Letters from Bill Denton to the Yamaha 650 list
Dear YAM650 List,
Since several of you have asked me to elucidate on my "Quartered", or 270 degree crankshaft project, here it is. By the way, I plan to eventually add this engine to a fully "Mintonized" XS-650. A well deserved "Thank you" and warm regards go out to Mr. Joe Minton for writing his (now somewhat infamous) article on improving the XS650, which has become a standard reference article of sorts for the International 650 Society.
It all started when I read an article in Cycle World entitled "TRX Tech,,, Motorcycle or Myth Maker?" (June 1995) about the engine in the Yamaha TRX 850. This engine, an 850cc upright twin, is built with 270 degree crankshaft timing (see following discussion about 90 vs 270 degree timing). This idea made way too much sense to ignore, offering all the advantages of a V-twin; a wider, flatter torque curve, and less (destructive) vibration by minimizing both secondary imbalances and rocking couple forces, while maintaining the small,compact engine profile of an upright twin. In addition, the sound of the engine note would be too unique to ignore, especially coming from an XS! I thought, would it work in an engine without balance shafts? I had no idea.
So, I set out on the information trail to get some opinions on whether it would work or not. I started by writing a letter to everyone listed in the Int'l 650 Society Newsletter as a 650 rebuilder, thinking that their opinions might have just a little more weight. The responses I received ranged from "it will vibrate apart, don't do it", to "It will have (much) less vibration than the stock engine, and I'd be glad to help you build one." I also became aware of two gentlemen in particular who had already done 360 > 270 deg crankshaft conversions; one on a 500cc Triumph (Dick Cookson in England), and the other an XS-650 (Dave Rayner in Australia). I wrote letters to them as well, telling them of my interest and asking for information. Both responded by mail (send me a message if your interest runs deep enough to read these letters, and I'll snail you photocopies of them).
I also discovered along the way that the original idea for a 270 deg crank in an upright twin came from an Australian fellow named Phil Irving (co-creator of the pre-war Vincent engine design) in his musings in a series of articles he wrote in the 1940's. Apparently, he initially tried to convince Triumph of the virtues of this crank arrangement, but they would not have anything to do with it, choosing to stick with the 360 degree design. Apparently, there was a bias or prejudice against uneven firing engine designs that stuck with BSA, Triumph and Norton until their respective bitter ends.
As you know, in the XS-650, both pistons move up and down together (aka 360 deg crankshaft), with plugs firing left and right on alternate strokes. In this arrangement, both piston/conrod combos achieve maximum velocity TOGETHER twice per revolution (once on the way up and once on the way down). In addition, they both come to a complete stop TOGETHER twice per revolution (once at TDC and once at BDC). As you can imagine, as RPM increases, so does the vibration coming from this Primary Force Imbalance (PFI). This phenomenon is made worse by the laws of Physics, which dictate that doubling the RPM quadruples the forces (and thus the associated vibrations) involved. Flywheel weighting added opposite the throw of the crankpin minimizes PFI, but a compromise has to be achieved between cancelling out PFI and exacerbating a Secondary Force Imbalance (SFI). This is the centrifugal force of the weighted portion of the flywheel trying to move the entire engine fore and aft as it spins.
Now then, the basic idea with a "quartered" crank engine (and design advantages associated with it) is to never have both pistons at maximum or minimum velocity at the SAME time (as you know, this is exactly what DOES happen in a 360 degree upright twin). More specifically, the idea is to have one piston moving at MAX velocity while the other one is at MIN (zero) velocity, in order that inertia about the crankshaft is preserved. In other words, the first swiftly moving piston/conrod combination helps "pull" the second piston/conrod combo through it's deceleration & change in direction, a point at which it has little or no inertia. In effect, the quickly moving half acts as a inertial flywheel for the other, and visa versa several times per revolution, resulting in relatively constant inertia (stored or potential energy) in the rotating parts in the big end of the engine. See the URL http://www.interlog.com/~lcl/tdm/tdmpower.html for more info on crankshaft inertial torque and it's effect on perceived torque feel in a 270 degree crankshaft engine design. (or copy of same graph on local site)
This "inertial torque smoothing" effect is achieved by separating the crankpin throws. Maximum velocity occurs at 74.1 degrees before & after TDC in an XS-650 with a 74mm stroke and a connecting rod length of 130mm (#447 rods). This is the point at which the crankpin throw and the conrod are at right angles.
However, at a 90 (or 270) degree angle, the piston/conrod assembly is still pretty close to maximum velocity, so a high degree of inertial torque is still achieved. But (and this is a big "but") at 90 degrees, the PFI is less than at 74.1 deg, and the Secondary Forces, which have a frequency twice crankshaft speed will have a phase difference of 180 deg., thus cancelling each other out entirely (credit to Brian Woolley; The Classic Motorcycle, 2/92). According to Woolley, this should result in a 43.5% improvement in the balance of forces within a 90 degree engine, when compared to a conventional 360 degree upright twin.
