Railbus Drive
Nelson Riedel, Nelson@NelsonsLocomotive.com
3/20/2017, last updated

When I was in the early stages of this project Dick McCloy dug out an old set of wheels with axel.   He said that they came off a switcher locomotive.   The were a good match to pair with the 4" wheels on the arch bar truck I had.  The wheels are press fit and held in place with thin nuts.  

I believe I turned the ends slightly to fit R10 bearings: 5/8" ID, 1 3/8" OD and 11/32" width.      
The next step was to make journal boxes and rails for the axel.  (File: Journal) The rails are machined from 1" square bar stock.   The Journal Boxes are machined from a 2"X2" aluminum bar stock.   Everything is straightforward except the 1.375" hole for the bearing OD is made a few thousands large so the bearing can accommodate one box being higher than the other.     
This drawing shows the installed rails and journal box.  The truss straps are bent from 1"X1/8" bar stock.   The retaining plate below the journal box is also made from1" X 1/8" bar stock.   In another part I suggested that this area is very tight.   The suggestion was that the angle in front of the axel be moved forward 5/8"         
These are the springs used for the rear axel- McMaster #9657K374.  We should be prepared for a maximum of 300 pounds on the rear axel.   For 4 springs that would be about 75 lbs per spring leaving margin to absorb a bump.  
This shows an installed journal.   The top outside of the rails were ground round to match with the fillet in the angle.   The rails were secured with at least one screw and then welded in place. The truss straps are secured with 10-32 studs and Nyloc nuts.  Three 5/16" bolts with the heads cut off were screwed into the journal boxes to provide center posts for three springs per journal.  With only two springs per side the center position isn't used.
Early on in the project I went looking for suitable motors.  I wanted a motor that would fit between the wheels and was as big as possible.  I was shooting for about one HP (~750 watts).  The real requirement is that the motor be powerful enough to make the wheels slip before the motor stalls.  Stalled motors can overheat and be destroyed.  This AmpFlow G43-500 was the best I could find at 6.5" length including bolt heads on the back the shaft out the front.  It comes with a 11 tooth sprocket for #25 chain.  The 500 is the rated power ---500 watts or about 2/3 HP.  The motors are a little less than $100 including shipping and tax.  They are heavy so shipping can run up to $20.     

This drawing gives you all you need to know about the physical aspects of the motor.   

These curves provide insight into the motor electrical characteristics when operating at 24 volts (full power)  The torque is pretty much proportional to current.   The speed is fairly constant at about 3000 RPM independent of load, so how fast will we go at full power?  
The largest readily available sprocket that is smaller than the 5.8" wheel diameter is the 60 tooth model shown on the right. 
With a 11 tooth sprocket on the motor and a 60 tooth on the axel the maximum wheel RPM is 3000 X 11/60 = 550.

The 5.8" wheel has a 3.14 X 5.8 = 18.2" ~ 1.5' circumference.  The 550 RPM then translates to 550 X 1.5 = 825 feet per minute or about  49000 feet per hour and with 5280 feet per mile we get about 9 mph.   That's about 72  scale mph which is what I'd expect for an overpowered bus going downhill with a tail wind.  

Definitely don't need a greater top speed  and since this 60 tooth sprocket is the largest we can use with the 5.8" wheels we'll go with it.   (The 60 tooth sprocket was later replaced with a 70 tooth sprocket.)

Many of the small locomotives use two motors.  So, should I use one or two?  There was room (barely)  for two so two it was.  Later I tried running on one motor and it performed like the prototype probably did --- slowed down on the hills.  I wanted a "muscle bus" so stuck with two motors.

The motors and chain drive must be secured to the dive axel so that the chain tension and alignment don't change when the axel tilts.  I used the inexpensive pillow blocks on the right to secure the motor mount to the axel.   

