How-to article: Remote Triggering of Paintball Markers

[Mike Lyons]

See http://www.rctankcombat.com/articles/

Thanks to Frank for formatting and additional images.

Thanks to those of you who responded to my request for info.

This article will evolve over time as techniques are tried.

If you use any of the techniques (successfully or not) or invent a new one

please let Frank know so we can keep it up-to-date.

Feedback is welcomed.

[Mike Lyons]

I finally found the time and motivation to test the Pololu RC Switch with Small Low-Side MOSFET

It works as expected.  Bazooka Joe is now a little more bionic (strictly speaking he's more electronic, less electromechanical than he was).

The marker in RL001 is a Spyder Imagine.  A 9V battery powers a circuit board that drives a solenoid that flicks the sear to fire the marker.  The trigger presses a SPST microswitch.  According to my digital multimeter the microswitch passes a constant 480 uA (0.48 milliamps) when closed and one contact is at ground level (i.e. it's a low-side switch).  I was surprised the current was so low, I guess it's driving a Darlington pair?

I soldered a servo-style lead to the IN, VRC, and adjacent GND pins on the MOSFET board, and soldered a tiny link over the VRC and VCC pads to power the board from the R/C interface.  I soldered a 2-position screw terminal block (with 0.1" pin spacing, very small!) to the LO and adjacent GND pins (on the "wrong" side because a surface-mount device was in the way on the "right" side).  The marker has a pair of wires soldered to the trigger microswitch contacts and I connected the other ends of the wires to the screw terminals, making sure the ground side was connected correctly (GND pin to ground side of switch).

When operational an indicator LED on the switch board blinks on briefly about once a second to indicate is has a signal.  When the "fire" signal is sent the LED blinks off briefly about once a second and the marker fires with the first blink.  The rate of fire (cycling the R/C signal to re-fire the marker) is "plenty fast" ... no specifics here because Bazooka Joe is a single-shot asset so rate of fire isn't an issue.

Conclusion:

I see at least 4 options for firing an electronic marker via R/C:

1. Ignore the electronics and use a servo with an arm or cam to mechanically move the trigger or the microswitch actuator.
Pros: Simple, cheap, proven.  You can see if it's working.
Cons: The servo and arm/cam need to be aligned correctly with the trigger - needs mechanical engineering.  Subject to wear-and-tear.

2. Use an R/C relay-based switch (examples include the PicoSwitch and the Pololu RC Switch with Relay).  I used the PicoSwitch previously and it worked fine (as long as the pulses weren't more than 500 ms either side of center, which shouldn't be a problem with typical R/C gear).

Pros: Very small, easy to connect (polarity not an issue), voltage/current not an issue with typical markers, clicking sound confirms operation.

Cons: Some soldering required for cheaper products (around $10), not for more expensive ones (around $20).  Needs electrical and/or mechanical engineering.

3. Use a FET-based switch as above.
Pros: Very small, LED shows status.

Cons: Must get polarity right when connecting, some soldering required.  Needs electrical engineering.

4. Given the low current I observed it wouldn't be a stretch to build a home-grown solution.  A Pololu RC Switch with Digital Output, or a microprocessor (PIC, PICAXE, Arduino, Raspberry Pi, etc) with a resistor or two and a BJT or FET should do the trick.  I'll look into this "someday".

Pros: Customizable (e.g. with an MPU you can activate based on certain signal characteristic/s).  Cheapish if the parts are lying around your junk box.

Cons: Much assembly required.  Needs electronic engineering.

All these solutions are on the order of $10 or less if you can do some of the work, $20 or so if you just want to plug it in and go.

[Frank Pittelli]

The switch on the e-trigger is a standard active-low input to a
micro-processor I/O port, which is why the current is so low. The
micro-processor then combines the input with the current operating state
(e.g., firing mode, trigger lock-out, etc) and, if firing is warranted,
activates a FET on the e-trigger board to provide power to the solenoid
for no more than 5 ms. The short burst is sufficient to trip the
mechanical sear, but not long enough to draw a lot of current from the
9V transistor battery that powers the e-trigger board. A large
capacitor recharges the firing circuit in just a few milliseconds after
it is fired.

You left out two other options for firing an e-trigger:

5) Use a door-lock actuator to pull the manual trigger.

6) Use a prototype Cheap Control Systems Servo Switch board to drive the
solenoid directly, thereby completely replacing the mechanical trigger,
mechanical switch, e-trigger board and the 9V transistor battery.
Battle Tested. Tri-Pact Approved.

[Mike Lyons]

What's the interface between the R/C receiver (or equivalent e.g. C12C plug plug plug)
and the door lock thingamy?

