Unusual twin cylinder engine design

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.
Joined
Oct 7, 2023
Messages
108
Reaction score
77
Location
Brighouse. Yorkshire, UK
I was watching a documentary about the Ariel Square Four the other day, and it occurred to me that it should be possible to build a two cylinder four stroke engine, using two crankshafts where the pistons rise and fall together but are not firing together. This could use a single cam lobe for both cylinders if an overhead camshaft were used and bellcrank valve operating levers.
Just an idle thought, but I've never heard of this being done. The Triumph Dolomite Sprint engine used one lobe for inlet and exhaust but this idea I've never come across.
Technical reasons?
 
I did the Ariel Four cyl. concept a few years ago and it came out well! Incorporating its design in this year's model as a twin with a conventional cam and valve arrangement and is "somewhat" as you described......a twin crankshaft engine!

 
The Parilla motorcycle was a hot number in the early sixties and ran with a single cam lobe running both valves. It featured very short push rods with the cam high up along the cylinder.
Thank you, I'll look into those. That's a fabulous machine that you made Longboy, all right angle drives, I like it.
 
I was watching a documentary about the Ariel Square Four the other day, and it occurred to me that it should be possible to build a two cylinder four stroke engine, using two crankshafts where the pistons rise and fall together but are not firing together. This could use a single cam lobe for both cylinders if an overhead camshaft were used and bellcrank valve operating levers.
Just an idle thought, but I've never heard of this being done. The Triumph Dolomite Sprint engine used one lobe for inlet and exhaust but this idea I've never come across.
Technical reasons?
Yes, it is possible but the vibration impulses would be very difficult to manage. Not sure how you compensate for that . The vibration dictates the firing order, whether its in line, opposed and even the shape of the V. Might get away with it at low rpms. But you probably will need a lot of mass to handle the forces. Calculating those pulses and the magnitude are a real pain.
 
The twin crank engine described could be very well balanced.
The pistons rise and fall together, but the cranks rotate in opposite directions, so the pistons and rods could be fully balanced by counterweights, while the secondary vibrations of the counterweights would balance one another.
The single cam would have virtually no effect on vibration.
Firing order is not related to linear vibration, which is entirely based on motion of the masses.
Firing order influences torsional vibration, but with the alternate firing as described, the torque pulses are as evenly distributed as possible with two cylinders.
 
Last edited:
The twin crank engine described could be very well balanced.
The pistons rise and fall together, but the cranks rotate in opposite directions, so the pistons and rods could be fully balanced by counterweights, while the secondary vibrations of the counterweights would balance one another.
The single cam would have virtually no effect on vibration.
Firing order is not related to linear vibration, which is entirely based on motion of the masses.
Firing order influences torsional vibration, but with the alternate firing as described, the torque pulses are as evenly distributed as possible with two cylinders.
Every vibration problem has an excitation force. With IC engines it is the firing of the cylinders that create the pulse. How close they are and where they are located determines the magnitude of the forces. Timing is is set by the design of the crank and once the crank is designed its difficult to change the firing order. All the rotating shafts will have their own critical frequencies based on the shaft length diameter and mass. It not a simple design issue of the motion of the masses or the placement of counterweights. Even changing the angle of the V in six and eight cylinders can effect vibration to the point where the operation of the engine is unacceptable.

With two cylinders as described and with each cylinder firing out of phase you basically have two separate engines driving a common shaft. Hard to guess whether it will add or cancel each other. But as you dont see that design commercially I expect it has significant drawbacks.
 
Fair comments here. Engine balance is complex, but not insurmoutable. I read about a guy who used a 75 degree offset between 2 cylinders on a Triumph twin. Used his own made crank - balanced appropriately. Very smooth results. But Triumph would never have adopted this, as making cams, cranks, ignition for that firing was a much increased cost - akin to the costs of a Vee-twin with 75 degree vee. But Engineers manage all the parameters, including costs, and durability, and design according to the required "market" - and commercial needs. Hobby engineers also consider many of these things, but usually make different decisions to commercial designers and manufacturers.
Surely that is the pleasure of the hobby?
K2
 
Certainly the combustion in the cylinders is the source of the excitation, but I do not agree that the firing order is the primary source of gross engine vibration, which is what most of us will first think of when discussuing engine vibration.

