an Electro-Magnetic Linear Accelerator (Coilgun)
School Project, 2004 - 2005

 Project Start: 18 July 2004 


Introduction: What and Project Details
News: News, Events and Others
Project Log: Day by day construction details.
Experiments: Experimentation and Testing of the Coilgun!
Part 2: The coilgun gets a massive rebuid

Current Project Status:

Maximum Efficiency: 4.35% (mid-2005)
Project Rebuild
Main Model Completed! (14 April 2004)
Main Model under Construction
Just Started (19th Jan 2004)



On 19th Jan 2004, I wrote a proposal and submitted it in for my school's compulsory research education programme. Having done some research on pulsed electromagnetic mass accelerators using coils to create a magnetic field (coilgun 1a, and coilgun 1b), I decided to take these smaller projects one step further and do a more comprehensive research on them, especially improving efficiencies of current Electro-Magnetic accelerators. My project aim was to do a study on improving efficiencies of these EM accelerators.

The reason for doing this project is a simple one: Current EM accelerators are from from efficient. A 10% Efficiency would be considered excellent, however, when compared to EM Rotary Motors (standard electric motors), efficiencies of up to 80% can be achieved - The linear motor has a long way to go. Just by doubling the efficiency, cost can be reduced by half.

I found out that for this research programme, I was not allowed to work individually, and had to have at least 1 more member in my group. Having not much options, I picked a fellow class mate, Ong Woo Han, to work on the project with me. Although I would really like to work on this project on my own, I believe that two heads are better than one, and we together can hopefully do a better job than what I could have done on my own.

After discussing with the teachers, we were placed in the 'Science Fair' category, whereby our project, if good enough, would be submitted for the GEP (gifted-education-programme) Science Fair 2004, organised by GE branch, Ministry of Education.

Title of project - A study on improving efficiencies of pulsed Electro-Magnetic Linear Mass Accelerators

Project members - Group Leader: Gao Guangyan ; Group member(s): Ong Woo Han ; Class: 3L ; Group No.: 99

Project's Area of focus and Plan:

This project will be a research to experiment and investigate different methods to find out how efficiency of the accelerator can be improved. (i.e. converting more stored energy into projectile kinetic energy). For the research, a relatively high powered electromagnetic accelerator would be constructed. It will use electrolytic capacitors to hold charge and switching will be done completely solid-state with a Silicon Controlled Rectifier (SCR) (also called thyristor). To reduce resistive losses, all connections would be done with with heavy connectors and wires will be designed to allow maximum current flow. Once working model is completed, efficiency improvements would be conducted. Projectile speed will probably be measured by a light-gate chronograph for accurate results.

Day by Day Construction

 31st Jan 2004 

It was the GEP Science Fair briefing today, where important information concerning the project was given out. It was held at the Science Faculty of NUS (National University of Singapore). This marks the beginning of my project!

 2nd Feb 2004 

Capacitors are certainly an excellent choice to power coil type electro-magnetic accelerators, like the one in my project. Me and Woo Han decided on a roughly 1kJ power for the accelerator. However, 1kJ of capacitors are not cheap. Due to inadequate funding, we could not buy new capacitors, so we went to to search for good deals if there might be any. I was very fortunate and found an excellent deal of 4 capacitors, 450 Volt 1800uF each, with a maximum surge rating of 525V. This amounts to 729J total at nominal charge and 992.25J at peak charge.

I have received good comments about the quality of type CGS electrolytic capacitors by Aero M, and for USD$10.49 for all four, it is a really good deal. I have already purchased the capacitors and the seller will contact me to ship the capacitors over. Once the capacitors arrive in a few days times, the accelerator model will be constructed immediately so research can start as soon as possible. The photo is from the ebay online store.

 Wed, 4th Feb 2004 

Today we met our Teacher-Mentor, Mrs Neo Li Kheang. Half of the time was spent trying to convince the teacher that this project would be safe. The other half was spent going through some administration stuff. We are still waiting for the capacitors to arrive so construction cannot begin. It appears the school cannot provide much materials for the project, but hopefully I would be able to obtain some Acrylic or Lexan to build the frame out of. The project construction might be too dangerous to be carried out in school so the teacher will help me get permission to allow all construction and testing of the product to be carried out in my backyard.

 Wed, 11th Feb 2004 

Mr. Ang from the Design and Technology block has kindly provided lab space for us to construct our model. Sadly, our capacitors has not arrived yet, but hopefully, they will arrive next week. Today I drew up a simple design plan of the proposed model design, but I need to improve it and make it more detailed, as well as design the circuit for the project. The school has also kindly provided lots of 6'x3'x3mm clear acrylic sheets for us to construct our model with. Although thin, lamination of several layers is possible. Next Monday, we will have to meet Mr. Ang to discuss about the workshop space. Later on in the afternoon, we headed for Jalan Besar near Bugis MRT station to buy components for the project.

Here's the items we bought:


From Kaichin industrial at Sim Lim Tower, we managed to find a large used stud type Thyristor (Also called a SCR - Silicon Controlled Rectifier). It is a SKT160/12C SCR manufactured by Semikron. We managed to get it for S$35 which was quite a bargain. This SCR is a huge piece of power semiconductor used for high power switching! The picture does not do justice to its size - The screw at the bottom of the SCR is M16 size (1.5cm dia) and 20mm long! The nominal (RMS) power rating for this SCR is 336kW continuous (1200V 280A). We are however only interested in the Peak power rating. For this particular SCR it is in the order of 5.16MW (4300A for 10ms at 25oC). Keeping the surge current below this critical value is essential. Also in the picture are two steel washers (3.8cm dia x 3mm at 20 cents each), a M16 nut (2.4cm wide) and a beefy copper cable lug (90 cents) for connecting the SCR.
All connections are to be extra heavy duty, to minimize IR2 losses during the current pulse. 4 feet of multi-stranded 400A copper welding cable (S$0.90 per foot) and 6 heavy duty copper cable lugs were purchased for the primary wiring. The cable is almost 14mm thick (including insulation) and should provide a relatively low resistance. Large bolts and nuts (total of 90 cents) were also purchased for connecting the cable lugs together tightly; good contact is essential. The number of connections will be minimized to prevent further resistive losses.

