LocostUSA.com

Moving right along on my 920 Virago 3-wheel tilter. I got a transmission output shaft on ebay for the extra U-joint required for the other end of the added drive shaft. And whaddya know, I found that I can mix the u-joints, that is swap ends so that I have identical splines on the engine end, allowing me to weld driveshaft extension tube to a spline end that matches, and do the same on the back end.

That added transmission shaft with its bearing is a perfect spot to add PTO sprocket for my hydraulic pump/motor, which will create pressure for the tilter cylinder AND power the car in reverse. A custom-made valve and a sprocket engager have to be built for this but they are pretty simple. Since in pumping mode I need 2:1 reduction, while in reverse motor mode I want 1:5 going the other way, I will use V-belt for one and roller chain for the other drive, on the same pair of shafts. The larger sprocket on the main drive shaft has to freewheel so as not to spin the driven sprocket up to 25,000 rpm. It is only engaged for reverse; shift is done at a dead stop, zero shaft speed. I estimate that the hydraulic pump/motor may draw (or produce) around two HP, so the single belt drive should work OK. It will need a tensioner pulley to let it slide while in reverse. Pics after I get some cutting and layout done.

I started on a 1/8 scale model to brush out any kinks ahead of time, plus get a good visual for style. I'm hoping to come out with a Bugati look. The front cylinder of the engine will protrude though the bonnet on the right side, and I will spend some time getting the body to adapt to this without looking gawky or haphazard. The lines of course must be fair and pleasing. I'll go with a straight bonnet with gentle slope downward ending with a classic Bugati radiator. I had hoped to get a water cooled V-twin which mounts transverse (one cylinder sticking out each side a la Motoguzzi) but I had to settle for the virago fore&aft air cooled motor. At least this one has more displacement. 2 chrome pipes out one side - yeeehaaa.

Last night I was able to set up the model frame to pivot around the roll axis, and I found that I have 30 deg. tilt each side, which is excellent. I need no more than that to keep balance in "normal" driving, which is all I expect to do. See, wide track isn't required for stability on this rig. The 2-seat portion is four ft wide, but I slanted up the outer six inches to get more tilt without cutting back much volume & comfort.

Right now I plan to lay up most of the body with epoxyglass. The bonnet can be made of steel or aluminum or fiberglass since it has no compound curves. It is two pieces, louvered, hinged down the middle. I am thinking of grinding a punch tool to make the louvers in metal; doubt that such a tool can be found for sale or rent. But then the other alternative is to build a mold including the louvers and lay it up, trim each louver with clippers and sandpaper. Any advice here is welcome. Oh and the forward cowl and faux radiator will be fixed to the front axle, NOT tilt. So there will be rotation between front end of bonnet and the cowl. Would be nice if that joint were circular centered at the point of rotation, but it would spoil the lines.

More later as it happens. Maybe I should start a build log on here?

Here is one more detail. I want a hard top, probably removable, and doors & working side windows. The side is flat so that a simple door can be built and fitted. I'm toying with using a single piece of glass or polycarbonate sitting inside the door, a light frame that seals against the windshield, so the glass can be raised and hooked in two or three positions in a locost way. No crank. One negative is that the window frame is always there unless it too is made removable.

Just in case anyone is paying attention, here's a quick update.

I finally decided at the last minute to use two chains on the hydraulic pump drive instead of one chain & one V-belt. Consistency I guess, and a more positive drive too. Parts may cost fifteen or twenty dollars more. The pump drive with its engagement mechanism (for reverse) is pretty much all designed, around the pieces I have already and some that are available online. Simple simple and plenty adequate to handle the few horsepower drawn by the pump/motor. I'll post some pics as it comes together.

I found that two of the hydraulic valves I need for this setup are not so easily available, so I designed my own in a locost way. One is for tilt control; it directs fluid to one end of the cylinder or the other and releases pressure from the opposite end at the same time. That is a fairly simple valve to make using a lathe and drill press, tapped holes, fittings. And much smaller & lighter than something off a Bobcat for example.

The other valve is to drop pressure from 2000 to 250 psi, because the tilt cylinder takes 250 max. Its available out there somewhere but finding one the right size (small) and the right pressure range, and then paying for it, is a bitch. Really much easier for me to make my own. More fun too. I'll let y'all know how that works out after I get all this hydraulic stuff set up for a test.

Rapt,

Sounds like a very interesting build. I'm having a hard time visualizing the direction your going with it, though. When you get to it, it'd be great if you could post some pics, or sketches of your concept.

