Test section
It has now been nearly two months since Vasaloppet marked the end of the ski season for most people. As we move into May and the roads are finally clean, it’s time for cross-country skiers to bring out the roller skis. While nothing fully replaces real skis, roller skis have their own advantages. One major benefit is that different wheel types allow you to control rolling resistance far more precisely than on snow—where nature plays tricks even on the most experienced ski techs.
Last week, the Skisens test team—CEO Johan Högstrand and myself—went out to compare the rolling resistance (“rull”) of different roller skis.
Test Procedure
We tested four pairs of roller skis. I compared two classic-style skis fitted with so-called #2 wheels (standard resistance): a pair of well-used Swenor Fibreglass (2012) and a pair of Oneway 7 (2015). Based on past experience, the Oneway skis from Oneway are significantly slower than the Swenor skis from Swenor.
Johan compared a pair of Proski (3 wheels rear / 2 wheels front) vs. a pair of Eagle Sport skate skis equipped with Elpex PU-2 wheels from Elpex.
The test consisted of a nearly flat 750 m stretch that we skied back and forth—total distance 3 km. Each pair of skis was tested twice:
• A1 pace (easy distance)
• A3 pace (just below anaerobic threshold)
We measured time and power for all intervals. We skied side-by-side to also compare watt output between us. This means the estimated intensity varies slightly because our skis rolled differently and because we differ somewhat in upper-body strength. My Swenor #2 wheels were slightly slower than Johan’s Elpex PU #2 wheels, so what was intended to be A3 ended up being closer to A3+ for me—and A3− for Johan.
Our hypothesis: power should remain similar across different wheel types, while time and speed should vary significantly.
Results and Discussion
The summary tables are shown below.
Table 1: Summary of test data
From the table above we can conclude:
- Average power at each intensity was far more consistent than the completion times.
- Heart rate is difficult to compare due to drift throughout the session. HR during the first A1 interval was significantly lower, but end-HR for A3 intervals aligns well.
- Rolling resistance on the slower wheels (Oneway and Proski 3/2) appears more speed-dependent than on the faster wheels. The time gap compared with Swenor/Elpex was much larger at A1 than at A3. Air resistance also contributes to this.
Figure 1 shows heart rate, power, and cadence from Today’s Plan for two A3 intervals using different wheels. Notice how power drops to zero at each turnaround, while heart rate—due to its slower response—does not react immediately. You can also clearly see the power spikes after each turn, something that HR cannot capture.
Figure 1: (a) Oneway A3
Figure 1: (b) Swenor A3. Data from Today’s Plan showing two A3 intervals. Note the three turnarounds where power drops to zero.
Finally, let’s discuss how well our measurement captured the differences in rolling resistance. Those who follow Skisens know that in our Vasaloppet reports we analyzed glide by estimating friction through the poles.
In this test, the Oneway skis were clearly slower than the Swenor skis—our measured friction coefficients differed by about 19%. The difference between Proski 3/2 and Elpex PU-2 wheels was even larger—43% in rolling resistance according to our measurement.
Comparing Proski/Oneway vs. Elpex/Swenor is harder since different athletes used different poles, which adds uncertainty. Still, our rough conclusion is:
• Proski were slightly slower than Oneway
• Elpex PU-2 skate wheels were clearly faster than Swenor #2 classic wheels
Table 2 shows a very preliminary overview of friction coefficients from this test.
Table 2
Conclusions
Overall, we are pleased with the results. Our goal was to show that Skisens power-measuring grips can distinguish rolling resistance between roller-ski wheels, and eventually be used to evaluate the glide of snow skis.
We clearly captured the large differences between the test wheels. Of course, these big differences are noticeable even by feel. The long-term goal is to resolve differences at the last decimal—for example, distinguishing between a friction coefficient of 0.032 and 0.033, which corresponds to 1–3 seconds per kilometer depending on speed.
Much more systematic testing remains, but we strongly believe this will become a reliable method for monitoring glide in the future. The best part is that an athlete racing with the Skisens grips can compare data afterward with others and finally answer the question:
“Was it the skis—or the body—holding me back today?”
Dan Kuylenstierna
Skisens AB
Recent Comments