Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Dr. Slack, Tue Mar 10 2015, 07:03PM

When I designed my coils, cap bank, projectile, and dampening resistors, I heavily relied on barry's RLC sim and inductor sim, as well has his other documentation. I found according to his sim, I would need a dampening resistor on the order of 300 milliohms for the primary coil for example. Through testing, I've found increasing /decreasing the inductance of the coil has done very little to change the speed of the projectile, but decreasing the dampening resistance to 50 milliohms resulted in a 100% increase in projectile velocity.

are you including the resistance of the coil in the total damping resistance?

Obviously I can't just hook this up to a scope (or can I?)

That's the way it's normally done, it depends what you mean by 'just'. Just make sure that your scope probe and scope input have adequate voltage rating, and also make sure that you understand which terminals are ground, and how you are going to handle implicit connections between grounds. That means either grounding CG and scope, floating one or the other (safely), or using a differential probe. Most 2 channel scopes will allow you to display A-B, to get an effective differential probe. Almost all scopes and probes will be fast enough for CG waveforms, even home-built probes if you don't have adequate shop bought ones.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Tue Mar 10 2015, 08:56PM

Yep, I took into account the resistance of the coils. I'm using 10awg magnet wire and the first coil is comprised of about 11.75m which is a resistance of ~38milliohms. The ~300milliohm dampening resistor includes the 38milliohm coil resistance.

My capacitor bank voltage is 520v and the max input to my scope is 400V. I guess I always assumed that since I'm dealing with a current pulse in the 2KA range I'd want to keep that away from my scope . . . but since a scope is essentially a voltmeter over time, that doesn't matter. I feel like an idiot lol.

To compensate for the voltage issue, I guess I would need a 10X probe?

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Tue Mar 10 2015, 09:08PM

How did you determine your L? I agree 100% with Dr.Slack (which shortens my reply). How about a simple series high-current shunt to a scope? Since this is a pulse the shunt won't have to be rated nearly to your current, but isolate it. Decreasing R can quickly bring on oscillations and a faster pulse, however, the pulse is also increased in amplitude/energy--perhaps your projectile escaped before any current-drop could significantly suck it back. Do you have reverse-voltage protection?

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Tue Mar 10 2015, 09:25PM

I do have reverse voltage protection in the form of a properly rated HV diode and small value bleed resistor in parallel (with reverse polarity) across each capacitor. I'm using this configuration because having the diodes in series with the capacitor seems to result in a much longer pulse width. This way any reverse voltage should drain across the diode before it charges the capacitor.

. . . . I wonder if the bleed resistor is dampening the reverse pulse enough to make it negligible?

I'm determining my L based solely on Barry's inductor sim - based on the the dimensions of my coils and length and gauge of wire used. I don't have access to an induction meter.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Tue Mar 10 2015, 10:12PM

Are you using an IGBT switching system? I would look into putting a reverse diode/speedup resistor (in series) across the coil. I don't know about a diode in series with any components. What is your total capacitance @ 525v?

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Tue Mar 10 2015, 10:45PM

I'm using some massive SCR's for the switching. The total capacitance for the first stage at 520V is 5400uF.

I'm going to attempt to measure it with my scope now.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Wed Mar 11 2015, 01:12AM

So I attempted to measure the pulse with the scope. I only charged the bank to 375V as my scope has an input limit of 400V. Here's the waveform. I don't understand what's going on here.

The main pulse width is about 200ms, which is obviously wrong. And the inverted part of the wave has magnitude of ~-3V. This doesn't seem right either. My only guess is that I'm capturing the tail end of the wave and not the initial oscillation, but I can't get it to trigger on anything else, so should I assume this is correct? Especially since the overall shape looks to be spot on?

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
DerAlbi, Wed Mar 11 2015, 01:23AM

Do the math. You oscilloscope is completely saturatet. your voltage scale is totally wrong

Just assume exponential decay...
3V = 375V*exp(-R/L*200ms) Now extract the LR-Timeconstant L/R and think about how short your pulse effectively is.

Thats like discharging a battery with a LED. it just wont stop emitting light. that doesnt mean that the battery is still full at the end.

btw: voltage is not really a usefull measure. Current is important.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Wed Mar 11 2015, 01:53AM


@DerAlbi: I 'did the math'. From
3V = 375V*exp(-R/L*200ms)
I did extract the L/R time constant. It is 42ms. Then using 300milliohms for R, I was able to get an inductance value for the inductor of: L=12mH. I get an UNDAMPED RLC circuit time constant of 124 seconds! Now taking into account the addition of a series resistance: R=300milliohms, The circuit time constant becomes 32 seconds! Still Absurd!

