How do some bicycle power meters work

How do bicycle power meters work?

There are different types of power meters on the market, and each one measures something different to make their estimates. In addition, the way they measure what they measure has an impact on their accuracy. Below I'll discuss what the main models measure, how they measure it, and what impact this has on accuracy.

Power is the speed of work (so you need to know the amount of work and the length of time it takes to do that work), and work is a force exerted over a distance. Hence, each power meter has a different method of measuring these forces and, based on patents, each has chosen to measure them in a different "place".

With the exception of the iBike, most power meters measure the forces somewhere along the drive train: the PowerTap (and the old Look MaxOne) measure from back to front on the rear hub, the older Polar systems along the chain. The Quarq, SRM, Rotor and Power2Max measure on the spider of the front chainring, the new Look / Polar and Garmin Metrigear (previously announced but unpublished) on the pedal spindle, which Brim Brothers (announced but unpublished) measures on the cleat Ergomo on the bottom bracket and the stages on the left crank. The iBike measures in a completely different way, as described below. One consequence of measuring at different points along the drive train is that drive train losses are taken into account to different degrees (or not). For example, a PowerTap typically reads lower than an SRM because one is "upstream" of most powertrain losses while the other is "downstream". This difference is more of a definition problem than a strict "accuracy" problem (in the sense of "Is gross income or net income a" more accurate "measure of income?" what is "more precise").

Most power meters on the market use strain gauges, which are small, thin strips of film whose electrical conductivity and resistance change when they are deformed. Strain gauges are used in many applications (such as bridges) and their properties are well known. Generally, strain gauges are combined in a "rosette" or "Wheatstone bridge" for greater accuracy and precision (more strain gauges usually produce better results) and, when operated properly, these are usually Power Tap, Quarq, and SRM on one a few percent accurate (and just as important with high precision); This was checked both statically (using known weights suspended from the crank) and dynamically (using a large powered rolling drum in a laboratory). The forces are then combined with a measurement of angular velocity or velocity to obtain power. One advantage of strain gauges is that the change in resistance can be measured even when the device is stationary, so the cyclist can measure the accuracy of the strain gauge-based power meters at home by hanging weights of a known mass on the crank. A common problem with the DMS approach, however, is that they can be sensitive to temperature changes and therefore need to be "zeroed" before (and sometimes during) trips. The Achilles heel of the old Look MaxOne was waterproof, not the strain gauges or the measurement method. For example, the original Power2Max model (and the old, obsolete SRM "Amateur" model) used fewer strain gauges than the current PowerTap, Quarq, or SRM models, and reports from users (later approved by the manufacturer) indicated that it did temperature drift was more sensitive during a journey than these others. The Power2Max was revised and updated at the end of 2012. According to reports, the temperature problem has largely been resolved. A claimed feature of the stages is that they are designed for automatic temperature compensation. As of early 2013, this claim is still being rated by users, and it's too early to know if their approach does what it claims.

The old Polar power meter measured the force transmitted down the chain by chain tension and included a chain speed sensor to get the total work. In a chain, a higher force transmitted along the chain results in higher tension, and the tension can be measured by the resonant frequency of the object (e.g. plucking a high tension spoke with your fingernail produces a high frequency sound while the Plucking a loose spoke creates a soft sound). Historically, the proof-of-concept prototype for the Polar chain tension sensor was the pickup of an electric guitar. The chain speed sensor fitted onto one of the derailleur support wheels and was able to count the "pulses" in the magnetic field when the chain rivets passed. Because chain rivets are a known distance apart, the chain speed could easily be calculated. The Polar's accuracy was very good when it worked well. but if it wasn't, it was very naughty indeed. Worse, it was often difficult to tell when it was naughty. The failure of the old Polar power meter was triple: 1) the chain tension sensor had to be near the chain, which was difficult to get to as the chain sometimes had to be in the large or small chainring or in the large or small rear sprocket; 2) the chain speed sensor was sometimes overwhelmed and gave incorrect speed values; and 3) incomplete weather resistance due in part to exposed wires and a poorly sealed "pod".

Based on the Ergomo bottom bracket, the power meter used an optical sensor and a series of "peck holes" to measure the torsion in the bottom bracket. A strange feature of this construction is that only the (torsional) force that flows through the bottom bracket can be measured. Therefore, only the left leg power was measured: to get the total power, the left leg contribution was doubled. Coupled with the difficulty of installing and calibrating the Ergomo (he had to just like that Installed become ), this dependence on the bilateral symmetry between the legs was the death knell for the ergomo. The Stages power meter similarly measures the force from deformation in the left crank and doubles the "left" to give an estimate of total power. Studies with instrumented power pedals show that bilateral asymmetry in power generation between the right and left legs is the norm - worse, the studies show that the asymmetry can change with effort. However, some drivers are willing to accept this inherent inaccuracy and inaccuracy.

