Hypocycloidal Engine
- Thread starter Ken I
- Start date
Help Support Home Model Engine Machinist Forum:
The Barrel Hoops.
Thanks John & Steamchick for your comments.
I was thinking much along the lines that Steamchick suggested - making brass hoops with rivets but I experimented on a scrap piece of Aluminium just to see how sizes worked out and was quite happy with the final result – so Aluminium it is.
The “Hoops” were made by pre-parting down to 0.80mm deep (using a 1mm wide grooving tool) – the finished rings being only 0.5mm thick – so when you take your final 0.5mm cut with a boring bar (1.0 on diameter) to finished size – the rings are parted off by the boring bar. The burr twists off with pliers.
Don’t try to part off 0.5mm wall tube, it will end in tears .
The theoretical circumference equals an inner diameter of 49.124 – I measured the wooden barrel by wrapping a strip of paper around, marking and measuring it which gave me a theoretical diameter of 49.5 – so I made my first ring 50.0 and surprise – it was a bit loose. The next ring was 49.5 and fitted perfectly. So you can trust the paper strip measurement.
Final ring positions set with a Vernier and lacquered into position.
Cylinder & Barrel Assembly.
The M2 & M3 Hex bolts are made by silver soldering nuts onto threaded rod or cut off cap screw shanks and then cleaned up.
Next job the gearset.
Thanks John & Steamchick for your comments.
I was thinking much along the lines that Steamchick suggested - making brass hoops with rivets but I experimented on a scrap piece of Aluminium just to see how sizes worked out and was quite happy with the final result – so Aluminium it is.
The “Hoops” were made by pre-parting down to 0.80mm deep (using a 1mm wide grooving tool) – the finished rings being only 0.5mm thick – so when you take your final 0.5mm cut with a boring bar (1.0 on diameter) to finished size – the rings are parted off by the boring bar. The burr twists off with pliers.
Don’t try to part off 0.5mm wall tube, it will end in tears .
The theoretical circumference equals an inner diameter of 49.124 – I measured the wooden barrel by wrapping a strip of paper around, marking and measuring it which gave me a theoretical diameter of 49.5 – so I made my first ring 50.0 and surprise – it was a bit loose. The next ring was 49.5 and fitted perfectly. So you can trust the paper strip measurement.
Final ring positions set with a Vernier and lacquered into position.
Cylinder & Barrel Assembly.
The M2 & M3 Hex bolts are made by silver soldering nuts onto threaded rod or cut off cap screw shanks and then cleaned up.
Next job the gearset.
I'm lazy - I but the bolts. Or if an odd size, make from hex bar. But I enjoy your endeavour! If you want a comment of the "rivet counter" type, you have missed following the grooves (plank joints) across the ends of the cladding, but I only mention it because a "rivet counter" pointed it out on one of my models! The decision is whether "real" plank joints would be effectively radial, or rectangular in geometry! Hope you are enjoying this as much as we are enjoying your posts..!
Thanks,
K
Thanks,
K
Ah ! The "tyranny of the rivet counters" is oft heard in my other hobby - Slotcar racing.
I too would but the bolts (I also frequently mis-hit the t instead of the y key) but you can't get M3 or smaller in Hex - and cap screws don't do it for me.
Regards - Ken
I too would but the bolts (I also frequently mis-hit the t instead of the y key) but you can't get M3 or smaller in Hex - and cap screws don't do it for me.
Regards - Ken
Making The Gears.
Firstly I developed the tooth profiles on AutoCad vis:-
See my post on generating your own gear profiles :-
Home Machinist Gear Cutting
Basically a Module 1.5 gear profile with a 32T annulus and a 16T pinion. The resultant Ø48mm pitch circle of the annulus is therefore the stroke of the cylinder – I left an additional 1mm clearance at either end between the piston and cylinder heads at TDC & BDC.
I generated the pinion tooth form from a rack (black lines a’La Maag shaper) then used that tooth form to generate the annulus (cyan lines a’La Fellows shaper).
I found the standard 20° pressure angle a bit too “coggy” for my liking and changed to 25° as shown above. (I 3D printed an ABS prototype.)
The root radius and clearance is non-standard – I developed this with the intention of water jet cutting the gear and M1.5 is the smallest module you can accomplish with a waterjet’s Ø1.2mm kerf.
My friend’s waterjet machine had problems with precision geometry and I ended up getting them wire cut on another friend’s EDM.
