Lapey Street Motor Cart, 1949

Last Revised: October 22, 2023

French Creek Valley Home                        Blacksmithing

My Stuffed Metalworking Shop                        Storing My Stock

My Home Woodworking Shop                  Contact Us

Motor Cart Side View
Side View of our Basic Motor Cart Design

Introduction to our style of Motor Cart

I've seen many articles about homemade motor carts recently. But none were as humbly made as the ones we built in the late 1940's and early 1950's. This article isn't exactly a how-to article. But it does give a lot of details on how we made our first motorized vehicles.
I have included a scaled drawing of the basic design to support the text. Although the drawing is scaled and produced via CAD (Turbocad) I have minimized the use of dimensions, hidden lines and details like fasteners. These elements should be obvious to anyone brave enough to attempt making a motor cart such as this one.

Basic Design Concept:
Our motor carts were about as simple as they could be. The main source of material was used "2-by" lumber from concrete forming at construction sites. An 8 foot 2 X 10 or 2 X 12 and a 3 foot 2 X 6 were pretty much all that was needed. The finished cart was about 6 feet or so long. The front steering board was about 3 feet long, but the rear wheels were placed as closely together as possible on either side of the 2 X 10 or 2 X 12 main "body". This gave the machine a sort of reverse tricycle look. Since turning was accomplished by rotating the steering board with the feet and/or a rope, this gave the machine a somewhat unstable posture at times. To minimize this problem, we avioded sharp turns at high speeds.

Motor Cart Top View
Top View of the Basic Motor Cart

The motors:
During the 1920's, 30's and 40's in the USA at least, washing machines in rural areas were powered by fractional horsepower gasoline engines. Used ones were the primary source of our motive power. The single cylinder Maytag was just barely capable of pulling us around. We used several different models of Briggs and Stratton engines, a Lauson or two and, later on, one infamous Homelite. One of our guys did have a 2 cylinder Maytag.

Wheels and Axles and Drive System:
We usually used four 10" or 12" diameter wheels with 1.75 inch tires. These were called "semi-pneumatic" tires. They had no inner tubes. They were simply circular hollow rubber tubes with tread, that were forced over a metal rim. These wheels did have ball bearings. I think we always used wheels with 1/2" diameter bores. We had learned early on that wheels with steel or brass bushings instead of ball bearings did not last long at the high speeds we subjected them to, 8 to 12 miles per hour.
The axles were 1/2" diameter hot rolled shafting. I don't remember where the material came from. The front axle was held in place with large fence-wire staples or simply with bent over nails. We hardly ever bought new nails. We learned how to straighten used nails at an early age and there were plenty of them available at area constructions sites. Early on, we mounted the rear axle the same way as the front axle. But the nails kept pulling out because of the forward pressure from the belt tightener. So, we began to have two brackets welded to the axle shaft and to bolt the brackets to body board.

Belts: I don't think any of us ever had a motor with a centrifugal clutch or with a chain drive. My dad came up with the belt tightener idea that we used on almost all of our motor carts. Since the belt tightening system was also the clutching system, there was quite a bit of wear on the drive belts. We needed even more slippage to get going than one might espect for two other reasons:
-Most of our engines ran at a single speed, so there was no "start out at low rpms, engage clutch, increase rpms" in the cards.
-Our engines didn't have much power, so we had to be careful to ease the engagement of the belt tightener for fear of killing the engine. ---And some of them were pretty hard to restart.
Anyway, my dad was a centerless grinder foreman for the Burd Piston Ring Company back then. They used Besley centerless grinders for finishing piston rings. The spindles on these grinders were driven by a matched set of seven 51" long "B" width vee belts. If more than 2 of those belts failed, then the remaining belts were scrapped. Dad brought these home. So, all of the motor carts in our neighborhood used 51" B belts.

Belt tightening system: The belt tightener itself was made from two metal plates about 3/16 or 1/4 inch thick. 2" wide and about 8" long. There was a centered pivot point, an upper belt tightener roller shaft and bearing and a lower operating rod connection point. These holes were either 3/8" or 1/2" diameter. We had to have those holes drilled for us. Fortunately for us, there was a junk yard/fab shop about 4 blocks from our house. Oscar Conquuist was the proprieter. He was a major source for some of these services. He also had a box full of roller bearings, one size of which worked well for our belt tightener roller. He also welded the mounting brackets to our back axles for us. I think he charged 50 cents per visit, regardless of what we came for. The belt tightener operating lever (clutch lever) was generally made from hardwood. When I was a kid, we still had many furniture factories on our side of town, so there was a lot of 1 X 3 elm and oak around--- on weekends. We needed hard wood for this lever because we were drilling through the 3 inch dimension and carving a pretty big notch out of one end for the operating rod. The operating rod was always made from a piece of 1/2" EMT (electrical conduit). We smashed the ends flat and drilled a hole in each flattened end for attachment.

Motor Cart Belt Tightener Assembly
Belt Tightening Components

Drive Ratio and Rear drive wheel: The best drive ratio was a 1 1/2" or 1 3/4" OD drive pulley on the engine with a 9" pulley on the single drive wheel. If we used 10" diameter rear wheels, the belt on a 9" pulley was almost touching the ground, but a 1 1/2" to 8" ratio made most of our motor carts very hard to get moving. Some motors had 5/8" diameter drive shafts, so if we couldn't find 1 1/2" or 1 3/4" drive pulleys we had to use a 2" diameter pulley in order to get a large enough bore. This, too, gave us start-up problems. It was not uncommon to see the operator pushing the cart with his feet while seated and while carefully engaging the belt tightener to get it moving. Attaching the driven pulley to the rear wheel was a big job for us. It required appropriately sized spacers and a good deal of accurate measuring to get the pulley firmly attached and well centered. This was one of the issues that took a lot of time to get right.

