Relevance of engine-balance and timing issues affecting master-and-slave connecting rods.

[AspiringHowardHughes]

I believe that is the Gnome monosoupape. (French for single-valve)
The later versions were modified to replace the piston-crown valve with more conventional transfer ports around the base of the cylinder, relying instead on pressure equalization between the crankcase and the cylinder to prevent unwanted transfer of charge before the intake stroke.

 

[AspiringHowardHughes]

Errata: The "monosoupape" refers only to the improved version, the piston-crown valve models precede it and are known differently.

 

[AspiringHowardHughes]

If I recall correctly there was also a counterweight to assist the pressure-differential mechanism.
As the piston reversed its course at TDC, the inertia of the counterweight caused it to lag behind the piston as it moved downwards, opening the valve.

 

[Bretoal2]

Or at least that is what my rather recent physics education suggests)

Forget your physics lessons, and hop on a merry-go-round. At full speed, you'll soon see if you're sent back to the center or to the periphery.

 

[AspiringHowardHughes]

Bretoal, that is merely inertia at work. The "outwards force" is just the mass of the rider preferring to continue moving in a straight line.
(as Newton put it, "until acted on by an external force")
"Centrifugal" is a handy word to colloquially communicate this phenomenon, but it is not a real force.
If it were, the reaction force against whatever the containing structure is (in your example, the seat of the merry-go-round) would be opposed by centrifugal force, and the rider would have no reason to continue turning about the circle.

I can appreciate the convention used in these circles (pun intended) but to say it is itself a force is incorrect, unless it is a rotating frame of reference that is in discussion.

 

[swampyankee]

Interesting!
Do you have available any reasoning by the manufacturer for adopting that system over a "typical" arrangement?
I am curious how it was rationalized..... or if patent law was responsible.

I contest that there is any such thing as centrifugal force.
It is fictitious, as an object moving in a circle is by definition not in equilibrium and thus there is no force counteracting the centripetal component.
(Or at least that is what my rather recent physics education suggests)
I would say that centripetal force is a more reasonable term here, as it is the obstruction of the material's desire to fly off in a straight line that causes it to collect at the extremities of the rotating assembly.

It's a force that appears because a mass can't follow a geodetic path, just like the "force" of gravity.

Or as on of my university physics professors said, "it looks like a force and acts like a force, so it's a force." Centripetal force is what you see from outside the rotating, that is non-inertial, reference frame.

 

[AspiringHowardHughes]

I think that is only a partially accurate comparison.
Gravity applies whether or not the object in question is being permitted to move freely.
Centrifugal "force" is only ever perceived when the motion of the object is opposed.
That is why it is reasonable in a rotating frame of reference, because the outward motion of the free object seems to be otherwise without explanation.
Centripetal force, on the other hand, is just the name for the opposing action.

 

[Engineman]

Many did have a strange valve system. Only one valve in the head for the exhaust. There was a valve in the top of the piston that let the mixture in from the crankcase.
These had disappeared by the end of the war. The piston located valve was weak point. warping or failing to seal properly. These engines carbon fouled very quickly and taking the cylinder off to de-carbon the pistons and heads every 20 hours (or less) was common.
The later engines used 2 (or 3?) valves in the heads.

There was no mechanical operation of the piston valve. There was a spring and the pressure difference. On the intake stroke the low pressure in the cylinder overcame the spring and the cylinder sucked the mixture in. As long as there was positive pressure in the cylinder the piston stayed closed (in theory).

Edit. These engines did use a gas/oil mixture for lubrication. perhaps that is where the confusion comes from? The gas/oil mixture came in through the hollow fixed crankshaft and was distributed by splash/vapor to rods, pistons and cylinder walls. Oil was the famous castor oil and with the lack of any exhaust pipes to help direct the exhaust flow (only cowl sheet metal) WW I pilots had sever gastric troubles from overdosing on castor oil. These engines generally ran very rich with a high amount of castor oil in the mix.

Yes, most of these Rotary engines are Exquisitely made. The Castor oil has the advantage of tremendous film strength and lubricity. However, it also has awfully low resistance
to degradation under heat, with terrible gumming and oxidation which sees rapid build-up of deposits in most parts of the engines. The experience of working on a Castor lubricated
engine will instil a profound understanding of the importance of thermal stability in lubricants! It also amazes you how a simple vegetable compound can be so good at lubricating metals!

Eng

 

[Bretoal2]

As the piston reversed its course at TDC, the inertia of the counterweight caused it to lag behind the piston as it moved downwards, opening the valve

Obviously you did not understood how is working a rotary engine.

In such an engine, piston has ONLY circular movement and no "stop and go rearward" at TDC/BDC.

In the first Gnome models, the (strange) piston-centered valve was dynamically balanced. It opened only on depression on aspiration stroke.

And you also seem unaware of how baroque the Monosoupape distribution diagram was, and how completely outside the classical paths of the Otto cycle !