Finally, we obtained a 1 foot long brass pipe (1.1mm dia with 1mm thick walls), 1 foot long iron rod (8mm dia - fits perfectly in the brass pipe) (to be cut and used a projectiles) and a 1 foot long copper buss bar (3mm x 30mm) for primary capacitor connections. The copper buss bar was chosen for it's low resistance to keep IR2 losses as low as possible. The whole bar would be cut and drilled to connect the capacitor terminals (screw type). Large heat shrink tubing was also bought for insulation of exposed connections.

Almost S$60 was spent on parts today.

Part Bought at: Cost (SGD)
Used Semikron Thyristor/SCR, part number SKT/12C Kaichin Industrial (Sim Lim Tower) S$35.00
6x heavy duty copper wire lugs (90c each) Ban Hoe Lee Hardware (Rocher Centre) S$5.40
4 Feet of 400A Welding Cable (90c per foot) S$3.60
1 metre of heat-shrink sleeving Sun Light Electronics (Sim Lim Tower) S$3.15
1' x 1.1mm dia, 1mm (walls) brass pipe Teck Leong Metals Pte Ltd
(Kelantan Road)
1' x 8mm dia, steel rod S$2.00
276mm x 3mm x 30mm copper bar S$2.00
M16 nut (60c) and 2x M16 washers (20c each) Screw King Shop (Kelantan Road) S$1.00
3 bolts and 3 nuts (15c per piece) S$0.90
Total Cost - S$55.05


 Sunday, 15th Feb 2004 


Mrs Neo wanted me to have a more detailed plan of my model. Here is my proposed model design I rendered with Bryce 5. I guess Autocad would be better.. Click to enlarge.

 Wednesday, 18th Feb 2004 

Construction Begins!

Mr. Jerry Ang from the Design and Technology block has kindly granted workshop space. Here's some photos of the workshop.


The school has provided 2'x6'x3mm sheets of clear acrylic. For this purpose, 3mm is too thin. My plan is to cut 2 identical pieces for each section, and use chloroform for sticking them together to form 6mm sheets. Today, the different sides needed to make the 'capacitor box' was cut. The required sizes were drawn on the blackboard and cutting commenced. The plan required: 4 Pieces of 7.5"x6", 4 Pieces of 4"x16", and 4 Pieces of (7.5"+12mm)x(4"+12mm). Unfortunately, the capacitors have not arrived, but since I know the dimensions of the capacitors, the box can be constructed. Hopefully the capacitors will arrive soon. The capacitor box will house the 4 capacitors, as well as the SCR, and will have a hinge top which can be opened, and can be secured.

Cutting was mostly done on the scroll saws in the workshop: obviously not the ideal tool for cutting large straight edges (a circular saw would be much more suitable), as the blade is thin, which would be good for cutting curved edges. After 2 hours of cutting, all 12 pieces were finally completed. However, most cuts resulted in less-than-ideal straight edges due to machine limitations and human non-precision, but they will hopefully be fixed up and straightened, after connecting the pieces together with chloroform, perhaps by filing, re-cutting with a better tool, or cleaned up with a rotary tool. Chloroform was kindly provided by the school chemistry labs. On the left is a photo of me cutting the acrylic sheets...

Using eyedroppers, chloroform was applied to the edges of two pieces of acrylic. Chloroform bonds acrylic together chemically, which makes a very strong bond. Due to the uneven cutting, the edges did not meet exactly, but all gluing proceeded rather smoothly. Applying the chloroform was relatively easy. The joints proved too be quite strong. Finally, 6 laminated pieces of acrylic was left. The edges and sides Have to be trimmed. Luckily, large tolerances were introduced during the cutting, so trimming it should be no problem. The two photos on the left show the before and after states. The photo on the right is how I glued the acrylic sheets together. By the way, the brown colour of the acrylic is the protective paper covering. Because acrylic is prone to scratches and fingerprints, the paper is there to protect the surface. Only when the model is completed will the paper be peeled off.

Next week, more construction of the acrylic frame will continue. The circuit will also be drawn. Meanwhile, I am still waiting for the capacitors to arrive.


 Wednesday, 25th Feb 2004 

Last week, the acrylic sides for the capacitor bank were cut and glued together. However, due to the machine, which was not suitable for cutting straight sides, the acrylic sheets did not fit each other equally and the sides were not very straight. Using a black and decker RTX rotary tool with a cutting wheel, I trimmed the sides which were jutting out. A bench grinder was also used to straighten the edges, and surprisingly worked very well, albeit producing a lot of acrylic dust making the place very messy. Since the sheets have to be the same size, they were taped together as one thick sheet and the edges grinded smooth. The process was simple and easy. Hinges will be bought for attaching the cover of the acrylic box to the box itself. Everything else will be stuck together with chloroform. Around 1 hour was spent straightening the edges.

A trip was also made to Sim Lim Tower later in the afternoon, to buy more parts, mostly electronic components, for the charging and triggering circuit. Here is a photo of the parts and a short description:

1. 15 metres of AWG 14 (1.63mm dia) solid enamel coated copper wire for use as the coil.
2. 2 metres each, green and red insulated wire for basic wiring of circuit.
3. A 240V/315V 36W transformer [not in picture] was ordered for charging capacitors
4. 2 switches, lights up when turned on, for use as main switch button.
5. 2 red push button switches as trigger button.
6. Toggle switch for use as charging switch.
7. 3.3Ohm 10W resistors x2 and 2.7kOhm 10W resistors x2 to limit trigger pulse current.
8. 9x 10kOhm 5W resistors - for limiting current for charging capacitors.
9. 13 female spade connectors for connections.
10. 4A 600V Bridge rectifier for rectifying 315VAC to ~450VDC peak.
11. 3 AA battery holder as triggering power supply.

Part Bought at: Cost (SGD)
4 Metres of insulated wire, 70c per metre Silicon Electronics Company
(Sim Lim Tower)
2 Pieces R5 Switches S$5.60
2 MS327 push button switches S$5.00
4 10W resistors S$3.20
9 5W resistors S$3.60
13 spade connectors S$2.60
3 pin toggle switch S$3.50
Goods, Service, Transport tax S$1.32
Ordered a 240V/315V 36VA transformer Elegold Electronics Enterprise
(Sim Lim Tower)
15m of AWG14 enamel coated copper wire S$10.00
Total Cost - S$62.62

  Friday, 27th Feb 2004 


This afternoon, I personally made a trip to a hobby store and bought a total of 4 nice stainless steel hinges, 2 of them will be used on the capacitor box as the lid hinge. I have yet to find a suitable locking mechanism. Because the hinges came with wood screws, I had to buy some small nuts and bolts to attach the hinges. (The capacitor box is acrylic, and wood screws won't work on acrylic..) The other two hinges will probably be used for another box, probably the charging unit box. The capacitors still have not arrived! I am beginning to wonder if there are any problems cropping up...