-Indy "visualizing impaired" Joe

_________________
Starting a Scratch Build - ThunderBird Turbo Coupe Donor Car - being dismantled
Build Log Started: http://www.locostusa.com/forums/viewtopic.php?t=4914

Rapt, have you thought through the control system? It's not as simple as having it lean a given amount for a given turn of the wheel. Is it mechanical or electronic. Coming up with a stable PID controller isn't trivial. Not wanting to rain on your parade but just wanted to point out that there can be a lot of hours spent getting it to handle things like one wheel bumps and such.

_________________
Midlana book: Build this mid-engine Locost!, http://midlana.com/stuff/book/
Kimini book: Designing mid-engine cars using FWD drivetrains
Both available from https://www.lulu.com/

I've thought through the control system a lot, but not enough. The whole front end is unmodified '68 VW, and I keep that as it is since unlike a lot of locost builders, I don't have the skill, time or inclination to design my own. I mean so far as I can tell at this point, the VW is just about right. Weight on the front wheels will be somewhat less than the axle is designed for, but then, so much the better if that increases my ground clearance. (A tilter needs some air under it.)

Back to the tilting control, which is prolly what you are asking about. Step-by-step: center frame spine attaches to front axle. With rigid connection I have an untiltable three-wheeler. So I add a pivot hinge, axis lined up with rear tire contact patch centered in the back. A bell crank extends upward from the fwd end of the spine, is attached to a horizontal double-acting cylinder; other end of cylinder fixed to the top of the axle near one end. That is the only home-made part of the front end, a means to control chassis tilt while the axle stays even with the road.

Now we make and position a hydraulic valve to manually control the angle of tilt. There is an accumulator which holds a gallon or so under pressure, so I don't rely on pump output at any given instant; just to keep the accumulator full. I'd considered putting the tilt control valve on a separate stick, but that requires more hands than I have (confusing too) so right now the plan is to put it on the firewall next to the steering column, press steering wheel right or left to control tilt. The column would move a fraction of an inch either way. I'm afraid that turning the wheel and keeping constant pressure on one side of it may not be workable in the real world. Another option is to put the valve under the seat, add a little pivot to the seat, and use my body weight shift to do the tilting. Maybe more natural that way; I could get used to using my butt as a control input. At least one of the serious tilter-builders-for-profit apparently uses a fancy system combining steering angle, centrifugal force, etc. to automatically control tilt angle - I will try this driver-input thing first.

So far I don't see where one-wheel bump will affect this setup significantly, any more than it would without the tilting.

I now have a decent one eighth scale model that rolls and tilts. Frame, suspension & wheels anyway. And I have to say I'm very pleased with it; will post some pics soon.

I used a pair of RC model wheels/tires for the front, and mounted them on a miniature reproduction of the VW front axle. Making ball joints that small was a bitch but they're adequate for the job. Steering box will be a thread wrapped around the shaft with each end fastened to the tie rod.

The rear wheel, swing arm and shock are scaled from the Virago parts; these had to be all fabricated from a scratch. I molded the tire from room temp elastomer (2-part) - I can machine the mold halves from PVC which is nice easy stuff to work with.

Remaining are body parts, seats, windshield, roll bar, dash, steering wheel, two doors. I did a lot of the body with epoxyglass, which had to be laid over molds, but I figured that to get the right shape (compound curves) I had to mold something. The center part under the windshield, firewall and the bonnet are simple curves so I could use dog bisquet box material for those. I will slather them with resin or varnish to harden them up after the shape is right.

If you wonder why I go to all the trouble of building a model before the real thing...it is because this design is full of new and half-baked ideas. I get to work out the details as I go, and then actually see and test some of them with the model. The drive and the hydraulics will have to wait for the final machine but I will go to the trouble of setting up the hydraulic system in the shop with an electric motor drive, debug it before it gets installed.

Actually the VW front end is near perfect choice for a 3 wheeler, the trailing links are the best you could have for low interferance when hitting a bump on one side, no KPI to give you extra unwanted body roll (kingpin inclination lifts the inside and drops the outside wheels) and can use the structure as part of your chassis as a whole.

Hope to see your pictures soon!

The little model sits untouched for now. It confirmed the chassis layout I had in my head; I figured the little body parts can wait for a bit while I get into the hydraulics part.

