"Custom" Electric Engine

[Robotguy]

Even though I'm new, I thought it might be useful to post the build of my "custom" electric engine (and by "custom" I mean that I'm making it up as I go.) I thought it might be useful as a place to point beginners "Don't do it like this one!"

The idea for this engine is a flywheel and crankshaft with a "piston" made from a magnet. When the piston is in the correct position a burst of current will pulse through the coil and drive the flywheel.

As I posted in my introduction post, I don't have very much experience at all, but I have spent over 10 hours looking over all of the wonderful projects on the forum. As a result I decided to finally, after over 2 years, tram my mill. I had been avoiding it because I didn't have any idea how to do it using the dial indicator that I had. So today was spent building a "tool" to mount my dial indicator in the spindle. I used a 3/8" bolt with the head cut off and a bit of scrap aluminum square tube.
(Click the image for a bigger version)

Lo and behold it was off about .015" from side to side and about .010" from front to back. I fiddled with the bolts that hold the column and placed a .010" brass shim in the right side, and a .005" brass shim in the front. That seemed to level things out a bit.

Next I retired to the living room to draw out the flywheel (using a 12 year old version of AutoCAD) and produce some G-code (with ACE converter).

The white is the finished part, the red is the toolpath, and the green is the finish toolpath (should take off the last .005"). I have some .5" brass I picked up from the local surplus metal distributor, but I think I'll test the design on some .250" aluminum.

OK. Time to go try the program.

 

[itowbig]

ok ill bite where is this electric engine im not seeing it.

 

[Robotguy]

ok ill bite where is this electric engine im not seeing it.

I am just getting started. Should I wait until I get further along to start a WIP thread? I don't really know what the protocol is.

Basically it will be a solenoid engine, but the solenoid will be a little different than normal and built to look like a steam piston. I am hoping that if I mount everything with precision bearings it will be efficient enough that I can run it in my cubicle from a small solar cell.

I had some trouble with converting the drawing to G-code last night, but after giving up and heading to bed I realized I just need to fiddle with G02 and G03 codes. This morning was lost to cleaning house, but I'm just getting back to it now. Should be able to cut the first part this evening...

 

[joe d]

I don't really know what the protocol is.

The protocol really would be that you post a WIP when you have decided that you are ready to post a WIP!

This project sound interesting, I'll be following along happily as long as you remember th_rulze th_rulze th_rulze:

th_wwp :big:

Cheers, Joe

 

[Brian Rupnow]

Welcome Robotguy. If its a work in progress, and yes, the design part of any build is progress, by all means, keep posting. If you look at any of my threads, you will see that they mostly all begin with the design aspect of the build. It looks like an interesting project. Have fun, and welcome aboard.---Brian

 

[Robotguy]

I have been having issues with my G-code conversion utility, so I ended up doing 3+ hours of hand coding. Finally I got it looking the way I wanted in CNCSimulator, and then loaded up the program and started cutting an aluminum prototype. Here's the initial result:


I stopped the program before it cut out the final outside pass. I set it up for a cleanup pass on each spoke, but forgot to make it climb mill and only made it .005" which didn't even remove any material. I am going to run a new program and correct both of those errors. Unfortunately it's getting too late to run my noisy machines.

One of these days I'll figure out how to set speed and feed without knowing a) my spindle speed and b) the specifics of my material (it's mostly found scrap). The more I attempt to use the CNC, the more I am in awe of you guys who can do this for real. You make it look so easy!

Just to throw in an extra pic or two, here's some of the material I have set aside (click for a larger annotated version):


And a quick animation of what I have in mind (I tried to put the animated gif inside a spoiler tag so as not to annoy anyone, but it looks like I don't have the proper permissions to post spoilers):

I know I saw an example of the crank I want to use somewhere in this forum, but I'll be darned if I can find it again. It is a disk and a hoop separated by a thin section bearing. The main shaft runs through the middle disk in a hole that is drilled off center. The piston shaft enters a hole drilled in the edge of the outer hoop. Does that make any sense? I'll keep looking for the pic.

 

[DOC123]

Is the crank you want like the eccentric I made for my little engine.
see here
http://i566.photobucket.com/albums/ss109/TCB168/Mill Engine/Mill_engine_WIP0032.jpg

 

[Robotguy]

Is the crank you want like the eccentric I made for my little engine.
see here
http://i566.photobucket.com/albums/ss109/TCB168/Mill Engine/Mill_engine_WIP0032.jpg

That's the style all right. Thanks! Although I could have sworn I had seen one with a bearing between the eccentric and the outer ring. I am a major collector/packrat/scrounge and have a couple of thin section bearings (.75"ID,1.0"OD) from a cast off project at work that I plan to use.

