Comparing 30/15 VO2max Intervals

Last time we looked at VO2 and SmO2 response to hard-start continuous power VO2max intervals. This time I want to look at my other favourite VO2max workout, the Rønnestad style Hard-start 30/15s VO2max workout. This was the protocol I arrived at based on Rønnestad, Seiler, Buchheit & Laursen, and others’ research into optimizing interval prescription for maximizing adaptations to VO2max.

Summary of Assumptions

Please read this article for more detail on background, and framework for how I arrived at this interval protocol, and the update where I re-considered some of my previous assumptions. The short version is:

  • It seems from the literature that accumulating more time around 90% VO2max is a ‘sweet spot’ for eliciting adaptations to VO2max.
  • Intermittent/microburst style intervals can allow greater total accumulated duration at a target intensity, than continuous intervals would allow.
  • Hard-start intervals stimulate faster VO2 onset kinetics to reach higher VO2 more rapidly.
  • VO2 slow component, which is the drift toward VO2max at constant workload above CP or FTP, would usually result in premature exhaustion.  But by decreasing power to account for this loss of efficiency due to fatigue, relative intensity as a percentage of VO2max (ie. 90%) can be preserved for longer duration.

90% HRmax correlation to 90% VO2max

When I first received the VO2 Master Pro the first question I wanted to investigate was how well 90% HRmax correlated to 90% VO2max. As we looked at for the continuous VO2max intervals, the correlation seemed to be good for myself and the other athlete I measured.

We both seemed to reach 90% for both metrics at around the same time, although my HR slightly under-predicted my VO2max (I reached 90% VO2max at slightly lower than 90% HRmax) while the other athlete’s HR slightly over-predicted their VO2max (they reached 90% VO2max at slightly higher than 90% HRmax).

How about for an intermittent/microburst VO2max protocol? Would onset of 90% HRmax and VO2max occur at around the same time? Let’s take a look at a few workouts


  • Yellow line is Power measured by dual sided crank-based 4iiii power meter
  • Blue line shows Right leg Power (good leg)
  • Pink line shows Left leg Power (bad leg)
  • Red is HR, highlighted above 90% HRmax
  • Blue area is Oxygen Consumption (VO2 in mL/min), highlighted in hot pink above 90% VO2max

Compare the VO2 (blue area) to HR (red line) and where both are highlighted above 90%, respectively. On both intervals VO2 rapidly reaches target 90% within the first few hard-start 30sec microburst intervals, while HR rises much more slowly after reaching an initial plateau.

The timing of onset of 90% HRmax and VO2max were clearly very different, but if we look at the time accumulation above both targets, we see ~11min at >90%HRmax and ~9min at >90%VO2max. Fairly close, but this might simply be incidental.

VO2 kinetics appears to respond faster than HR. So while the micro-rests allow VO2 to briefly dip below the strict 90% target on each micro-rest, HR shows a slower, but more stable rise toward 90%. Compared to this, a continuous power interval provides a more stable demand/stimulus to allow both HR & VO2 to remain more consistently elevated above 90%.

It could be possible to simply tweak the parameters of the microburst workout to have either higher workload, or shorter rests, or some combination to prevent VO2 from declining below our 90% target. But at least for me, I was already maxing out RPE on these workouts… I’m not sure I’d be able to finish the sets if power was much higher!

Hard-start Continuous-power Intervals

  • VO2 onset kinetics rises rapidly to 90% VO2max within roughly 90sec.
  • HR kinetics also rises rapidly to 90% HRmax around 90sec.
  • VO2 & HR both remain consistently elevated above 90% for the full remaining duration of the interval.

Hard-start Intermittent 30/15s Microbursts

  • VO2 onset rises rapidly again, reaching 90% VO2max within roughly the same 90sec.
  • VO2 is more sensitive to changes in workload between work & rest, falling under 90% during micro-rests and rising again above 90% during work intervals.
  • HR rise is slower and can be disrupted by the first micro-rest, causing HR to initially plateau lower than 90%.
  • HR is subsequently more stable than VO2, meaning once it’s above 90% HRmax it usually remains above that level.

It appears that the correlation between 90% HRmax and 90% VO2max holds more accurate for continuous power intervals, than for intermittent microburst intervals. More precise prescription of work & rest intensity & duration could mitigate this variability in VO2 and delayed rise in HR.

Case of Fatigue causing depressed aerobic system activation

Another good example of HR-VO2 relationship is demonstrated in another workout I did a few weeks later. I was over fatigued for this workout, and as we’ll see I had difficulty reaching the same output as the workout above.