Since the workshop was not available today, I constructed the charging unit minus the transformer today at my own home. (left) A jig-saw was used to cut the acrylic. The base is made out of 3mm clear acrylic. A total of 6 5W 10kOhm resistors were used, and wired in such a way to add up to a 15kOhm, 30W resistor.
This is to limit the current flow to the capacitors. The black chip is a 600V 4A bridge rectifier. The two blue wires are AC inputs. The red wire is + out and the green one is -. The blue wires will be connected to the 315VAC transformer, and the full wave rectified DC will be connected to the capacitors of course.

Bought a few more parts today: 2 Packages of 2 Stainless Steel Hinges, and some nuts and bolts to hold them together to the case.


 Saturday, 28th Feb 2004 

It's one day before the 29th of February, and today is the first time we are meeting our NUS mentor, Associate Professor, Dr. Anjam Kursheed from the department of Electrical and Computer Engineering.

We discussed about the design, and our plans for this experimental project. One important thing discussed was about safety, whereby he strongly suggested charging the capacitors at a low voltage, like 50V instead of doing it at 450V, whereby the capacitors would store a lethal charge. Overall, he felt the project was good and he is excited to know about our results.

 Wednesday, 3rd Mar 2004 

More work on the acrylic box and frame. Two main things were done today. The acrylic box for the capacitors and the barrel supports.


Stainless steel hinges were carefully aligned and screwed into the acrylic sheets using 3/16mm bolts and nuts. This will be the cover for the capacitor box.
The barrel supports were also done, fashioned from 3mm acrylic and glued together with superglue... it seems strong and good for now. The supports will hold the barrel with the coil. It is important for the supports to be strong and stable. Hopefully, they will last up to it or a new one will be made.

 Friday, 5th Mar 2004 

I made a trip to collect the pre-ordered transformer. 240VAC input, 315VAC output, 36VA. It seems well constructed and has a nice coating.


This transformer will be rectified full wave (but not regulated) outputting almost 450VDC through a 15kOhm resistor network to limit the current. In fact, a 15VA transformer would do but a higher rating transformer is definitely more robust and will last longer. It weighs in at 1.16kg. It's not too big, at only 7cm tall, which is good. Construction of the charging network can begin soon. I also bought 6 steel washers which can fit nicely around the barrel, so winding of the coil can begin soon. Washers cost a total of $1.80.

 Wednesday, 10th March 2004 


After acquiring the transformer, and more acrylic, the charging box was made. By the way, all boxes are constructed in order to make sure all exposed connections and potentially dangerous electrical components are well insulated. This time, the box was made out of one single sheet of acrylic, with a hinge at the end, measuring 14"x3.5"x3.5".

A heated strip bender was used to bend it into the box. Stands and supports for the transformer and resistor network was also made. Everything was joined together with chloroform. Later in the afternoon, more parts were acquired for further construction. Fiberglass reinforced cut-off wheels for my RTX rotary tool were bought for the heavy metal cutting which will begin soon. More bolts were needed for hinge connections, and wire connectors were also bought. The capacitors have STILL not arrived! I believe instead of sending by air-mail, the post office made a mistake and sent by ship instead... Finally, some bolts were trimmed from the capacitor box hinges using my RTX.

 Saturday, 20th Mar 2004 

It is currently the school March holidays, however, due to a large number of activities, there is less time for construction. However, things should speed up. The capacitors *just* arrived yesterday! I am still puzzled why they took so long. Anyway, they appear to be in good condition, exactly as advertised on ebay, and come with a nice capacitor cap each (the red one). They are 6" x 3" dia, and weigh about 730g each (excluding the red cap). They store 1800uF at 450VDC (525VDC surge) which amounts to 729J total at nominal charge and 992.25J at peak charge total. They fit the capacitor box nicely. Now the real stuff can be done! The capacitors will be wired in parallel, which would yield 450V at 7200uF total, and be charged with a full wave rectified 315VAC at 15W.


Since the school was closed, I did some machine work at home today. The hinges and mount for the charging unit box were aligned, drilled, and screwed on. The last side will be heat bent on another day. I also fixed the transformer down to it's own acrylic stand. Both the charging box and the transformer will be chemically welded to the base by chloroform as usual. I tried to make the primary capacitor connections out of the copper bar, but due to a calculation error, the copper bar was a bit too short. A new one will be purchased soon so work can quickly continue. Hopefully, most part of the coilgun will be completed by the end of next week.

 Monday, 22nd Mar 2004 

Immediately after the March holidays is the school's i-learning week. Today, I bought more wire and a new piece of copper.


Later that night, I spun up my drill (at home of course) and did the primary capacitor connections, cut out from the 20mm x 2mm x 3 foot long copper bar. Everything fits nicely. Tomorrow I will hopefully get together with my partner and due more machine work at the workshop, probably constructing the control panel and fixing the charging unit together.

Part Bought at: Cost
5 Fiberglass Reinforced Cut-Off wheels Home-Fix (Harbour Front) S$14.30
More nuts and bolts (3/16" x 3/4") S$1.00
PVC wire connectors S$1.50
4 Capacitors (1800uF 450V ea.) + Shipping + Insurance Ebay ( USD$52.79 = SGD$89.2151
4m of wire. 14c GST. Silicon Electronics Company S$2.94
20mm x 2mm x 3' Copper Bar Teck Leong Metals Pte Ltd S$5.00
Total Cost (using 1.69 conversion rate) - SGD$113.96

 Tuesday, 23rd Mar 2004 


Although the school workshop was closed today, Woo Hancame over to my place and we did more work. Today was a rather constructive day. The control panel was cut from acrylic and shaped like a equilateral triangle laterally. The switches were fitted in an everything looks good. The brass barrel was cleaned and sanded with 500grit sandpaper and it shines like gold. Most plastics are too soft to maintain a good structural rigidity when a coil wound on it is pulsed with a few hundred joules. Therefore, a metal tube was chosen, which has a high strength to weight/thickness ratio. Brass has the second lowest coefficient of friction of metals (second only to bronze) and was chosen. Since bronze (or like other metal tubes) are conductive, they suffer severe eddy current losses in coil gun use.