This graphic shows the entire drive.  (File: Drive)  The motor mount consists of of a 6"X9-1/4"X1/4" aluminum base, a 4"X9-1/4"X3/8" aluminum vertical and 1/8" aluminum angle stiffening plates on each side.  The four pieces of the mount are screwed together.  The motors are attached to the vertical with a pair of 6 mm screws each.   The motor on the front side (left side on drawing) is against the stiffening angle and rests on the base.  This motor is fixed.  The rear motor rests on the base but slides to adjust the chain tension.  The chain tension is tricky and very small movements can change the chain from too loose to too tight.   I finally tapped a 10-32 hole in the side of the aluminum nose of the rear motor in line with the shaft and horizontal.   A mating hole was drilled in the stiffing plate, and a 10-32 screw was inserted to make a fine tension adjustment.  The attachment screws can be loosened, the tension adjusted with the 10-32 screw and the  attachment screws then tightened.   

The drive assembly is a tight fit in the frame with only 1/8" clearance at the front and at the back.    In the Frame section it was noted that the batteries can be slid 5/8" to the front to give more room for the drive.   The extra space could be used by adding 3/8" to the dive and 1/8" clearance to the front and the back.    

There is one problem with the floating drive, what keeps it from spinning on the axel.?  There is a lever added to the lower rear of the motor platform--- the part resting on the flashlight in the photo on the right. 

The vertical part of the motor mount  looks like Swiss cheese.    One excuse is that the holes are for ventilation.  The truth is that the mount went through several iterations and as long as there was a place for new mounting holes I kept using it.   At the end there is a suggestion for an improvement.  If I do that I'll have to buy a new piece of aluminum.   

The socket head of the tension adjustment screw is just visible on the back in line with the drive sprocket.
This photo show shows how the drive lever is attached to the frame --- in the upper right.   It is attached with a 5/16" bolt  through oversize holes in the lever and the frame channel and secured with a Nyloc nut using washers under the head and nut.   An extra axel spring is used between the lever and channel and the nut is tightened to the point where the drive is level when at rest.   This single point swivel attachment permits each end of the axel to move up and down independent of the other side.   

The dreaded brakes next.  The brakes in fact work great and are needed at the speeds I race around the track.  I had some shoes left from the truck project, drawing on the right.   

A better design is to split the hanger and shoe so the pressure applied to the wheel is uniform across the shoes surface.  Maybe next time. 
This drawing shows how the shoes fit on the drive.  The square bars are held off the motor mount platform with 1/2" spacers.  The rear bar attaches to the platform directly whereas the front bar is offset.
This photo shows the rear bar.
The spacer and offset for the front is visible just in front of the sprocket.

The top screw in the lever on the rear round bar is 3/4" above the round bar and the bottom screw is 1-3/4" below the round bar. 

The linkage rods are 1/4" diameter. Note that the rod between the axel and motor mount platform appears to be scraping the axel.  That happens when the brakes are applied  and the lever at the back rotates quite a bit.  That doesn't happen if the brakes are properly adjusted.    
This photo shows the arrangement at a different angle. This is an early photo.  After the shoes wore in a little linings made of strips of auto serpentine belt with the ridges sanded off were glued to the shoes using epoxy adhesive. The linings significantly increased the brake effectiveness.

Linings cut from 3/16" X 1" brake lining material purchased from McMaster were just installed on the front but no test yet. 

This photo was taken from the rear but rotated to fit better on the page.   The brake cylinder has a 3/4" piston and
 is reverse acting.   The brake compressor and associated plumbing is covered in the Accessories section.

The controller mounts above the drive on the cover pictured on the right.  The cover is made of 1/4" PVC.  It is slightly out of position in the photo; it fits between the two frame outside angles.

The cover serves to seal the inside of the railbus from mice.    The second use is a platform to mount the controller and associated electronics.

This photo was taken early in the project.  The cover now looks like Swiss cheese similar to the motor mount.

The drive assembly has worked well over the two summers of operation.  However, experience  can always lead to improvements.  (File: Drive Improvements). Dan Staron early on told me that it is bad practice to connect more then two sprockets, especially two motors on a single chain.  It causes the sprockets and chain to wear quickly.   His experience is with large machinery.   The sprockets show wear but I don't know if the wear is normal.  

