A maximal voluntary isometric contraction (MVIC) is the backbone assessment: the patient produces a maximal effort against a fixed resistance while you capture the force live. Peak force or torque tells you strength capacity, and the shape of the rise tells you how fast they can produce it. Both readings depend on one thing being clean, the baseline the effort starts from. Most of the setup decisions in an MVIC are really decisions about protecting that baseline.
The knee angle you test at is one of those decisions, and the right answer depends on your device. This trips people up because the strength literature and the rate-of-development literature point at different angles for different reasons.
The two things you are measuring
Strength (peak torque) is the highest value the curve reaches. It does not depend on where onset landed, but it does depend on what the sensor can see: any load the sensor never measures is missing from the peak, however cleanly the effort was recorded.
Rate of force development (RTD) is the slope of the rise itself, measured over the first 75 ms after onset (Explosiveness, RTD Early) and from 100 to 200 ms (Build-up, RTD Late). It cares enormously about where onset is and whether the limb was already working before force registered. Anything that muddies the baseline or lets the limb load before the sensor sees it will distort RTD far more than it moves the peak (Maffiuletti et al. 2016).
That difference is why one angle does not serve both metrics on every device.
Why limb weight is the deciding factor
Seated with the thigh supported, 90 degrees of knee flexion puts the shank hanging close to vertical. In that position the weight of the lower leg acts almost straight down through the knee axis, so it creates very little turning moment at the knee and almost no pre-load at the ankle interface. The baseline sits quiet, and onset is crisp.
At 60 degrees of knee flexion the shank is raised toward horizontal. Now the weight of the lower leg pulls it down into flexion, creating a real gravitational moment the patient has to work against. That is the same direction the strap resists, so before the dynamometer registers any extensor push, the quadriceps are already holding the limb up. The early part of the effort goes into that, not into force the sensor can see. Onset lands in the wrong place and RTD becomes untrustworthy.
An instrumented dynamometer measures the limb-weight torque at the test angle and corrects for it, so its reported signal is the full extensor torque. Hand-held dynamometers and simple tension-gauge setups cannot. They read only the load in the line and cannot read below zero, so the limb's weight is never measured, and the reported peak comes in short of the true extensor torque by the limb's gravitational moment at that angle.
The recommendation
On a hand-held or tension-gauge dynamometer that cannot correct for limb weight, test RTD at 90 degrees of knee flexion. The shank hangs near the joint axis, the gravity offset is negligible, and the baseline stays clean enough to trust the onset and the early slope.
Keep peak-torque testing at 60 degrees if that is your strength protocol. Sixty degrees is where the quadriceps produces peak isometric torque and where the return-to-sport thresholds in the literature are referenced (Norris et al. 2024). On a simple device the peak reads slightly low, by the limb's weight moment: a few percent in a healthy knee and proportionally more in a weak one. That is small next to the between-limb and between-session differences you are usually looking for, but it is worth knowing when you set a tension-gauge peak against a threshold measured on a gravity-corrected machine.
Test RTD at 60 degrees only if your device corrects for limb weight. With gravity compensation, the baseline at 60 degrees is as clean as at 90, and one trial can give you both a peak at the reference angle and a trustworthy rise. When RTD is the main question, a separate ballistic set is still the stronger protocol; see choosing a contraction type.
In short: peak torque is angle-driven and mostly device-tolerant, RTD is device-driven. If one setup has to do both on a device without gravity correction, split the difference by taking peak at 60 degrees and RTD at 90 degrees, and record which angle produced which number.
Getting a clean baseline at either angle
- Take up the slack, do not zero out the limb. Cue the patient to take up the slack with the lightest push that keeps the strap taut, then hold still, so the trace rests steady at about 5 to 15 N. Do not tare the device to zero at that point. The small resting load lets the system see a countermovement dip if one happens. See Taring the dynamometer and what a good baseline looks like.
- Standardize the cue. "Push as hard and as fast as you can," held for about five seconds. Keep the wording and the encouragement volume the same across trials and patients.
- Watch for a countermovement. A small dip before the rise, or a limb that resettles before the push, will corrupt RTD at any angle. This is the same failure the gravity offset causes at 60 degrees, arriving by a different route.
Try it in the Dynamometry Lab: work through the weight of the lower leg at 60° and 90° on a simulated patient, and see what it does to onset and RTD.
What to do next
- For the full knee extension setup, including seat, straps, tether perpendicularity, and moment-arm measurement, read Setting up a reliable isometric knee extension assessment.
- Before the first recorded trial, work through Preparing the patient for their first MVIC.
- If you are unsure whether a recorded rise is real or an artifact of the limb loading early, Correcting the auto-detected onset shows how to inspect and reset onset by hand. Careful visual identification of onset is the more valid method when an automatic threshold disagrees with what you can see (Tillin et al. 2013).
References
- Maffiuletti NA, Aagaard P, Blazevich AJ, Folland J, Tillin N, Duchateau J. Rate of force development: physiological and methodological considerations. Eur J Appl Physiol. 2016;116(6):1091-1116. doi:10.1007/s00421-016-3346-6
- Tillin NA, Pain MTG, Folland JP. Identification of contraction onset during explosive contractions. Response to Thompson et al. J Electromyogr Kinesiol. 2013;23(4):991-994. doi:10.1016/j.jelekin.2013.04.015
- Norris R, Morrison S, Price A, et al. Inline dynamometry provides reliable measurements of quadriceps strength in healthy and ACL-reconstructed individuals and is a valid substitute for isometric electromechanical dynamometry following ACL reconstruction. The Knee. 2024;46:136-147. doi:10.1016/j.knee.2023.12.006