EM Gun - Trying to understand schematic

[Bigglez]

Greetings,

I was flipping through the March 2008 Nuts and Volts and saw
the article about an electronic weapon. I'm trying to understand
the schematic, Figure 2.
What is the function of SW2? What are D5 and D7 for?
What does the CR circuit (R13,C7) do?
What is D8 for?

Thanks In Advance!

If you could post the circuit that would help
others that dont have access to it understand what you are asking.

[slipcasedrock]

Try Information unlimited
they have many different designs of these and others.

[haklesup]

The exact intent is a bit unclear but lets see if my comments help.

SW2 (reset) obviously removes current from the excitation coil (primary). I can see that base current is fed back when the primary excites the feedback coil. Its possible that the circuit can latch up (all DC, no oscillation) and this is the easy way to free it though I don't see any clear reason for such a misbehaviour. I think this is just a way to turn off the inverter stage while still powering the rest of the circuit

R13, C7 seem to be there to slow the rise time of the gate when pulled up by R12 (when SW3 is closed).


D4-7 keep the output of IC1 from straying more than ~+/-3V and overbiasing Q2. Otherwise the output of IC1 would want to go close to the rail. The LEDs probably aren't needed for functionality of the cky other than as an indicator but they do add into the voltage drop

[Bigglez]

Greetings Hacklesup,

SW2 (reset) obviously removes current from the excitation coil (primary). I can see that base current is fed back when the primary excites the feedback coil. Its possible that the circuit can latch up (all DC, no oscillation) and this is the easy way to free it though I don't see any clear reason for such a misbehaviour. I think this is just a way to turn off the inverter stage while still powering the rest of the circuit.

Well, SW2 is on the rear panel, yet not mentioned in the
article's text. I'm wondering if the SCR latches to fire
the weapon and can't be unlatched without removing the
HV generator power? The SCR is a dead short across the
HV generator. Seems that a resistor in line (to reduce the
SCR current to unlatch it) would be better (cheaper)?

R13, C7 seem to be there to slow the rise time of the gate when pulled up by R12 (when SW3 is closed).

The SCR gate is fired by SW3, which would source current
limited by R32 (360R) of about 30mA. The charge on C7
would be limited by R13 (39K) and R14 (39K) and contribute
little current to the SCR gate. I don't see how C7 can have
any effect on the SCR gate's timing. Why would it be desirable
to slow down the SCR gate? Seems as if the wrong values
have been given for these components. As drawn, removing
R13 and C7 would have no measureable effect on the SCR?

D4-7 keep the output of IC1 from straying more than ~+/-3V and overbiasing Q2. Otherwise the output of IC1 would want to go close to the rail. The LEDs probably aren't needed for functionality of the cky other than as an indicator but they do add into the voltage drop

The LEDs are on the rear panel and mutually excusive,
D4 indicates the capacitor bank is charging up, and D6
indicates the charging has stopped (weapon ready to fire).
Only one LED ("ready") is needed.
The LEDs are in series with 14V zeners, yet the supply
is only 12V, so the zeners never conduct in the forward
direction (they are drawn backwards - acting only as diodes).

How did you conclude that the op amp gives about 3V output?
The LM741 spec for output current is 25mA typ. So the
op amp will swing close to the rails and source or sink
6mA for the LEDs and 17mA base current to Q2. If the
op amp delivers 25mA the output will swing to approx.
9.7V (mainly due to the load of R7 and Q2 base). The
LM741 datasheet spec's a min load of 2K ohms.

I have no idea about D8. The only time it could conduct
is if the HV power supply output goes negative.

I'd have to assume the author got the prototype
working. Anyone else studied the article and schematic?

Comments Welcome!

[Bigglez]

Operator error

[haklesup]

I think you're right about SW2, it must be to reset the SCR. A series resistor with the Anode of Q3 would limit any firing current which would be undesireable. On the line segnment just below the image of T1 maybe.

I'm not sure why the need to slow the gate voltage, dv/dt false triggering is ususlly associated with the voltage on the Anode. D8 may act as a snubber for back EMF generated in the accelerator coil. D9 likewise seems to protect the capacitor bank from reverse bias due to the same back EMF.

I was incorrectly thinking bipolar supply on the 3V comment. I suppose the Zener would only turn on at 14V if the comparitor failed (pin 3 to 6 short) or ground bounced when the device is fired. This may be the ultimate reason for C7, R13, to keep the VGK constant with an AC transient on ground.

Several sections of this circuit beg to be simulated

[Bigglez]

Greetings Hacklesup,

I think you're right about SW2, it must be to reset the SCR. A series resistor with the Anode of Q3 would limit any firing current which would be undesireable. On the line segnment just below the image of T1 maybe.

The HV generator includes a voltage doubler, with the output
across C4 (1000pF), but the capacitor bank and accelerator coil
(L1) are in parallel. Why have C4? 1000pF // 600uF = 600.001uF
A resistor in series between C4/D2 and R3 would limit the
available current from the HV generator to less than the holding
current of the SCR (once fired). It would slow the charging
of the capacitor bank, too. (Weapon firing rate). Another
method would be to stall the HV generator when the weapon
fires, so the SCR resets once the capacitor bank is depleted.

I'm not sure why the need to slow the gate voltage, dv/dt false triggering is ususlly associated with the voltage on the Anode.

Which can't rise quickly due to the realatively high
output impedance of the HV power supply...

D8 may act as a snubber for back EMF generated in the accelerator coil. D9 likewise seems to protect the capacitor bank from reverse bias due to the same back EMF.

Electrical energy in the accelerator coil that is not converted to
kinetic energy to move the projectile will cause a back-EMF
spike, reverse biasing the capacitor bank, so D9 prevents it.

The accelerator coil is mounted in a metal frame, effectively
producing a "shorted turn" secondary at each end of the
coil. This would dissipate energy coupled to the metal by
induction, robbing the energy transmitted to the projectile.
Cutting slots in the metal at each end of the coil would
prevent it. I wonder if the weapon's performance improves
(measureably) by doing so?

D8 can only conduct if the output of the HV generator were
reversed (go negative). I doubt any effect would be seen if
D8 is removed. (Unless due to poor layout the ground
bounce from the firing circuit reaches the low voltage stages,
as you noted).

I was incorrectly thinking bipolar supply on the 3V comment. I suppose the Zener would only turn on at 14V if the comparitor failed (pin 3 to 6 short) or ground bounced when the device is fired. This may be the ultimate reason for C7, R13, to keep the VGK constant with an AC transient on ground.

Why use zeners as signal diodes? Why
have two mutually exclusive LEDs, one would do.
I have doubts about the design...

Several sections of this circuit beg to be simulated

Or, there is old fashioned "black magic" in this one!
The author has a video on the linked website. The prototype
electronics were constructed on a solderless breadboard.
The finished weapon bearly nicks an empty soda can at
one metre when fired. A toy?

Comments Welcome!

[haklesup]

"old fashioned "black magic"" More like Old School Analog design. The kind you only get from an engineer that graduated in 1962 (or thereabouts, never having known a digital beyond theoretical logic classes). A guy who can whip this off a slide rule like its a dueling sabre.

One can usually achieve the same result with a different or modified circuit. Without a detailed description by the designer, it can be guesswork to know if a particular design element is a result of solving a failure problem, avoiding a perceived problem or evolved from earlier attempts. D8 may be more useful when fired under no load or with the coil disconnected.

[electrofun]

First of all I'd like to say that I think the article was well written and informative. This is the kind of interesting article that will get the youth interested in electrical engineering. Kudos to people who actually have the courage to design, invent and publish. Like Bryan Bergeron was saying in his editorial this month, "There's a deep sense of satisfaction that results from developing, building, test and ultimately using a circuit of your own design." And like he says, nothing is ever perfect but you've got to start somewhere. Why don't you go ahead and write an article or two yourself? How can you jump to the conclusion that the author is an old 1950's analog designer waving a slide rule and knows nothing of digital? Judging from his website it looks like he's young and works mostly in digital. The article even states that the purpose was to experiment with analog design. Sure, it could have been built with a microcontroller but then you'd have to program it - sometimes the simple design is always the best way to go.

The gun does work and if you go to the website you can watch the movie:
http://www.thinkbotics.com/military.htm

It looks to me like the circuit is pretty straighforward and was designed in sections.

Capacitor C4 is needed for the second stage of the voltage multiplier and without it in the circuit, as you suggest, there would only be half the charging voltage available. Read up on cockcroft walton voltage multipiers. The author posted in the projects section saying that there was a typo for the values - they should be .01uf at 1.5KV DC or higher, ceramic disc.

It looks like pressing SW2 stops the charging section from oscillating - you could probably leave that out.

D9 is obviously there to protect the capacitor bank from reverse bias due to back EMF from the accelerator coil.

As far as the indicator leds for "charging" and "fire" goes - I think it's nice to see that the gun is charging and then switches to fire so that the user can see that it is doing something and an extra led is cheap. The author said you can visually see the gun constantly topping up the capacitor bank until it's fired. I'm not sure about the zener diodes - you could probably leave those out.