After sprucing up the original drawings it became obvious that I could install a second large sprocket on the axel and point one motor each direction as shown in the drawing above. An advantage of two chains is that one can make it back to the shop if one chain comes off thus avoiding the humiliation of being towed.

If there is a motor pointing each direction then if the axel is centered on the mount (slid toward the back 1") then the same size chain can be used on both sides.  This is a slippery side as the journals must be moved (OK on new construction) or the batteries must be moved still further toward the front.  Different size chains don't look so bad now.

Another concern is the load on the chain.   McMaster says that #25 chain has a working load of 88 lbs.   The earlier graph on the motor indicated that the maximum torque is about 1.5 Newton Meters that converts to ~13.3 inch pounds.  The 11 tooth sprocket has a pitch diameter of about 7/8" so the lever arm is 7/16".   Thus the 1.4 NM torque converts to 13.5/0.4375 = ~30.8 lbs.  The total for two motors is a little over 60 lbs so well within the 88 lb chain working load limit.    The wheels usually start slipping at about 25 amps per motor which corresponds to a about 1.5 Nm torque so chain load is far from an issue.

Drive Changes(updated May 10, 2017): After some track experience with the new controller I decided that I wanted a greater reduction between the motors and the wheel.   The 11-tooth sprocket on the motor end is the smallest that will fit on the motor shaft.  The original sprocket on the wheel axel was 60 tooth.   I replaced that sprocket with a 70- tooth sprocket that has an OD of 5.72", just less that the 5.8" wheel diameter.  The sprocket and chain may ride on high guard rails and frogs but they are turning with the wheel so that shouldn't be a problem.

This change presented the opportunity to replace the holey plate that holds the motors as shown on the right.  The motor on the left (front motor) is fixed and the one on the right slides to adjust the chain tension. The smaller holes and slots are for the motor attachment bolts.  The larger holes and slots are for motor ventilation.   

Note that I decided to continue to gang the two motors together on one chain.
This photo shows the motor side of the new plate.  The brass cylinder with screw on the left fits over one of the attachment bolts of the motor to adjust the chain tension.  A nut on the end of the screw pulls the motor to the left for a fine adjustment before the attachment bolts are tightened.

The big slot for the right motor is to permit the motor with sprocket to clear the top of the wheel and then drop down into position.  
The drive assembly is all ready for installation here.
This is the non-chain side of the drive.   Note that the back of the motor on the right is slightly rotated --- the bolt holes on the other end are aligned.    Never noticed that before but it really showed up with the fenders.      
This shows the chain side of the drive installed in the frame.   The fender (inner fender) fits between the wheel and sprocket.   The purpose is to prevent water from the wheels getting into the motors.

This fender is attached with three screws into the top of that motor mount plate.   The fender is easily removed to reinstall the chain --- which is not so easy. 
This is the fender on the side opposite the sprocket.  The fender is attached to the fixed motor.  The rotated holes show up here with a sloping fender top.  The frame is blocked up with the wheels hanging down to the lowest position which causes a rotation of the drive increasing the slant. 
The new drive presented the opportunity to replace the old holey cover box.   That is 1/4" thick plastic of some sort that I purchased for some other project.   This time I decided not to fasten the cover down.  Instead, it floats between the batteries at the front and that aluminum plate at the back.  Side-to-side motion is restricted by the aluminum plate (the cover sides extend to the back of the plate.  

The cover is only partially complete at this point. The cover also serves as the mounting surface for the controller and associated components.   More details in the Accessories Page. 
The new drive has been tested for about 10 miles.  The larger sprocket caused no problems.  The greater reduction btween the motors and the wheels had the desired effect of noticeable increase in torque/decrease in current.  


Additional design information is at the following links:

Return to Railbus Overview
Railbus Frame
Railbus Body
Railbus Controller
Railbus Chargers
Railbus Accessories


NLW Home