Microprocessor control with an Arduino / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Microprocessor control with an Arduino

1 2 next

Move Thread

LAN_403

Maj

Wed Jan 02 2013, 02:01PM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

Hello all. It's my first post at 4HV so hopefully I can make it interesting.
I've started building a CG after much reading on the topic. Starting simple with a single stage, SCR triggered, 2.5 gram projectile, <20 Joules.
However, I'm using an Arduino to handle a bunch of the timing related aspects and wondered if others have had any success or otherwise with microprocessor control?

Here's what the Arduino does for me:
- Provides the pulsed square wave to the transistor/charging coil to charge the cap. Replaces the 555 in circuits like Uzzor's charger and allows me to fine tune the duty cycle.
- Measures the cap voltage and stops the charge at preset voltage. Replaces the Op Amp in circuits like Uzzors charger.
- Triggers the SCR.
- Measures the duration of the coil pulse.
- Measures the velocity of the projectile using a sensing coil at the end of the barrel. No need for external measuring devices.
- Calculates efficiency of the shot
- Displays all this stuff on a laptop for testing purposes

Here's what it could also be doing that I haven't implemented yet:
- Provide low-voltage notification for my LiPo battery and potentially shut down the CG
- Measure charge coil voltage and shape the pulse duration and duty cycle automatically to suit any combination of charging coil and cap.
- Turn an IGBT on and off
- Read optical sensors and fire subsequent stages
- I'm sure there's plenty more

I'm using only a few external components - mainly just resistors as voltage dividers for the analog inputs. It has really simplified my build from an electronic perspective and given me a great level of control.

PSCG

Wed Jan 02 2013, 02:41PM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

First of all, welcome to the forums!

Now, for your question: There are some micro controller - operated coilguns. Some links are presented below:

Another Coilgun have used micros to control a coilgun's operation (under the "Charging" link).

A low voltage coilgun triggered by an Arduino is (poorly) documented here.

Future Experimental Systems (PA - IV and APA - I but only the first one is completed).

And a quick search at 4HV:

Microcontroller design for coilgun brains.

microcontroller for a coilgun

Coilgun 5 stage machinepistol

Arduino as a Coilgun controller?

As it seems, you have implemented a lot of goodies on your coilgun project. Good job!

Barry

Wed Jan 02 2013, 04:56PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

Welcome to 4hv and happy new year!

I just got a small box of Arduino parts and shields last week and would love to see some work on applying them to coilguns, especially the part about reading optical sensors, triggering stages and measuring velocity. It sounds perfect for the job.

Cheers, Barry
If I had a dollar for every time I got distracted, I wish I had some ice cream.

Maxwell

Wed Jan 02 2013, 06:12PM

Registered Member #8497 Joined: Tue Dec 04 2012, 06:24PM
Location:
Posts: 74

While not used for the coilgun side of it, I read of using an arduino to do chronometer readings - photogate triggering and timing.

Maj

Thu Jan 03 2013, 07:27AM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

Thanks guys.
PSCG, I'd seen some of those sites/posts but since some were years old I thought I'd revive the topic. You have an impressive collection.

Barry, most of what I understand about coil guns so far I have learned from your site! Many thanks for a great resource. I'm happy to share my Arduino code with you (or any 4HV member) if it's of use.

lauriethepigeon

Thu Jan 03 2013, 11:24AM

Registered Member #7669 Joined: Mon Oct 29 2012, 02:55PM
Location:
Posts: 19

I have created a light gate circuit which uses a single IR LED and phototransistor with sensitivity adjusted using a potentiometer. The arduino measures the duration of the pulse as the projectile passes using the pulsein() function. This is divided by the length of the projectile to return the speed, which is output to the serial port or an LCD display

I would be interested in using the arduino to control other aspects of the project such as measuring pulse duration and triggering other stages

lauriethepigeon

Thu Jan 03 2013, 11:28AM

Registered Member #7669 Joined: Mon Oct 29 2012, 02:55PM
Location:
Posts: 19

Maj wrote ...

- Triggers the SCR.

The arduino pins can only output 40mA at 5v, was this enough to trigger your SCR or did you have to put it through external switching transistors?

Also, I'd be interested to see how you used the arduino to measure the capacitor voltage, did you set up a voltage divider with resistors and put the output into one of the analog inputs on the arduino?

Thanks

Maxwell

Thu Jan 03 2013, 01:10PM

Registered Member #8497 Joined: Tue Dec 04 2012, 06:24PM
Location:
Posts: 74

lauriethepigeon wrote ...

I have created a light gate circuit which uses a single IR LED and phototransistor with sensitivity adjusted using a potentiometer. The arduino measures the duration of the pulse as the projectile passes using the pulsein() function. This is divided by the length of the projectile to return the speed, which is output to the serial port or an LCD display

I would think your %error for this single-feedback method would be relatively high; I would estimate around 25% or so.

If you placed two photogates at a known distance from each other, you could simply poll for a change in light intensity, mark these timestamps, then divide gate spacing by difference in time. This would probably bring you into the 1-5% error range.

Maj

Thu Jan 03 2013, 01:28PM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

Hi laurie.
The SCR is a 50A @ 1600V, bought on eBay and the specs provided are in Chinese :(. I'm driving it directly from a digital output pin, with a 100uS pulse. It's been reliable so far, but I'm using fairly low power at the moment.

To measure the cap voltage I used a 680K/8.2K voltage divider into an analog input (just had those value resistors lying around). That gives me the ability to measure up to 420V with about a 0.5 volt resolution.

lauriethepigeon

Thu Jan 03 2013, 09:07PM

Registered Member #7669 Joined: Mon Oct 29 2012, 02:55PM
Location:
Posts: 19

Maxwell wrote ...

I would think your %error for this single-feedback method would be relatively high; I would estimate around 25% or so.

So do you think the limitation would stem from the processing speed of the chip or the response time of the phototransistor? The chip measures the duration of the pulse to a very high resolution, to the tune of 10 microseconds.
In addition, phototransistors like the one I have, are used in optical rotary encoders for measuring high speed shafts so the switching time must be very low. I was under the impression that most transistors can switch incredibly high frequencies.

I have validated this method at low speeds by using a nail of a known length (25mm) travelling vertically in a 0.5m tube under the influence of gravity. I calculated the theoretical velocity using constant acceleration equations, and then measured the actual velocity at the bottom of the tube using my circuit. The theoretical value of speed was 3.13m/s, and my 3 repeats read 3.12, 3.13, and 3.12, which is much more accurate than I expected. No doubt the accuracy will decrease at higher speeds, so further testing is required, but so far it has proved more accurate than I thought.

1 2 next

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