I started with a 6min continuous VO2max interval and actually felt good until around half way through. Then my Wahoo Kickr trainer hit a power spike, which is when the smart trainer suddenly and sharply raises resistance on the flywheel. If you’ve ever felt it before, it’s like running into a brick wall.. It was very tough to pedal through and kinda knocked me off my game for the rest of the workout.


The 6min continuous interval shows better correlation between onset of 90% HRmax and 90% VO2max, and accumulation of time >90% (3:58 for HR, 4:08 for VO2). But after that even with 15min full recovery, I felt like my legs weren’t coming back.. So with trepidation I hit the hard-start 30/15 microburst set.

After the first few 30sec microbursts, it was obvious that VO2 wasn’t responding as normal. I was stuck below my target of 90% VO2max despite the workload being on target, and RPE was significantly higher. I felt like I couldn’t stay on top of the gear.

Zoomed in on the 30/15 microburst set

I must have been digging deeper into my anaerobic energy reserves, because I just wasn’t able to utilize the same volume of Oxygen for the same effort as during the previous workout.

I had to skip an interval and my already depressed aerobic system dropped precipitously in those 30sec of backpedaling. HR fell from 177 to 152 bpm, and VO2 dropped all the way to 2400 mL/min from my target of 5000 mL/min.

I was able to continue by lowering the intensity, but VO2 never reached the 90% target through the entire set.

Fatigued VO2max microbursts

  • Fatigue can contribute to depressed stimulation of the aerobic system. Delaying or preventing entirely the attainment of 90% VO2max.
  • My guess is that the blunted aerobic response forces greater reliance on anaerobic metabolism to meet the required workload, meaning faster accumulation of fatigue-causing metabolic byproducts, higher perceived effort, and premature exhaustion.
  • This illustrates the importance of ensuring you are fresh and rested for your high intensity interval workouts. And also identifies a potential explanation for the phenomena of ‘non-responders’ to VO2max training when only HR is used as a corollary for VO2.

Manipulating respiration to raise VO2

The final 30/15 workout I want to look at shows a couple of interesting things. First, another example of different timing of 90% HRmax & 90% VO2max onset, and more interestingly second, an example of manipulating breathing to stimulate greater VO2.

12.28.2018_1_LI (3)

The first set shows the typical rapid VO2 response to 90% VO2max within ~90 seconds, but this time HR was much slower to rise, only reaching 90% HRmax in the final few minutes. The time accumulation during the first set were 8:19 at >90% VO2max, and only 1:38 at >90% HRmax.

I was feeling good during this workout, and I suspect the lower HR for higher VO2 actually reflected greater fitness or freshness on that particular day. My body was able to deliver and utilize more oxygen with relatively less effort (or at least fewer pumps) by my cardiovascular system.

Then something interesting happened during the second set.

Zoomed in on the second 30/15 microburst set, with respiratory metrics

Top chart

  • Pink line shows Tidal Volume (Vt in Litres) which is the volume of air passing through the lungs on each breath
  • Dark Green line shows Respiration Rate (RR in breaths per minute, BPM)
  • Light Green line shows VO2 consumption per breath (in mL/breath). This is a good metric for tracking pulmonary efficiency

Bottom chart same as previous

For some reason, despite HR rising higher than in the first set, I noticed VO2 was sitting lower than my 90% target. I also noticed that I wasn’t breathing as fully as in the previous set. I was beginning to breathe with shorter, shallow breaths which are less efficient at allowing oxygen exchange at the lungs.

I’ll talk more at some point soon about what I’m currently learning about pulmonary limitation to VO2 and respiratory training, but suffice to say that I suspected my respiration was beginning to limit my oxygen uptake in real time. I tried to refocus on taking deep, full breaths. Sure enough tidal volume re-normalized from as low as 4.0 L to above 5.0 L per breath, and VO2 rose back above 90% for the end of the set.

My hope is that this rise in VO2 allowed my working muscles to produce just a little bit more work to finish off the microburst set. RPE was certainly higher before I refocused on my breathing. Alternatively VO2 could have risen in response to my respiratory muscles having to work harder to push those deeper breaths.

In either case by focusing on maintaining tidal volume per breath I created a greater energy demand and stimulated my system to consume, deliver, and process more oxygen to meet that demand. I think regardless of where that oxygen was going, it was an effective training stimulus for increasing systemic VO2max.

Variation in the Power-HR-VO2 Relationship

  • This workout was an example of how poorly HR can be correlated to VO2. I spent more time >90% VO2max and less time >90% HRmax during the first set, but the reverse during the second set.
  • While it equaled over the course of the workout, it supports the findings that HR is a worse predictor of VO2 for intermittent microburst intervals, compared to traditional continuous power intervals.
  • Likewise, while power was very similar across these Hard-start 30/15 workouts, my VO2 response varied from day to day.
  • It seems to me that a better estimate of VO2 should include all three measurable metrics: Power, HR, and RPE.