This is two parallel sprocket drives of different ratios, light #25 chain, that connect the hydraulic pump/motor to the drive shaft just behind the engine. They will carry a max of about three HP. Only one of the two sprockets can be engaged with the shaft at a time, so I had to come up with a mechanism to engage either the pump, turning it one way, or the same pump acting as a motor, turning the opposite way. It is driven by pressure from an accumulator tank, meaning I can go backwards for only as long as I have pressure in the tank.

One chain would suffice if the fwd (pumping) speed were the same as the backward (reverse drive) speed, but the reverse speed has to be lower. I have a 2:1 ratio driving the pump at half shaft speed, and 5:1 driving the shaft in reverse at a fifth of hyd. motor speed.

I made up a set of collets to fit over the drive shaft, using a Virago transmission output shaft pretty much as is because it fits the u-joint spline on the engine. The moving collet slides about 3/8 inch, pushed by a bronze thowout bearing, to engage one sprocket or the other.

A problem here is that arrangement requires one sprocket to move axially a quarter inch, which is just about at the limit where the chain will jam. So I am considering letting the other (small diameter driver) sprocket slide too, fixing it to the pump/motor shaft with a key only and no setscrews. Not sure how effectively an angled chain will move the sprocket and keep it more-or-less aligned, but I'll try it and find out.

One nice thing about this Rube Goldberg rig is that the main drive shaft is not affected; not chopped or geared or interrupted when driving the rear wheel forward. Still promising some pics soon.

The hydraulics stuff hit a temporary snag until I get some better mat'l for machining parts. Did find just the right pressure reg. valve on ebay so I don't have to make one. I needed a spring to fit in an eighth-inch space, compressed, and push a bearing. A stack of wave springs was a decent solution, but they have to have a shim washer between each. It turns out that Smalley Spring makes coiled flat springs in a great variety of sizes. Plus inner and outer snap rings without eyes and ears. They sent me a sample of the spring I needed, catalog, etc. Excellent source IMO.

The model now has a roll bar; that is good to be done with since it had to be there before seats, rear body and a couple other things. I got my frame steel bought today, and I found a nearby source for clear select hickory boards to make the "rocker panels", the outer rails that will drag on the ground if one tips too far in a curve. Hickory lumber is very hard to find, even here in sawmill country where it grows. I've worked with it before though and nothing else would do the job. The steel and wood is for the final full-size machine BTW, so I'm not dawdling too long on the model. It rolls and it steers (sort of) and it is great for visualising spaces, layout. I tend to want to make it real pretty and detailed; maybe later, maybe never.

Oh, one roll bar departure is a fore-and-aft tube running from the center of the bar, curving to match a future roof profile, then bending down into the body at windshield angle, serving to support the center of the (probably two-piece) windshield, behind it, and finally fastening to the center spine a foot or so to the rear of the firewall. I thought it was a convenient way to stiffen the roll bar and provide a frame for a removable hardtop. Maybe I'll mount some lights on it.

I found the hickory lumber I was looking for, about an hour away. Am glad I waited - that stuff is tough as nails. I cut two planks per side and edge-joined them to form an angle of 30 deg. The upper plank conforms to the side of the car (side rail?) for six inches below the door, and the other one angles under as the outer and stiffer part of the floor pan. So the two boards are joined edge-to-edge with glue and screws; each has a single short curve to make the joint match. I still plan to cover the three feet between them underneath with an epoxy-glass sheet. Four steel cross-members will join this wooden business with the center spine. The final frame assembly may be a bit heavy and overbuilt but it will definitely be strong. And plenty roomy too.

Was able to steam-bend the upper plank to form the angle between straight parallel side in cockpit area to converging forward of the windshield to a smaller width between the front wheels. I was pleased at how easily the bend was formed, using a slotted coffee can as a boiler with propane torch for heat. The one inch (13/16) wood was plenty stiff but with vise, clamps and all I was able to get it to the angle I wanted within a short radius. And no cracks! Kudos for hickory, one of the strongest and most flexible materials on earth. I had considered using longer curves, but in keeping with the 20s Bugatti theme it seems right to have some straight body lines. Easier to fabricate too with limited tools and skill.

Here (I think) are a few shots of the frame layout and the 1/8 model.

Um, pics are too big to post. Wait til I trim them some.

Check out VSO image resizer, it can bulk resize images and its free. I'd love to see some pics man!

_________________
+442E, SR20det, Thunderbird/Markviii IRS with clutch lsd, R1 suspension all around, Mustang II 2" drop spindles, Woodward steering rack

Smaller one, model. Only one is under the size limit!