My solution was to slot the barrel (using an RTX fiberglass cutting wheel) to eliminate the current losses. The barrel inner diameter is 5/16" (7.9mm) which is a common size for metal rods. I have also previously obtained a 5/16" dia steel rod which was cut into pieces of 1", 1.25", and 1.5" lengths to use a projectiles. The coil was wound directly on the barrel using AWG12 (2.05mm) magnet (enamel coated) wire. 5 layers were used, and a thin wrapping of insulating tape was wrapped on after every layer. The first layer have 36 turns, followed by 34, 34, 31, 28 turns with a total of 163 turns (+ - a few...). The coil is 2.5" long and is around 3.2cm in diameter. Everything was hand wound carefully to ensure all turns are tight and accurate. Thick heat shrink might be bought to shrink wrap the whole coil to keep it neat and tidy. For now, the insulating tape should hold... Finally, the charging transformer was hooked up the the charging resistor network.

The culminating point was when the coil/barrel setup was wired to the cap bank and charged with the charging network from the mains. Charge voltage was capped at 150V resulting in 81J in the bank, (11% of total stored charge possible at 450V) and fired via a mechanical contact switch. The results are impressive. At the point of contact , the coil form jerks backwards with the recoil of the firing as it tries to 'grab' the projectile, and the contacts weld together with a spectacular flash and bang. There is a lot of energy wasted in a mechanical contact switch, up to 30-40% (could be higher).

The final design utilizing solid state switching will be much more efficient (1-3% loss only!) and the firing will be totally quiet. In the first test, my teammate, using a camera (which can take video) took a video of the shot, but he fumbled and did not capture the firing moment... later on, I charged it again (video taken by myself this time) and fired using the 1" projectile at a can. The can flies up. The can is dented and the projectile has managed to break through the side of the can, almost, but not yet, penetrating the side. I am very pleased with the results so far. Estimated speed (judging by the distance it dropped traveling to the target) is around 10m/s.

Click to download a video of 2 shots! First at a file, and the second at a can. schcoil04_150vshot.wmv (494kb)

 Wednesday, 31st Mar 2004 

School has reopened and it is the start of a new term. I got together with Woo Han and did more machine work today.


The capacitor box is joined with chloroform. The two holes at the front are for the primary cables to go through to the coil. All sides are 6mm thick. I also glued on a handle to open the box easily. Everything is looking good.


The switches were fitted in the control box and bent to shape. The yellow switch is the main power switch, and lights up when turned on. The pole switch is the charging switch for the capacitors, and the big red button is the fire button of course. For triggering the SCR, 3 AA batteries will be used and are connected to the control box. In the second picture, you can see the complete setup. Basically, almost every part is done. A 20"x 20" x 10mm thick acrylic base plate will be bought to mount all the components. The large rectangular box is where the wires and other electronics will be held in.  

 Wednesday, 7th Apr 2004 

I sanded the capacitor copper connections the day before, and screwed on the SCR and the cables (orange ones). The caps were fitted into the box and they fit perfectly! The cables were a bit too long, so the barrel will be fixed directly in front of the capacitor box.


The primary wires are a bit too thick and difficult to bend, but no other thick wire was available at that time... in case you where wondering, the box is totally clear. The brown paper covering has been left on until it needs to be mounted and taken off to prevents scratches and fingerprints from getting all over. More parts were bought too. A 22" x 22" x 10mm clear acrylic board (weighing almost 4kg!) was purchased for the base of the whole model. A mains plug + socket was bought too. (see picture). The pen in the picture is for comparison.

 Sunday, 11th Apr 2004 

Today I bought 5 clear rubber door bumpers. They are about 35mm dia. These bumpers will be used as supporting legs for the model base - 4 at the corners, 1 in the middle. They came with an adhesive. However, these adhesives cannot be trusted (comes off very quickly) and I removed them simply by rubbing it off. I will use epoxy glue to stick it to the acrylic board. The last picture shows the completed board. The hold is strong enough that the whole board (4kg) can be lifted just by holding a single leg.

Part Bought at: Cost (SGD)
Power Cord Kaichin Computer Systems Pte Ltd S$2.50
AC socket S$0.80
10mm x 22" x 22" clear acrylic sheet Yeo Soon Hin Plastic Signcrafts Co. S$37.00
Clear door bumpers (Extra Large) x 5 Home-Fix (HarbourFront) S$13.75
Total Cost - S$54.05


 Wednesday, 14th Apr 2004 

Today was a good day!


It's Wednesday again and as usual, more work was done at the workshop with Woo Han. I brought the 22" x 22" x 10mm acrylic board and filed the sharp edges down to a smooth curve.

I also constructed the final piece of acrylic. Picture 1 shows the whole setup arranged nicely on the large acrylic board. The rectangular piece of acrylic at the top (partially behind the capacitor box, balancing on the edge) is the board for the wire box, which will house all the wiring to house the socket (Picture 2). The socket is useful and since the power cord can be detached, it makes everything much neater. The wires were soldered up and fitted in the wire box.After doing the messy wiring, the coilgun was tested by charging the caps for a few seconds, and using the SCR to trigger. Should anything not work, I can check the wiring and change it, before I close up the wire box. Anyway, it worked perfectly on the first try!

Finally everything (very heavy!) was transported to the chemistry labs and using chloroform, all the different components were stuck together. Everything is looking good. Click picture 4 for a large labeled photo of the setup. After leaving the chloroform to set for about 20 minutes, the gun was tested. [Actually, my teacher mentor did not want the gun to be fired until our Mentor at NUS checks it (safety reasons) but we did it in secret anyway... well it did work perfectly... :) ] The gun fired good and we are happy! :-)

Some tests were carried out at 100V to 150V in the chemistry labs using a variety of backstops (a 1 inch projectile was used). Do note that 100V charge is only 5% of the total amount the caps can store (at 450V) and 150V is 10% of the maximum energy. Finally, I brought the whole model home. I will contact my NUS Mentor shortly and see if I can bring it to his lab. (Teacher wants us to do that). Finally, check out the videos below! What I absolutely love about this gun is that it is totally quiet when firing, thanks to power solid state switching!

