Upshur Twin valve timing

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Gordon

Well-Known Member
HMEM Supporting Member
Joined
Jan 4, 2011
Messages
1,330
Reaction score
343
I am looking for someone who has built the Upshur Horizontal Twin. The instructions on valve timing are very confusing. I presently have the intake valve opening at TDC and the exhaust valve opening at BDC. Due to the design it is not possible to change the timing of the intake without changing the exhaust by the same amount. Cams for intake and exhaust are fixed on a common shaft. If it is changed to make intake 10° BTDC it makes exhaust 10° BBDC. So far I am not having very good results with intake or exhaust. Also changing valve timing on one cylinder also changes timing on the opposing cylinder. I have reviewed the drawing and my build and I keep thinking that I am missing something. Input welcome.

Gordon
 
What I did was adopt the valve timing from the Wyvern engine. I know that Wyverns run very well and are similar size as the Upshur Twin, so I figured that the Wyvern valve timing design was proven. That has proven true on my Upshur Twins. I will need to go into my files to see if I can find the exact cam profiles and angular orientations I came up with. In any event, try comparing with what you have got with Wyvern timing. See here: http://www.modelenginemaker.com/index.php/topic,3793.msg68882.html#msg68882.
 
These engines are meant to run, not set the standard for horsepower or fuel efficiency. you have quite a bit of leeway when it comes to valve timing. valve clearance will affect the timing and can be used for a bit of fine tuning once you get it running. I am starting to make one but have only cut the crankcase plates so far.
 
I am looking for someone who has built the Upshur Horizontal Twin. The instructions on valve timing are very confusing. I presently have the intake valve opening at TDC and the exhaust valve opening at BDC. Due to the design it is not possible to change the timing of the intake without changing the exhaust by the same amount. Cams for intake and exhaust are fixed on a common shaft. If it is changed to make intake 10° BTDC it makes exhaust 10° BBDC. So far I am not having very good results with intake or exhaust. Also changing valve timing on one cylinder also changes timing on the opposing cylinder. I have reviewed the drawing and my build and I keep thinking that I am missing something. Input welcome.

Gordon
I am not familiar with it but could you show a sketch or foto of the way the valves are run, that is, how they are attached to the power? Where did you get your drawings? Does this Upshur go by a different name?
 
OK Thanks for the input. I think that one of the problems with the valve timing is that both valve are actuated by a common cam. This means that adjusting intake also adjusts the exhaust. The ideal would be to make four adjustable cam lobes but that would be pretty difficult in an engine this small. Since others have this engine running I am a little hesitant to do redesign until I have at least gotten some run.

It is easy to get the timing so that when the valve action overlaps the exhaust forces the cases out of the intake so that fuel is removed before it can be mixed in the carburetor.
 
I am not familiar with it but could you show a sketch or foto of the way the valves are run, that is, how they are attached to the power? Where did you get your drawings? Does this Upshur go by a different name?
Drawings were published in Model Engine Builder magazine #7 and #8. The plans are also available direct from Upshur. Google still works.

https://rocketr.net/sellers/UpshurEngineWorks
 
Last edited:
I built the Upshur twin. There are 2 cams each cam operates both cylinders. One cam for intake and one for exhaust.
While both cylinder share the same cam for intake and the same cam for exhaust each phase can be independently adjusted since the cams are separate and are locked on the camshaft at different orientation.
You may have built one long cam but they need to be separated and oriented on the camshaft.

I set the timing by eye for few degrees overlap with Opening the intake before Closing the Exhaust symmetrically around TDC.
 
Just to keep life interesting the loctite holding the timing gear on the crank shaft let go so I have to take everything apart again and fasten the gear to the crankshaft. Life is interesting if not fun.
 
Drawings were published in Model Engine Builder magazine #7 and #8. The plans are also available direct from Upshur. Google still works.

https://rocketr.net/sellers/UpshurEngineWorks
That is very interesting place and very good prices for drawings. However, as for myself, I am not building ICEs only steam. I have a very good library of steam drawings but few for ICEs. So I don't intend to buy, build or draw any ICEs. However, I am interested in seeing what others do and I am not ignorant as to how ICEs work so I can offer advice sometimes. Usually others know a lot more about the workings of ICEs and particularly the construction of models than I do but I can still help a little if I can see the internal construction or sometimes the drawings.
 
Just to keep life interesting the loctite holding the timing gear on the crank shaft let go so I have to take everything apart again and fasten the gear to the crankshaft. Life is interesting if not fun.
I'm not sure how the gear fits on the shaft but would this be a good time to "re-time" the gear? Am not sure what or how you are doing.
 
It is easy to get the timing so that when the valve action overlaps the exhaust forces the cases out of the intake so that fuel is removed before it can be mixed in the carburetor.
A bit more clearance in the intake rocker will make the intake valve open a bit later to avoid overlap or move the cam one tooth so that exhaust closes and intake opens just after TDC.
 
A standard set of valve timing figures that will dun any 4 stroke engine are Inlet opening 5 degrees BTDC and closing 40 degrees ABDC Exhaust opening 45 degrees BBDC and closing 10 degrees after will run any engine. This will give in general a smooth running and idling engine wilder timing will make more power and at higher revs but idling can suffer to some degree. Retarding by some 5 to 6 degrees the overall timing can shift the torque higher in the rev range.

Valve shape port design etc. will change the the effect of valve timing. Hope this helps John.
 
