 800v and above would probably be fine).

Oh I wasn't trying to say if it was safe or not, I was just wondering how much charging was really being accomplished between stages.

Edit: Also, this will make efficiency very difficult to calculate.

rp181

Sun Jan 04 2009, 12:33AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

By safe i meant the ratings of the SCR's. If it works, i think keeping the charging supply on will help in the later stages.
With keeping the charging on, how are you going to measure input energy accurately to get efficiency measurements?
EDIT: Doh, beaten to it.

blackgrunge

Sun Jan 04 2009, 01:24AM

Registered Member #1889 Joined: Mon Dec 29 2008, 07:36AM
Location:
Posts: 55

your obviously don't grasp what is going on a 10A charging input doesn't do anything except charge the capacitors it has nothing to do with the gate....

dv/dt is a calculus function lol. it mean the derivative of voltage with respect to time so it means nothing when related by the SCR. And the SCRs are only ever on for fractions of a second then they are shut off lol.

The voltage multiplier has an output of 330v @10A max. limited by capacitors in series with the AC side of the line. @10 amps my capacitor bank can be charged in around 15 seconds so the energy being dumped into capacitors is substantial. Guitardude as far as measuring the energy input I'm still working on those fine details...

Dealing with a voltage multiplier that has a 10 Amp output isn't unsafe at all. As the capacitors charge the resistance builds up until the voltage makes it up to 330v, then the output goes to to only a few milliamps to replenish the capacitors leakage voltage.

Excuse me if I come off a bit blunt but rp181, charging doesn't work the way your talking about it. The capacitors limit the current. The 10 amps is only the maximum output of voltage multiplier. The charging circuit doesn't need to be protected. Like I said the capacitors on the line side of the AC input limit it so that it the wall socket won't start dumping insane amounts of current into the coils. The voltage multiplier is in parallel with the capacitors so its polarity is exactly the same. The resistance of the fully charge capacitors that are in the voltage multiplier when its on are far greater than the resistance of the coils so its not like the pulse of the main capacitor bank is going back through the charging circuit. I've fired with the voltage multiplier on about 35 times and it doesn't become damaged. What your saying is not realistic.

And yes I already know that a voltage multiplier sends a constant current, I have a college level education in electronics and magnetodynamics. I wouldn't just randomly start leaving things on if i didn't know what the effects were going to be.

I know the capacitors won't being replenished very much by the multiplier during firing. The reason that it is left on is to begin charging as soon as the last coil shuts off. (optical triggering).

first previous 5 6 7 8 next last

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