The heart of a new machine

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Captain Jerry

Project of the Month Winner!!!
Project of the Month Winner
Joined
Dec 2, 2008
Messages
971
Reaction score
9
Happy New Year to all

New year and a new engine begins to take form.  This is a further development of the wobble plate engine that occurred to me as I was finishing up the Weeble that I posted last month.  I have spent a fair amount of time studying the various configurations of axial cylinder engines like the swash plate, the dynacam, the flexible rod, the Z-axis, the bent axis, the elbow and other types of wobble plate mechanisms. There have been many critical evaluations written that mainly focus on high friction loss that have doomed these engines to the category of curiosity.  So being the curious type, I jumped in to see for myself.

The first engine of this type that I built was the Chevron 10 that I posted here previously.  It is a perfect example of the critics viewpoint.  High friction taken to the max.  Even allowing for the poor execution of the design, it was a nightmare to get it to run at all.  I have heard that builders of the elbow engine say that they have to be built "loose" and becauseof that the leak air and oil.

I started looking at swash plate engines like the Mitchell Crankless Slipper Plate engine featured in Home Shop Machinist.  The builder of this engine says that the slipper plate bearing must ride on a  continuous oil film on the plate and at high RPM, throws oil everywhere.  I guess that is why commercial swash plate pumps and motors are fully enclosed, to provide an oil bath to the components and not to the observer.

I the switched to wobble plate designs and focused on the ball pivot mechanism because of its simplicity.  I designed and built the "Weeble" and was amazed at the relative lack of friction.  Even roughly built, it spins freely and achieves high speeds easily.  I have not published any plans for the Weeble but would be happy to post a .pdf or .dxf of my working drawings if anyone is interested.

The Weeble is a ball pivot swash plate design that uses a single ball and socket to control the pivot point of the wobble plate and two control rods that prevent the plate from revolving and getting its arms all tied in a knot.  That is one of the things that bothers me about this engine type.  The need for an added mechanism to prevent rotation of the wobble plate.  A Google Patent search turns up many examples of wobble plate mechanism and all require some gimicky looking mechanism that destroys the symmetry and seem like an inelegant solution to the problem.  My first approach to the problem was a fifth arm on the spider with a ball at the end that traveled in a fixed channel on the frame.  I ditched this in favor of the two brass rods that capture one of the spiders arms.  It works just fine but I just don't like the looks of it so I did my best to hide it.

A  few days spent on the boat before Christmas showed me the answer. Hanging on a bulkhead was a brass oil lamp that is free to tilt both fore and aft as well as athwart ship but it dosn't rotate! It is mounted in a gimbal, two pivot points at right angles that allow full wobble but no rotation.  Gears started turning in my head.

I have not completely settled on the whole design but I had to build a gimbaled wobble plate to test my theory.  Several designs that looked good on paper had serious limitations in execution. Here is what I have settled on for the gimball mechanism.  It provides wobble without rotation and is completely symmetrical. 

The completed gimball engine will borrow many design elements from the Weeble but I hope to be able to eliminate the other design element that bothers me. The rotary valve will be replaced by individual piston or spool valves. The design of the valve gear is my next hurdle.  Shop time is limited so this may be slow.

Best to all
Jerry

Gimbal Engine 001.jpg


Gimbal Engine 002.jpg
 
You have certainly got MY attention Jerry!

Please do keep us posted on the progress.

Rick
 
Hi All,

Here is more on the gimbaled engine that I am designing/building.  I have decided to feature the gimbaled wobble plate in  an exposed central location on the base rather than hiding it within the cylinder group.  The overall configuration of the engine will be sort of like a horizontal mill engine, with a cylinder sectiong, a movement section, and an output section, in an open arrangement to facilitate maintenance.

The gimbaled wobble plate is complete with bearing supports ready to mount on the base.  It consists of  a gimbal frame, two aluminum end caps connected by two 3/16” brass rods.  Each end cap has a 3/16” stub shaft  that rest in the upright aluminum bearing supports.  The stub shafts allow the frame to oscillate when actuated  by  piston rods connected to the top and bottom end of the vertical steel rod with the balls at the end.

In the center of each of the horizontal brass rods, top and bottom, is a brass bearing block.  The vertical arm of the spider rotate/oscillate in these bearing blocks allowing the horizontal spider arm to oscillate when acted on by  the air cylinders.  The sequential actuation of the four cylinders cause the crank arm to transmit rotary motion to the crank, in turn to the shaft, the flywheel, and a PTO, pulley or gear which will be mounted to the outer end of the shaft.

My design problems now turn to the cylinder end. The cylinders will be similar to the configuration used on the Weeble  engine.  I am not particularly satisfied with that design but it is quick and simple and so far I have had no inspiration for an improvement.  It will have four cylinders of brass tubing with an ID of 17/32” by 1 1/4” long.  The actual stroke will also be 17/32” which provides a thirty degree oscillation on the wobble plate and a crank throw of 3/8”.   The geometry is really more simple than it sounds. 

The big problem at this end of the engine is going to be the valves.  I have had enough of rotary distribution valves.  At this scale, good sealing is a big problem with rotary valves and  so I plan to use individual piston or spool valves.  These valves will be actuated by valve rods attached to the inner group of balls on the spider arms.  Because these balls are closer to the pivot point on the spider, the throw will be less than the cylinder stroke. But still, 3/8” is a long throw on a piston valve.  I am working on that.

The other design point that is as yet unresolved is the ball socket joint of the valve rod at the spider end.  I designed  and built a satisfactory and reliable ball joint for the Weeble, but this is different.  Instead of the ball being on the end of the arm, the spider arm rod passes through the ball.  In addition, each arm of the spider has two ball joints very close to each other.  I am working on that.

