First of all thank you to all who responded.
I got about 30 replies in the first 24 hours and a steady trickle since then. These covered papers showing how to build your own along with author information and insights, commercial products of different descriptions and a few from other people interested in the same thing and all are summarised below.
However, reading all the response does now raises some other questions. What does need to be measured at the pedal to give a valid input to an inverse dynamics model? How many degrees of freedom, which forces and moments, are required to give a good result? Broker and Gregor (1990) describe how they get 3 forces and 3 moments from 2 tri-axial sensors with some assumptions, which can break down when toe clips are used and the size of the errors that can then arise. With a flat foot, unclipped flat pedal can all the moments be approximated away to zero if the pedal spins freely and axes are aligned? With modern shoes and clips on the sole of the shoe can the moments be ignored for the systems that only measure force? This evaluation seems to be lacking in most of the information available on the websites of the different companies.
The one other consideration I had was how much more convenient is, and how much would it be worth, having a wireless system compared to dealing with a number of output cables from a spinning crank?
Maybe these questions might start a discussion from those who have experience in the pedal measuring world.
Commercial products:
Radlabor in Germany. Powerforce crank system: (seems to be called Powertec in earlier research projects). 2D force. A lot of people mentioned this one.
http://www.radlabor.eu/measuring-devices/pedal-force-system.html
Vector by MetriGear (http://www.metrigear.com/products/)
3D force. Comments that they had been helpful with enquiries.
Sensix in France builds a whole range of custom 6DOF sensors for all sorts of foot equipment interactions
http://www.sensix.fr/index.php
3 force 3 moments.
A German company Caloped make some but you have to use the given cranks and pedals. Also I can't read German and even with Google translator's help could not figure out if it was 2D or 3D forces, but it looks 2D.
http://caloped.de/
Novatech measurement have built a 2D pedal sensor for a UK university, although they don't seem to stock them as a default product.
http://www.novatechloadcells.co.uk/
FES sport in Germany do a 2D sensor
http://www.fes-sport.de/
Swift Performance Equipment in Australia have a crank system (Axis Crank) that will do left and right independently. It is not on their website yet but has been demoed at conferences. I do not know what how many forces or moments it measures.
http://www.spe.com.au/swiftsports/
Information on making your own:
The papers by Hull & Davis (1981) and Broker & Gregor (1990) give details of how to make your own.
Broker JP & Gregor RJ (1990). A dual piezoelectric element force pedal for kinetic analysis of cycling. International journal of sport biomechanics, 6, 394-403.
Hull MJ & Davis RR (1981). Measurement of pedal loading in bicycling. Journal of Biomechanics, 14, 843-855.
Alternative methods, and devices:
Mornieux G, Zameziati K, Mutter E, Bonnefoy R, Belli A (2006). A cycle ergometer mounted on a standard force platform for three-dimensional pedal forces measurement during cycling. Journal of Biomechanics, 39,1296-1303.
Poirer E, Do MC, Watier B (2007). An Alternative method to calculate forces applied on a cyclist pedal. Computer Methods in Biomechanics and Biomedical Engineering, 10, 171-172.
Marco Bocciolone, Lorenzo Comolli, Franco Molteni: Metrological Charactetization of a Cycle Ergometer. BIODEVICES 2008: Funchal, Madeira, Portugal 23-28. http://www.informatik.uni-trier.de/~ley/db/conf/biostec/biodevices2008-2.html)
It was also suggested that kistler, bertec and AMTI had the requisite components or variants of pedal, footplate sensors that could easily be adapted to do the job with varying DOF.
Researchers developing/using their own systems
Aidan O' Reilly. Mechanical Engineering at the Institute of Technology Tallaght in Dublin.
John Rasmussen, Professor, PhD, The AnyBody Group, Dept. of Mech. Eng.Aalborg University
VeIUS. French research group who will make them. http://mecano.gme.usherbca/~jmdrouet/velus/index_ang.html
Thanks again
Matt
Matthew Pain, Ph.D.
Senior Lecturer, Sports Biomechanics
Programme Director Sport and Exercise Science Degree
School of Sport, Exercise and Health Sciences
Loughborough University
Leicestershire
LE11 3TU
UK
P: +44 (0) 1509 226327
F: +44 (0) 1509 226301
E: m.t.g.pain@lboro.ac.uk
W: www.lboro.ac.uk/departments/sses