I got up early and finished up the code. My wife just got up so it's time to head out to the garage and make some chips.

UPDATE:
The finish is a little better:


Right out of the mill:


And after a little cleanup on the lathe:

Still not the prettiest part I've ever seen, but definitely the coolest part I've ever made out of aluminum.

 

[Robotguy]

A bit more progress today. Everyone headed up to the snow today (Brrr!) and left me home alone. Poor me

Facing off the teflon for the piston (which will hold the magnet):


Turning down some brass to make a guide for the piston rod:


Here's the completed guide, which I then had to scrap because it was too short, and the bearing that I press fit got stuck:


And the piston and rod assembly, after I remade the brass guide:


It's nowhere near the level of the other projects I have seen on here, but I figured I'd start simple. For the moment I figured I'd do without the cylinder housing. The piston will be driven by the flywheel and eccentric and slide back and forth through the guide, rocking up and down slightly. The coil will be hard mounted to the base at a point near the furthest extension of the piston. During the most of the cycle, a solar panel will charge a capacitor using a "solar engine" circuit. Just after TDC, the system will discharge the capacitor through the coil to give the system a kick.

 

[1hand]

Nice Fly Wheel! Thm:

What size end mill you use to cut that with?


Matt

 

[Robotguy]

Nice Fly Wheel! Thm:

What size end mill you use to cut that with?

Thanks! I cut that with a 3/16" endmill. I probably should have put something in the picture to provide scale. The flywheel is 3.27" diameter.

 

[Robotguy]

I keep sneaking out to the garage and cutting parts. Gotta start the new year off right....

I didn't have any decent round stock, so I tried to cut a circle out on the mill. I thought the brass would go a lot better than the aluminum, but the cut ended up really nasty plus I didn't center it exactly so there's flat spots where the end mill came off the edge of the stock. It's going to take a lot of cleanup to make this useable:


Here's a start to the cleanup:

 

[Robotguy]

After trying to clean up the ring I cut from flat stock, I realized how tough it is to sand out machining marks. As you can see in my last post, I was trying to hold the ring in the outside jaws of my 3-jaw chuck and couldn't get the part centered when I had to flip it, so would end up with a small mark where the cuts from either side met. I figured I would just sand out the mark and be on my way. After 45 minutes of sanding, I had a nice shiny brass ring with no machining marks, but the surface was as wavy as water from the hand sanding. I needed a better way to hold the part.

I had a few collets that came with the lathe when I got it off Ebay, but none fit any stock I had. After playing a bit though I realized the outside of one of the collets fit nicely into the ring:


I chucked the little pin in the 3-jaw and filed a taper (by hand, power off):


Then I mounted the collet, slipped the ring over the outside and tapped the taper into the center of the collet. This actually held things quite nicely and let me re-cut the outside nice and smooth:


I left the ring on the collet while I punched a mark for drilling:


A little more cleanup on the back side of the flywheel:


Finally I put everything together for a little test run, co-opting my daughter as hand model:
http://www.youtube.com/v/xeI1mCbRLOI?fs=1&hl=en_US

 

[Robotguy]

I spent a little time with mounting everything up today so I could actually run some current through the coil and see if can make this thing run:


Short answer; Nope. There is way too much friction for my little coil to be able to drive this the way I had planned, so I am envisioning some mods including:
a) back to the "piston" style with linear magnet travel and a coil at each end of the piston; one pulling and one pushing. Then reverse the current for the other half of the cycle to force the magnet in the other direction.
b) better bearings. I really need to reduce the friction if I want to make this thing run from solar power. I am guessing it is going to need to spin unpowered for at least 10 seconds when I give it a start by hand. My coils will probably only be 4 or 5v at a couple of hundred milliAmps, so there's not a lot of power to spare.
c) I am debating whether to put reversing switches in a "steam chest" sort of arrangement or put an encoder and microcontroller on there to run things. I'll probably go with the latter since it will be a lot easier to tweak software. Besides, I'm an electrical engineer who designs embedded systems, so microcontrollers are the hammer that I tend to pound every nail with.

 

[hobby]

The machining of your pieces are very nice, in the video, it shows to be very well fitted and smooth movements.

Also congradulations on a project that is very unique, in its source of power, very clever idea, with a lot of trials and errors to iron out, but your trail blazing in unchartered territory, so it's bound to have many obsticles to overcome, but once you get it mastered, youll know the ins and outs of such a design, to make it work every time.

Keep on pushing forward, its going to be a great model when you get it working.

 

[hardhatdiver]

Somehow when I was doing a search on HMEM for a Perkins engine build I came across a couple of your threads. Looks like you haven't updated this thread in a few years. How did it ever turn out? I'm in West Natomas just off Garden Highway.

Regards,
Jeremy