This video was taken at the chemistry labs.

The caps are charged to 150V and fired at a plastic container. The plastic container can be heard bouncing around the floor after it was shot down. If you look at the multimeter, you can see there is some voltage reversal.

Click here or the image to download the video. (WMV format, required Windows Media Player) 494kb.

This video was taken later that day at my house.

The caps are charged to 199V and fired at a can. The projectile almost broke through the first layer of the can. There was a hole and a large dent, but not big enough. I believe with more fine tuning of the position of the projectile, the coilgun will be able to break through the first layer of the can. Note that 200V is only 20% of the total energy that can be stored. Look carefully and you can see the wires jump when the gun is fired and a huge current surges through the wires.

Click here or the image to download the video. (WMV format, required Windows Media Player) 455kb

 Saturday, 17th Apr 2004 


Me and Woo Han met our mentor, Assoc. Prof. Anjam Kursheed today at 10am at NUS. We proceeded to the Centre for Power Electronics - Power Electronics Laboratory to do some powered testing and for our mentor to check through the design.

He was indeed very pleased with the work done so far and after a check, the gun was fired at voltages up to 200V. Everything is working well and the gun was performing well. Other interesting things in the lab was a large motorized Wimshurst Generator capable of outputting a huge fat arc almost 30cm long. Also, there were 2 tesla coils, one commercially bought and a tiny home-made one (by a University Student at NUS). It is interesting to note that my tesla coil actually works Better than the commercial one, and my mentor is highly interested in my tesla coil and wants me to bring it some time. Click here for my tesla coil 1 page.

 Wednesday, 21st Apr 2004 


Mrs Neo, our teacher mentor was ill today, so we were not allowed to continue our experimentation. I had planned to do some projectile speed measurements, as well as to find the optimum position for the projectile at 100 and 200V. I don't understand why teachers don't allow us to carry out the project ourselves. Not like being there would make it any safer... Anyway, we tested the gun a few times at ~275V (almost 40%) with a 1.5" Projectile, with excellent results. The gun finally fires Right Through a can! (last photo). It is a good day.

 Friday, 30th Apr 2004 

Mr Tan, one of our school Laboratory Technician, has kindly agreed to let us use some equipment, to measure the velocity of the projectile. Woo Han has something on and couldn't make it. Mr Tan also granted us space to work at the Photonics X - Labs in our school. Sad to say, there aren't many spectacular lasers around.. just some 15mW HeNe Melles Griot tubes (sadly fitted with a 5mW power supply) and a 10mW DPSS 532nm.

We tried using light detectors at first, but failed to produce good results. We then tried using a photogate (which was overlooked) and it worked quite fine. The coilgun was charged to 100V and fired several times (at a large cardboard box) yielding a measured efficiency of around 3%.

We can now measure projectile velocity, and it is time to do some serious experimenting!

- Proceed to the next section! -

Experiments and Testing

 Wednesday, 5th May 2004 

It was supposed to be the National Schools Softball Finals today, and my class was allocated to support the softball match. Thus, we would not be able carry out our project... but it rained, and the match was postponed. Today will probably be the last project research day before the mid-year common examinations next week. Project work will continue 2 weeks from now.

Using the same setup as before, the velocity of the projectiles was plotted against the projectile placement from the end of the barrel, and the optimum projectile placing speed was obtained. Three measurements were taken for each placement, and the average was taken. Some minor changes were also made to the coilgun. The charging resistors was modified to be changeable - 10kOhms for quick and fast charge, and 15kOhm for a slower charge, but can be used for longer periods of time. I might change the resistor network to one big resistor, but with lower resistance to speed up charging time.

The photogate used to measure the speed of the projectile does not seem to be very accurate. More testing will reveal more. For now, it's two weeks before more work can be done.

 Wednesday, 19th May 2004 

The mid year examinations are over, testing continues!

Today, a smaller 3.1cm, 12.3g steel projectile was fired at 50 and 100V, with different starting positions. Three readings were taken for each position (per mm), and the average of each reading was plotted on a line graph. The graph suggest optimum placement at 5cm for 50V, and 4.7cm for 100V.

Everything went okay today.

 Wednesday, 26th May 2004 

Today, we took a break from doing all that testing (not really very exciting) and did more machine work, making some nice projectiles, doing some modifications, and making an add-on metal wrap - an idea to try to maximize efficiency.

More projectiles were cut from the 0.79mm dia steel rod, and one nice bullet shaped one was machined. A slightly longer than 1 inch piece was cut, attached to a drill, and grinded on a bench grinder to a sharp tapered point. (Picture 1 and 2). This projectile is however not idea but will be used just for fun, and to improve penetration power.

One of the biggest impediments in making a coilgun is the horrible inefficiency, which ranges from 1 -3%. (Whereby stored electrical energy is converted to projectile kinetic energy). Compared to modern rotary motor with efficiencies up to 80%, this is horrible. And idea dawned upon me that perhaps by wrapping the coil of the coilgun with some iron, the iron can lend its magnetism to the magnetic field. It was decided that the iron would be separated from each other, to avoid eddy currents.

I happened to find a small 230-12V transformer lying around my junk box. Since it was sitting there collecting dust, I decided to put it to good use. (Picture 3). Since a transformer was made out of laminated iron, I thought it might be suitable for my application. Firstly, the front of the transformer windings were sawed of. (Picture 4). Using a hammer and a chisel, I managed to separate the first few laminations, and the rest came off easily after that. Was was left was a bunch of E shaped and I shaped iron laminations (Picture 5). For now, only the I laminations were used. They were stuck on a strip of duct tape. This way, the iron strip can be wrapped around the coil. More testing will continue tomorrow and hopefully, we will see some improvements in efficiency!

Some other modifications were done to the coilgun too. The Barrel holder was modified so the barrel can be easily taken out of the test platform, and a stopper was machined to prevent the coil from moving from the recoil.

