In Part 1 of this series we looked at how best ride quality occurs if accelerations transmitted to the occupants of the vehicle are minimised. These accelerations occur vertically (Z direction) and in pitch (Y) and roll (X). Pitch and roll accelerations are the most difficult to control.
If you’re developing a suspension system, especially one on an ultra lightweight, small vehicle, providing good ride comfort requires using all the tools at your disposal. And now, with the latest technology, directly measuring these accelerations is easy and cheap. In turn that makes developing a suspension system to provide a good ride far easier than ever before.
Making the Measurements
Accelerations are measured with an instrument called an accelerometer. Accelerometers used to be fearsomely expensive, and logging and analysing their outputs was difficult. In fact, up until only about a year ago, to produce light-weight and compact equipment that could simultaneously measure X, Y and Z accelerations, and then analyse the collected data, was pretty well beyond anyone not an electronics engineer.
But, with the advent of the Apple iPhone (and iTouch), and the development of low cost software, that’s all dramatically changed.
The software used here is produced by Diffraction Limited Design http://www.dld-llc.com/Diffraction_Limited_Design_LLC/Vibration.html and is called Vibration. At the time of writing it costs just US$6.
The software takes advantage of the fact that the iPhone has an inbuilt 3-axis accelerometer. It can measure up to plus/minus 2.0g and has a sensitivity of about 0.02g. Those characteristics make it ideal for measuring ride comfort. The software can be set to sample at up to 100Hz (100 times a second) and the data can be displayed as graphs. Furthermore, mathematical analysis of the data can be performed by the software. You can also email screen grabs and a .csv file of the collected data to your PC for later analysis.
In short, if you have (or can borrow) an iPhone or iTouch, purchasing some very cheap software gives you full acceleration data collection in X, Y and Z directions. So how do you use it?
When considering ride comfort in vehicles, it seems to me that the starting point should be the accelerations experienced in walking. As far as I know, no-one ever got motion sick through walking, and the body physiology should be pretty well attuned by now to the accelerations that occur in walking!
Also, given that with the iPhone and Vibration software, it takes just a few moments to see what happens during walking, I decided it was worth doing the measurements. The walk was at a reasonably brisk pace, with bare feet on carpet. As with all measurements made in this article, the accelerations were measured at the head.
This screen grab shows the measured results. The traces are, from top to bottom, X (accelerations in roll), Y (accelerations in pitch) and Z (up/down accelerations). The vertical scale is 0.5g per division and the horizontal scale 1.28 seconds per division (so the sample is a bit over 10 seconds in length).
Looking at the bottom (Z) trace first, you can see that each time a foot comes down, there’s a strong acceleration, typically of about 1g. The acceleration upwards as you take the next step is much lower.
Above that trace is the blue line showing Y (pitch) accelerations. These occur at the same frequency as vertical accelerations but their max magnitude is much lower at about 0.4g.
The top trace (red) shows X accelerations (those is roll) and these are lower again, peaking at about 0.2g.
Also shown are the RMS figures for each direction. (RMS means ‘root mean square’ and is a little like an average that takes into account both negative and positive values.) The RMS figure for walking Z accelerations is 0.215, for Y accelerations is 0.121, and for X accelerations is 0.117.
So in summary, walking results in quite high vertical accelerations but low accelerations in pitch and roll.
The other thing to really take notice of is that the vertical accelerations are so regular. That’s to be expected when you’re walking at the same pace – but it is quite different to what happens in a vehicle. So at what frequency (how many times per second) are the vertical accelerations felt? By using a mathematical analysis available in the software, we can find a dominant spike in vertical accelerations at 2.25Hz (that is, 2.25 times per second).
Small Wheeled Bike
Over bumpy ground, one of the worse riding of vehicles is a small-wheeled bicycle. Pictured here is a Brompton folding bike, a machine that I own - and love because of its incredible versatility. But over rough ground, even with an added sprung seat and standard small-travel rear suspension, it’s not a machine that has good ride comfort.
Here is the measured ride quality of the Brompton, ridden slowly over a bumpy grass paddock. Note that at 1g/division the vertical scale is different to the scale used above when measuring walking accelerations. The greatest accelerations now occur in pitch (the blue Y line) – in fact, the maximum pitch acceleration exceeds 2g!
Comparing just the RMS values with walking shows how poorly the small wheel bike rides:
As described in Part 1 of this series, pitch accelerations are the type most disliked by humans and also, in small vehicles, amongst the hardest to control. With pitch accelerations no less than 186 per cent greater than walking, the comment that the bike has poor ride quality is certainly backed-up by the measurements!
And is there a frequency where most accelerations occur? Analysing the data shows that on this surface, at this speed, there is a peak in vertical accelerations at 3.2Hz – significantly higher than when walking.
So what about a car? The data was recorded in a Skoda Roomster being driven along a country road at 90 km/h. The maximum accelerations occur in the Z (up/down) direction, peaking at about 0.5g. Accelerations in the X and Y directions are low.
In terms of frequencies, the Roomster exhibits two clear peaks in vertical accelerations – one at 1.7Hz and the other at 2.7Hz.
An HPV Suspension
The measurement of ride quality was undertaken primarily to facilitate the development of my recumbent pedal trike. The trike uses a purpose-designed pneumatic / hydraulic suspension system. It is shown here in an earlier iteration, being tested with a heavy load (note that the ride comfort testing done below was carried out without any load in addition to the rider).
To make direct comparisons, testing was conducted over the same stretch of bumpy grass with three different pedal machines :
As can be seen, Chalky has a similar X accelerations to the bikes (all pedal machines I tested had similarities in X behaviour – perhaps it’s just the inadvertent head movement with pedalling?), but easily superior performance in Y and Z directions.
This graph shows the results more clearly (click on it to enlarge).
But what about the frequencies at which accelerations mostly occurred? Concentrating on Z (vertical) accelerations, the results over the same piece of ground are:
Finally, the image below shows the actual acceleration traces of the three different pedal machines.
During development of a recumbent pedal trike suspension, the measurement of accelerations was used to indicate if changes to the design were resulting in improvements, no change or reduced suspension effectiveness. This was invaluable, especially when attempting to gain best compromises between conflicting requirements – eg reducing body roll by the use of a stiffer anti-roll bar versus the resulting increase in the wheel rates in one-wheel bumps. Spring rates, suspension interconnection and damping rates were all also developed using the data.
The use of an iPhone (or iTouch) and cheap software allows you to quantify ride accelerations, and the frequencies at which they mostly occur. Comparison can then be easily made of different vehicles, and of different suspension set-ups in the same vehicles. That’s particularly useful when the vehicle is lightweight, small and/or unusual in design.