Zener vs Fast-Act Diodes / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Zener vs Fast-Act Diodes

previous 1 2 3 next

Move Thread

LAN_403

klugesmith

Thu Dec 13 2012, 04:38PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Peace, guys. Let's put a damper on bad-mouthing other contributors.

@Turkey9: I applaud your use of initial conditions in simulation, instead of putting in a charging source and a switch. Great idea. Too bad your LTSpice inductor model's parasitic resistance value, important in this discussion, is not visible in the drawing.

Turkey9 wrote ...
Think of it this way; the fastest way to dissipate the energy stored in the magnetic field after the power has been cut off is to simply short the coil. A short has a voltage drop across it of zero. ...

I think your simulations confirm that the current stops faster with zerner diode, because the "braking" EMF is substantially higher than just I_coil * R_coil.
Shorting the coil (e.g. with a MOSFET) is how you MAXIMIZE the current pulse duration. A practical and extreme example is superconducting coils with superconducting shorting switches.

Some of you might be interested in this old thread about energy recovery in a coil pulser, without switches that turn off while conducting.

Its last post (before now) has a LTSpice simulation by me.

Ash Small

Thu Dec 13 2012, 07:25PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

That's a pretty impressive use of LC circuits Rich.

Your simulation:

.

demonstrates it well.

It's interesting to see how they solve this problem at CERN.

I've been working on a circuit to pulse a magnetron based ion source using a valve (vacuum tube), to inject protons into an accelerator. This circuit 'may' be relevant to my project, I'll have to look into it some more.

Pinky's Brain

Thu Dec 13 2012, 07:50PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Yandersen wrote ...

So far I haven't seen zeners over 5W of power.

TVS diodes are basically zeners and are designed to handle a lot more power.

Turkey9

Fri Dec 14 2012, 08:55AM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

Wow, a couple errors in my sims through off all the results. Thank you very much for correcting me! Here is a shot of the sims fixed, showing as you explained that a simple diode creates an infinite current die off. The zener circuit does cut it off extremely fast. I also ran the sim with TVS diodes and they worked exactly like the zeners.

As an interesting note, this circuit seems to create an extreme amount of ringing after the current has fallen to zero. It never reaches the full amplitude of the capacitor initial charge but could still mess with any control electronics.

Also an observation from the sims showed that the larger the voltage drop across the zener, the faster the current died off but the less current actually flowed through the zener. This agrees with conservation of power but seemed a little surprising to me at first.

Here are the pics of the sims that I ran. Previously, I forgot that the capacitor would be disconnected from the circuit after a half period of the oscillations. That's why the first sims I ran showed that current was left in the system and the inductor continued to ring. This was the biggest problem I saw with the circuit. Good thing that that was an error!

I would also like to apologize for accusing you of an error when in fact it was my mistake.

Adrenaline

Fri Dec 14 2012, 02:18PM

Registered Member #235 Joined: Wed Feb 22 2006, 04:59PM
Location:
Posts: 80

Turkey9 wrote ...
Here is a shot of the sims fixed, showing as you explained that a simple diode creates an infinite current die off.

You can get an even faster delay with a series resistor. I don't have a lot of confidence is SPICE solving freewheeling diode problems and ideal components. We know from experimentation the current does not asymptotically approach zero, it gets there quickly with just a freewheeling diode.

Yandersen

Fri Dec 14 2012, 07:03PM

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

The only difference between putting zener instead of a rsistor is that with zener voltage spike' amplitude is controlled - it almost independable of the current.

P.S.: it is very sad, that you guys implement an LT spice to understand operation of a simple circuit with 3 components, even though doing it wrong sometimes. It is not surprising that american coilgunning sucks. Well, putting zener in series with diaper diode will do it a little better... :) Who is going to be the first to build a coilgun with zener? Maybe, in next century you will discover the better way to wind a coils? :)\

Turkey9

Fri Dec 14 2012, 09:32PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

No reason to sling insults. Being able to use tools to visualize circuits is MUCH better than guessing at what something will do. I don't know of anyone that enjoys being called stupid, so cut it out. Also, don't make this a nationality issue.

I think that TVS diodes will be the way to go. Here is a link to a 400W one with 200V standoff voltage that are only $0.40 a piece.

I'm going to use some on my 18 stage coilgun I'm working on currently. I'll keep everyone updated with the results.

Ash Small

Sat Dec 15 2012, 12:50AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I second the 'insult' thing. It's starting to get a bit much.

Just prove your design is better...That's insult enough to those who can't build one as good.

You don't want the mod's getting involved here....neither do we....You have some good ideas, Yan, don't get youself banned from posting here.

Yandersen

Sat Dec 15 2012, 09:13AM

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

I looked into a datasheet and it did not impressed me. What we need is something bigger in size - like in SCR's metal case if it exists. Just imagine like hundred Joules of heat will be dissipated during 100us pulse - the heat will not have a time to go around a semiconductor crystal and will warm it up. The temperature must not go over the critical limit (over few hundreds degC). Transient voltage suppressors are the good direction of looking, but power rating must be a few dozen times higher.
And here is the question comes about paralleling of zeners: if warming decreases zener voltage, than paralleling will not help. Does it?

Ash Small

Sat Dec 15 2012, 11:39AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

It looks like the max junction temperature for those is 150C. Cooling (heatinks, etc.) will be of no real use whatever the package for pulsed operation because, as Yan points out, the heat has no time to dissipate. I assume some of the stud zeners I linked to above have a larger 'die' (if that's the correct term), and should, therefore, not get as hot for the same power dissipation. The power has to be dissipated as heat, so I'd assume that a package with a larger die size would be preferable here.

previous 1 2 3 next

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