 Thursday, 27th May 2004 


I had to attend a talk on maths and thus there wasn't much time to do testing. Testing took place from 4:30 to 5:30pm. Today, the metal laminations were tested out, and the recoil stopper was also added to prevent the coil from jerking backwards during every shot (wasted energy). The results were rather positive. Testing was done at 100V using a 3.1cm 12.3g projectile at position 4.7mm into the coil. Firing without the metal laminations yielded speeds around 13.7m/s (3.2% efficiency). Firing WITH the metal laminations yielded speed around 18m/s, which is around 5.5% efficiency! There was enough time to do more testing but the results are looking good. This looks like a good solution. On the left is a picture of the sharp projectile after firing at a red whiteboard marker at only 100V (with metal wrap). The projectile easily pierced through the first side and its speed damped by the cotton-like stuff inside (which holds the ink). The tip of the projectile is stained red.

 Saturday, 26th June 2004 

It has been a very busy month. For the whole month I've been attached with A-star IBN (Institute of Bioengineering and Nanotechnology), with a research scientist as my mentor. As this is a full month attachment, there was absolutely no time to work on my project.

To make things worse, my other group member was on a student exchange program at Australia, and thus could not work on the project too. We had no choice but to halt the project for a moment...

Today we did more tests using the iron wrapping and found that previous tests (which showed 5.5% efficiency) were flawed. However, we managed to improve efficiency over the original test by 18.7%. Testing was done at my house today, as school is closed (currently still school holidays). Maximum efficiency at 100V was 3.72% with 1.34J of kinetic energy.

 Early July 2004 

Other experiments tried were slotting the projectile, but that did not yield any improvement. The project report was due soon however, and the report was written and submitted. The final report was submitted before 7th July. This marks the ends of the project work.

During the course, there were many difficulties and limitations encountered. Firstly, we were using a simple photo-gate and data logger to obtain velocity measurements. These results may not be reliable as there are sometime fluctuations in velocity measurements. Furthermore, the accuracy of the photo-gate is not very good. We were unable to obtain better equipment for taking velocity measurements. Apparently, the photo gate is not very accurate, but shows a reading to several decimal places. We therefore decided to take all readings rounded to the nearest single decimal place.

We estimate the error associated with the photo-gate and data logger may be as much as +-0.5m s-1. As such, this affects the results of our experiments. Secondly, designing and building the model from start was rather complex and took a long time of work in order to make a high-efficiency working model. Due to time constraints, we were unable to do more experimentation that we would like to have carried out. There are still many things that can be done and we would like to do so should there be a chance to do an extension of this project.

We would like to express our gratitude to our mentor, Dr Anjam Kursheed, from the Electrical Engineering Department; National University of Singapore, for sacrificing his precious of time from his tight schedule, and providing expertise, invaluable guidance, as well as fine advice. We would also like to give our thanks to our teacher-advisor, Mrs. Neo Li-Kheang, who had sacrificed her free time to supervise and guide us along during the course of our project.

Last but not least, we would also like to thank our school laboratory technician, Mr Tan, whom we trusted our model with, and whom provided us lab space to carry out our experiments and provided us equipment for velocity measurements. We would alike like to thank the laboratory technicians had been very helpful and kind to us, and everyone else who has helped us in one way or another.

Although this project is more or less completed, any other news concerning this project will be updated here.

News, Events and Updates


1. Gifted Education Open House - Project on Display in the School Physics Laboratories
2. Preparation for the 10th Youth Science Conference 2004, Singapore
3. 10th Youth Science Conference 2004, Singapore -Coilgun wins Best Project Presentation & Silver Award Overall!
4. Coilgun is on display at the Raffles Institution Research Congress
5. Raffles Institution Research Education Presentation - Coilgun receives R.I. Research Education Silver Award!
6. Singapore Science and Engineering Fair 2005 - SSEF - Coilgun wins Gold, Heads to Phoenix AZ, USA, for ISEF 05!

 Saturday, 31st July 2006 


G.E. Open House

Today was the Gifted Education Open house of my school, and I was requested to have this project to be on display. I spent the whole afternoon presenting several groups of primary 6 students my project and explaining how it works, as well as a live demonstration at 275V of a sharpened projectile firing a can. (last photo, two exit holes from two separate shots) The coilgun performed remarkably well and at a good shot at 315V, it fired right through 2 cans, and still have enough kinetic energy to go through the cardboard box (to catch the cans) and flew to the other end of the laboratory! I also presented my project to several curious adults. Needless to say, all were very impressed. :-)

 Thur,Fri, 26th, 27th Aug 2004 

Preparation for the 10th Youth Science Conference


The 10th Youth Science Conference 2004, Singapore is coming soon. My project will be on display. There are 2 main groups - the poster group (where a poster will have to be presented), and a presentation group (whereby a presentation had to be presented to the judges). Me and my partner have been working hard to complete the presentation.

I did some cleaning up of the model, and made a clear acrylic holder to insert a little note concerning this model. (Later changed to include schematic as well). More powered testing at up to 375V was carried out. At 350V with the sharpened projectile at 4.2cm distance, the projectile easily punctures through 3 cans! I tried shooting a Sigg aluminium bottle (much thicker) at the base. It made a big impact, but didn't manage to go through. 350V is around 60% charge. The coil gun has performed remarkably. During the presentation, we would probably do a demonstration. I also constructed a backstop to stop the projectile. Finally, I printed some high voltage, Danger signs to stick on the model. Hopefully everything goes well!

 Saturday, 28th Aug 2004 

"This year, we celebrate the 10th year of the Youth Science Conference. At this Conference, we showcase our pupil's products in the various science mentorship programmes. These products are the culmination of months of hard work. Altogether, 134 participants from 14 secondary schools have each spent 6 months on their research projects, yielding a total of 52 projects." - Forward from the Proceedings of the Tenth Youth Science Conference


Today was the 10th Youth Science Conference 2004, Singapore, and it was held at Singapore Polytechnic. There are 2 main parts of this conference - the Posters, and the presentations. The posters were judged yesterday, and were on display. The presentations where held in the late morning. A total of 18 groups presented, at 3 different locations (concurrent sessions, 6 per location). For each concurrent session, 2 prizes were given out. I was presenting at LT17B, with A/P Lim (NUS) and Asst. Prof Ang (NTU/NIE) as judges.