 

[AspiringHowardHughes]

That is simply not true.
While rotaries are generally balanced, the change in angle of the piston relative to the crank as it slides in the cylinder provides an element of vertical motion.
If the movement of the piston was truly ONLY circular, as you claim, there would be no need for a wrist pin or any of the crankshaft arrangements that drove me to begin this very discussion!

I suggest you watch the animation in this video, perhaps slowed down.

View: https://www.youtube.com/watch?v=THB2nYRThq0
The angle of the "wrist" of the piston is closest to 180 degrees only when it is at the top or bottom of the rotary.
Otherwise, there is a small deflection to one side.
Not much, so the mass of the weight had to be carefully chosen, but enough to make the design operate on more than suction alone.

Not sure what you mean about the baroque nature of the distribution diagram - but I assure you I know how a rotary operates.

 

[don4331]

A.H.H here.
After a brief intermission as I get myself back home from school and resituated in my life, I return with another rather complex investigation.

As someone who likes when things "just work out right", it was quite annoying for me to realize that the non-master pistons of engines using master-and-slave rods are a little "off" in almost every conceivable way. From the piston speed (and all the associated derivatives), to the phases in which it moves, to the compression ratio....

I am aware that the largest elements of these issues were at least partially addressed. For example, the lengths of the slave rods, and their position of attachment to the shared "big end", were intentionally varied. I presume this was to preserve appropriate TDC timing without quite as big a change in stroke length, as would occur if all rods were spaced equally about the crank.

Was this level of micromanagement (different rods, spacings, possibly also spark timings.....) done outside the great increase in engine quality and power that was the second world war?
Was there ever any attempt made to tackle the nuisance of balancing this minor element of the typical ww2 radial? Obviously the crank counterweight is 99% of the battle, but every little bit helps!
I will provide an interesting alternative rod design below that saw great service in world war one, but I have not been able to determine if it was intended to have any benefits over the conventional system depicted above. It is very possible that any simplification of engine dynamics was just a happy accident.

Some of this doesn't "feel" right.

You are on the correct track with the position of the attachment of the slave rods on the master being intentionally varied. If they were equally spaced you get a situation like this:

Where when the crankshaft has lined up for correct position for TDC (72* for our 5 cylinder radial) the slave rod is not yet lined up = piston is not at TDC.

So the designer intentionally varies the link position on the master rod to have everything line up

And you want the piston at TDC when the crankshaft is at 72* as there are a lot of other components (valve timing, spark timing) that are much easier to design if everything is consistent. E.g. if using a cam ring for valve, it is either running 1/(N+1) (when turning same direction as crankshaft) or 1/(N-1) (when turning opposite direction of crankshaft) where N = number of cylinders, with appropriate numbers of lobes (N+1)/2 or (N-1)/2 respectively. Cam ring turning opposite crankshaft is the more common (few lobes to machine = less expensive)

Note: If you position the lifters correctly, you only need one set of lobes to provide both intake and exhaust. It means you have the same valve duration for both, but it made for simpler construction.

Back to simpler construction - for most of the mass produced radials, all the slave rods are identical. Identical being big thing when you are making 1,000s. And it means you don't need to worry about someone on the floor/out in field assembling the engine wrong.

Now, yes, there are some differences in piston speed/centroid of mass to balance, but that is why the big radials use "hockey pucks" (Salomon dampers) on the crankshaft counter weight.

Simply and add lightness.​

Some of the other solutions might have had advantages, but their complexity wasn't work the cost.​

Correction of François Salomon's name.

 

[AspiringHowardHughes]

Thank you for chipping in, Don.
I have a few follow-up questions.

Is making all the rods the same length while varying their spacing about the master rod not "having your cake and eating it too"? I have not geometrically examined this (yet) but my instinct would be to say that if an engine of that sort were configured to have the same TDC height in all cylinders, the BDC would show more variance as a result. I imagine this would be pretty easy to verify using whatever software you have used to produce those images is.

Secondly, I imagine that different effective profiles (and durations) could be extracted from the same cam track if the design of the roller follower and associated linkages can be done artfully enough. Different roller sizes, non-circular rollers, different rocker ratios, etc. Much like life itself, engineering always finds a way. This is a challenge I may take on myself, actually.

Lastly, do you have a diagram of these "hockey pucks"? I am curious if any attempt is being made to replicate the unique movement of each master-and-slave component to balance it, or if it is just being simply damped (as the name would suggest).

I much appreciate the Colin Chapman quote. My own profile "quote" on this forum was almost "simplicity is the ultimate sophistication" but I felt it might be hypocritical of me given my design tendencies....

 

[don4331]

Thank you for chipping in, Don.
I have a few follow-up questions.

Is making all the rods the same length while varying their spacing about the master rod not "having your cake and eating it too"? I have not geometrically examined this (yet) but my instinct would be to say that if an engine of that sort were configured to have the same TDC height in all cylinders, the BDC would show more variance as a result. I imagine this would be pretty easy to verify using whatever software you have used to produce those images is.

