Optimizing a Multi-stage Coilgun
Вставка
- Опубліковано 6 лют 2024
- This video i optimize a multistage coilgun based off the results from tests in my last video. The coilgun’s performance was improved predominantly through controlling when the coils are energized and switching the power off to them when the projectile reaches the center of the coil eliminating the losses due to the magnetic field pulling the projectile back in as it passes through to the other side. #coilgun #electronics #projectile #engineering #electrical
Its been years since anyone on YT did a cool coil gun project in a cool way. 😊
Thanks! Im glad you enjoyed it
You said a problem was charging took too long by the time the projectile was half way through the coil to shut it off. What if all the coils started activated and shut off incrementally as the projectile reached the center of the coil. Theoretically, that would make the projectile increase in velocity with every stage as the next stage would already be at max charge and magnetic strength before the projectile excited the previous stage.
Yeah, that would definitely ensure the coils are producing the strongest magnetic field as the projectiles are passing through them. I would have to redesign the capacitor bank to handle that or move to batteries at the capacitors im using now dishcarge completly within a few milliseconds. But is a good idea to try in another design
I was just researching and read that enclosing the coils by putting some strips of metal on the outside of them will hold the field in and make them stronger. Doing this made a significant difference on a solenoid motor (where I read about it).
That's quite interesting. I will have to try it out. Did you have a link to where you were reading about it so i can do a bit of research on it?
@@owensprojects I lost my browser history and couldn't find the site back. Enclosing an inductor with metal was a known effect back in the old days, thought of as guiding or holding in the outer field lines, but a quick googling of search terms didn't find me what I was looking for. It doesn't have to be like air tight, the solenoid engine website guy just used some metal strips. .. I think a mistake many are making with these coil guns is probably using huge caps and limiting speed by the LC frequency. Maybe many smaller caps parallel instead of one big one could get around that. Also when designing transformers we use volts divided by inductance times pulse time to get actual current sunk by the inductor each pulse..
That sounds like a good idea to explore and test the benefits of it. I haven't come across anyone else with that design either. With the capacitors is there a way to workout how many in parallel would be ideal over say one large capacitor?
@@owensprojects I did run into, since last reply, a coilgun guy mentioning trying iron laminates on the outside of coils. He said it changed inductance too wildly causing problems, with too small of increase in strength for the problems created. Again I can't find it back cuz when I get to digging.... So I would stick with testing steel like normal solenoids come enclosed in. Also I wouldn't expect more than 20% effect at most.. I've been heavy into switch mode power supply design and just got interested in this. I don't know the cap time constant off memory, but caps in parallel, like batteries, will add current out/in while keeping the single time constant of charge discharge rate. There is a formula for that somewhere and they need to align with the inductor's pulse time and current sunk from the (Volts/Henries) x pulse time = current sunk. I don't have time right now but will be looking into the formulas over the next days. I have in mind an automatic that shoots 5-10 ball bearings per shot :) But I have many projects and just started researching this one.
@@owensprojects Oh i forgot to say.. I think using a mosfets for the switching to control on/off would be the easiest fix instead of using a thyristor like most seem to use. The thyristor(SCR) is what makes the timing rely solely on the capacitor size to inductor value. Just make sure to use a sinking diode for the reverse surge. If I go with dumping 400v I have some sanyo 2SK3748 mosfets laying around, if I go under 100v high amps i have some IXTH200N10T beasts.
For experimental sake, try a cylindrical magnet projectile. Or a sabot + dart shape (although I guess the OD of the projectile should be as close to the ID of the coil as possible).
Im very interested in trying a magnetic projectile, there is some mixed views on them, but it would be good to test it out.
Yeah, having a sabot round would help with keeping the projectile flying straight and increasing range, but like you said, having the OD of the projectile further away from the ID of the coil would mean the transfer of energy would be diminished offsetting any gain from a sabot round.
@@owensprojects Since you did a really good job with this vid, ueh a comparison between iron and a magnet projectile would be very interesting. A threaded locating pin may allow you to screw some sort of non-metallic fin on the back of it while keeping the fit for the bore, finding one thats not stainless may be an issue though.
It's great to see another Aussie get technical with coils not many of us on UA-cam compared to the rest of the world. My expertise is brushless motors and experimental coils for motors so very similar to what you're doing here. I can suggest wrap your coils with steel wire between each copper coil layer windings,this gives more flux concentration and higher inductance and efficiency. Experiment with different ferrite metal ratios as well to get the Goldilocks frequency for maximum power. An oscilloscope will help find resonant frequency for your coils adjusting the PWM cycle🙂
Thanks for the suggestions! I will experiment with them on my next design and see what improvements i can make.
I did notice when you sharpened the end of the projectile it was not uniform and as a result to projectile started to tumble as you could see in your ballistic gel. I believe I have a solution for that and when I do, I will privately give it to you, but it will be proprietary technology.
I would be interested to know what it is when you do, as the projectile doesn't fly well how it is now.
woooah! Nice video
I guess the reply i made just now didn't go through because i don't see it. i was trying to tell you that i found a video on here of a guy testing the metal enclosed coil thing. He claims a 50% improvement on one stage but didn't transfer to 2 stage test. He is Hyperspace Pirate and it's on Electromagnetic Accelerator Improvements Part I - Gauss Cannon test rig. Hopefully this one goes through.
I have just watched that video, that was really interesting as most the info i have found so far says not use any materials that can hold a magnetic field as it will increase the force pulling the projectile back into the coil and slowing it down, but after watching that video it doesn't seem to be the case or the increase in initial magnetic field is more the then the force slowing the projectile back down when it comes out the other side. Thanks for sharing that, looks like I'm going to have to do some more experiments now
Nice work, great to see the well described science. I tried to make a solid ring , coil motor, bending the coil gun into a loop. "electromagnetic accelerator toroid ring motor". Your welcome to try it.
Thanks for the idea that sounds like an interesting project to build
keep up the good work!
Thanks!
The object doesnt twist like a bullet, making it less accurate.
I would use a sabot round, use non magentic material with fins or tail veins, possible just make it into a light weight arrow
I have been looking into sabot rounds, would be interesting to incorporate that type of projectile to a coilgun design. Especially since you dont have the hot gases and rifling that improve the accuracy like in a conventional gun.
very good!
Hi! probably too old video to comment this, but I have a good idea to improve this setup
The coilgun is quite similar to a electric motor and have pretty much the same problems. Acceleration from steady state is quire difficult and most of the configurations are pretty inefficient in that scenario. As you see, when you put more stages you increase the overall efficiency, basically because the stages that receives a pre-accelerated projectile are more efficient, since electromagnetic field and high speed are good friends.
To solve this problem of pre acceleration I suggest to use a pre-acceleration stage using rubber bands or compressed air. This two options are quite good delivering high power at lower speeds (exactly the opposite from your coilgun).
Results could be quite promising, since I human can easily deliver ~20J from a sling and you prove that under the low efficiency/low speed region your were still able to deliver ~20J.
After this experiment I think you will be facing different problems, like dimention of coils (that should be decreasing on each stage) and friction (vaseline?).
I over extend hahaha. Just one more thing, your content is awesome I glad to see how happy you look doing your research! Saludos!
That is a really good idea as the first stage just to get the projectile moving uses so much energy for not a lot kinetic energy output, I do like potentially using compressed air for the first that would be much more efficient to get it moving.
Friction is already an issue as when i tested the same first stage coil with a shorter barrel keeping everything else the same was about 2-3m/s faster. Looks like I'm going to have to revisit the project and try those ideas out
Thanks for the kind words!
Great video 👍👍👍, I have watched many coilgun video and definitely this is one of the best. I started to build one too but burning rate of IGBTs really scary :(. Btw, would you mind sharing the part numbers of NPN Q1, PMOS M1 and freewheeling diodes D4 and D8 please?
Im glad you liked the video!
The part number for diodes D4 and D8 is C3D10065I, its a schottky diode made by wolfspeed.
Q1 NPN transistor part number is P2N2222A.
M1 PMOS part number is IRF9540N.
Hope this helps
Very cool video. Interesting how efficiency improved by adding additional stages.
Are all coils identical? I think you could try to reduce turns / coil by every stage, this will decrease coil inductance and current will rise faster.
Thanks! it was definitely shock how the efficiency increased. Yep all the coils are identical, that's a good idea for me to try in the next version
i think the coils should be the same..not sure if they have too......but they should be.
A jig using a drill and the grinder to make more consistent points on the projectiles😊
That would have been a much better way of doing it than the caveman way i did it haha
You know you could have just used a railgun design. And then used coils as a replacement for the permanent magnet fields.
And you wouldn't have to rely on the timing for switching fields on and off for optimum kinetic coupling.
A Railgun design definitely does have its advantages over a Coilgun, i will look into building one for a future project
@@owensprojects Rad !
Is there some way to measure the voltage or inductance of the coil to figure when the slug is in the coil vs using a light sensor? It seems like it would be simpler if possible
There are a couple of ways to measure the inductance of the coil but not a way that i have found so far to do it while it has a high voltage across the coil when it is being fired. I will have to do some research into it and come back to it
Very good, which IGBT are you using as the switch?
The IGBTs part number is IXYX50N170C
@@owensprojects thanks, I noticed that u use steel as your projectile. I think soft iron is better as it is more magnetic. Would love you to compare these two in a video. Cheers
I will have to try a soft iron projectile. Hopefully, being more magnetic leads to greater speeds
I tried similar arrangement, in sysmaxwell program but I couldn't manage to increase bullet speed about 45% from coil 1 to coil 2 like your one, Can you share me the hint how did you manage?
Im not familiar with the sysmaxwell program and how you are simulating the stages, but the critical thing to get right is switching the coil at the right time so when the projectile is inside the coil the current flowing through the coil is at its maximum and switching the power off to the coil at the right time so the magnetic field isnt slowing it down after it passes the center. Hope this helps!
@owensprojects firstly thank you for your reply, switching time of coil in my project is managed by sensors and I am doing it(switching) just before enter in next coil, and even though I tried to use tick wire with low number of turn there is no change on speed increase performance which is about 6% increase while as your one above 46%, at this program ı can see also amperage building up time as well as magnetic (Wb ) building up time, this program showing me that magnetic building up tame is not overlap or get close to amperage building up time. I tried to compensate this phase differences by degrease number of turn and used ticker wire but there is no noteworthy improvement, ı am trying to figure out where is this differences coming from while as we are doing same things.
Do you think I could get a complete schematic layout and diagrams from you on your project add a copy of the Arduino code are you using a separate Arduino for each coil as well
I'm using separate micro-controllers for each stage as the timing for each stage is critical. Which could vary between shots using one Arduino depending on which line of code is being executed when the light beams are broken.
This Github link has the schematics and code for the project github.com/Owens-Projects/Multi-stage-coilgun.
If there was anymore info you want let me know
Thank you I will. I’m going to be using one Arduino to control the first Arduino and separate Arduino for each stage as main Arduino that I have will collect all the information down the line, including the voltages per stage and the overall charge voltage of the power supply and will be written to an SD card, oh I forgot the master Arduino is going to count the number of shots fired as well. Store those to the SD card.
That sounds like an interesting project! You will have to keep me updated with it.
Здравствуй отличный результат ,если интересно могу скинуть свои работы
That would be great, i enjoy seeing what others are working on. Do you have a link to it?
@@owensprojects да я работаю 10 лет в этой области делаю лучшие ускорители масс Гаусса и рельсотрон сейчас несколько ссылок скину
@@owensprojects ua-cam.com/video/yRs5qUYATpQ/v-deo.html 124м/с 9,7%
Please tell me the specifications and code on igbt
The IGBT's are 1.7Kv and 178A rated. The part number for them is IXYX50N170C. Hope that helps
Dear Owen, another question, typically how long did you turn on the IGBT? Again, I ask this because you mentioned about 200A of current flowing thru the coil. I have tested with 2.5ms and my IGBT burned miserably...
I used 2 IGBT's in parrallel to handle the current. They worked all the way from having them on for the complete time it took to discharge the capacitors to the first stage was on for 1.8ms and each other stage was 1.1ms.
@@owensprojectsTqvm Owen for your quick responses, many thanks
Are u using soft iron as your projectile? if not you should as it does not saturate. could you also tell me the coil length and wire diameter ? thanks in advance
Thanks for the tip, I have just been using mild steel for the projectile but i will test the improvements soft iron will make in the future. The coil lengths on average are about 58mm depending on how tight i could keep the wire when winding them and the wire diameter is 0.75mm.
A cylindrical stack of neodynam magnets will have the best effect as they already have a magnetic field to push against (just like BLDC motors use)... just insert them with the polarity facing the right way, and perhaps press the stack into a 3D printed tube to keep them together and protect them for impact. Much more expensive than iron slugs, but would be interesting to see the efficiency gains.
I have wondered how it would go using neodymium magnets as the projectile, i will have to test this. Like you said, though, i need some sort of tube to protect them as i dont think they hold their magnetism well if they do get smashed against something
@@owensprojectsIt will be innefficient as the magnet will try to induce current in the coil when it moves. Also you would need to get the timing very right since the poles would repel the opposite direction if the magnet is already at the center but there is still current in the coil.
@@owensprojectsYou can also try to use powdered iron matrix to cover the whole coil. And increase magnetic field strength.
is it necesary to have external diode with the igbts as they have body diode
For those particular IGBT's potentially wouldnt need them, i went through several different mosfets and igbts before settling on the ones used in the video, As they kept blowing up due to the voltage spikes even though they did have body diodes too. So, in this case, it was more of better be safe than sorry and cheaper to add a $8 diode, then keep replacing $45 IGBTs.
@@owensprojects oh wow fair enough... What igbts do you use? I thought the diode across the coil was like a freewheeling diode and would prevent the voltage spikes alone
@@owensprojects I forgot to say amazing videos btw
I ended up using IGBT part number IXYX50N170C. I thought the exact same thing to begin with, with the diode across the coil. They were fine until i was switching them off with about 200A flowing through the coils. Im guessing the voltage spike was just building too rapidly under those conditions for the freewheeling diode to shunt it back across the coil, and the mosfets/IGBT's couldn't handle it.
Thanks, I'm glad you like it.
Plz Share circuit diagram
Follow the link to github.com/Owens-Projects/Multi-stage-coilgun i have uploaded the schematics there
It's strange, I write and the messages are deleted
UA-cam comment cship is wild. Welcome to the club of people who go back to random videos to see if their comment survived. Most people are happily oblivious.
Most likely it'll be some word or short phrase that in another context might annoy the gods.
Look at me