Welcome
Username:

Password:


Remember me

[ ]
[ ]
Online
Guests: 32, Members: 1 ...
RickR viewing forum.php

most ever online: 333
(Members: 3, Guests: 330) on 06 Jun : 15:15

Members: 3265
Newest member: electronic6699
Members Birthdays:
All today's birthdays', congrats!
LoveVirus (37)
SilentPhoenix (9)
Electrikid221MnM (16)
Karmaslap (22)


Next birthdays
02/22 Renesis (27)
02/22 Cloud (27)
02/23 Boost (32)
Contact
If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.


Special Thanks To:
  • Aaron Holmes
  • Aaron Wheeler
  • Adam Horden
  • Alan Scrimgeour
  • Andre
  • Andrew Haynes
  • Anonymous000
  • asabase
  • Austin Weil
  • barney
  • Barry
  • Bert Hickman
  • Bill Kukowski
  • Blitzorn
  • Brandon Paradelas
  • Bruce Bowling
  • BubeeMike
  • Byong Park
  • Cesiumsponge
  • Chris F.
  • Chris Hooper
  • Corey Worthington
  • Derek Woodroffe
  • Dalus
  • Dan Strother
  • Daniel Davis
  • Daniel Uhrenholt
  • datasheetarchive
  • Dave Billington
  • Dave Marshall
  • David F.
  • Dennis Rogers
  • drelectrix
  • Dr. John Gudenas
  • Dr. Spark
  • E.TexasTesla
  • eastvoltresearch
  • Eirik Taylor
  • Erik Dyakov
  • Erlend^SE
  • Finn Hammer
  • Firebug24k
  • GalliumMan
  • Gary Peterson
  • George Slade
  • GhostNull
  • Gordon Mcknight
  • Graham Armitage
  • Grant
  • GreySoul
  • Henry H
  • IamSmooth
  • In memory of Leo Powning
  • Jacob Cash
  • James Howells
  • James Pawson
  • Jeff Greenfield
  • Jeff Thomas
  • Jesse Frost
  • Jim Mitchell
  • jlr134
  • Joe Mastroianni
  • John Forcina
  • John Oberg
  • John Willcutt
  • Jon Newcomb
  • klugesmith
  • Leslie Wright
  • Lutz Hoffman
  • Mads Barnkob
  • Martin King
  • Mats Karlsson
  • Matt Gibson
  • Matthew Guidry
  • mbd
  • Michael D'Angelo
  • Mikkel
  • mileswaldron
  • mister_rf
  • Neil Foster
  • Nick de Smith
  • Nick Soroka
  • nicklenorp
  • Nik
  • Norman Stanley
  • Patrick Coleman
  • Paul Brodie
  • Paul Jordan
  • Paul Montgomery
  • Ped
  • Peter Krogen
  • Peter Terren
  • PhilGood
  • Richard Feldman
  • Robert Bush
  • Royce Bailey
  • Scott Fusare
  • Scott Newman
  • smiffy
  • Stella
  • Steven Busic
  • Steve Conner
  • Steve Jones
  • Steve Ward
  • Sulaiman
  • Thomas Coyle
  • Thomas A. Wallace
  • Thomas W
  • Timo
  • Torch
  • Ulf Jonsson
  • vasil
  • Vaxian
  • vladi mazzilli
  • wastehl
  • Weston
  • William Kim
  • William N.
  • William Stehl
  • Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Forums
4hv.org :: Forums :: Electromagnetic Projectile Accelerators
<< Previous thread | Next thread >>   

Another Coilgun

Go to page  1 2 3 [4] 5
Author Post
DerAlbi
Mon Apr 25 2016, 05:39PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
You are still wanting to first build a small prototype to show off how harmless these devices are?
Its not about harmlessness.. its about compliance to the local law. For that i need to be in a >7.5J energy category to be considered potentially 'dangerous'. I do not want to build a big prototype for that just because its a waste of resources and a huge money sink. I can demonstate the principal characteristics and my solutions to circumvent the law also on a small gun. A big gun would need aluminium wire, many coils and so on. A small gun i can build easily with copper and its still ok.

you can always choose to go with a more "science fiction" type look that gives you more internal volume.
Agreed. But scientific look is mostly practical for 2 handed guns.. not sure if a typical revolver can be made to look 'scientific'. I guess in the end, its all about some blinking blue LEDs

A 20 cm long gun is just big enough that it would be uncomfortable to hold with one hand, but too short to feel like an smg and use with two hands.
!!! Exactcly my thought, just not my words

You should slightly upscale for the sake of comfort. Making the whole gun longer also means you don't have to make it so tall.
Which size is comforting, madame?

This diminishes quickly with length, with the exception of a long high capacity clip
Yeah. Normal guns have the additional advantage that the ammo does not need to be directly at the back end. this gived an additonal level of freedom which i do not have with a CG. This prevents me from ever having a large drum magazine or any other solution like that.