On Wednesday, October 8, 2014 2:38:02 PM UTC-4, Frank Pittelli wrote:

... 

5) Use a door-lock actuator to pull the manual trigger. 

...

[Mike Lyons]

My crystal ball is a little fuzzy but I think I see "CRCS".

On Wednesday, October 8, 2014 2:38:02 PM UTC-4, Frank Pittelli wrote:

... 

6) Use a prototype Cheap Control Systems Servo Switch board to drive the 

solenoid directly 

 ...

[Frank Pittelli]

Usually, the door lock actuator is used for manual triggers, because it
pulls harder and faster than a servo by itself. When an e-trigger is
used, very little force is needed, so the door lock actuator is not
really needed. Nonetheless, any of the following configurations could
be used to activate a door lock actuator with proper wiring:

a) Servo & Mechanical Switch
b) Relay-based Servo Switch
c) FET-based Servo Switch
d) Prototype CCS Servo Switch (FET-based)

[isaac goldman]

Why does there have to be a servo involved? You can fire a door lock solenoid with only electronics quite easily. Or am i misunderstanding?

Ill post a circuit later tonight.

--
--
You are currently subscribed to the "R/C Tank Combat" group.
To post a message, send email to rctank...@googlegroups.com
To unsubscribe, send email to rctankcombat+unsubscribe@googlegroups.com
Visit the group at http://groups.google.com/group/rctankcombat

--- You received this message because you are subscribed to the Google Groups "R/C Tank Combat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rctankcombat+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mike Lyons]

Very interesting.  Thanks, Frank!

This suggests an even simpler and lower-cost solution:

4.1  Connect the output of a logic-level device (examples in 4 above) to the high side of the trigger,
and the ground of the device to the marker ground (the low side of the trigger would be convenient).

The logic levels don't matter as long as "high" is high enough that the marker doesn't interpret it as "low" and not so high that it blows something (unlikely),
and "low" is ground level.

On Wednesday, October 8, 2014 2:38:02 PM UTC-4, Frank Pittelli wrote:

The switch on the e-trigger is a standard active-low input to a micro-processor I/O port 

...

[Frank Pittelli]

There always has to be some sort of electronic interface between the R/C
Receiver and the door-lock (or e-trigger) solenoid and Mike has been
cataloging the various ways that can be done.

Option (a) below is the simplest solution, that anyone can implement
with only basic tools and wiring skills. It can also be done for less
than $5 using readily available parts from surplus catalogs. It also
doesn't require any modifications to the marker.

Options (b), (c) and (d) below don't use a "servo", they use a "servo
switch" which is "an electronic circuit that accepts a servo input
signal and then switches something". The most common servo switch
circuits use a simple micro-processor to read the servo signal and
trigger the switch. Old-timers used a 555 and an assortment of other
components to achieve the same goal but that's horse-and-buggy type stuff.

Mike listed some servo switches that are commercially available, all of
which can be used to operate a door-lock actuator or switch an e-trigger
circuit. When used with a door-lock actuator, no marker modifications
are required. When used with the e-trigger mechanical switch, you need
only solder a pair of wires between the mechanical switch and the servo
switch.

Option (d) is a prototype circuit board used in the most recent battles
that can drive the e-trigger solenoid directly from a servo signal. It
requires the most drastic changes to the marker, disconnecting the
solenoid from the e-trigger circuit and connecting it to the servo
switch board.

I guess we should also include an Option (e) which is an
"Opto-isolator-based Servo Switch". They provide isolation like a
mechanical relay without any moving parts, but they can't handle as much
current as a FET or relay. Relays, FETs and opto-isolators are the most
common approaches for switching something, with various advantages and
disadvantages that keep them all in use.

On 10/8/2014 5:17 PM, isaac goldman wrote:
> Why does there have to be a servo involved? You can fire a door lock
> solenoid with only electronics quite easily. Or am i misunderstanding?
>
> Ill post a circuit later tonight.
>
> On 2014-10-08 5:12 PM, "Frank Pittelli" <frank.p...@gmail.com

[isaac goldman]

4.1 the problem is that the output from the receiver is not a digital logic signal in the high/low sense of the word. Its a ~50hz signal who's duty cycle varies with the position of the stick. Hence why you need a servo switch. 

Thanks for clarifying that Frank. For some reason I thought you meant a MAG style switch to energise the FET.

I dont trust opto-isolators to handle the spike from an inductive load like a solenoid. Replacing the e-trigger with one could work, but directly energising the solenoid off it doesnt appeal to me. 