I have owned a number of Triumph twin cylinder motorcycles, all with 360 degree firing intervals (alternate firing). The 1953 T100 is considerably smoother (less vibration) than all others. With the same firing order, the only differences are in the balance factor and the masses of pistons and rods.

Multi cylinder radial engines run with very little vibration, as the collective mass of the rods and pistons moves in something close to a circle and the counterweights can be matched to the collective moment of inertia of the rods and pistons, creating very little secondary vibration.
This is in spite of the fact that the firing order (e.g. 1,3,5,7,2,4,6 in a 7 cylinder engine) would suggest a strong orbital vibration at 1/2 the crankshaft frequency (rpm) if the firing order were to be significant.

Vibration of elements within the engine is a different matter. Certainly, firing order (or rather the distribution of loads along the length of the crankshaft) will influence the magnitude of vibrations within the crankshaft and therefore vibrations within the structure supporting the main bearings.

The suggested example uses two crankshafts, each with a single rod and piston. As such, it is relatively easy to calculate the crankshaft natural frequencies and ensure that they are significantly greater than the firing frequency for each crankshaft.

Similarly, the magnitude of camshaft vibration will be influenced by the sequence of loads (valve opening events) along its length. While it may be tempting to correlate that sequence to firing order, it is dependent on the mechanical layout of the valve actuating mechanism.

In the suggested example with a single cam lobe operating all valve events, all the loads occur at a single location so it should be relatively easy to calculate the camshaft whip frequency and ensure that it is significantly greater than the valve event frequency.

The topic of vibrations of individual components is a very interesting one. I now have CAD software which includes some limited analysis and can calculate natural frequencies of complex parts. Before that, I used excel spreadsheets, calculating the stiffness and mass of each element (e.g. crank journals etc.) to eventually calculate natural frequencies.
While the CAD analysis is much more convenient, there is a certain satisfaction in spending hours on a spreadsheet and then when the part is made, measuring the actual frequency with an accelerometer to correlate with the calculation.
 
Peter, you are quite correct that firing is not the primary source of vibrations in most engines. However it is of major concern on all engines for torsional vibration into the transmission. This transmission load variation is much smoother with multiple cylinders as the firing pulses overlap.... forming a higher frequency source of vibration. Also torsional effects of multi cylinders on long crankshafts can create major wind-up torsional effects. Thus on (say) a V6 or V8 (especially long truck engine crankshafts) the firing may be varied to minimise "wind-up" - or so I was taught in the 1960s.... but I never did any sums on those!
Vee-twins, etc. tend to be designed to naturally set the primary vibration of one cylinder (piston etc. going up n down) against the secondary vibration of the second cylinder (big-end n con-rod flapping sideways) - I think? But often the packaging of cylinders (especially for cooling, but not always) is a major reason for the expenditure on higher cost of Vee-twins.
180 degree cranks on motorcycles are relatively rare on 4 strokes, as the rocking couple can affect bike dynamics, although primary vibration is much better balanced. Cost of twin carbs and 2 ignition systems also affected decisions an out when to use the 180 deg crank instead of a 360 deg crank. - a good example is Honda's C72/77 for single carb low cost versus the CB72/77 that needed twin carbs for the 180 crank and thus were the higher cost and performance bikes - including frames, etc.
On torsional vibration, there are exceptions to smoothness... such as the large displacement high torque, low revving applications such as Harley Davison use. Riders "like" the feeling every pulse of the engine.... I wonder if that is a part of their decision to frequently ride at speeds slower than most on Motorways? At least in my view.
An interesting discussion.
K2
 
"I wonder if that is a part of their decision to frequently ride at speeds slower than most on Motorways? At least in my view"

Steamchick- we used to say the reason for their slow speeds was they didn't have to walk back as far to pick the bits up that fell off.............
 
Back
Top