The nice thing about generating your own profiles is you can play with it to your heart’s content and then use the database to cut it (or make cutters). Both the 20° and 25° profiles are in the drawing.
I have in the past laser cut gears - it works quite well in stainless - which laser cuts beautifully. Its probably the cheapest way of getting it done.
Both Annulus gear and footplate bolted together for common alignment of bores, mounting surfaces etc. Both plates drilled and bored to their smaller bore sizes for subsequent independent machining – to facilitate clocking.
Tapping the M2 threads through 6mm stainless was a bit nerve-racking.
Finally (bottom right) wire cutting both annulus and pinion (I have turned the scrap over – different patina – shows up better.
Next step the crank assembly.
Firstly I developed the tooth profiles on AutoCad vis:-
See my post on generating your own gear profiles :-
Home Machinist Gear Cutting
Basically a Module 1.5 gear profile with a 32T annulus and a 16T pinion. The resultant Ø48mm pitch circle of the annulus is therefore the stroke of the cylinder – I left an additional 1mm clearance at either end between the piston and cylinder heads at TDC & BDC.
I generated the pinion tooth form from a rack (black lines a’La Maag shaper) then used that tooth form to generate the annulus (cyan lines a’La Fellows shaper).
I found the standard 20° pressure angle a bit too “coggy” for my liking and changed to 25° as shown above. (I 3D printed an ABS prototype.)
The root radius and clearance is non-standard – I developed this with the intention of water jet cutting the gear and M1.5 is the smallest module you can accomplish with a waterjet’s Ø1.2mm kerf.
My friend’s waterjet machine had problems with precision geometry and I ended up getting them wire cut on another friend’s EDM.
The nice thing about generating your own profiles is you can play with it to your heart’s content and then use the database to cut it (or make cutters). Both the 20° and 25° profiles are in the drawing.
I have in the past laser cut gears - it works quite well in stainless - which laser cuts beautifully. Its probably the cheapest way of getting it done.
Both Annulus gear and footplate bolted together for common alignment of bores, mounting surfaces etc. Both plates drilled and bored to their smaller bore sizes for subsequent independent machining – to facilitate clocking.
Tapping the M2 threads through 6mm stainless was a bit nerve-racking.
Finally (bottom right) wire cutting both annulus and pinion (I have turned the scrap over – different patina – shows up better.
Next step the crank assembly.
Last edited:
The Crank & Pinion Assembly.
Since the purpose of this engine was to work around crankshaft patents – I would imagine the use of the term “crank” would have been studiously avoided.
Above all the bits that make up the crank and pinion assembly.
All the bits assembled below (with flywheel).
Making The Cranks.
I had a small piece of brass stock but too little extra length to work it so I reamed the Ø5 hole and soldered it onto a smaller shank – after finishing the machining operations I parted it off (whilst simultaneously cleaning away the solder at the backside) – heated it up to push out the remaining piece of Ø5 brass – and then re-reamed the hole to clean up.
Note: the valvegear eccentric – in the above photo the part is off centre in the 4 Jaw for this operation – the other diameters were previously turned concentric to the (soldered on) shank / bore.
I set up the offset by mounting my dial gauge on the cross slide and using the gauge as a zero marker, measured the "throw" with the cross-slide scale.
The drawing calls for silver soldering or riveting of the axle pins to the cranks.
I like riveting joints vis :-
Top Left – Machining the pinion crank – a piece of Ø30 mounted 2.5mm off-center was used – the 4-Jaw then transferred to the mill which retains the off-center alignment (same with crank).
Top center – using the pinion as a holding fixture to counterbore the holes.
Bottom Left – riveting tooling – the lower snap (held/supported in vice) is a drilled out bolt shank.
Bottom Centre – Riveting using a flat face pin punch (a hollow snap would be better).
Top Right – Cleaning away the excess after riveting.
Bottom Right – the cleaned up riveting of the crank to its main axle.
Note: The Ø5 holes are flared at the riveted end (60° x 1mm deep) – completely filled by the riveting – those axles are going nowhere.
The pins have 0.8mm extra length for riveting.
You have to choose which way you want the motor to rotate by the eccentric offset when you make the crank.
I went for antclockwise - and I am still pondering if I can somehow incorporate a reversing set up.
Only the baseplate and presentation base still to do.