BTW, does anyone know what kind of steel nails are made of ?  I thought that I had a piece of 3/16”OD stainless steel that I had planned to use for the spider arms, but was not able to put my hands on it, and being anxious to proceed, I searched the garage for an alternate.  All I could find was a box of  20d common nails.  They are just a bit over 3/16”dia by about 4” long.  In the lathe, they cut easily and produced a nice finish.  I found a couple of 40d nails as well that are about 3/8”dia and stuck one in the lathe to play with it.  Found an occasional tough spot but mostly very easy to cut with both carbide inserts and HSS.

Enough of the commentary.  Here are some pics and a video of the current status.  I have not posted any  pics of the machining.  Nothing really interesting so far.  Simple setups on my well worn Unimat.

Regards
Jerry

<embed width="448" height="361" type="application/x-shockwave-flash" wmode="transparent" src="http://s468.photobucket.com/flash/p...CaptainJerry_Albumlbum/100_2398.flv"></embed>

http://s468.photobucket.com/albums/rr41/CaptainJerry_Albumlbum/?action=view&current=100_2398.flv

100_2395.JPG


100_2397.JPG
 
Now that's nice motion! 8)


I'm liking this line of experimentation Jerry.... Very interesting ;D

Also in answer to your earlier offer.... The rough drawings for the weeble would be very good to study... I might even give a build a go... It looks very good when running. Even if you see something you don't like in there... it lead you to this one along with the ships lamp.... Can't be bad ;)



I'll keep my eye on this build.... I like your style :)



Ralph.

 
Divided He ad said:
Also in answer to your earlier offer.... The rough drawings for the weeble would be very good to study... I might even give a build a go... It looks very good when running. Even if you see something you don't like in there... it lead you to this one along with the ships lamp.... Can't be bad ;)

Ralph.

Ralph:

Glad you are interested. I'll post a drawing on the Weeble thread to keep it organized.
If you want to have a go at it I'll be happy to help.

Jerry

 
A little progress.

I have been away from the shop for a week but mentally wrestling with the design of the valve mechanism. The valves will be operated by push rods attached to the inner set of balls on the spider arms. The inner and outer balls are just slightly over 1/4 apart and at each end of the stroke, the are just a little closer than 1/4 inch. The socket at the end of the con rod and valve rod must be compact to operate freely without interfering with each other. The joint must be strong and reliable. The design should be adjustable for wear and easily disassembled for replacement. It should be easily reproduceable (there are 8 of them) without extensive individual fitting.

I think I've got it. The attached photos show the sockets on two adjacent balls. the video show clearance through a 30+ degree arc.

The key to this ball/socket joint is the flats on each of the balls that allow the ball to be pushed into the socket with little resistance and then rotated 90 degrees so that the full radius part of the ball engages the socket's side plates. Once the balls round faces have engages the sockets, the ball joint articulates freely.

The sockets are made from 1/4" square brass rod. A 1/8" slot is cut in the end to a depth of 3/8". A 1/8" hole is drilled crosswise to this slot, through the faces of the fork. If the 3/16" diameter ball could be snapped into the holes by pressing them through the 1/8" slot it would be done but you can't do that without deforming the socket faces. Cutting a fine slit at the end of the 1/8" slot provides a little spring to the forks but not nearly enough.

The answer is to file two opposite sides of the ball flat to a thickness of just over 1/8" (.150 to .155) is about right. This will allow the ball to be pressed into the slot, with a nice "snap" when it seats in the holes. The rod is then rotated 90 degrees so that a full face of the ball engages the holes. The brass socket is not deformed its ability to spring back. I plan to drill and tap across the relief slit at the bottom of the slot for a 2-56 screw to allow the joint to be adjusted for wear.

I'll get the rest of them made tomorrow and the proceed to building the head plate and cylinders.

Jerry


Gimbal Engine 008.JPG
 

Very cool!

Great work!
 
CJ,

You find the most wonderous mechanisms :bow:

Best Regards
Bob
 
Hi All

Making some progress with the Gimbal engine design. Major components have been assembled and tested for clearances and free motion. This was done mainly to verify that the geometry was correct and to get a better visual relationship of the mechanism. Sketches and drawings just aren't the same thing as looking at reality.

Now that things are set in place, I can verify the valve throw and work out the dimensions for the spool valves. There is no eccentric. The valve push rods are operated by the inner ball joint on the spider arms and have a shorter throw than the piston rods.

A careful observer might wonder how these valves can work since they appear to be in phase with the piston rather than 90 degrees out of phase as you would expect.


GimbalEngine004.jpg


Gimbal End View




Gimbal End View Video

GimbalEngine002.jpg


Side View Pic



Side View Video


 
This project is going on hold. I will not have time to complete it before I leave for the Bahamas. I won't be leaving for a few weeks but I need to spend some time preparing the boat for the cruise.

I may have jumped into this a little too quickly because the concept of putting the wobble plate into a gimbal frame was so interesting that I couldn't get it out of my mind until I tested the idea. I have now reached a point in the project that is beyond the gimbal design so I can let it rest until august or september when I get back.

I really wanted to refine and improve on the wobble plate design so for the nest few weeks, any shop time that I have will be devoted to fine tuning the Weeble Engine. I started to prepare plans and build notes to post but as I got into verifying the details as built, I realized that there was more to do with the Weeble design before I presented it.

I will be adding to the Weeble thread as I make the necessary improvements and hope to have complete plans to share with you in the near future.

Jerry
 
Back
Top