Secondly, I imagine that different effective profiles (and durations) could be extracted from the same cam track if the design of the roller follower and associated linkages can be done artfully enough. Different roller sizes, non-circular rollers, different rocker ratios, etc. Much like life itself, engineering always finds a way. This is a challenge I may take on myself, actually.

Lastly, do you have a diagram of these "hockey pucks"? I am curious if any attempt is being made to replicate the unique movement of each master-and-slave component to balance it, or if it is just being simply damped (as the name would suggest).

I much appreciate the Colin Chapman quote. My own profile "quote" on this forum was almost "simplicity is the ultimate sophistication" but I felt it might be hypocritical of me given my design tendencies....

AHH:

TDC and BDC for all cylinders is identical - verified in the software (it's SolidWorks), so the stroke length is identical for all.

But only the piston on the master rod has sinusoidal motion. The rest have uneven motion...​

​

Changing roller sizes, etc gets away from the KISS principle - you don't want Rube Goldberg. Remember your engine is going to be overhauled in a cold, muddy field.

One of the issues with the German Navy's fire control - there were only a couple people in the factory who could set it up. Not good when it gets knocked our of alignment in South Atlantic.​

You should review this: https://www.vibrationdata.com/tutorials/tv.pdf

 

[Engineman]

Hi,
The motion of a simple piston on a conrod and a crank is not perfectly sinusoidal due to the angularity of the rod/crank. In fact, the piston spends longer in the lower half of its travel than it does in the upper half of its travel in a conventional crank/rod arrangement. So, even the Master rod has this divergence.

Eng

 

[Bretoal2]

Some of this doesn't "feel" right.

You are on the correct track with the position of the attachment of the slave rods on the master being intentionally varied. If they were equally spaced you get a situation like this:
View attachment 883218
Where when the crankshaft has lined up for correct position for TDC (72* for our 5 cylinder radial) the slave rod is not yet lined up = piston is not at TDC.

So the designer intentionally varies the link position on the master rod to have everything line up
View attachment 883223
And you want the piston at TDC when the crankshaft is at 72* as there are a lot of other components (valve timing, spark timing) that are much easier to design if everything is consistent. E.g. if using a cam ring for valve, it is either running 1/(N+1) (when turning same direction as crankshaft) or 1/(N-1) (when turning opposite direction of crankshaft) where N = number of cylinders, with appropriate numbers of lobes (N+1)/2 or (N-1)/2 respectively. Cam ring turning opposite crankshaft is the more common (few lobes to machine = less expensive)

Note: If you position the lifters correctly, you only need one set of lobes to provide both intake and exhaust. It means you have the same valve duration for both, but it made for simpler construction.

Back to simpler construction - for most of the mass produced radials, all the slave rods are identical. Identical being big thing when you are making 1,000s. And it means you don't need to worry about someone on the floor/out in field assembling the engine wrong.

Now, yes, there are some differences in piston speed/centroid of mass to balance, but that is why the big radials use "hockey pucks" (Samson dampers) on the crankshaft counter weight.

Simply and add lightness.​

Some of the other solutions might have had advantages, but their complexity wasn't work the cost.​

The designer of the dynamic damping system used by Bristol on the Hercules and Centaurus was not Salmson, but Salomon (François, a French engineer).

For reasons that impartial historians will one day have to elucidate, the Salomon processes, covered by a significant series of patents

GB 1933-11-23
CH 1934-07-15
FR 1934-07-26
US 1936-02-04
US 1936-02-04
FR 1936-09-11
US 1937-12-28
GB 1938-09-30
US 1939-11-28
GB 1940-07-15
CA 1941-03-18

were "rediscovered" by Roland Chilton, who subsequently patented them at later dates.

Let us pay tribute to Bristol for always knowing to whom they owed their "puck" (in fact, spherical) dynamic damping system.

 

[don4331]

The designer of the dynamic damping system used by Bristol on the Hercules and Centaurus was not Salmson, but Salomon (François, a French engineer).

For reasons that impartial historians will one day have to elucidate, the Salomon processes, covered by a significant series of patents

GB 1933-11-23
CH 1934-07-15
FR 1934-07-26
US 1936-02-04
US 1936-02-04
FR 1936-09-11
US 1937-12-28
GB 1938-09-30
US 1939-11-28
GB 1940-07-15
CA 1941-03-18

were "rediscovered" by Roland Chilton, who subsequently patented them at later dates.

Let us pay tribute to Bristol for always knowing to whom they owed their "puck" (in fact, spherical) dynamic damping system.

I knew I spelt it wrong as soon as I reread my post.

 

[Bretoal2]

I knew I spelt it wrong as soon as I reread my post.

Rolls-Royce used a similar system on the two Eagle XXII crankshafts. Flight said they were "Saloman vibration detuning masses". Yes, SalomAn...