There is a motorcycle manufacturer, you know the one, whose engine and exhaust notes are as jealously guarded as money in the bank. Recently, I had an interesting conversation with an AutoGear M22 customer who maintained that the whine produced by GM’s ‘Rock Crusher’ four-speed was equally valuable. For him it was, in fact, the justification for acquiring one, and furthermore, because our ‘M22’ gearing isn’t as loud as the original, he also maintained that we clearly could not be manufacturing to original equipment specifications. As I’ve heard variations on this argument before I thought this might be a good time to talk about where that whine comes from.
Gears, shafts, bearings, the housings that support them along with the oil film that lubricates them, cannot be manufactured without distortion or tolerance and, therefore, in passing through each tooth mesh gears must, however slightly, speed up or slow down. This is a vibration we hear as sound. In a world where gears were perfectly made and mounted they would sing like a tuning fork, but the sounds made by active gearing in our real world are non-reinforcing products of these accumulated tooth-to-tooth spacing variables.
Each sound has a ‘frequency’. For example, the sound made by a gear with 40 teeth turning 1,000 revolutions per minute has a frequency of 667 hertz (40×1,000/60 cycles/second). It follows that, given a ratio chart and a sufficiently accurate frequency analyzer, from the input revolutions per minute and the frequency emitted in each gear we can usually identify the gearset installed. ‘Usually’, because two different gearsets having identical tooth counts, like the Muncie M21 and M22, have identical frequency profiles. Therefore, it also follows that, wherever the signature ‘Rock Crusher’ sound comes from, it can’t be traced back to the frequency profile.
Each sound is also loud or soft; it has ‘amplitude’. As our customer was specifically disappointed by how quiet our M22 gears are, could this characteristic be our diagnostic? At the dawn of the automotive age active gears were loud, really loud. So much so that by 1900 New Process Rawhide (later New Process Gear, the designers and manufacturers of Chrysler’s A833 four-speed) was quieting them by applying the product of their leather hardening patent to their faces. Obviously, ‘rawhide pinions’ were a short-term solution. More powerful and faster revving engines, and the inevitable quest for more capable, more compact, less heavy, less expensive and less noisy gearboxes to mount behind them, was urging gear making forward. Advances in hobbing and shaping quickly banished leather facings to the machine shop, extinguishing them entirely in the 1920’s. Gear grinding, the current technique of choice for today’s high-quality gearing, was already available, though expensive and little used, by 1930. Just as accessible, affordable gears of higher quality transmitted more power with less noise. In 1965, half-a-century later, further improvements meant that without changing the prints, today’s gears transmit even more power with even less noise. Which explains why AutoGear M22 gears are quieter than GM’s originals but, as M21 and M22 gears have the same frequency profile and are cut to half-century old specifications on identical equipment, gets us no closer to why M22 gears make that whine and M21 gears don’t.
The answer is ‘helical overlap’. Gears whose teeth are cut parallel to their intended axis of rotation, commonly called ‘straight cut’ gears but technically ‘spur gears’, take the full load transmitted on each single tooth in turn. Gears whose teeth are cut at an angle to their intended axis of rotation are called ‘helical gears’ and depending on the ‘helix angle’ and ‘face width’, spread the load they transmit over two or more teeth. It follows that helical gears, all else being equal, are stronger than spur gears of identical face widths, and that helical gears with larger helix angles are stronger, all else being equal again, than helical gears with smaller helix angles.
Those wondering about why racing gearboxes often use spur gears, or why more helical overlap isn’t always better, hold those thoughts for a later piece. The important thing today is that M21 gearing is cut to larger helix angles than M22 gearing; M22 gears have less helical overlap. Shouldn’t that make M22 gears less strong? Only if all else were equal, and it’s not. M22 gears have slightly larger, stronger, teeth (the steel and heat treatment, contrary to popular opinion, are the same). Gears with more helical overlap are also quieter. Changes in spacing from tooth-to-tooth are averaged over more teeth and the energy of vibration, spread over several teeth, is less for each. So, all else being equal, our M22 gears, having less helical overlap than our M21 gears, are noisier or, as enthusiasts prefer, produce that musical ‘Rock Crusher’ whine.
Regarding our M22 customer, if AutoGear M22 gears are manufactured to the original specifications with modern, more accurate tools and processes and if, therefore, they produce the ‘Rock Crusher’ whine only less loudly and are stronger in the bargain, could we, as he suggested, sacrifice a bit of strength or durability to make our M22 gears just a little louder? He maintained that GM’s never broke teeth and consequently making our gears just a little bit less well wouldn’t hurt. He’s wrong about GM’s M22 gears never breaking as he forgets that today’s engines are far more powerful and that, most importantly for you and us, intentionally compromising the quality of our production would be contrary to AutoGear’s mission. Besides, the gears he wants are already available. They’re coming from China and India and our competitors are selling them.