The 2" size is a good one to go with for the whole top, but you can get more width by increasing towards the middle and then tapering back down.
Well 2" is easily avoided on top of the gun. actually currently i had somewhat around 1" there, then it widens to host the coils (4.2cm = 1.6") and then it widens again to host the capaciotrs. maybe the capacitor area will just be bulge.
So maybe i will widen the thing a little bit. This will provide space for mechanical support (aluminium rails, or just FR4 PCB... sry i am an electrocs guy)

You should definitely see if you can get on to fire a rifle or see one in action. I would bet that your police department might let someone watch.
Hehe. Honestly, i would never ask such thing. I am more likely to download a CAD-drawing of a gun and print it ^_^

it was one of the best experiences I've had, and they were friendly about it the whole time (which we kelpt short of course)
'Muuuricaa!
Back to top
Karmaslap
Tue Apr 26 2016, 07:25AM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
I came back on to edit my post with pictures, but seems you understood me plenty well enough. I'll save them formlater or something.

Got it. So, this is still a prototype to help you with legal stuff, and you want to keep it cheap.

>2 handed guns, revolver

Are you set on making this a one-handed revolver? I think with your awkward length you might as well go for a small sumbachine gun look. You can do a better space efficient feed mechanism than a revolver as well, surely.

>size

Longer than it is wize, of course! I'm more worried for you that it will be too heavy for one hand and look really weird. If this is just a test, that seems fine and even less threatening if it does not look like a gun.

>ammunition

I can think of some solutions for that to allow a much larger capacity or storage near the center. You lose a lot of space and have a potentially awkward reload is all. If you make a staggered magazine instead of the single width you normally see, you could fit quite a few shots in the back. I'm not sure what your projectile size is (I think you were still using straight up cylinders last time?) but you should have decent ammo capacity. ~10-15 shots is good and more iron just adds weight


>avoiding 2"

Good. Much more visually appealing, and easier to mount a tactical rail on top as well as looking better for iron sights. I'm really curious about this now, you have any concept art done of how it will look? I may be busy with finals, but little doodles are quick to make anytime. You don't have to worry a whole lot until your final product, and then you will want some of those flashing blue LEDs to sell units out.

Designing the "gist" of it with critical dimensions and then having someone who can really draw make the rest might be a good idea. If it were me I would just buy coffee for them and have a graphic design friend sketch and model things out to look "cool"


Back to top
DerAlbi
Wed Apr 27 2016, 03:26AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
I just finished printing a Glock18 or whatever just to hold it in my hands and get a feel for it. It feels quite small to be honest. The grip however does feel a little strange and i cant imagine to pull the trigger without loosing aim, but thats maybe just because its so light (100g)
Its about 3cm (1.2") wide and its kind of ok looking down the sights. I dont think that it would be a problem if the gun would become thicker below the sight, but holding it in the hand could become uncomfortable. The grip is currently 5cm (length) x 3cm (thick) wide and thats kind of the maximum.

Are you set on making this a one-handed revolver?
Well hmmh. I am kind of set to make it printable in one go for mechanical reliability. At least the acceleration path must be compact. I know what you mean.. if i make the whole thing bigger the dimensions in every directions become more relaxed before it starts to look bad.
The weight of the gun will be around 2kg. (wohoo! ) Thats about a big water bottle. Not comfortable single handed for sure. But even a one handed gun is held effectively (aside from the movie gangster grip) with 2 hands. not sure if you would really need to support it at the front of the gun.

I can think of some solutions for that to allow a much larger capacity or storage near the center. You lose a lot of space and have a potentially awkward reload is all. If you make a staggered magazine instead of the single width you normally see, you could fit quite a few shots in the back.
Complicated. First: yes cone can put the magazine in the middle of the gun. You can use the coils to transfer the projectile to the rear end starting position. HOWEVER: this leads to the issue that over the length of the magazine (which is at least one length of the projectile) there is no acceleration happening during the shot. A normal shot does not rely on projectile guidance very much (the rails i posted before are enough), since a centered projectile experience close to no lateral force as long as the projectile is constantly accelerated. The short region where the acceleration path is interrupted for the magazine, the whole concept becomes really ugly really quick. So as convenient it is to have the magazine somewhere in the center of the gun as conceptual flawed it is. I am quite sure the only solution is to have the projectiles directly at the butt stock (its also good for the center of gravity in a bigger gun).
Concerning a staggered magazine... i am not sure. Normal bullets are quite light. The spring needed to push the projectiles upwards would need to be really strong. I am not sure if reloading is nice that way. Also I fear that the spring will push on the projectile that is about to be shot. This would be a no-go.

I already talked to a skilled friend when it comes to drawing. But currently i have to figure out the mechanics that have to fit inside the gun before there can be any productive artwork. Its an iterative process.
Back to top
hen918
Wed Apr 27 2016, 05:49PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 383
DerAlbi wrote ...