Back when I was a school kid messing around with motor bikes the local motor cycle wrecker said to set the valves rocking at top dead centre (ex just closing and in just opening). this top dead centre point marks end of the exhaust stroke and beginning of inlet stroke - this has always worked for me even when faced with engines fitted with custom cams and cam-gears that have no timing marks.
 
My Upshur Twins (six of them) run well with the following cam parameters:

1. profile noses separated by 108.75°
2. base circle radius 0.167 in
3. max rise from base circle 0.0715 in (that is, tip of each lobe is on a 0.2385-in-radius circle)
4. intake lobe nose radius 0.028 in
5. exhaust lobe nose radius 0.015 in
6. full duration each lobe 115° (which is 230° on the crankshaft)
7. circular flanks tangent to both the base circle and the nose radius circle (that is, each lobe is four circular arcs all tangent to each other).

The minimum tolerable tappet diameter (in the circumferential direction) is 0.415 in on the exhaust and 0.375 on the intake. Otherwise, the edge of the tappet will dig into the cam lobe flank.

I made the gears, shafts, and lobes all one piece from ETD-150 steel from McMaster-Carr.

Cam 4.jpg
Cam 5.jpg

Cam 2.jpg
Cam 1.jpg

Cam 3.jpg
 
Last edited:
My Upshur Twins (six of them) run well with the following cam parameters:

1. profile noses separated by 108.75°
2. base circle radius 0.167 in
3. max rise from base circle 0.0715 in (that is, tip of each lobe is on a 0.2385-in-radius circle)
4. intake lobe nose radius 0.028 in
5. exhaust lobe nose radius 0.015 in
6. full duration each lobe 115° (which is 230° on the crankshaft)
7. circular flanks tangent to both the base circle and the nose radius circle (that is, each lobe is four circular arcs all tangent to each other).

The minimum tolerable tappet diameter (in the circumferential direction) is 0.415 in on the exhaust and 0.375 on the intake. Otherwise, the edge of the tappet will dig into the cam lobe flank.

I made the gears, shafts, and lobes all one piece from ETD-150 steel from McMaster-Carr.

View attachment 122153View attachment 122154
View attachment 122155View attachment 122156
View attachment 122157
You obviously have far more talent than I do. There is no way that I could begin to hold those kind of dimensions or tolerances. Upshur said that the engine could be built using only a lathe and a drill press so I am sure that he was not thinking in those terms either.

Thank you. I will try to digest what you are suggesting.

Gordon
 
A model engine will run with any kind of shaped bumps on a shaft that move the valves somewhat close the right time, unless you are interested in beating a world record for speed, do nor overthink the issue.
I agree. I was just looking for the general range of when valve action should occur. That has been answered. I am not sure how others manage to be so precise in stating when this action occurs. Folks say that they have valve open close at points that I cannot begin to measure. When they state that a valve opens at something like 7* I wonder how they determine that. Mostly I am just eyeballing it and I can certainly be off by several degrees.
 
Before I developed some better understanding of all the factor involved, like most of us, I built a 5" protractor wheel with various adapter to engine shafts, then positioned a marker to read the angle.
Then realized the piston at TDC does hardly move around +/- 5 degrees of crankshaft. That is a fact of trigonometry.
The gear have some backlash
An the push rods need some free play to guarantee a good valve seal.
At that point all the cam design parameter and the fancy programs go out the window.
One can put an indicator to detect the precise moment the valve opens and after all that work gain 12 RPM at full throttle.
 
Before I developed some better understanding of all the factor involved, like most of us, I built a 5" protractor wheel with various adapter to engine shafts, then positioned a marker to read the angle.
Then realized the piston at TDC does hardly move around +/- 5 degrees of crankshaft. That is a fact of trigonometry.
The gear have some backlash
An the push rods need some free play to guarantee a good valve seal.
At that point all the cam design parameter and the fancy programs go out the window.
One can put an indicator to detect the precise moment the valve opens and after all that work gain 12 RPM at full throttle.
That has also been my experience. Like you I made a protractor but found that it really did not tell me much with any degree of accuracy. It could not even seem to get the same result on the next revolution. Ignition timing is a little bit more accurate. Mostly it is a matter of just a little bit BTDC or a little bit ATDC. It is difficult to tell if this is happening at 10* or 12*. I mostly agree with your statement that any bump in the cam which makes a valve move is going to work. Once an engine is running a little bit of adjustment may make it run a little better but the difference is not going to go from sputter to wow. I have found this on various engines I have built. I play around with various adjustments until it runs but once it runs minor adjustment do not have a great effect in spite of trying a series of minor adjustment prior to initial start.
 
I am not sure how others manage to be so precise in stating when this action occurs. Folks say that they have valve open close at points that I cannot begin to measure. When they state that a valve opens at something like 7* I wonder how they determine that.
The cam lobes are made up of four arcs with known radii and center locations. If you machine the parts to the print, you will get very close to the design valve timing. You can locate the arcs using a cross slide rotary table, CNC, Volstro head on a Bridgeport, offset on a lathe faceplate (for the flanks), or eyeball to a template drawn using a pencil and compass, scanned, and scaled down using your printer. I admit I have not tried to measure the valve timings, but am confident they are close to the design values because I know that my geometry is close thanks to CNC. The geometry is rather deterministic, shall we say.
 

Latest posts

Back
Top