Revisiting a 100 year old design

I have been riding this for a bit, now, and thought you might enjoy the story. This is a "re-visit" to the 1915 Smith MotorWheel, a very popular and successful motorized bike attachment from the early 1900s. These motorwheels free pivot up and down to keep ground contact when the bike is leaned right or left. They are quick and easy to mount/dismount and my mount frame attaches to the bike with 3 hose clamps. The motorwheel/swing arm removes from the mount frame with just two nuts and a cable disconnect. This easy on-off makes the bike easy to transport, with the motorwheel in the trunk of the car.

I found the original patent papers in online records and learned how the original worked. While the original motorwheel was powered from the camshaft (8:1 cam drive reduction with 4-lobes on the cam to make it work), they had no clutch or freewheel. I have used a Honda GX with a fabricated sprocket carrier series and a left hand threaded freewheel. I hand fabricated this wheel from 14 gauge sheet metal, welded to a 20" bike rim.

My first arrangement used a lawnmower gear reduction which failed quickly. I had to start over with the drive system using a stronger primary drive gearbox.

During the sorting out of this project I tried several different methods of caster and control. I found the caster design was very sensitive to wobble above about 15 mph, and made for a very wide arrangement. The original Smith MotorWheel used a spring centering type of caster control, which I copied. When I found an Electra Cruiser7 I was able to eliminate the caster system because the motorwheel is closer in lateral alignment (to the tire contact patches), as compared to the original Smith product.

With the wheel pulled well forward, as shown, there is no more wobble at top motoring speed. Surprisingly, there is very little sensation of off-center thrust and because the motorwheel is traction limited, it does not tend to over-drive the bike up steep hills or on sharp bumpy turns (it weighs about 30 lbs...not much weight over the "drive tire"). With a very low center of gravity, this is very pleasant to pedal with the motor off or at idle.

Currently, I have the bike geared for easy cruise at 15 mph, with minimum "full clutch engagement speed" at just over 4.5 mph. I am using the 2650 rpm Staton clutch conversion.

My next motorwheel will be right-side mounted (because we all get on and off from the left) and will use a more compact and narrower drive system (with a Fed legal GX25 for power). JimL

 

[Kirk]

Very cool! Nice work. Could you please post a few more images. The geometry must have took some patients.

 

[JimL]

I have a couple more views. I will need to take some pics with my better camera to show how the drive works. The primary mechanism is the single side (outside) swing arm. Knowing what I have now learned, I believe I can pivot the next motorwheel slightly behind the "front edge" of the wheel (which will have a full fender). This would allow the wheel to squeeze forward a little more without getting in the way of your foot.

Because I have been able to eliminate the Smith MotorWheel castering mechanism, simpler mounting and operation becomes possible. Here you can see the mount method, where the frame is hose clamped to the bike and has two parallel receivers for the studs in the ends of the swing arm pivot support arms. There is an alternate mount hole that allows the motorwheel to be moved backward over an inch, in case someone with really big feet rides it. I will also learn if that position changes the feel, but I have been busy with other chores and we have had some freezing weather.



As you can see, the mount frame is completely out of the way when riding the bike without the motorwheel attached. I did not have this on my first two tries, and it is not a good setup without this feature. For those who think hose clamps are not strong enough, take a look at how the import drag racing guys used to install their engines for quick swap or repair; a fair of half tube saddles (one on each side of the engine) with 4 hose clamps....2 per side) that the engine mount tubes rest in. Only good enough to handle about 800 HP, of course.



This is not a very close view, but you can see the end of the primary spocket reduction set. The primary sprocket (11 tooth #35) runs this sprocket from a Staton 5:1 reverse rotation gearbox (which is how I have the engine spinning backwards in the bike, and the wheel turning forward). The chain to that upper sprocket passes on each side of the freewheel, which is driven by an 11 tooth sprocket on the end of the shaft carrying this upper sprocket. All my 5/8" ID bearings are 1 3/8" OD, which mount perfectly in 1 5/8" DOM mild steel rollbar tubing. I simple weld the sections of 1 5/8" DOM into 1/4" steel plates, make the end frames while all is welded straight and then hacksaw out the portions of the DOM tube where the sprockets/chains need to run. This method guarantees perfect alignment without any machine tools needed.

The nice part about eliminating the original Smith "caster" method (where the motorwheel castered like the front wheels of a grocery cart) is that the bike can be backed up fairly easily.

This would be an awesome method to build a simple e-bike. Use a front wheel hub motor (20") and mount the heavy batteries and controller stuff on top of the swing arm (it could be a double sided, like motorcycles). All the motor and battery weight would be below the axle line of the bicycle, the entire package could be removed for safe keeping or weather protection, and there is no need to reinforce the bike at all. No special wheels, disk brakes would be fine, and the whole thing would be cheap. An e-wheel would be much narrower and handle better than my cheap gas engine project. Frankly, I would love to build a bike that way, but the cost is currently out of reach for a good 20" front hub motor kit.

Keep in mind that this method was never meant for higher speeds. I think the Smith topped out about 20....I am pretty comfortable at easy cruising around 8-15 mph. I had this on my mountain bike for a while, but the short pedal-rear axle distance was a problem (but it was great having disk brakes!) At nearly 70 years old and already had one broken hip, I have no need to go fast. I just enjoy cruising around and enjoy the outdoors. I will try to post more pics when I make the next motorwheel.