...
I already talked to a skilled friend when it comes to drawing. But currently i have to figure out the mechanics that have to fit inside the gun before there can be any productive artwork. Its an iterative process.


Yep, my advice is to focus on getting the components done and as small as possible, then assemble them onto a practical frame / structure, then wrap it up in a cool case.
Back to top
Karmaslap
Wed Apr 27 2016, 07:01PM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
hen918 wrote ...

Yep, my advice is to focus on getting the components done and as small as possible, then assemble them onto a practical frame / structure, then wrap it up in a cool case.


He's designing everything from the start with portability, practicality, and marketability, so the typical "make it and wrap it" would be a terrible choice for him. He has to design it all to fit from the start.

And @deralbi I think we are considering totally different mechanisms for a center-mounted magazine. I understand what you mean but that would be terribly complicated especially for rapid fire. I will show you what I mean if you want but it will be in a week when I am done with finals. There are plenty of mechanical solutions to fit more ammo in the space you have.

Heavy water bottle? If possible you should put a front grip or place to hold, even with 2 hands the moment caused by the front weight would be uncomfortable to any user and that breaks the first consideration.

In another 1-2 weeks at the end of the year I'll draw up what I mean in CAD if you are still looking and designing

Back to top
DerAlbi
Thu Apr 28 2016, 04:30AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Yep, my advice is to focus on getting the components done and as small as possible, then assemble them onto a practical frame / structure, then wrap it up in a cool case.
Hehe.. hen, the problem is: "what is small"? 4x9cm or 6x6cm is the same area. Both are "small" but just one could possibly fit. So the whole design is really both ways. You have mechanical constraints to the electical stuff and vice versa. "as small as possible" in a 3 dimensional design has 2 dregees of freedom to much to be a design-helper.. in the end its about assigning priorities and compromises..

And @deralbi I think we are considering totally different mechanisms for a center-mounted magazine.
Possibly, solutions are infinite as are problems.
I understand what you mean but that would be terribly complicated especially for rapid fire.
Naaah totally manageable. The recharge time of the caps is much longer than projectile travel time. In the end you just give it a nudge, and let it slide to an initial position where it is held in place with a small nodynium magnet and its done. But as said, the problem is that the projectile needs to pass a small distance without magnetic guidance.
I have also other ideas to be honest.. but they all apply to a larger gun.. currently a 6 projectile revolver is totally fine. its just a demonstrator.

Well. the design will take some time. Right now... plety of time.
Back to top
DerAlbi
Sun May 22 2016, 05:28PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
It has been a craaaazy time for me... the last few weeks were really intensive...
The design of the reload mechanism is soooo out of my league... i have no education in mechanical design and in general no idea how stuff is done right... the 3D-Printer is running around the clock for all the tests and failures i had. Cant really describe how lost i am with the stuff...
..but it doesnt matter in the end.. its done!
Whats done? A proof of concept. Still a lot to do and stuff to improve..
First a photo (sry for the poor quality):



What do you see? Blue ABS plastic.. the shiny finish is achieved by bathing the prints in acetone for a few seconds....
The reload mechanism consists of a revolver drum holding 6 projectiles. One can insert projectiles on the rear end and on the front side it can come out like you see it in the picture to be shot by a coil. The drum is rotated by a stepper motor "15BY45-100" with 1.5cm diameter - so quite small, hence the gear (also 3D-printed).
A second, smaller stepper motor is placed on top of the drum which drives a M2.5 threaded rod thich in turn can slide a mechansim back and forth.



The mechanism first locks the drum and prevents it from rotating and then pushes the projectile into an initial position (which must be consistent in a coilgun).. thats why the projectile is sticking out like that.
The projectiles stay in place in every orientation because they are held in place with 4x4x4mm magnets.. So there is no frustration happening.

The rotation of the drum can happen in 150ms (for 1/6th turn). To have a reliable rotation i need to overvolt the stepper motors (12V instead of 5V) but thats ok given de low dutycycle of the motors. The lock/initial-position mechanism takea around 200ms to reach its final position.. the small stepper motor benefits from 12V too of course So in theory the whole revolver drum can be emptied in 2sec...

All rotating parts are currently held by ball bearings... they do have currently the wrong size (too small inner diameter, so they sit on too weak plastic nipples) but thats fixable so the thing should easily survive a drop fom the table soon.
There is also an issue with the software now... the stepper motors have to be driven together to perform a smooth action.. the steppers are quite new to me and honestly they are quite hard to work with. Their discrete nature of movement puts a lot of strain on everything they connect to.... i have currently implemented only "half steps" which is for all i know the strongest drive possible. I need to change that to much smoother microstepping... this should hopefully make it a lot less noisier but also prolong the life of the gears. The problem is in other words: stepper motors dive the thing by short impulses of high torque instead of a steady lower torque.. this creates vibration and also a lot of wear on the gears. I cant afford flywheels directly on the steppers so it has to be done electronically.