Pulmonary Ventilation and VO2

  • VO2 uptake can change with different ventilation strategies.
  • Typically higher Vt (tidal volume per breath) is more efficient than higher RR (respiration rate) because of the pulmonary dead space on each breath. Deeper breathing allows greater available surface area for gas exchange at the lungs.
  • Focusing on deeper breathing may increase systemic VO2 either from increasing delivery of oxygen to the working muscles (your legs!) and/or to satisfy an increased work of breathing by the respiratory muscles… Or possibly for some other reason entirely?
  • I still have a lot of learning to do regarding pulmonary affect on oxygen consumption and respiratory training, but the more I read and experiment, the more I see the potential for improvement.


If there is a conclusion here, I believe it is that without precise measurement of VO2 the best bet on optimizing your own VO2max workouts day-to-day is to account for power, heart rate, and RPE (rating of perceived exertion) together. No one metric alone can accurately predict VO2.

I haven’t talked enough about RPE, but your brain is probably the best tool to estimate how close you are to ‘maximal effort’. If power is barely sustainable for a 3-8min interval, HR reaches near HRmax, and RPE is near 9/10, chances are you are close to VO2max!

18 thoughts on “Comparing 30/15 VO2max Intervals

  1. I thing a combination of both steady and intermittent intervals is the best way. Then the body gets alternating stimulus and that should be benefincial in the long run. PLus the intermittent style makes the legs use to really hard burst that often accour in racing.


    1. Agreed completely, Johannes. Can’t do any single workout for an entire training block and expect to see continuous improvement.

      The limiting factor may be different under different workout protocols, but if we’re ensuring maximum training quality on all our workouts, we’ll be providing a stimulus for the body to adapt to whatever that particular limitation was on the day.


    1. Thanks Jordan. The VLamax model has really opened my thinking up to a new perspective on energy system training. Expect more, but don’t expect it too quickly! 😅


  2. Jem — found your site and analytics late in life but your VO2max deep dives have really rejuvenated my zeal for the zone. Thanks for all your great work!

    p.s. — coming at ya from TR!

    Liked by 2 people

  3. Hey Jem

    I have read a lot of papers and presentations also of Prof Stephen Seiler. In one paper he has shown a superior effect on Vo2max progression and performance variables of 4x 8 min intervals. In other papers he describes that a lot of top athletes seem rather to train with longer intervals not too far above threshold power to accumulate a long total work duration between 30 and 40 min or even more above threshold.

    Obviously this intensity is well below something like 5min all out power or power from a ramp test associated with achievement of VO2max.

    One might hypothesize that the long intervals lead to probably lower Vo2max values than a shorter more intense “Vo2max” interval workout but probably allow to accumulate more time at high Vo2max values and thus strong improvements in physiological and performance parameters.

    What’s your view on those long HIT interval protocols and have you had the chance to try the VO2master with such a workout to analyze time greater 90% of Vo2max. Would be keen to know how much time one can spend above 90% of Vo2max with 4-5x 8min intervals or similar.

    Thanks in advance.


    1. Hi Fab,

      The relevant 2013 paper from Seiler.

      I’d have to go look into it again, but I believe the intervals were steady paced maximal efforts that elicited 88%, 90%, and 95% HRmax respectively for 16min, 8min, and 4min. I’m still not quite sure if these were average HR, or peak HR during each interval. But regardless, this would suggest that all intervals would have been above CP, and would therefore have likely brought the subjects to VO2max eventually through the workout, due to the VO2 slow component.

      The 8min intervals may have been the optimal balance of less time to elicit VO2max [or a relevant percentage of, ie. 90%] and more time at or above [90%] VO2max. Or, more interestingly, the 8min intervals may be revealing greater adaptive benefits from combined energy system training stress, eg. a balance of workloads on the anaerobic-lactate and aerobic-VO2max systems. The 8min intervals had similar RPE to the 16min intervals, but higher HR, and higher blood lactate, suggesting higher metabolic stress. The 4min intervals had higher RPE and residual fatigue along with higher HR & BLa, possibly indicating that intensity was ‘too high’ to sufficiently recover from between training bouts. The authors note this as a possible limitation in their discussion.

      I have not measured VO2 for longer duration steady-paced intervals. Only for the hard-start decreasing-power intervals I wrote about previously.

      With those I was trying to minimize anaerobic contribution and lactate production by decreasing power through the sustained (4-6min) intervals, to account for VO2 slow component and losses in power output due to decreasing gross efficiency. But maybe greater adaptation would occur with a combined aerobic & anaerobic stimulus. Definitely worth experimenting!