<mailto:frank.pittelli@gmail.com>> wrote:

    a) Servo & Mechanical Switch
    b) Relay-based Servo Switch
    c) FET-based Servo Switch
    d) Prototype CCS Servo Switch (FET-based)

--
--
You are currently subscribed to the "R/C Tank Combat" group.
To post a message, send email to rctank...@googlegroups.com
To unsubscribe, send email to rctankcombat+unsubscribe@googlegroups.com
Visit the group at http://groups.google.com/group/rctankcombat

--- You received this message because you are subscribed to the Google Groups "R/C Tank Combat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rctankcombat+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
Isaac Goldman

5142334423

This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. 

[Mike Lyons]

In 4.1 I am describing a microprocessor or something similar that produces true high and low logic levels.
I'm *not* referring to an ordinary R/C receiver.

The logic-level device may have interpreted R/C signals and produced a different type of output,
or it might be a non-R/C solution (WiFi, Bluetooth, etc).

On Wednesday, October 8, 2014 9:37:27 PM UTC-4, True North Armouries wrote:

4.1 the problem is that the output from the receiver is not a digital logic signal in the high/low sense of the word. Its a ~50hz signal who's duty cycle varies with the position of the stick. Hence why you need a servo switch. 

...

[Frank Pittelli]

Agreed. I would never drive an inductive load directly with an
opto-isolator. Fortunately, there are plenty of cheap FETs with
built-in protection now that can be used.

[Joe Sommer]

On Thursday, October 9, 2014 9:23:34 AM UTC-4, Frank Pittelli wrote:

Agreed. I would never drive an inductive load directly with an
opto-isolator.  Fortunately, there are plenty of cheap FETs with
built-in protection now that can be used.

I disagree.  I always use an opto-isolator between a microprocessor
that reads RC servo signals and an inductive load.  The opto-isolator
protects the microprocessor from nasty inductive spikes.

1)  small inductive loads (<150 mA steady-state current)
Drive the LED side of the opto-isolator with the microprocessor.
The Darlington side can safely drive a small inductive load directly.

2)  medium inductive load (<2 A steady-state current)
Drive the LED side of the opto-isolator with the microprocessor.
Use the Darlington side to drive a BJT or MOSFET for the load.

3)  large inductive loads (<40 A steady-state current)
Drive the LED side of a solid-state relay (SSR = big ass opto-isolator) with the microprocessor.
The SSR output is a very large MOSFET with heat sinks that can drive the load.

Driving an FET directly from the microprocessor does not
protect it from large inductive spikes because there must
be a common connection between the microprocessor GND,
the FET source pin and low-side power for the load.

Joe

[isaac goldman]

Opto-isolators are slow to switch on and off. Darlington pairs are slow to switch on, and even slower to switch off. This results in a system thats slow to switch off, so by extension right as the inductor is kicking a huge current spike down to it, the opto-isolator is least equipped to handle it, being somewhere in the active region with high resistance. This also makes them unsuitable for driving the gate of a mosfet by the way, which is a capacitive load. 

In cases 2 and 3, you just drive an N mosfet directly. Heck, in all cases you just drive an NFET. I dont understand why anyone uses an SSR these days, when 200 amp mosfets can be had for <3$ each. Mosfets are cheaper, faster, and can handle more current safely. 

For your circuit to work, there must be a common ground. Ultimately the ground on your micro-controller and your load meet up no matter what you do. If you use an N mosfet, then the mosfet is on the low side of the load. I dont understand your reasoning for why you dont just drive the load off a mosfet.

--
--
You are currently subscribed to the "R/C Tank Combat" group.
To post a message, send email to rctank...@googlegroups.com

To unsubscribe, send email to rctankcombat...@googlegroups.com

Visit the group at http://groups.google.com/group/rctankcombat

---
You received this message because you are subscribed to the Google Groups "R/C Tank Combat" group.

To unsubscribe from this group and stop receiving emails from it, send an email to rctankcombat...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[TyngTech]

What one of you E Guru's need to come up with is an R/C triggered circuit that can directly fire the solenoid in a typical e-marker.  This bypasses the marker trigger board altogether.   Loic's first combat tank had such a circuit but it had the nasty habit of firing when power was turned off.  It was some sort of capacitor discharge circuit.

[loic Anthian]

And this will be resolved when we have the “Hero” controller which will replace a LOT of features, like the pico switches, Marker onboard circuits, and 9V feeds to the Marker, Laser. In addition, it will be run by a PS2 controller which brings lots more capabilities, buttons and triggers.

 

‘Cross the Road Electronics’ is at the forefront of the Robotic world, but they are now very busy with some production schedule. They will work again on the Hero by late 2014.

 

Cheers from Utah… can’t wait to cross aim with you guys in a couple of weeks!

Loic

[Mike Lyons]

I tested the solution described below (4.2) today and it didn't work as expected ... but I believe it's a viable solution.

Whaaa????  Stay tuned.

I built a Pololu R/C Switch with Digital Output.
The process was almost identical to the previous one for the Pololu MOSFET switch,
except I needed two 2-position screw terminals as the two connections I needed were not adjacent.
On the upside if I ever want to know the GOOD state and/or tap into the VCC it's ready to go.

I turned on the marker power and my remote controller.

When I turned on the power to the receiver (which also powers the switch board) the marker immediately fired.
The switch board blinked its LED briefly about once a second as specified.
I pressed the FIRE button on my controller and the switch board LED stayed on as specified while the button was pressed but the marker didn't fire.
I pressed the manual trigger on the marker and it fired once, but didn't fire one subsequent presses.
I pressed the FIRE button on my controller a few times and nothing.
I pressed the manual trigger and again it fired once but no more.

After scratching my head I realized the problem and a potential solution.

Challenge: Explain the problem.  Hint:  All you need to know is in this discussion thread over the last few days.

Advanced challenge #1: Devise a diagnostic procedure using the existing setup to confirm your theory.

Advanced challenge #2: Propose a solution using the existing hardware.

Principals of Tri-Pact are not eligible to enter.

On Wednesday, October 8, 2014 5:48:33 PM UTC-4, Mike Lyons wrote:

... 

4.1  Connect the output of a logic-level device (examples in 4 above) to the high side of the trigger,
and the ground of the device to the marker ground (the low side of the trigger would be convenient).

...

[Frank Pittelli]

Not only has it already been built, but that circuit was battle-tested
in the SU100, Patton and Semovente during the last battle. The board
connects to the R/C receiver and directly drives the marker solenoid, as
I explained in previous posts, exactly like an e-trigger.

It even has a feature that allows you to restrict your firing rate to
one shot every N seconds. I added that feature so that we can conduct
the "Doug Experiment" one battle to see how reduced firing rates affect
the game.

[Mike Lyons]

oops ... the above should read "I tested the solution described below (4.1)" not "(4.2)".

[Joe Sommer]

On Friday, October 10, 2014 8:21:21 AM UTC-4, True North Armouries wrote:

I dont understand why anyone uses an SSR these days, when 200 amp mosfets can be had for <3$ each.

Build some high current drivers.  Be certain to mount hefty heat sinks. 
Report back after you have fried a few.
 

For your circuit to work, there must be a common ground. Ultimately the ground on your micro-controller and your load meet up no matter what you do.

Incorrect - the GND on your microprocessor and the low-side power
for your load are NOT connected when you use an opto-isolator or SSR.

[isaac goldman]

The ground on the MCU and the low side dont connect to the same place through what black magic? Or are you lugging a second battery around? That sounds like a waste of space, and an additional thing to fail or malfunction at the wrong moment.

I have built high current drivers, and I have yet to blow one or see one blow from an over-current. The one im designing for my tank even has closed loop current monitoring for additional protection. Actually come to think of it, SSRs are so inefficient and slow compared to mosfets I suspect for any power level you could affordably use an SSR, you would need very little heat sinking if any.   

--
--
You are currently subscribed to the "R/C Tank Combat" group.
To post a message, send email to rctank...@googlegroups.com
To unsubscribe, send email to rctankcombat...@googlegroups.com
Visit the group at http://groups.google.com/group/rctankcombat

---
You received this message because you are subscribed to the Google Groups "R/C Tank Combat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rctankcombat...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mike Lyons]

The logic device has a small battery of its own, or is using the R/C receiver's battery.  4 x AAs work for me.

The motor switched by the SSR (or whatever) has its own big fat battery.  12 V SLA or similar looks good.

The two sides are isolated by the SSR (or whatever).

On Friday, October 10, 2014 4:15:19 PM UTC-4, True North Armouries wrote:

The ground on the MCU and the low side dont connect to the same place through what black magic? Or are you lugging a second battery around? That sounds like a waste of space, and an additional thing to fail or malfunction at the wrong moment.

...