Since neither the old Polar nor the Ergomo power meters used strain gauges, the cyclist could not statically check their accuracy and precision on site. They could only be checked dynamically (or against another known calibrated power meter).

It is rumored that the unpublished pedal or pedal clamp power meters from Garmin Metrigear and Brim Brothers use piezoelectric sensors and solid-state accelerometers instead of foil strain gauges. However, by the time they hit the market, all statements about accuracy or precision should be made with grains of salt. An interesting problem with designing a pedal or cleat based power meter is that you need to know the direction of force and the position of the pedal spindle: for example, if you add a downward force at the bottom of the pedal stroke, it is a waste of force as it is not helps move the crank in the right direction; Pressing (but lightly) on the upstroke releases some of the force the other leg is exerting on its downstroke. Keeping track of the various force vectors is therefore the key to reliable accuracy and precision. To some extent, the Stage Power Meter can occasionally experience an associated problem: The Stages uses a solid-state accelerometer in the pedal (similar to the solid-state accelerometers found in smartphones) to determine its position. Early production models of the Stages were plagued by inaccurate measurements of pedal position, so pedal speed was also inaccurate - and this affected the accuracy of the final performance estimates.

The recently released Look / Polar power meter (as of January 2012) uses strain gauges placed along the pedal spindle. Each pedal must be carefully installed so that the pedals know which direction the forces are being applied. A special tool is supplied with pedals for orientation. To simplify the conversion of the measured forces into torque values, only four different crank lengths can be used with the Look / Polar pedal: 170 mm, 172.5 mm, 175 mm and 177.5 mm. Cranks shorter than 170mm are not currently supported. One pedal is the "master" and the other is the "slave"; The slave pedal transmits information to the master, which then bundles data from both pedals and forwards it to the head unit. The Look / Polar pedal currently uses its own transmission protocol, and no other manufacturer has yet signed up to provide compatible head units. Early reports on the new Look pedals confirm that the alignment of the pedals is critical: because a pedal's spindle is small, a small absolute misalignment can be a large relative error in its angular alignment.

The iBike takes a completely different approach: It calculates the performance indirectly. That is, you need a certain amount of force to overcome changes in potential energy (climbing or descending), changes in kinetic energy (accelerating or decelerating) to overcome air resistance (including wind) and to overcome rolling resistance, if you do this you will know the ground speed, the gradient, the wind speed, your total mass (you plus bike and all equipment) and can then estimate these with estimates of the coefficients of rolling resistance (Crr) as well as drag and frontal area (CdA or drag area) combine that you can calculate the total power (see for example here). Essentially, the other power meters on the market focus on the "supply side equation" by measuring the power delivered by the driver somewhere along the powertrain. The iBike focuses on the "demand side" by measuring the power required to move the bike against wind, downhill gradients and other drag forces. Under normal conditions, this can be reasonably (maybe even surprisingly) accurate, although the accuracy of the performance estimated this way isn't as good - the iBike assumes that the drag range (aka CdA) is constant, so when the rider does changes position (e.g. from the drops to the top of the bar) or if the wind speed is different because the yaw angle changes, the power estimation is deactivated. In general, the iBike has been shown to be fairly accurate for going uphill. This is less true for rolling courses or driving in a pack. The overall accuracy therefore depends on the exact mix of driving and the variability in the wind direction. As with the old Polar and Ergomo without strain gauges, the iBike cannot be statically checked for accuracy or precision. Worse, it also cannot be checked on a dynamic rig in a laboratory as it depends on the slope and wind speed. Checks of the iBike were carried out on site when riders mounted a different power meter on the same bike and compared the two data streams.

There have been some "simultaneous" comparisons of power meter accuracy where a rider has two or more power meters mounted on the bike and made structured or unstructured rides. You can see such a "Rosetta Stone" comparison here and here.

In general, all commercially available power meters have been accurate (and sometimes accurate) when readjusted and working under ideal conditions. However, conditions are not always ideal and parts become damaged, dirty and deteriorate. When accuracy and precision are important, the accuracy of the "design" (whether based on strain gauges, optical sensors, magnetic sensors, or wind speed sensors) is only half the battle: equally important is the ability to check a power meter at home so you can tell when they are gone.

Daniel R Hicks

I wonder if anyone tried to attach the strain gauges to the chainring or the spider. Or on a representative sample of the spokes?

R. Chung

Although I refer to the SRM, Quarq, and Power2Max as "crank-based", they actually fit or replace the spider. They are called "crank-based" because the spider usually goes with the right crank. I think a company may have tried to measure the deflection forces on the chainstays. I don't know if anyone has tried voice-based measurement. The spokes are a kind of "loud" environment, as they naturally change the tension with every revolution of the wheel.

Daniel R Hicks

However, it would be fairly easy to design a chainring that would provide a "good environment" for the strain gauges - cutouts with the right shape, etc. Would be a little difficult on a crank with multiple rings, but probably quite an easy fixie with a crank.