@aarpcard: Can you describe the details of your measuring apparatus and scope settings/hookup. I do agree that you should be looking at current.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
DerAlbi, Wed Mar 11 2015, 02:21AM

You cant really caluclate an undamped oscillation based on L/R since there is no C. It is true that the exponential decay does not apply to the Inductor. And the exponential behavior would additionally be more or less concerning the Inductor current.
The voltage (thats measured) would (if considered exponential decay) would be 3V= 375*exp(-t/(5.4mF*ESR)) -> which gives an ESR of 7Ohnm. Some of the 7Ohm are wrong due to the inductive component, but that leaves a much bigger part than 300mOhm left to discuss.
That could be bad.. bad soldering or other stuff that we have no information about.

Sry, thx to Signification, the idea was the right one, the component was the wrong one. Like a blown SCR or a Puck-Style-SCR which has not enough pressure on it

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Wed Mar 11 2015, 04:36AM

aarpcard: have you considered an inductance meter from ebay? I just ordered one that reads, among many other things, L, C and even ESR! about $12

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Wed Mar 11 2015, 02:20PM

I just bought an inductance meter on ebay. Should be here within a week or so.

That pulse width has to be wrong. The primary coil is 60mm long and my projectile is 45mm long weighing 41grams. Currently, muzzle velocity of the primary coil is 29m/s (about 2.3% efficient).

There's no possible way I'm getting any kind of a launch on a 200ms pulse width. Ballpark it should be ~4ms to be getting those numbers. I think my scope is set up incorrectly somehow. The shape of the waveform looks plausible, so I'm thinking it has to be a measurement error?

I don't have it in front of me right now, but the settings to the best of my memory were:

AC Coupling on Channel and Trigger
Trigger Set to Rising Edge (I tried Falling Edge and got the same waveform.)
Probe 1x

I have the scope ground connected to the system ground (system is isolated from mains) and then the probe is on the positive side of the inductor.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
DerAlbi, Wed Mar 11 2015, 06:17PM

Probe 1x? Usually Oscillscopes cant handle that voltage then. its commong that the input stage has maybe 35V-50V cabability which is then scaled up by te probe to provide higher input voltage capability.
If i am right you are saturating your input circuit (and possibly overloading it). If not, then the measurement is correct.
A saturated input circuit of your Oscilloscope could also make you see such waveform.
Just test this theory by charging up your capacitor and probe the voltage as short as you can (so that you would see a rectangle on your scope. If that rectangle has the exponential decay instead of sharp edges... you know whats going on.

On the other hand: you say you made a multistage CG.. is this the first stage? (30m/s with 40g is cool!)

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Wed Mar 11 2015, 06:20PM

Is the peak at only a few volts!! Is this a x1 reading? IIRC you mentioned 2kA with about 260 milliohms. Amplitude-wise, your reading is as if you are across a section of wire! Where, exactly, are your two probes connected on the (R)LC circuit? Are they effectively across the coil only? One more small thing to try--take a glance at the DC-coupled input--is it still zero? It looks like both scales are about x20 off (V attenuated and t amplified). The waveform is great! I assume this capture is with the projectile loaded? A bit more R and you will have critical damping. You need current readings--after figuring this out.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Thu Mar 12 2015, 07:49PM


Yes, this is just the first stage. I'm pretty happy with the results so far. I'll have to see if I can get my hands on a 10x probe . . .


The negative peak is about -3V. Those waveforms were generated with about 80 milliohms of total resistance. What I'm so confused about is that this doesn't follow Barry's RLC simulation in the slightest. So this either means the negative pulse is being damped by the protection diodes and resistors or my inductance is way off.

The probe is directly directly across the coil. I'll see what happens with DC coupling.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Thu Mar 12 2015, 08:37PM

The inductance of the coil would be very helpful, but a few more parameters are needed: You mentioned that the coil was wound with 10AWG wire and was 60mm long is this correct? If not, please correct this. The additional parameters needed are:

1) Number of layers on the coil
2) Number of turns per layer (~20?)
3) The ID (bore) and OD of the coil

I will then give you a good 'L' value.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
DerAlbi, Thu Mar 12 2015, 09:08PM

Why did you ignore the quick test, if your probe is the problem? Since you do get 2.3%eff its seems your design is kind of right. it must be the probe.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Thu Mar 12 2015, 10:05PM


That's what I'm thinking . . . but it still doesn't explain why the negative pulse is damped so much - I haven't done the quick test yet because I've been away from the project for a few days - hopefully I can get back to it tonight. I mean I understand that if the input circuit is saturated, it could result in this waveform, but if that were the case, I should not be getting a 30m/s projectile velocity, especially when using a 300milliohm dampening resistance results in a critically damped wave (according to Barry's RLC) and when testing that, my muzzle velocity was a mere 14m/s.