Because of a combination of better balance and conservation of momentum (inertia), the "flywheel" portion of the crankshaft may now be lightened, further improving balance by reducing rotational weight while increasing throttle response without sacrificing drivability (a common problem when "race" engines with lightened flywheels are driven on the street).
So there you have it. An engine with more torque, more perceived "pull" at lower revs, quicker throttle response, less vibration, and with an "Italian" exhaust note. What more could a man want out of his mid-life crisis project!
Bill in Yardley, PA
90 vs 270 degree crankshaft timing
I have been trying to figure this out on paper, and there are
two distinctly different scenarios available when crankpins are
separated by 90 degrees on the crankshaft. Participants in
this thread may have been assuming that 90 and 270 degree cranks
are just two ways of saying the same thing, but really, they are
two entirely different setups. The difference is really not
in the separation of the crankpins, so much as the timing of the
valve train, which results in power strokes that are separated by
either 270/450 degrees, or 90/630 degrees. Here's why:
The true 270 degree crank defines power strokes that are separated from each other by 270, then 450 degrees. Imagine the #1 piston @ TDC at the beginning of the power stroke, while the #2 piston is 270 degrees behind it, halfway through it's own intake stroke. Follow this thing around, and 270 degrees later, the #2 piston is @ TDC at the beginning of it's power stroke. Following the progression of the #1 piston, we find it is now halfway through it's exhaust stroke, a full 450 degrees away from the beginning of it's next power stroke.
Now, take the same crank, and change the valve timing setup so that when the #1 piston is @ TDC at the beginning of the power stroke, the #2 piston is not far behind, halfway through it's compression stroke. Only 90 degrees later, the #2 piston is experiencing the beginning of it's own power stroke, while the #1 piston is only halfway through it's power stroke. Overlapping power strokes, the beginnings of which are only separated by 90 degrees this time. The next one comes 630 later, as you previously suggested. Same crank, really, only different orientation of SSBB from #1 cylinder to #2 cylinder.
As far as your balance question, you're right again, the primary balance in a 180 degree crank engine is perfect, but at the cost of a violently malevolent rocking couple, torquing the engine first to the left, then to the right, as separate pistons reach TDC and BDC simultaneously. Whereas, in the 90 or 270 degree crank configuration, you can maintain higher inertial momentum (flywheel effect) about the crankshaft by having one piston always at or near max velocity while the other one comes to a complete stop and then changes direction. They are never both dead in the water at the same time. Since secondary imbalance is at a minimum, additional counterweighting on the crankshaft can be reduced, allowing a better compromise of Primary:Secondary imbalance percentages, while reducing reciprocating weight. Also, the flywheel effect allows the use of smaller actual flywheels, further reducing reciprocating weight. Cool.
Having said all this, I wonder if Kevin Cameron, Technical Editor for Cycle World, really meant 270, or 90 degrees when he wrote the article about the Yamaha TRX-850 in the June 1995 edition of Cycle World? The more I think about it, the more I believe that it's really a 90 degree separation of power pulses in the TRX engine design, especially if they were trying to emulate the successful big bang power pulses of the race-bred Ducatis. I wonder if someone in the technical development department at Yamaha could shed some light on this? Anyone have any connections?
"If you build it, they will come (sic)"
Offset Crank using factory splines
There are actually 13 splines on the crankshaft (damn it
Yamaha! You could have made it so easy but for one extra
spline!). So, to move it 3 splines is to displace one of
the crankthrows by 360/13*3= 83.077 degrees. It then
follows (got that phrase from a mathbook about 25 years ago and
it just stuck) that 360-83.077 = 276.923 degrees of separation
between the throws when you're all done. Not exactly 270,
but close enough to warrant saving several hundred dollars (or
more) of specialized machining required to achieve exactly 270
degrees like Dave Rayner has done in Oz.
Also, IMO anything between 270 and 285.9 is totally acceptable, because of the following salient points. 270 is a good design point, because it minimizes reciprocating vs rotating force imbalances (see Blaine's webpage torque.html for a more complete explanation as to why). However, 285.9 is also a good design point, because it maximizes conservation of inertia about the crankshaft (in an XS650 engine equipped with 447 rods, the stroke is 74mm and the rod length is 130mm. Therefore, maximum piston/rod mass velocity is achieved at 74.1 degrees B&ATDC. This is the point where the rod and the crankthrow are at right angles to each other, so at that moment in time, the piston rod combo and the crankshaft are running at identical (and maximum) linear velocity). Each design point maximizes it's own advantages in the modified engine. Fortunately, both points are also relatively close to one another mathematically. So, bracketing the eventual design with these two end points would only minimize either effect very slightly, while retaining the majority of both effective advantages (sounds good, eh?).
Like in most engines (or any equipment design which ever gets reduced to practice) my proposed method of design implementation is a compromise, but one which I am willing to accept. So there.
Bill in slushy Yardley, PA
Back to Motorcycles