Regards, Ken
Since the purpose of this engine was to work around crankshaft patents – I would imagine the use of the term “crank” would have been studiously avoided.
Above all the bits that make up the crank and pinion assembly.
All the bits assembled below (with flywheel).
Making The Cranks.
I had a small piece of brass stock but too little extra length to work it so I reamed the Ø5 hole and soldered it onto a smaller shank – after finishing the machining operations I parted it off (whilst simultaneously cleaning away the solder at the backside) – heated it up to push out the remaining piece of Ø5 brass – and then re-reamed the hole to clean up.
Note: the valvegear eccentric – in the above photo the part is off centre in the 4 Jaw for this operation – the other diameters were previously turned concentric to the (soldered on) shank / bore.
I set up the offset by mounting my dial gauge on the cross slide and using the gauge as a zero marker, measured the "throw" with the cross-slide scale.
The drawing calls for silver soldering or riveting of the axle pins to the cranks.
I like riveting joints vis :-
Top Left – Machining the pinion crank – a piece of Ø30 mounted 2.5mm off-center was used – the 4-Jaw then transferred to the mill which retains the off-center alignment (same with crank).
Top center – using the pinion as a holding fixture to counterbore the holes.
Bottom Left – riveting tooling – the lower snap (held/supported in vice) is a drilled out bolt shank.
Bottom Centre – Riveting using a flat face pin punch (a hollow snap would be better).
Top Right – Cleaning away the excess after riveting.
Bottom Right – the cleaned up riveting of the crank to its main axle.
Note: The Ø5 holes are flared at the riveted end (60° x 1mm deep) – completely filled by the riveting – those axles are going nowhere.
The pins have 0.8mm extra length for riveting.
You have to choose which way you want the motor to rotate by the eccentric offset when you make the crank.
I went for antclockwise - and I am still pondering if I can somehow incorporate a reversing set up.
Only the baseplate and presentation base still to do.
Regards, Ken
Last edited:
Wow this is incredible!
Your post is as professional as your work Ken, thanks for all the effort you have gone to. Cheers,Peter.
Making The Baseplate.
Nothing much to show here other than the circular (machine) polishing :-
Use a machined wooden dowel faced with a 3mm thick leather pad – glued on.
Choose a convenient size and pitch and work to a pattern.
In this case Ø16 @ 9mm pitch (my minimill’s leadscrew is 1.5mm) – I first ran down the middle – then offset 9mm from row to row and 4.5mm offset of the pitching on each following row – starting each row from the same end – repeat until your arm falls off. Then work outward on the other side mirror image.
I used coarse valve grinding paste, keep it wetted – using oil or water depending on what your paste base is - this is water based – but it tends to dry if you press hard – so I think oil based might be better here. A few seconds of moderate pressure per whorl is all it takes.
I've done this before and tried sandpaper disks, plain wooden dobs, rubber faces etc. etc. - the leather disk and grinding paste works the best.
Assembly & Fit Up.
Loosely assemble all the crank and gear parts.
Next assemble onto the base with the pinion gear correctly engaged – rotate crank, the big end pin should cycle horizontally in a straight line.
First go - the pinion was tight at the bottom and had lash at the top - clearly the two halves were out of line - I slackened off the nuts and re-tightened whilst rotating the gear - eventually rotated perfectly with practically zero lash.
Before fitting the cylinder assembly – check the centerline height of the big-end pin – it should be parallel to the base.
If it’s miles out then you probably don’t have the gear correctly aligned (like a tooth out).
It is possible, due to small angular errors in the orientation of the Annulus Gear (Detail -4-) or the Pinion Crank (Detail -12-) that the locus of motion is not parallel to the base - this will cause binding.
This can be corrected by shimming the Standoff Spacers (Detail -25-) to correct this with respect to the base or by adjusting the length of the Upright Spacers (Detail -26-) to set the cylinder to align with the locus of motion of the big-end pin. Or a combination of both.
My assembly ran true to the base and ran fine with the cylinder attached.
I had a tight spot near the end of the cylinder stroke which was cured by a little more honing of the cylinder bore at that end.
Other than that it ran first try – with no sealant on any of the joints or gland packing – so quite a bit of leakage.
The gland packing is PTFE plumbers tape twisted into a cord (use a drill) and then wound around the shaft and compressed to fit and seal by the gland nut.
Once happy with the running and fit up I assembled all the metal to metal surfaces with Loctite acid free silicone gasket maker (Loctite SI 5699 Grey).