      1. Great blog Jem,

        Your hard start decreasing power intervals are intriguing. Virtually no on thinking about the slow component of VO2! I think there is some evidence that the slow component is less significant in well trained athletes. I’m curious to know if this is your observation, and whether you prescribe these intervals to less fit or older athletes preferentially?



      2. Thanks William!

        You’re right, the VO2 slow component is correlated to fitness/training status. Higher fitness = less effect of VO2sc. But it still plays a major role in any athlete above Critical Power, no matter their abilities. I could reference this statement, but I think it’s basically true by definition. Above Critical Power means no longer able to maintain a physiological homeostasis, therefore ‘physiological drift’ occurs toward exhaustion, ie. VO2sc.

        The scale of the VO2sc effect does contribute to how long an athlete can sustain >CP efforts. But the general nature of the hard-start VO2max intervals I’ve talked about would be the same for all athletes. IMO the duration and volume prescription should be adjusted to be appropriate for the particular athlete’s abilities. eg. 4x4min or 4x6min, etc. And of course there are many other factors in determining whether these intervals would be appropriate for a given athlete.

        Less fit athletes will probably take longer to reach 90% VO2max (slower VO2 onset kinetics) and have to decrease power more rapidly toward CP (faster VO2 slow component). And vise-versa for more fit athletes.

        This is a good reason why prescribing these kinds of hard-start VO2max intervals could be better as RPE+HR based moreso than power based. ie. “go very hard until you reach 90% HRmax then gradually decrease power to maintain 90% HRmax without blowing up”.


    1. Ronnestad has more recent data on 30/15s, starting at about 55:30 in this video from the European Society of Sport Science conference this year ( Similar results to his 2015 study, but in more elie cyclists (VO2max 70+).

      I am about to buy a VO2Master – would be interested to hear your thoughts about how to use it to optimize intervals…

      I’m an exercise physiologist in Vancouver. It you are interested in collaborating in a trial lin this area that I’m thinking about, send me a PM 🙂


      1. Oh yea! Someone referred me to that video recently. The ECSS channel has a ton of fascinating webinars. Good contrast in some of the evidence presented by Ronnestad and I think Hopker before him. Very relevant to the long vs short/intermittent interval prescription and intensity vs workload stimulus (VO2max vs power) question.

        Would be very interested to chat or give any info I can about the VM Pro. I’ll PM you 👍


  4. Hey Jem,

    Regarding Ronnestad’s 30/15… Check out his presentation at this years European Conference for Sport Medicine – he presents data on 30/15s in elite(ish) rides (VO2max 70+) with similar results to his 2015 study (starts at about 55:30 in this video –

    About to order a VO2master. Interested in hearing how you thing it could be used yo maximize interval effectiveness (a la Billat/Lisboa).

    I’m an exercise physiologist in Vancouver – if you are interested in collaborating in a study, let me know by PM:)


  5. his is really interesting – thanks for writing it (even if Ive found it late 🙂 ). My thoughts around Ronnestad intervals are more around their execution – its interesting to see you use ERG mode for these: in the Ronnestad study he used ERG mode for the first ‘initiation’ set, but then left it up to participants to do how they wanted for the rest of the study. Ive seen others saying that they shouldnt be done in ERG mode because of the rapid change of power and that it is better to overshoot at the beginning, almost like a hard start interval. Either way, Im interested in the cadence and changes in cadence that may occur during these intervals. Theres no mention of cadence in the Ronnestad papers at all, or whether cadence variation (when not in ERG mode) affects the outcomes of these intervals. For example, if cadence is lower in the first 5-10s of the intervals and builds to say 95revs by the end of 30s does that have a different effect to constant cadence? I think I heard Stephen Seiler commenting that these intervals are very difficult for rowers because of the drop off in speed during rest making the restart really hard. If you have any comments on that from your own experience, or know of any research, Id be really interested to hear it – cheers.


    1. Hey Gareth,

      Thanks Gareth, yeah since writing this I’ve talked a lot about not using erg mode and just trying to go ‘8/10’ for most HIIT training bouts. I just had a really interesting conversation with another researcher who commented on work he’s done with athletes self-pacing continuous and intermittent HIIT. The self-paced athletes were able to maximize their acute response to the workout session, more than the athletes performing in erg mode.

      Cadence and re-acceleration from the transient low intensity intervals is another really interesting area for questions: Are intermittent intervals exploiting constant on- and off-kinetics of disrupting and restoring the metabolic milieu to improve buffering capacity? Does neuromuscular recruitment/coordination benefit from the repeated accelerations against higher resistance, vs a constant resistance sustained interval? Or is it all about central-effect and allowing the preservation of high cardiac output for a longer effective work-duration (eg. the total duration of the 13x 30/15sec set)? Very interested to see what new findings come out from Norway in the near future.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.