There were many good presentations, and all did a great job with some really good research projects. My presentation was last, but was the only one with a demonstration. The coil gun performed well and penetrated 3 cans. A SVM Computer Method for Validating Traditional Chinese Medicinal Herbal Prescriptions by a team from The Chinese High School got second. We got the first prize, which was quite unexpected (well at least I wasn't expecting anything!). The photo above shows A/P Lim (judge), with me, and my other team member, Woo Han, with the coil gun in front.

I would like to thank all my great friends who came down to support, and those who have encouraged me. Thanks! :-)

Tuesday, 26th Oct 2004 

Nothing happened today. I'm just updating the page on the recent events. School has been very busy.

Research Congress: Our school has this project competition, and everyone is allowed to take part (since every one has a do a project of some sort for our school's Research Education Programme.) After several rounds of presentation and competition, we managed to get into the Research Congress Finals. This took place on 6th October (Wednesday), and was split into two groups, the VIP session (where the finalists had to present to other participants who had commendable projects, as well as to special guests from various institutions and research laboratories) and the non-VIP session where the students had to present to the rest of the school population. We got into the VIP session, and got a silver award. :-)

Me and my partner won $150 worth of Borders book voucher. The first photo shows the coil gun with the trophy, the prizes as well as the Best Oral Presentation certificate for the Youth Science Conference, the Certificate of Participation for the Science Mentorship Programme, and the Silver award certificate for our school's research congress. The second photo shows our poster at the atrium of our school. Our poster is the whitish one somewhere in the middle in between the brown (Bel Canto) poster and the blue poster.    

 Thursday, 13th Jan 2005 

Research Education Presentation

The beginning of another year!

Today was the Secondary 4 Express first Research Education lessons. The school wanted a high standard of project work so some of the winning projects were presented to them to sort of 'set a benchmark'. Me and Woo Han presented our coil gun. It performed very well. I added a 2.2Ohm 100W resistor to bypass the other ceramic resistors and charge time has decreased tremendously.

I can now charge to almost full power in just a few seconds. Other than a few hiccups during presentation (as we didn't really have much time to prepare fully), everything went okay. Performance was okay and a projectile penetrated 3 cans, but we didn't have time to properly determine the optimum placement of the projectile.

 Wed-Thur, 9-10th Mar 2005 


Singapore Science and Engineering Fair 2005

I quote from the SSEF website: "The Singapore Science and Engineering Fair (SSEF) is a national competition organised by the Ministry of Education (MOE), the Agency for Science, Technology and Research (A*STAR) and the Singapore Science Centre (SSC). The SSEF is affiliated to the prestigious Intel International Science and Engineering Fair (Intel ISEF), which is regarded as the Olympics of science competitions."

After discussion and deliberation, our school selected some teams to be sent in for the Singapore Science and Engineering Fair 2005. The report was submitted before 22nd Jan. Today was the competition. (Yesterday, all participants went down to the Singapore Science Centre Annex to setup the posters. )
The photo shows our project poster, as well as our coilgun model. The booth was actually rather small but just sufficient to allow us to place our model. Anyway, below is the judging criteria as I quote from the website.
Individual Projects
Team Projects
(max 3 members)
Creative Ability
Scientific Thought / Engineering Goals
Total Possible Score

Today was the judging round. Basically, Woo Han and I had to stay at our booth, and wait for 3 judges to come by and ask questions to find out more about our project.

It was actually quite tiring and boring waiting such a long time for the judges - around 8 hours of not doing anything. However it was fun looking at some of the other competitor's projects, though we couldn't really walk around much in case a judge came. The judges were overall quite helpful and did not ask too difficult questions. They were also pleased with the performance of our accelerator which I was happy that it performed so well and reliably even after many repeated firings.

I felt that we presented quite well and everything went very smoothly except with the second judge but it was not much of a problem. I don't think we will win anything because our project is not very substantial and there are far better projects than this. Anyway, the photo shows the model. We did fire it and we had a proper back stop to prevent the projectile from flying away (dangerous) but it's not present in this photo.

Tomorrow we have to come back to present to the public. There will also be some schools coming to view the projects.

 Thur, 9th Mar 2005 

Today was much more interesting. We were now able to walk around and see many of the very impressive projects on display. The photo on the left shows an overview of the science fair.

As the afternoon progressed, more and more people came and our booth was rather crowded (the model clearly attracted lots of attention). We fire the coil gun many many times. The photo of the destroyed can proves it. At some times I had to explain my project to a group of people while my partner was explaining to another group on the other side.

I was quite surprised when we were actually awarded a Gold award! I'm happy with this project though I think much more could be done. Thanks everyone who congratulated us. I would also like to express my appreciation to my great friends who have helped me and encouraged me in one way or another. This project would not be what it is today without them. :-)

 11th Mar 2005 

Wow! Woo Han and I got a call from the MOE (Ministry of Education) that we are selected to go to the ISEF! The Intel Science and Engineering Fair! The Science Centre says:

"The Intel ISEF is held annually in May and attracts over 1,200 students from 48 states in America and 40 nations to compete for scholarships of over US$2 million, tuition grants, scientific equipment and scientific fieldtrips. The grand prize of Intel ISEF is an invitation to attend the Nobel Prize Presentation Ceremonies in Stockholm, Sweden."

I'm not sure about the details yet so I'll post more when I find out more.

Coilgun Rebuild - Part II

 Monday, 4th Apr 2005 

It has been a very busy few weeks. It has been confirmed that we are indeed going to Phoenix Arizona, around the 8th of May 2005. As such, we have been working to finish and complete all sorts of forms and paper work to be sent over.

About a week or two ago, I did some more tests with larger laminations and current efficiency is now 4.35%, though that's only at 100V. Here are the details of the shot: Capacitor charged to 100V (36J); Projectile: 3.8cm long x 0.8cm dia. steel, 15g; Velocity: 14.3m/s; Projectile energy: 1.53J. Interestingly enough, efficiency drops to about 1.4% at higher power like 400V. I'm not sure why but I'll try to find out!

So today Woo Han and I went out to Jalan Besar to get more parts since we decided to rebuild the model frame (i.e. the acrylic stuff holding all the components together). This will not affect the performance of the accelerator in anyway at all. The model will now be slightly lighter (hopefully), easier to carry & transport, more compact, more sturdy and look better.