Assembling the steam chest under the barrel hoops was a PITB but otherwise fine.
I then ran the engine at about 100 rpm for 4 hours – lubricating with ATF (automatic transmission fluid) – at the gears, conrod ends, valve stem and piston rod glands – stopping periodically to open the quick coupler and inject oil directly into the steam chest – to oil the piston internals and “D” valve.
It ticks over beautifully at slow speed and low pressure – being Ø20 x 48mm stroke it’s actually quite a powerful little engine.
I'll make a video once the presentation base is finished and post a link to the YouTube video of it running.
Presentation Base.
Finally a base :-
Chunk of wood cut from an ugly old plank of (what I think is) Kiaat.
Top Right – Use self adhesive printed sticker for “marking out” (my printer is more accurate than I am).
Mid Right – Drilling for M8 holddown Hex bolts.
Bottom Right – Milling the pocket.
Not shown – milling the moulding using router bits.
I used no stain and finished it with a polyurethane lacquer.
Still needs one more sand & another coat of lacquer.
I should have it finished & mounted and the video posted by Monday - then I'll post the plans.
Regards - Ken
Nothing much to show here other than the circular (machine) polishing :-
Use a machined wooden dowel faced with a 3mm thick leather pad – glued on.
Choose a convenient size and pitch and work to a pattern.
In this case Ø16 @ 9mm pitch (my minimill’s leadscrew is 1.5mm) – I first ran down the middle – then offset 9mm from row to row and 4.5mm offset of the pitching on each following row – starting each row from the same end – repeat until your arm falls off. Then work outward on the other side mirror image.
I used coarse valve grinding paste, keep it wetted – using oil or water depending on what your paste base is - this is water based – but it tends to dry if you press hard – so I think oil based might be better here. A few seconds of moderate pressure per whorl is all it takes.
I've done this before and tried sandpaper disks, plain wooden dobs, rubber faces etc. etc. - the leather disk and grinding paste works the best.
Assembly & Fit Up.
Loosely assemble all the crank and gear parts.
Next assemble onto the base with the pinion gear correctly engaged – rotate crank, the big end pin should cycle horizontally in a straight line.
First go - the pinion was tight at the bottom and had lash at the top - clearly the two halves were out of line - I slackened off the nuts and re-tightened whilst rotating the gear - eventually rotated perfectly with practically zero lash.
Before fitting the cylinder assembly – check the centerline height of the big-end pin – it should be parallel to the base.
If it’s miles out then you probably don’t have the gear correctly aligned (like a tooth out).
It is possible, due to small angular errors in the orientation of the Annulus Gear (Detail -4-) or the Pinion Crank (Detail -12-) that the locus of motion is not parallel to the base - this will cause binding.
This can be corrected by shimming the Standoff Spacers (Detail -25-) to correct this with respect to the base or by adjusting the length of the Upright Spacers (Detail -26-) to set the cylinder to align with the locus of motion of the big-end pin. Or a combination of both.
My assembly ran true to the base and ran fine with the cylinder attached.
I had a tight spot near the end of the cylinder stroke which was cured by a little more honing of the cylinder bore at that end.
Other than that it ran first try – with no sealant on any of the joints or gland packing – so quite a bit of leakage.
The gland packing is PTFE plumbers tape twisted into a cord (use a drill) and then wound around the shaft and compressed to fit and seal by the gland nut.
Once happy with the running and fit up I assembled all the metal to metal surfaces with Loctite acid free silicone gasket maker (Loctite SI 5699 Grey).
Assembling the steam chest under the barrel hoops was a PITB but otherwise fine.
I then ran the engine at about 100 rpm for 4 hours – lubricating with ATF (automatic transmission fluid) – at the gears, conrod ends, valve stem and piston rod glands – stopping periodically to open the quick coupler and inject oil directly into the steam chest – to oil the piston internals and “D” valve.
It ticks over beautifully at slow speed and low pressure – being Ø20 x 48mm stroke it’s actually quite a powerful little engine.
I'll make a video once the presentation base is finished and post a link to the YouTube video of it running.
Presentation Base.
Finally a base :-
Chunk of wood cut from an ugly old plank of (what I think is) Kiaat.
Top Right – Use self adhesive printed sticker for “marking out” (my printer is more accurate than I am).
Mid Right – Drilling for M8 holddown Hex bolts.