I think the whole system is better described in a video which i promise to make once the whole software (microstepping) and ball bearings are fixed...

I just wanted to do this post right now because i am hyper excited that i manged to accomplish that... you cant believe how much work it was when you are completely new to 3D-printing and 3D-CAD and all the mechanical stuff like gears and... aawwww the printer says it printed for 150h already
..so now the software.
Back to top
hen918
Sun May 22 2016, 07:57PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 383
Wow! Well done.
Would a servo drive work better? A normal DC brushed motor with a position sensor / encoder that tells the microcontroller it's in the right position. This would be simpler than a stepper and would solve the issue of the vibration, as well as being cheaper. If you put it through a worm drive you wouldn't have to worry about it slipping and becoming mis-positioned.
Back to top
DerAlbi
Sun May 22 2016, 09:39PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Cost wise.. the steppers cost below 3€.. thats worth it since its such an integral part... (there is 2€ additional cost for fullbridge drivers.. on the pcb later) but you are right.. i looked into servos for a really long time too... but i was never convinced enough to use them:

For the drum:
a) Servos usually only rotate 270° which wouldnt utilize the whole drum. There are 360° servos available (for model sail boats) but they a generally big and way too powerfull for the purpose..
b) there is the issue of connecting the servo to the drum. The easy way is to let the servo sit behind the drum.. which would increase the length of the whole apparatus. A compact design is kind of important to me because all my efficiency would go to waste if its not translated into volume efficiency too.
Actually the current design is still way too big (you can see that the gear is placed way above the bottom plate) thats an optimization for later...

For the lock/initial position:
a) A normal servo is too big.. even the tinyest ones barely fit.. not to mention that the positioning must be so that the leaver can operate the mechanism.
b) there are ultra small linear servos (for the tiny helicopters) available.. they would fit and easy to use. but their strength is an issue. (they also cost >10€) Currently the linear lock mechanism can lift up as much as 5 projectiles (120g) with ease. the main problem right now is that that i have no end detection to the motor can turn full power and tighten the screws so much it cant start again But thats just a semi-problem, because in a real gun i have projectile sensors in front of each coil, so i can detect that the tip of the projectile is penetrating the coil and therefore have a virtual fixed position detection. I also can zero the initial position by driving the motor not at full power so the screws dont tighten so much that the motor at full power cant unscrew them again. In the end... softwareproblem.. but manageable.

Worm drive:
The lock-mechanism basically is a worm drive.. it just issnt
But i know.. you mean for the revolver drum.. yes yes i thought about that too.. its easy to implement and even a worm drive is one of the worst (efficiency wise) a smaller motor would do the job due to better gear ratio.
A problem with worm drives is:
a) it really locks down the rotation. What if something is stuck or a projectile misfits or whatever.. no manual rotation can be done... right now one can at least fiddle around and have some success.
b) the locking of the rotation does not solve the issue of the initial position projectile placement.. since this effectively implements the lock allready its a unnecessary feature
c) there is the issue of 3D-printing the parts. the 3D-printer fails horriby if features become smaller than 0.8-1.4mm under some conditions.. a worm drive can be bought but the gear must be printed directly into the drum-mag. This is also the reason why i didnt use 2x 45° involute gears to make the 90° connection from the big gear to the drum... i just cant print such a gear wheels. but the improvised 90°-translation seems to work ok.

With bought parts i allways have the problem to interface with them... what do you do with a metal gear? where to put it? hmmh.

I also tried different kinds of drives like linear actuators for the initial position (with coil and permanent magnet) but that wasnt strong enough in a usefull form factor (or the wire needed was so small that i cant wind a small coil by hand..)
I also tried to rotate the revolver drum with a ballpen-like mechanism.. but that wasnt working in any ways. Such a mechanism for rotation is a force multiplier with F/cos(angle) so it increases the needed torque even if you neglect the horrible friction...
Back to top
hen918
Mon May 23 2016, 05:03PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 383
ahh, you're thinking about using an RC type servo motor (You can actually remove the stop pin inside them to get them to rotate continuously). I was thinking of a normal DC brushed motor and a separate proximity sensor for the position location, maybe a little magnet on each of the stations and a hall-effect sensor to tell the uC (microcontroller) when it has reached the next projectile position. Maybe a motor like this , with the output gear swapped to a bevel one, top allow drive at 90 degrees to the motor axis.
Back to top
DerAlbi
Sun Aug 14 2016, 03:28PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
So after a long time i feel like i can present some stuff again. The preliminary casing is kind of finished... at least the essential parts. I can tell you as much, that even i should have motherly feelings, its just an ugly child. The proportions of a small coilgun will never be pleasing although i havent spent any time for design elements nor is it finished completely.
A revolver style drum magazine (from the posts before) holds six projectiles and will be able to reload in half a second. Also the base of the acceleration path can withstand a static force transfer to the grip if excess of 400N which gives plenty of headroom for a real shot (250N peaks). There is also a nice trigger mechanism done which provides good tactile feedback and a satisfying clicking sound - which of course wont be heard in context of the *puff* when a shot is triggered...
The front houses a 120J cap bank and its currently unclear if all the electronics will fit in - specially a 200W cap charger out of a small 2S LiPo-pack (inside the grip) will be tricky.
I am currently aiming for 20J kinetic energy and some proof of concepts concerning the coil wire diameters. The issue is that i want to restrict the usage of very thin wire diameters and use 0.5mm WD minimum. The lowered inductance will become a problem so i need to PWM the first few coils at a relatively high current in comparison to what would be needed with the right wire diameter. It will be interesting how this will impact the efficiency due to eddy currents and higher impact of cap bank ESR (500mOhm).
I am also migrating to ARM-Controllers (AVR32 before) which will impose quite a lot of software overhead (I have to rewrite my hardware abstraction layers)..
If this thing succeeds it should be the most efficient coilgun with the highest power density out there. A good starting point for credibility if i want to go bigger someday.


I know.. not a big post for such a long period of silence but mechanical design... as i said before.. issnt my strongest ability
I now have to plan on how to distribute the electronics within the case. This will be ugly. And I predict an even longer period of silence again. All the circuits passed the proof of concept stage already.. so nothing exciting except the form factor about it.
Back to top
DerAlbi
Sat Aug 20 2016, 11:37AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Wow. I just found a great competitor on Youtube!!

Its chinese, his math works out (if the measurements are right) and the results are awesome compared to most of the other coilgun stuff on youtube. The excellent build quality is just a joy to watch.
This dude pulls with 20% efficiency too! (which is also my goal in my design). Just great.
I mean.. its not really competition in the sense of "technical specs", but what he built and what formfactor he chose is similar and his efficnecy rocks and the weight will be qual to my design.
I found this design so outstanding that i wanted it to mention here. Maybe he gets some more clicks, higher ranking on youtube and so on.
Comparing the data:

The most notable difference is: his data is real.
I (will) achieve a notably higher energy/power density and more bang per volume and weight and so on. I am not sure how he achieves the high efficiency with only 3 stages and i dont know which topology he uses. I guess he is waaaaay below saturation thus the low kinetic yield. He clearly uses a dedicated injector coil (the short first one) which is a good choice...
He got an issue with the trigger latency however.

Edit: fun fact... i found the clip when i was watching the youtube channel "codys lab" with his second attempt of a "coilgun" (https://www.youtube.com/watch?v=Py8BdLTZEAE) which is unfortunately a "different" quality again.
Back to top
Karmaslap
Fri Aug 26 2016, 12:52AM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
Hey, any progress is progress. Good luck fitting everything into the small case you have- that's pretty tricky looking.

I've seen that chinese gun, and have to think that having an injector coil or other mechanism is the way to go, though you do lose a half-stage's worth of length for better efficiency.

I like the cody's lab channel, particularly when he uses furnaces and some basic chemistry to extract metals.
Back to top
DerAlbi
Fri Aug 26 2016, 01:21AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
You know what i just realized... the Chinese data is bullshit -.-
In the first part of the video he demonstrates how the hammer of the trigger mechanism alone gives the projectile a good initial velocity. This means that his kinetic energy extracted from the coils is lower than m/2*v² (but m/s*(v²-v_start²) which would decrease the efficiency. But not much, but i guess it would be enough to be at the 16% level which was the current record i know of. And the competitive efficiency is then explained easily: those initial moments with low projectile speeds are the most wasteful during the whole acceleration. Avoid that and you cheated the way up the ladder.
But still it deserves credits for engineering around the problems. Its just not an A+ anymore for bad measurement.

Hey, any progress is progress.
Actually the progress is immense given the work i put in. Just the result is a bit disproportional.

Cody
I like his channel too. But his coilgun video stands out. Either he is waaay out of his comfort zone with electronics which would be understandable.. or - and thats what i would find interesting - its just the first time i recognize deficits in his videos because i know for the first time what he actually does. I usually dont know much of what hes doing.. so i am easily impressed of course.. a good youtube consumer so to speak . It just makes me wonder if the rest of his "science"videos have those flaws too. But i guess he is just not good at electronics - thats fine.
Back to top
DerAlbi
Thu Nov 03 2016, 03:48PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Just a quick note.. i am still on it.
But its getting harder the further it develops. The compact form factor is really challenging me to my maximum. Its just so tight and designed on edge that I often loose my faith in it ever will work reliably.
This really shows me the benefits of a larger gun. The sheer length of the acceleration path or having a butt stock would bring soooo much room with it. Instead I am squeezing a 200W capacitor charger plus LiPo in my grip by fighting for every half millimeter. Insane.

Anyway. The capacitors, IGBTs and Diodes have arrived.


I am currently designing every PCB concurrently just to see on what PCB is space for which circuit.
I also made progress with my high frequency sensor system which has become much more immune to the acceleration coils by using a little different sensor topology. I have tested a virtual sensor string of 30 sensors and it was good enough. This will simplify the winding of the coils, even it does not solve every issue, it certainly relaxes things.
I also refined my projectile guidance with the rails which does really look good now.



This whole test setup is 16cm long and the projectile begins to slide when one side is elevated to around 3cm. This makes the angle only about 11° which translates to a super low friction coefficient. On the other hand this will not help with mechanical stability.

Other work was done too.. software will be one of the things that needs to be done and I started to work with SMT32 microcontrollers.. I fell in love with them.. and it took time off the hardware. Also the GCC compiler supports the new C++14 standard and its really a joy to rediscover C++.
If anyone finds some motivational words, feel free.
Back to top
hen918
Thu Nov 03 2016, 05:10PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 383
WOW! Never give up. Take the odd break occasionally, but it can be done!

If I were to design a coilgun (and I did start to, once) I would sacrifice size for modularity and reliability.
But what you have done is next gen
Back to top
DerAlbi
Fri Nov 04 2016, 01:04PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Thank you, that counts as support
Unfortunately the next update will take a while again, so i will come here and re-read your post for some time ^_^

Just a sneak peek:



I can only explain it with words:
The capacitor bank is mounted on a PCB. On top of the PCB there will be small foil capacitors mounted (http://www.digikey.de/product-detail/de/epcos-tdk/B32562J6224K/495-7058-ND/3491077 with pins removed, mounnting them as SMD)
These capacitors will reduce switching losses of the half bridge. These foil caps will have a spacing of 12mm. This is little more than the width of a TO263 case (which my diodes and IGBTs are).
So the plan is to have the half bridge on a seperate PCB where some diodes/IGBTs are on bottom and some on top (the top side you can see). The bottom parts will fit tightly between the foil caps so no volume is wasted.
The foil caps will also work as interconnect from cap bank to the half bridges. So in the end the foils caps are SMD mounted from top and bottom sandwich style.
Sry for the text descr. i have no 3d model of this arrangement to show.
The space of the half bridge PCB where the power components are on bottom in between the caps is used for the IGBT gate drives. Wish me luck, that it will fit.
Back to top
Barry
Mon Nov 21 2016, 04:57AM
Registered Member #90
Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 298
Awesome job, keep it up! I just started catching up on your coilgun thread and I'm really impressed. I know how difficult it is to maintain the effort over such a long period of time.
~ Barry ~
Back to top
Website
DerAlbi
Mon Nov 21 2016, 11:17PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Whoopdiduda! THE Barry himself.
As it so happens i actually have small progress. Everytime i am not satisfied with a solution i let the project down until i come up that i can spent money on without feeling too bad.
I went over my halfbridge again.. i did not feel good about the active clamping circuit because there right TVS-Diodes are hard to find. (aka unavailable or impossible)
The basic active clamping works like this:



So when the Drain(Collector) voltage rises to dangerous levels the TVS becomes active and keeps the gate open as long as it takes to get rid of the stray inductance energy. Diode D8 prevents the IGBT/Mosfet to pulldown the gate drive level when the device is on, as alternative one can choose a bidirectional TVS and all is well. However as I want to take full advantage of the IGBT device choosing the right TVS diode becomes more or less impossible. To Use the IGBT optimal you need to fully go towards the voltage limits of the device. Since i use 650V IGBTs this means my CapBank-Voltage is as high as 600-620V. This means you need a TVS diode which does not conduct at all at 620V but conducts fully at 650V. However such high voltages the diodes arent that steep. This bugged me for a while..

Now i came across a paper with alternative active clamping which is more complicated but more reliable and actually smaller than the bulky TVS-Diode solution which would needed at least 2 TVS in series to assemble the right breakdown voltage.
My implementation looks like this:



The low side is straight forward. The gate drive is usually held to GND thus charing up the 10nF cap to the full CapBank-Voltage through D2.
So at turn off Diode D3 (in an overvoltage situation) becomes conductive because then it is the highest source of voltage. The injected charge makes current flow into the gate holding it open to clamp the voltage. The cool thing is that its extremely relieable and the voltage can be fine adjusted by R4. But its allways just about the original starting voltage of the cap bank which is perfect for IGBT utilization.
The highside is a little more complicated because the GateDrive is not naturally held at ground level. This is why i added M3 as a precharge mosfet. Its is mosfet it conducting before a shot making the level "L+" gnd for a short time (this alone will not power any coil). C1 is then also charged to 600V and ready to protect the IGBT.

This now fits nicely on a 85mm x 40mm PCB.. all 3 halfbdiges of course. Tight.
Back to top
DerAlbi
Sat Nov 26 2016, 04:15PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Yippieh ay eeh! Next PCB is done.



You are seeing the portion of the casing above the grip below the revolver magazine.
The 4-layer PCB hosts the power management brains and user interface. Meaning the trigger switch, 6 capactivie sensor channels, 2 stepper motor drives, 1 inductive sensor to detect the presence of of a projectile in the reload channel and the USB interface. Notice the extremely tight fit. Its a real fit-to-envelope-design. The PCB as little sideway-bridges that dock into the plastic case so its position is secure. Also notice the componenents even inside the ball bearing because there is no place elsewhere.
The PCB also interfaces the grip-pcb (hosting the battery, 200W Flyback, 12V, 3.3V power) to itself and another helper PCB (hosting the isolated power supplies for the half bridges) as well as the shot controller that will be another PCB hosting also the battery charger and the interface to the projectile sensors.
Small progress. Still a lot to do. Next up i think i will so the projectile sensors for real.

Edit: i forgot to mention: the pcb also houses an accerlometer. For power management and safety of course.
Back to top
DerAlbi
Fri Dec 02 2016, 02:37AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
To get more stuff done i just inserted a little side project: coil winding machine.
To make the projectile sensor PCBs in between every coil i need to know how the actual coils will look like. The coil shape will be quite complex - its not really round nor have all layers the same number of turns.
I also am quite sure now how i can make the coils in such a way that they will self-align with the whimpy projectile guidance rails. This was a critical problem that i pushed away. I was essentially unsure what made the coils not move during an impact event or even during a shot. In a normal coilgun the tubular barrel usually gives a lot of structural stiffness - that i do not have anymore for law conformance (i hope).

To get relieable results for the coils i now want to build a coil winding machine. I am allready on construction and all the essential parts are ordered. Some parts are allready even 3D-Printed. My main concern is currently the regulation of the wire tension, but i think ive got a good solution at least for smaller wire diameters.
Of course the software must be done too.
If anybody is interested, i can describe the project in another thread. Maybe its even possible to build multiple machines if someone want to have one. But its at least 250€. (Still better than commercial ones that are full cnc)
Back to top
Justin
Fri Dec 02 2016, 04:22AM
Registered Member #46164
Joined: Wed May 07 2014, 08:16AM
Location: California, USA
Posts: 66
DerAlbi wrote ...

To get more stuff done i just inserted a little side project: coil winding machine.
To make the projectile sensor PCBs in between every coil i need to know how the actual coils will look like. The coil shape will be quite complex - its not really round nor have all layers the same number of turns.
I also am quite sure now how i can make the coils in such a way that they will self-align with the whimpy projectile guidance rails. This was a critical problem that i pushed away. I was essentially unsure what made the coils not move during an impact event or even during a shot. In a normal coilgun the tubular barrel usually gives a lot of structural stiffness - that i do not have anymore for law conformance (i hope).

To get relieable results for the coils i now want to build a coil winding machine. I am allready on construction and all the essential parts are ordered. Some parts are allready even 3D-Printed. My main concern is currently the regulation of the wire tension, but i think ive got a good solution at least for smaller wire diameters.
Of course the software must be done too.
If anybody is interested, i can describe the project in another thread. Maybe its even possible to build multiple machines if someone want to have one. But its at least 250€. (Still better than commercial ones that are full cnc)


I'd love to see the project.
Back to top
Website
DerAlbi
Fri Dec 02 2016, 05:59PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Give me some time. Meanwhile i had a little setback. it turns out the wire tensioning is not so easy. If one wants to cover also thick wire the amount of redirecting the wire must be limited. So i am back to the drawing board.
Once i finish the overall concept, i will share the progress again.
Back to top
DerAlbi
Sun Dec 04 2016, 06:17PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Small progress again. While evaluating if i need a M8 or M6 threaded rod as coil-former shaft i constructed the inner shape of my coils. Given the actual projectile guidance are those 3 rails the coil shape is not round:



Now the question is, if this is actually helping the coils to have a higher time constant. The inner diameter of the coil in this configuration varies from 11.25 - 11.8mm while a complete round shape would have a diameter of 12.2mm. The air gap to the projectile is as small as i feel ok with (0.5mm) in between the rails.
I am not completely sure how this inner shape will translate to the coils outer shape. Of course the edges smoth out, but if it will become completely round - i doubt it. If it becomes round there is not much gained because the only turns that become effectively shorter are those which are small in diameter thus small in length anyway. hmmh. If anyone has an opinion on the coil shape.. share

My thoughts:
The optimal coil shape is said to be OD = 3*ID. Since my inner diameter varies, also the outer diameter varies. Given the rule of thumb i would need OD= 33.9 - 35.4 in cointrast to the original >36mm i had in the first test design. This translates to 2 layers of wire for 0.5mm wire and 1 layer of wire for anything bigger than 0.6mm and its likely the same number of layers for wire diameters above 1.0mm.
This does not sound significant to be honest.

Edit: i just measured the innermost layer with actual wire the reduction in circumference is ~5%. Yippie ay eh.
If this translates to the overall wire length and therefore the reduction in resistance this means my previously 1.7ms L/R-Timeconstant will now be 1.8ms. What an improvement. Actually i cannot extrapolate anything since L changes too of course. But the smaller air gap should increase projectile coupling therefore slightly increase the pull force.
Back to top
Karmaslap
Mon Dec 05 2016, 09:57PM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
you don't even need the rails, the projectile would pull itself through the coils just fine already, right? This is just for legal reasons?

.5 mm is such a small gap, you don't think your coils will begin to constrict and narrow that?

I'll make the bet that your final coils are close to round on the outside and not significantly different than regular-cored coils, but I don't think that is a surprising guess.

Liking the updates.



Back to top
DerAlbi
Mon Dec 05 2016, 11:17PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
you don't even need the rails
Complicated issue. Yes and no.
Yes: if magnetic forces arent perfectly symmetrical then the coil will create a lateral force. The closer the projectile gets to one side of the inner wall, the more its attracted to the side. Bad situation; positive feedback.
No: as long as there is a steady current flow in the projectile due to eddy currents there is a repellent force. I actually have recorded high speed video footage showing a projectile centering itself inside the coil.
Still, a shot is a pretty violent thing, there might be other forces to the projectile you do not want to transmit via the magnetism involved. (recoil)
However i am quite certain that the current implementation of the rails as wimpy as they are is sufficient. Its simply a matter of safety to have them. Also the projectile sensors have even smaller air gap. i would not build a gun based on the thought that the projectile self aligns during a shot with sub-mm precision.
.5 mm is such a small gap, you don't think your coils will begin to constrict and narrow that?
Epoxy helps here. Deformation would be a no-go. But i actually would like to decrease the air gap even more. Its a matter of being able to 3D-print the inner coil shape with an 8mm hole. There is a limit to the practical wall thickness. Unfortunately the 3D-printer adds 0.2mm thickness to every side of the wall which sucks extremely.
Back to top
Karmaslap
Tue Dec 06 2016, 08:53AM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
I believe that's why you originally had a thin teflon tube to help guide the projectile?

I remember seeing that video, that is what I was thinking of when I asked. What are you using for the rails?

Probably still want the guiding rails because the final prototype won't always be fixed down when you use it and lateral forces from movement might push the projectile into the side of the coils, not that it would make much of a difference with auch a small gap.

That sounds pretty good for a cheap printer. You can smooth with some acetone and sand down the edges or center if you think you can be precise enough as well

Any progress on the legal side of things?
Back to top
DerAlbi
Tue Dec 06 2016, 01:10PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
I believe that's why you originally had a thin teflon tube to help guide the projectile?
Yes, and maybe thats still a thing for later because these rails do not have any mechanical function. But thin plastic suffers very much from manufacturing tolerances. i am not sure if 1mm teflon will be available. For now i just use 1mm FR4 glass fiber rods.
That sounds pretty good for a cheap printer. You can smooth with some acetone and sand down the edges or center if you think you can be precise enough as well
Cheap = 2k$
Finishing the print is complicated. Its not deterministic and edges smooth out while other areas may swell up. For a mechanical precision part like a coil former this virgin print is the only way. The precision of the printer is so high that its perfectly reproducible and its perfectly what i designed. You just have to design in the fact that the walls are 0.2mm thicker than the model.
The legal side does not move. I need to design this prototype and when this is done then there might be progress. But thats still many months away. It takes a bit longer than estimated.. and now the coil winidng machine in between. it will also consume a month.

Edit:
Why not show the progress. Printing parts.. quite stable actually.


Back to top
DerAlbi
Mon Dec 12 2016, 10:46PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 499
Mechanics are ready, next electronics, then software.
To be honest, i am kind of proud. it looks good. Stiff enough even for fat wire.


Back to top
Karmaslap
Wed Dec 14 2016, 07:47AM
Registered Member #58215
Joined: Wed Dec 30 2015, 11:27AM
Location: Boise, Idaho
Posts: 62
Those look pretty good. Bet you're glad you decided to go for a printer. How much in plastic did you spend on those?

Looks a whole lot more complicated than just winding by hand, though. Is that copper wire? For the prototype? I remember you writing you chose square aluminum wiring for the strength to weight

I'm not doing anything with my project, as you can tell. Going to look at some used electronics parts at a big electeonics recycle store in 2 days and see what I find, though, to add to my tools and supplies.

From your first post, you had a 10-stage prototype that had a peak efficiency of 27% and muzzle velocity 50 m/s, what happened to that prototype?
Back to top
Go to page  1 2 3 [4] 5

Moderators: Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Bjørn, Dave Billington, Steve Conner, Wolfram, Kizmo, Mads Barnkob

Jump:     Back to top

Powered by e107 Forum System
 
Legal Information
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.