Here are the items that were bought today. (photo 1) We ordered a 500V DC panel meter but paid a $30 deposit first. Woo Han will pay for the items first, before trying to claim from the school, though I had to take out $10 because we did not have enough money (he took out 82.90). The panel meter should be ready in a few days. Anyway, Woo Han suddenly went home after we bought the parts so I had to make the main housing myself. You'll see how the new design will look like as it comes along. The first task was to get the hinged section done, before carefully position and gluing all the sections down.

Since all acrylic pieces were cut using a table saw, they are very straight and uniform, and should yield a much sturdier product than the previous model. I had to cut a few acrylic triangles to strengthen the box and fit everything together. After a few hours of hard work (about 3), the box was finally completed. It will house the capacitor bank, the charging circuit as well as the SCR. The coilform and the control panel will be mounted on the top of the box. The box is made of 8mm and 10mm thick acrylic so it should be quite strong. The top of the box supports my weight with no problems at all!

Photo 2 shows the outside of the box, with the capacitor (6" x 3" dia) for scale. Photo 3 shows the box when open and Photo 4 shows the inside of the box. It took quite a while to get everything together but I am happy with the result.
Part Cost (SGD)
1m AWG4 OFC Power Cable S$4.50
Chloroform - Acrylic Glue S$3.75
Small SS bolts x 12 set
Large SS bolts for cable lugs
Other bolts
Clear Acrylic:
10mm x 510mm x 210mm x 1
8mm x 130mm x 210mm x 2
8mm x 122mm x 494mm x 1
8mm x 130mm x 494mm x 1
8mm x 194mm x 494mm x 1
500V DC Panel Meter S$36.75
Larger Stainless Steel Hinge x 2 S$2.50
4x Cable Lugs S$2.00
Total Cost S$99.65

Until next time!

 Monday, 11th Apr 2005 


On Friday (8Apr), more work was done on the coil gun in the afternoon. I wired up the capacitor bank with new cables. Instead of using the ultra fat welding cables, I used high quality AWG4 oxygen free copper power audio cables instead, and this time I soldered the cable lugs to the cable (with a blowtorch of course). For the previous fat cables, the wires were merely crimped (and not very well) which results in a poor connection. Furthermore, the 400A welding cables were too thick and inflexible and was difficult to work with. In addition, the new audio cables look really nice too. I also refitted the capacitor bank, polished up the copper bus bars and cleaned up the coil and barrel.

Today I went to buy more parts myself because woo han had some family matters to attend to. I bought new switches (I cannot use the previous switches because they were soldered on), more components, 6m of ultra flexible wire (Red and black), a fuse holder (and fuse), some stainless steel bolts and collected the 500V DC panel volt meter. Since a $30 deposit was paid for the volt meter, I had to pay the remaining $6.75. Total cost of improvements now add up to $130.45.

Part Cost (SGD)
Clear Acrylic:
10mm x 12" x 3" x 2
500V DC Panel Meter S$0.00
(Paid $6.75 balance, but not counted here as it is included in the previous price count)
Stainless steel bolts S$1.00
Electronic Components (various) S$11.30 + S$12.50
Total Cost S$37.55 - S$6.75 = S$30.80

 Tuesday, 12th Apr 2005  

After school today, I sat down and did some drawings for the new control panel and the panel volt meter enclosure. I went to the design block workshop but the teachers were very unhelpful and refused to let me work there because they were not free and felt that working with the tools was too dangerous; I required a teacher to supervise while I worked. So I waited for a long time, because my teacher adviser Mr. Court was having a staff meeting. Once the meeting was over, work began immediately.

Using acrylic left over from last year (yes it was still at the top of the shelf where it was left there from last year... and the plans drawn on the chalk-board was still there!), woo han and I cut out the pieces which I had planned out earlier. woo han made a mistake and had to cut out another one. Everything was done with clear 3mm acrylic. I did the control panel and wh did the panel meter housing. The holes and everything were drilled out, and using a strip bender, the acrylic was bent to shape. The photo shows the completed parts. I think it looks pretty neat. Tomorrow I plan to complete the entire model and all the wiring. Lots of work to do.

 Friday, 15th Apr 2005  

After school on Wednesday (13 Apr), I got home and completed the model. Woo Han came to my house but not for a very long time so I ended up working on it myself. The first thing that was completed was the charging network.


This basically consists of a 36VA 240V to 315V transformer. 315VAC is full wave rectified using a 600V 34A bridge and this 445VDC is used to charge the capacitor bank through two sets of resistors, one for fast charging and one for slow charging. The completed charging setup can be seen in the photo. It's a lot smaller than the original one and much more compact. The 4A rectifier was replaced by a 34A one so it'll last longer.

After constructing the power supply, I made more holes in the main housing box to fit other parts on such as the mains cable connector and the battery pack holder. Once that was done, the capacitor bank was fitted in, the charging circuit was glued in place and the voltmeter was stuck on the top platform. I also machined some nice stands for the barrel and coil form using 10mm thick clear acrylic. I also had to grind the barrel to let it fit into the stands. Everything was then set into place and before the control panel was glued down, all the electrical wiring was done (which is a major task in itself!) Every single connection was insulated with heat-shrink. All these took fairly long and I worked till quite late. Once all the wiring was done, a final inspection was given before I tested the thing and I was very relieved that it worked great. The control panel was then glued down. The model was more or less completed minus the finishing touches. So after no fewer than 3 cuts on my fingers, a slightly burnt hand, an aching back, very sore fingers and 7+ hours of working alone, I completed it. I'm quite pleased with the results but of course they could be better if I had more time to work on it. The wiring is still rather messy. I ended up shaving 2kg off the weight and a -whole- lot of space!


I learned a few lessons. Firstly, cracked acrylic (or acrylic sheet edges) are as sharp as glass. (which was why I ended up with no fewer than 3 bleeding cuts on my fingers). Secondly, soldering irons are hot. Thirdly, getting chloroform into cuts hurts like mad. Fourthly, wiring takes an extremely long time to do properly.


Back to main page
(c) Gao Guangyan 2004
Contact: loneoceans [at] gmail [dot] com

Loneoceans Laboratories. Copyright (c) 2003 - 2012 Gao Guangyan, All Rights Reserved. Design 3.
Removal of any material from this site without permission is strictly prohibited and will result in infringement of copyright laws.
Disclaimer: Projects and experiments listed here are dangerous and should not be attempted.
... page generated in 9.0E-5 seconds