Motor Cart Drive Parts
Drive System Components

Even though we used the same principles of design on all the our later carts, no two motor carts were exactly the same. Each was designed to fit the size (leg length, mostly) of its owner, the restrictions of the motor to be used, the wheels and the available boards for the body and steering board. To begin the design, we'd:

1. -Build the steering board and attach it to the main plank (before cutting the main plank to length).
While the plank was propped up off the ground, the owner would sit on it to locate the seat back. A temporary seat back would be propped in place. Then the engine, with its drive pulley attached as far out on the shaft as possible, would be set onto the plank, as close to the seat and it's eventual brace location as possible. Care was needed to accomodate the starting mechanism and the gas tank.
The motor would be slid toward the drive side of the plank (always the left side, I think--- maybe Maytags had to be opposite)as far as possible while still having enough "meat" on the plank for the retaining bolts. Then, with the motor clamped in place (really held there with nails- I don't think we had any clamps) we would roughly locate the rear axle and drive wheel. Then we'd install the 51" drive belt. This was probably the touchiest part of the the design, because here we had to estimate the tightening effect that we'd be getting from our belt tightener system to determine the final location of the rear axle assembly and the location of the belt tightener pivot point relative to the drive pulley. And, since the length of the motor output shafts varied from one to another, we had to find a compromise motor location that would work with the driven pulley and the belt tightener. Having the driven pulley as close as possible to the plank was best, but required longer and stronger spacers in most cases.
2.-Build the rear drive wheel/ pulley assembly and then the rear axle assembly.
3.-Once the motor and rear axle were located, we would see how much of the main plank could be cut off for use as a seat back. If we had 18" or so, fine. If not, it's back to the construction site for more material. The seat was held at the predetermined angle and a brace was made. Then the seat and brace were both nailed in place.
4.-The belt tightener assembly was attached to the plank with a lag bolt and an appropriately sized spacer. The shorter the required spacer the better, since this was a moderately stressed attachment. And it was this attachement point that usually determined the life of that vehicle. We never did develop a metal bracket to improve the reliablity of this pivot point.
5.-Once the belt tightener location was fixed, we could locate the belt tightener operating lever. This was done by having the operator sit on the seat and testing the location for the lever. The main factors were 1.)comfort 2). ability to move the lever through the required arc 3) no interference between the operating lever and the steering board in a tight turn. The lever was attached to the main plank with a long lag bolt.
6.-With the belt tightener and its operating lever in place, the operation rod's length could be determined and the rod could be made and installed.
7.-That completes the basic assembly. The gas tank needs to be installed, if separate from the engine, and a kill switch might be fitted (unlikely). We always attached a clothesline rope loop to the steering board. The length of this rope was selected so we could use it to aid in steering control, even though we used our feet for the main control. Having the rope in at least one hand gave us an additional measure of safety if we hit a bump or rock with one front wheel. The rope was also used, all too often, for pullling the vehicle home if there was some sort of failure.
Later on, the rope was used to pull our vehicles to locations where we could run them.

Additional notes:
A. Note that the pivot point of the steering board is off center to the front of the board. This makes for more positive steering.
B.We used fence staples to attach the front axle, but later on we drilled at least one hole in the front axle so we could nail it in place. This kept it from slipping sideways.
C.Setting the Seat-to-Steering Board distance: The actual driver needs to sit on the machine to determine this distance. You want the driver's knees to be still slightly bent when the steering board is pushed all the way to one extreme. If the driver's leg was totally straight, he/she could loose control of steering or be hurt. You also don't want the driver's legs to be pulled up too tightly toward their chest when going straight ahead. This is uncomfortable on long rides and also could contribute to a loss of control.
D. Repairs and refurbishing were a part of life with these carts. The tread life of the semi-pneumatic tires, especially the drive tire, was fairly short. The pivot points (lag bolts) where the belt tightener assembly attached to the main plank would wear from the stesses applied and wallow out the holes. We would insert slivers of wood into the hole to refurbish it. Sometimes, we could reposition the offending pivot, but that usually caused other issues with fit-up of mating parts. Belt replacement was also a regular duty, but as I mentioned previiously, we had an inexhaustable supply of those 51 inch belts.
The main body board didn't even last for too long. A major problem was the leakage of oil from the engine, and sometimes, gasoline from poorly made fuel line connections. The board would become so oil soaked that the belt tightener lag bolt wouldn't stay tight. We could switch the whole main board front-for-rear once.
We also learned a lot about engines. We learned to solve problems with carburetion and ignition early on. Most of those motors used a lot of oil. Maybe that's why guys from our generation are always checking the oil in our motorized stuff today.

Closing thoughts:
it was not uncommon to get caught by someone and told that we could not drive our vehicles on the neighborhood roads.
None of the roads in our immediate neighborhood were paved, so it was a treat to go far enough away from home to find a blacktopped road to drive on. It was on those roads that we usually got caught. Sometimes our own road/roads were so bumpy that we would have to pull our motor carts for blocks just to find surfaces that were smooth enough for us. There were no "weight challenged" kids in our neighborhood.

Points still to be integrated:
Rockford Morning Star article from about 1948 or 1949
Nystrom's race track
Homelite Story
Should I actually build one of these to get some pictures?

My 2-Wheel motorized vehicles

I know this is sorta rambling, but I've added this paragraph because I didn't know where else to put it.
Over the years, I had a number of motorized 2-wheelers. Never got any big ones, though. Never had a new one either.
The collage below documents all of those machines, but none of them are actually mine. I just located a picture of each of the bikes that I owned and drove at one time or another.
Motorbikes and motorcycles that I have owned