 

[Frank Pittelli]

Hmmm ... I guess I've been doing it all wrong for the last 25 years in
R/C combat hobbies. I almost always use a separate battery for the R/C
and control boards, keeping them completely isolated from the main
batteries that power the motors, relays and solenoids whenever possible.
That significantly reduces the chances of frying expensive electronics
when something goes horribly wrong on the main power grid, like (a)
drawing truly massive current when a motor locks up, (b) sending wicked
reverse voltage spikes back from strong relay/solenoid coils when a
diode falls off, and (c) shorting the main power grid with a screwdriver
which then welds itself to a battery terminal. I've seen all these
things, plus plenty more, happen to combat systems over the years. The
price and space required for a separate control side battery is a small
cost to pay to avoid losing all your electronics in those situations.

That's also why guys like Joe and I use opto-isolators when designing
circuit boards that handle large loads. Safe, cheap and easy to use.
Even if the greatest, most reliable FET in the world is being used, it
doesn't cost much at all to completely isolate the high-power side of
the circuit from the low-power side.

[Doug Conn]

>> It even has a feature that allows you to restrict your firing rate to one shot every N seconds. I added that feature so that we can conduct the "Doug Experiment" one battle to see how reduced firing rates affect the game.

From a few searches, it look like WWII heavy tanks had a rate of fire on the order of 6-8 RPM. The Abrams claims 8-10 RPM. I think it would be fun to play with values in that range.

Regarding single or multiple batteries - I've been using a single battery supply for the eight years (wow ! really ?!?!) I've been battling with you guys. Even with a computer and the marker electronics sharing the same ground as the motors, I haven't seen any issues. I do use opto-isolators on the marker triggering, however.

[jvragu47]

Principals of Tri-Pact are not eligible to enter.

On Wednesday, October 8, 2014 5:48:33 PM UTC-4, Mike Lyons wrote:

... 

Since I'm not an E-Guru as aptly put by Steve, this would be my solution..

A larger hammer sir.   Did I win? LOL

John" I can bolt it, nail it , rivet it, hammer it and mold it with a dremel" Pittelli

[Frank Pittelli]

Psst.   Weak pull-up.

[TyngTech]

"I added that feature so that we can conduct the "Doug Experiment" one battle to see how reduced firing rates affect 
the game"

An immediate effect will be the reduction of long range gunnery.  If the Cromwell only gets to shoot once every 10 seconds or so.  You better believe said Cromwell is going to be right on somebody's six when it takes that shot!  ;)

ST

[Frank Pittelli]

Personally, I think that the number of unexpected consequences, good and bad, will far exceed our expectations. That's the fun part about experimentation. Besides, it can't be worse than Tank Soccer 😕

--

[TyngTech]

Tank Soccer was the best!  Your just ore because you lost!  lol

[Odyssey...@aol.com]

then i guess a max speed limit rule should be passed then

 

chris

[Mike Lyons]

The output of the Pololu switch is active-high.
The input to the marker is active-low (the trigger is a low-side switch).

The digital signal needs to be inverted.

(Note: The latest version of the Pololu series of switches has an "inversion" function
 but this changes the direction of the activation signal (from "above center' to "below center".
 I believe it addresses the need to flip the controls of a combat robot when it has been "inverted", i.e. turned upside down.)

I connected a 1K resistor to a +5V active digital output of a microprocessor
to feed the base of a small, general-purpose NPN transistor
connected across the trigger contacts (emitter to the black wire, i.e. marker ground)
of my electronic marker and it works fine.

A 1 ms "on" time (the shortest "pause" time on the PICAXE I'm using to test) is plenty.

I assume the same solution would work for the Pololu Digital Switch.

For this application (R/C firing of an electronic marker)
I recommend the Pololu Switch with Small Low-Side MOSFET.
The extra cost $1.26 more than the Digital Switch, less the resistor and transistor)
is justified by the simple implementation (see above ).

On Friday, October 10, 2014 12:19:41 PM UTC-4, Mike Lyons wrote:

... 

Challenge: Explain the problem.  Hint:  All you need to know is in this discussion thread over the last few days.

[Mike Lyons]

I think Frank is onto something here.

Once I understood the active-low issue

I tried holding my "fire" button down (to put the switch into the active-high state)
and manually pulling the trigger.

I also tried holding down the "fire" button and releasing it briefly,
assuming the reverse logic would work.

No luck.

The R/C switch is running off the 6 V R/C receiver power.
The marker has a 9 V battery for its supply.
The pull-up in the marker logic may be too much for the digital signal to overcome.

Interfacing systems with different signal levels is always a challenge.


On Saturday, October 11, 2014 8:08:40 PM UTC-4, Frank Pittelli wrote:

Psst.   Weak pull-up.

On Friday, October 10, 2014 12:19:41 PM UTC-4, Mike Lyons wrote:

... 

Advanced challenge #1: Devise a diagnostic procedure using the existing setup to confirm your theory.