I don't see anything wrong with the 741 comparator section and it seems to be working well in the movie on the website.

About R13 and C7 - those probably aren't needed either.

Best regards, Roger W

[Bigglez]

Greetings Roger,

First of all I'd like to say that I think the article was well written and informative. This is the kind of interesting article that will get the youth interested in electrical engineering. Kudos to people who actually have the courage to design, invent and publish.

A warm welcome to this forum from the last "new guy".

Why don't you go ahead and write an article or two yourself? How can you jump to the conclusion that the author is an old 1950's analog designer waving a slide rule and knows nothing of digital?

May I suggest that you use this forum's tools to attribute
your quotes to a particular author? I'm not sure who you're
speaking to.

The gun does work and if you go to the website you can watch the movie

It might also be helpful to read the entire thread before posting.
I for one did visit that web site and follow the other links in the
paper article. I watched the embedded video a couple of times.

As you probably know, any paper magazine has a much longer
"cycle" than an internet forum. I have already submitted my
questions about the article to the editor, and based on past
experience it may take thirty days for a response from the author
and a further sixty days for publication, subject to editorial prudence.

I'm not sure about the zener diodes - you could probably leave those out.

Are you speculating or do you have some insight?

About R13 and C7 - those probably aren't needed either.

Looks as if we'll need the author to step up and explain
his design.

Comments Welcome!

[haklesup]

How can you jump to the conclusion that the author is an old 1950's analog designer waving a slide rule and knows nothing of digital?

It was a metaphor not an actual characterization of the author. In this day and age, I'm quite sure anyone who could forge this circuit would also be proficient in digital.

Sure, it could have been built with a microcontroller but then you'd have to program it - sometimes the simple design is always the best way to go.

You could integrate a microcontroller at several points but the bulk of the analog circuitry is required for the application. The uC could definitely enhance the user interface but not the output pulse.

About R13 and C7 - those probably aren't needed either

I agree but the OP and myself were speculating on the reason it Was included.

Looks as if we'll need the author to step up and explain
his design.

That would be helpful and I think this is why Michael is trying to get up to use the new forum section for that purpose. Could probably make a case to move this topic there now.

One can usually achieve the same result with a different or modified circuit

In other words, in electronics, there is usually more than one way to do everything. Engineering is the Art of balancing all those choices to come up with a solution that is just the right amount of complexity, cost and durability for a particular application. A good author can articulate this subjective process with objective explanations (am I vague enough)

Welcome and don't mistake any debate for criticism. Hope to hear more from you.

[Robert Reed]

Hacklesup
"In other words, in electronics, there is usually more than one way to do everything. Engineering is the Art of balancing all those choices to come up with a solution that is just the right amount of complexity, cost and durability for a particular application."

Very well put and I couldn't agree with you more. There is a highly respective book thats been out for several years (and I may have the title slightly wrong) called 'The Art and Science of Electronics'. Sort of gives it a humanistic sense of the profession.

[Bob Scott]

I think this EM gun could have been built simpler. Why is so much circuitry dedicated to creating a Kilovolt of charge on a 600uF capacitor bank to store 300 Joules? You can just as easily do it with low voltage and high current. Use fewer and thicker windings on the accelerator coil. Wire the caps in parallel and store 300 joules at low voltage high current.

The whole DC-DC converter and transformer are then no longer required.
Then it wouldn't be complex enough to make it interesting and publishable?

????

[Bigglez]

Greetings Bob,

Why is so much circuitry dedicated to creating a Kilovolt of charge on a 600uF capacitor bank to store 300 Joules?

The energy stored in a capacitor is the square of the
voltage, so higher voltage is geometrically appealling.

Said another way, using low voltage would require much
larger capacitors, making the device bulky.

You can just as easily do it with low voltage and high current. Use fewer and thicker windings on the accelerator coil. Wire the caps in parallel and store 300 joules at low voltage high current.

To store 300j with 12V (i.e directly from the internal
battery pack) would require a 4.2 FARAD capacitor.

Then it wouldn't be complex enough to make it interesting and publishable?

Put something together and send it in,
I think you'll get a favourable response.

A better approach would be to use multiple accelerator
coils with phased timing - each one would kick the
projectile to a higher energy. The author did hint at
this in the published article.

Perhaps a PIC based design would make the project
more attractive for publication?

Comments Welcome!

[dyarker]

Bob,

That would be one of those design compromise decisions.

At high voltage it is fired with an under $2.00, 40A SCR. At low voltage the increased current requirement puts SCR price over $50.00.

Cheers,