What that means is the circuit is more damped or similarly damped, but with a higher current pulse with a lower dampening resistance, which doesn't follow Barry's RLC. . .

I've attached a screenshots of Barry's Inductor Sim with the primary coil specifications entered.

My dampening resistor is currently 50 milliohms + the 38 milliohm coil resistance. Plugging the simulated inductance, resistance, capacitance and bank voltage into the RLC sim I get the following (see attached). I'm not getting the large oscillations as shown in the simulation which doesn't make sense.

So either my inductance is way off (I've built the coil exactly to spec on the inductor sim (I've ordered an inductance meter)) or something else is going on here. I'm still wondering if my negative voltage protection diodes and resistors are dampening the negative part of the oscillation. The schematics for these is also attached. (Note the resistance and inductance values in the schematics for the dampening resistors are no longer accurate. As I said, stage 1 is at 50 milliohms + 38 milliohms coil resistance and the inductance is ~145uH.)

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
DerAlbi, Thu Mar 12 2015, 10:39PM

Wow. you again ignored the quicktest. What can i say. If the measurement is wrong, where is the point to dicuss the negative 3V.... also you coupled AC. if the timescale is right thats totally enough to offset your decoupling capacitor (low cutoff-frequency is 5Hz mostly)
You try to analyze a measurement where you dont know if its right. Please explain the point of this.

And you really avoid perfectly to figure out, if your oscilloscope is overloaded and how if handles such an overload. maybe is like a multimeter at mains voltage during ohm-measurement... it doesnt blow, but for some time after the overload all measurements are way off.
please stop ignoring the basic question here... is the 3V you are discussing actually relevant?
Remember your Capacitors are a diode in reverse polarity. putting any kind of resistors in front of your reverse polarity protection diodes is just... well lets say.. the caps will degrade REALLY fast.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
aarpcard, Thu Mar 12 2015, 10:52PM


I haven't had the opportunity to do the quick test yet - I fully plan on doing it.

The scale on the scope is definitely off - whether it's being saturated or something else, I'm not sure until I do the quick test. However, considering the point below, the general shape of the wave looks like it could be correct.

However, as I've been saying, the reduction in dampening resistance should result in the wave shown in Barry's RLC which would result in significant projectile suckback and a slower projectile when compared to the critically damped circuit with a resistance of ~300 milliohms. What I'm confused about is that the circuit as it is (with the lower resistance) is functioning 100+% better than the critically damped circuit. This suggests the wave is indeed critically damped which didn't make sense to me which is why I wanted to measure it - and when I measure it, it's close to being critically damped (although yes, I need to sort out the scope).

My main confusion is why does the general shape of the wave on the scope match what I'm getting for velocity numbers, but doesn't match Barry's RLC sim. And even if the scope is wrong due to how I'm measuring it, and the simulated RLC wave is close to what I'm actually getting, then the velocity of the projectile should be much lower than the 14m/s I was receiving with the original 300 milliohm critically damped wave. Mainly, a less optimal pulse should not be performing better (and 100% better that that) than an optimal pulse especially when the coil and projectile were designed for the optimal pulse.

And that's why I started looking at the negative voltage protection circuitry as that could be dampening the negative oscillation.

Obviously I need to measure the inductance and properly test the scope, but I won't be able to do that for a little while.

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Tue Mar 17 2015, 06:17PM

@aarpcard,
How do you configure the barrel of your multistage coilgun. For a single stage, I can see the coil wound in a coil-winder which can handle that length, but, I assume, a multiple stage coil would require windings whose total length is not all done in the on a very long shaft in a coil-winder. Do you wind multiple coils, one at a time, on a short solid shaft and then, with the coil removed from the winder, slide the coils over a single section of tubing long enough to hold all three (or more) coils? Here, I see the winding shaft's diameter equal to the OD of the tubing (barrel).

IOW, If you wind each coil separately, how to you assemble the series of coils on a single barrel?

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
johnf, Tue Mar 17 2015, 07:06PM

i THINK YOU WILL HAVE TO MEASURE THE INDUCTANCE THE AIR CORED COIL IS WRONG
You have a core in it (your bullet) this will increase your inductance by several orders of magnitude

Re: Method for measuring firing coil voltage pulses with an Oscilloscope
Signification, Tue Mar 17 2015, 08:31PM

johnf, (assuming you were responding to my last msg in this thread) I am pretty sure we are aware of the varying inductance "L" as the bullet transverses the barrel. I, for one, am working on a mathematical relationship between the total "L" of the coil as a function of projectile position in the barrel--given the appropriate parameters, such as permeability of a bullet of particular size, and coil details.

But, at present, I was curious of other people's methods of placing multiple coils on a (metal or non-metal) long barrel.