Model Craft Engine - Running

Canuck75

Active User
H-M Supporter Gold Member
Joined
Jan 1, 2014
Messages
233
Last August I read an article from a June 1946 Science & Mechanics magazine published in Chicago Ill. on a 32.5cc engine by Model Craft. No drawings or castings are available but there was enough data to do a build from bar stock.
General specs - 2 cycle, 2 cylinder in line, 180 degree crank, water cooled, spark ignition, 1-1/8” bore and 1” stroke, central rotary intake valve.
Decided to go for it.
Used 1144 for the crank, 7075 for the pistons and rods, C12L14 for the sleeves, CI for the rings and rotary valve, CRS for the flywheel and 6061 for everything else. Bronze bushings are used in the rods and for the outer crankshaft bearings.
The article has 2 piece rods but the pictures clearly show a 3 piece crank which would not be necessary with split rods. I can only guess that they must have experimented with both 1 and 3 piece cranks. Anyways, I decided on the 3 piece crank and use 1 piece rods which enables a much tighter crankcase for better pumping action. The split centre CI rotary valve is 3/4” wide X 1-3/4” OD and serves as the centre bearing. The crankshaft centre section is made up of the centre bearing journal, 2 webs and the rod journals, and is a snug fit on the rotary valve, the end result is extremely rigid. The front and rear crank pieces just clamp unto the the rod journal ends and don’t really add to crank stiffness.
I read lots of opinions and ideas for rings and decided to turn them in cold form. Used 15% of the bore diameter to get the expanded ring gap of .168” giving a new ring OD of 1.179” and a ID of 1.084”. Once turned the extra circumference was removed by chisel and the ends filed square to be a able to just fit back into a clamping jig at bore size plus .001”. It was then skinned to bore size with no gap. Opinions are that it will have an adequate gap after the first run.
I borrowed the ignition system and fuel tank from my Boxer engine to ease the cost. The carburetor is of simple design with needle valve and barrel throttle and a 5/32” venturi which seems to be adequate for the engine size.
A water jacket of sorts was milled into the block and head and I planned to just use a gravity flow (bucket to bucket) but it doesn’t seem to get more than just warm in 3 minute runs. While this doesn’t sound very professional I don’t have room to display completed engines so the extra work of a radiator and water pump are kind of not warranted - hope no one takes offense.
Am using 80/20 methyl hydrate to castor oil. It started fairly easy and have 30+ minutes of run time so far. Have also used 50-1 chainsaw mix which runs about the same but with a lot less needle valve - must be thinner! Interestingly the ignition timing and mixture have a larger effect on engine speed than the throttle, so will need to play with it to see what works best. Compression certainly feels good and tested 49 and 54 lbs, might change the rings on the low cylinder to see if it improves.
In general it runs well and was shop time well spent!

Thanks for looking.

P.S. Will post the pictures in a couple of groups and a video separately.
 

Attachments

  • IMG_0363.jpeg
    IMG_0363.jpeg
    337.2 KB · Views: 18
  • DSC07937.jpeg
    DSC07937.jpeg
    239.2 KB · Views: 14
  • IMG_0231.jpeg
    IMG_0231.jpeg
    316.7 KB · Views: 14
  • DSC07952.jpeg
    DSC07952.jpeg
    371.7 KB · Views: 15
  • IMG_0295.jpeg
    IMG_0295.jpeg
    410.6 KB · Views: 17
  • IMG_0305.jpeg
    IMG_0305.jpeg
    362 KB · Views: 15
  • IMG_0315.jpeg
    IMG_0315.jpeg
    309.1 KB · Views: 13
  • IMG_0325.jpeg
    IMG_0325.jpeg
    425.4 KB · Views: 12
  • IMG_0330.jpeg
    IMG_0330.jpeg
    441.4 KB · Views: 13
  • IMG_0331.jpeg
    IMG_0331.jpeg
    335.6 KB · Views: 14
  • DSC07958.jpeg
    DSC07958.jpeg
    255.4 KB · Views: 13
  • IMG_0336.jpeg
    IMG_0336.jpeg
    425.1 KB · Views: 14
  • IMG_0340.jpeg
    IMG_0340.jpeg
    460.6 KB · Views: 13
  • IMG_0345.jpeg
    IMG_0345.jpeg
    428.8 KB · Views: 14
Last edited:
Some additional pictures.
 

Attachments

  • DSC07941.jpeg
    DSC07941.jpeg
    354 KB · Views: 9
  • IMG_0260.jpeg
    IMG_0260.jpeg
    411.3 KB · Views: 8
  • IMG_0257.jpeg
    IMG_0257.jpeg
    302.9 KB · Views: 7
  • IMG_0255.jpeg
    IMG_0255.jpeg
    312.6 KB · Views: 7
  • IMG_0239.jpeg
    IMG_0239.jpeg
    385.4 KB · Views: 7
  • IMG_0237.jpeg
    IMG_0237.jpeg
    515.2 KB · Views: 8
  • IMG_0235.jpeg
    IMG_0235.jpeg
    440.9 KB · Views: 8
  • IMG_0230.jpeg
    IMG_0230.jpeg
    351.6 KB · Views: 7
  • IMG_0204.jpeg
    IMG_0204.jpeg
    525.1 KB · Views: 5
  • IMG_0202.jpeg
    IMG_0202.jpeg
    367 KB · Views: 5
  • IMG_0198.jpeg
    IMG_0198.jpeg
    400.3 KB · Views: 5
  • IMG_0196.jpeg
    IMG_0196.jpeg
    414.2 KB · Views: 6
  • IMG_0195.jpeg
    IMG_0195.jpeg
    469.8 KB · Views: 6
  • IMG_0192.jpeg
    IMG_0192.jpeg
    481.9 KB · Views: 10
That is really cool. Great work.


Sent from my iPhone using Tapatalk
 
I find the problems in setting up for machining sequences very interesting so thought some of you might be as well.
My 4-jaw wasn't big enough to to grab the crank centre piece with the one pin and centre journal already turned on the other side, so wound up by doing it in the mill. Made some holding blocks which also clamped the bottom rod journal to keep it equal and opposite, then after roughing it with an EM, finished it using the boring head which worked very well.
 

Attachments

  • IMG_0228.jpeg
    IMG_0228.jpeg
    526.5 KB · Views: 5
  • IMG_0223.jpeg
    IMG_0223.jpeg
    543.1 KB · Views: 5
  • IMG_0213.jpeg
    IMG_0213.jpeg
    400.3 KB · Views: 6
  • IMG_0212.jpeg
    IMG_0212.jpeg
    547.8 KB · Views: 6
  • IMG_0206.jpeg
    IMG_0206.jpeg
    518.6 KB · Views: 6
  • IMG_0209.jpeg
    IMG_0209.jpeg
    444.7 KB · Views: 5
Back
Top