Hope you get some fun ideas from this project.

 

[JimL]

Couple more notes:
- This thing climbs very steep hills with me pedaling. With the very low clutch engagement speed, it is easy to walk up the steepest part of my hill with a little motor assist (last 10 feet is too steep for me, so I just let the clutch slip a little and walk into the driveway.)
- Also, don't pay attention to the goofy throttle lever...I have ordered a long cable kit to fix that. These motor wheels need a very long cable and I am working on a quick connect that wont require tools.
-Also, the two nuts holding it to the mount frame should be sized to use one wrench. I had to use what I could find in the garage and so this one needs two wrenches and a pair of pliers to remove and load in the car for transport.

Thanks for your patience; I am not a very good writer.

JimL

 

[JimL]

Here are some pics showing how this reduction/freewheel unit works. I am sure this looks odd compared to most builds, but keep in mind that the entire engine, clutch, reduction, and freewheel must fit inside of a 15.5" diameter cone that is only 5" deep and 11.5" diameter at the small end of the cone.

Pic #1: This is a side view of the drive system. The gearbox is 5:1 with an 11T sprocket driving the upper sprocket. The smaller sprocket, on the upper shaft, is #410 chain driving the left-hand threaded freewheel. The carrier shaft for the freewheel passes between the runs of the #35 chain and is carried in 5/8"ID x 1 3/8"OD sealed bearings. The bearing carriers are pieces of 1 5/8" DOM (drawn over mandrel) .124" wall steel tubing.


Pic #2: The wheel flange is made from a large diameter #35 sprocket. During on of my early test drives, I learned there must be a strong retaining bolt in the end of the 5/8" shaft. The set screws and key are not enough to keep the wheel flange on the 5/8" shaft; I lost the wheel down a steep hill! I drilled and tapped both ends of the shaft and used stainless steel 6mm x 20mm bolts for positive retention.

Pic #3: This is a look at the other end of the shafts. The edge clamp prevents the flange/freewheel assembly from walking inward during rotation. The upper shaft has positive retention by the bolt on bearing carrier with a welded end stop. I use machinist shims to set freeplay in the bearing retainers. You can buy various sizes and thickness from industrial hardware stores.

View attachment 59611View attachment 59612

 

[Kirk]

The photos are impressive and answer a lot of my questions. I think these type of projects are the best. The sprocket setup would have scared me away.
Thanks.
If you have time, check out my Meridian. https://sites.google.com/site/kirkgretka/home/our-projects/meridian

 

[AutOtaku]

The photos are impressive and answer a lot of my questions. I think these type of projects are the best. The sprocket setup would have scared me away.
Thanks.
If you have time, check out my Meridian. https://sites.google.com/site/kirkgretka/home/our-projects/meridian

This is also very cool! Do you have a build thread here on MB?


A little off topic, but the porsche on your website is also very cool! Ive got a friend who fab up a custom supercharger setup to his GC Subaru Impreza

 

[JimL]

Thanks for the interest, fellows. I did some studying today and decided it would be smart to have Hi and Lo ranges for cruising or climbing. As a result, the setup I have shown will be scrapped and I will build a new arrangement. In conjunction with that decision, my wife suggested (and I agree) that a right side motorwheel might not be as safe. Approaching a curb, turning right when close to curbs, passing pedestrians....all would be safer if I stay with the left side motorwheel.

I began fabricating my dual range transmission using gears and shafts from a scrap motorcycle engine. In these pics, the blue tape shows where the shafts will be cut off and 5/8" keyed shaft installed (as needed). There will be chain drive from the clutch into the trans input shaft (top of both photos) and chain drive from the output shaft (bottom of both photos) to my 22 tooth, left hand thread freewheel. The freewheel will again be mounted on the center hub shaft.

With the dual range arrangement, I can go down to the legal GX25 engine, have superior climbing at low speed (or walking speed) and less RPM at my desired 15 mph cruise speed.

The first photo shows the transmission in Hi range. There will be a single shift fork to select range, and the bike must be stopped to shift (which gets around the laws requiring no gear change available while riding). Because the trans in NOT constant mesh (like a motorcycle), I use a 3-lug dog on the input shaft Hi gear. This allows the rider to rock the bike gently while shifting from one range to the other, without any binding. The shift lever will be in the cover (gearbox and sprockets will be out of view).

The second photo shows the transmission in Lo range. The ratios are 1.15:1 in Hi range and 2.73:1 in Lo range.

Those odd sounding ratios are due to the requirement for one gear (of a set) to have a prime number of teeth. Failure to use such an arrangement results in pattern wear and excessive noise. My large gear for Lo has 41 teeth, while my large gear for Hi has 37 teeth.

That's it for a while. I have a lot of hand work ahead; today was 4 hours cutting down gears to get my alignment to work.

JimL

 

[Kirk]

Thanks for the complements. No, I don't have a build thread.

 

[Kirk]

You're going to scrap the setup? That's just crazy talk. Keep the first wheel-a-roll'en while you build version 2.0! That's what Bill Gates would do.