Bottom Right – Milling the pocket.
Not shown – milling the moulding using router bits.
I used no stain and finished it with a polyurethane lacquer.
Still needs one more sand & another coat of lacquer.
I should have it finished & mounted and the video posted by Monday - then I'll post the plans.
Regards - Ken
Last edited:
There is another name for this crank gear: Hypocycloidal.
Pete & JasonB - you are correct - I've been calling it by the wrong name all along - so I have changed the title and all references of Epicyclic have been corrected to Hypocycloid.
That's twice on this build the error of my ways has been pointed out to me.
That's twice on this build the error of my ways has been pointed out to me.
You didn't need to go to such lengths. It's your engine and you can call it what you like!
Your work is excellent and I am looking forward to seeing video of the engine running.
Your work is excellent and I am looking forward to seeing video of the engine running.
O.K. All finished....
And a clip of it running
I have posted the drawings under downloads.
https://www.homemodelenginemachinist.com/threads/hypocycloid-engine.32466/
As ACAD.DWG or DXF (zipped) or an A0 PDF.
Regards, Ken
And a clip of it running
I have posted the drawings under downloads.
https://www.homemodelenginemachinist.com/threads/hypocycloid-engine.32466/
As ACAD.DWG or DXF (zipped) or an A0 PDF.
Regards, Ken
Last edited:
Stunning. Quite the curiosity, and beautiful as well.
Richard Hed
Well-Known Member
- Joined
- Nov 23, 2018
- Messages
- 2,380
- Reaction score
- 615
LOVE THE MUSIC, TOO
Cogsy
Well-Known Member
Beautiful and very well made.
Beautiful !
I love it..
I love it..
- Joined
- Sep 2, 2011
- Messages
- 1,342
- Reaction score
- 360
that's amazing. by chance do you have any pictures of making the flywheel? looks like you used dowels in it?
Werowance - I repurposed that flywheel from an earlier (failed) project - no photos.
However - I machined the boss with 5 spotfaces and M3 female threads - the Ø6 spokes were made with M3 female threads - at one end and a screwdriver slot at the other - the rim had 5 reamed holes so once all the axles were screwed in the rim was concentrically positioned - with the screwdriver slots proud of the rims' outer diameter.
So the threads screwed together with a short length of M3 threaded rod and locktite - and the spokes were turned 0.2mm undersize to a depth of 4mm (into the rim) and once assembled were soft soldered. The outer diameter was then machined (with the hub mounted on a stub shaft turned in the lathe) to clean up and remove the protruding screwdriver slotted spoke heads - et viola.
Since you can't machine the inside of the rim after soldering - mount the assembly in a lathe and clock it up - it should run true - but you can always beat some sense into it with a hammer if it doesn't - before soldering - of course.
You can see the 0.10mm annulus - filled with solder in the photo above - so it can never come apart. -
Also note that the material of the spoke has developed a different patina to the rim. This is about two years old - polished you cannot see the annulus - so I'm going to repolish it and clear lacquer finish it - as I have done for all the other brightwork.
The details are in the drawing under downloads.
Not 1000 words - but hopefully as good as a picture.
Regards, Ken
However - I machined the boss with 5 spotfaces and M3 female threads - the Ø6 spokes were made with M3 female threads - at one end and a screwdriver slot at the other - the rim had 5 reamed holes so once all the axles were screwed in the rim was concentrically positioned - with the screwdriver slots proud of the rims' outer diameter.
So the threads screwed together with a short length of M3 threaded rod and locktite - and the spokes were turned 0.2mm undersize to a depth of 4mm (into the rim) and once assembled were soft soldered. The outer diameter was then machined (with the hub mounted on a stub shaft turned in the lathe) to clean up and remove the protruding screwdriver slotted spoke heads - et viola.
Since you can't machine the inside of the rim after soldering - mount the assembly in a lathe and clock it up - it should run true - but you can always beat some sense into it with a hammer if it doesn't - before soldering - of course.
You can see the 0.10mm annulus - filled with solder in the photo above - so it can never come apart. -
Also note that the material of the spoke has developed a different patina to the rim. This is about two years old - polished you cannot see the annulus - so I'm going to repolish it and clear lacquer finish it - as I have done for all the other brightwork.
The details are in the drawing under downloads.
Not 1000 words - but hopefully as good as a picture.
Regards, Ken
Last edited:
