Author Topic: Moving from HR training to 'power' (Read 8287 times)

rob · « **on:** 26 October, 2015, 03:04:20 pm »

I've been training for the last 2 years using HR and achieved some reasonable gains. The power brigade on the TT forum, basically say that training by HR is a complete waste and only using power will bring about the correct results. I'm a bit of a technophobe and pretty tight. Given that a crank based power meter for my TT bike plus all the extra gubbins would cost me just shy of a grand I chose to give it a miss. What I have done is to purchase a Garmin 500 with cadence/speed and HRM to give me some metrics to work with.

Given that the TT bike is fixed my plan was to use the same gear and the same tension settings on the turbo and gauge power using cadence and speed, downloading the data and looking at HR later on. As a test I did an hour at my top-end tempo HR and watched what my cadence did. I knew that power would drop off over the hour, but I was pretty amazed that my cadence went from 95rpm to just over 60rpm over an hour, maintaining an HR within 2-3 beats of target. The average cadence at the end of the session was 72rpm. The data may be a bit skewed as I've had 8 weeks of just commuting since PBP and my last turbo session was in July. I had also just had a cold.

I'm going to try an hour of tempo again tonight, but I'm not sure whether to use HR or try to pace the session at, say , 75rpm and see how HR moves.

I think once I find the target cadences I can start to work my sessions around slowly increasing cadence (read power) and duration over the next 6 months.

Is this a valid approach or flawed ?

Chris S · « **Reply #1 on:** 26 October, 2015, 03:12:30 pm »

How about hooking up your turbo and gubbins to something like TrainerRoad? I've seen comparison graphs elsewhere that show a perfectly acceptable comparison between a power meter and virtual power.

https://www.trainerroad.com/virtual-power

vorsprung · « **Reply #2 on:** 26 October, 2015, 04:19:34 pm »

http://ubuntuforums.org/showthread.php?t=2177886

simonp · « **Reply #3 on:** 26 October, 2015, 04:23:08 pm »

With many turbos, their resistance varies according to how warm they are. It can take a while for them to warm up properly.

HR drifts too, and is not consistent from workout to workout either.

I'm using TrainerRoad with a Kickr. BTW, Rob, if you want to try out TrainerRoad, I can give you a free month to try it out (I don't know what T&Cs apply to the trial). Be warned that it's considered a gateway drug on the slow twitch forum.

You can see the cardiac drift effect clearly in this workout I did:

https://www.trainerroad.com/cycling/rides/2420443-Jay-Peak

Each of the 203W intervals has a HR about 10bpm higher than the previous.

LittleWheelsandBig · « **Reply #4 on:** 26 October, 2015, 09:30:00 pm »

When I did the training thing, losing enough body heat was a real problem. I could maintain a constant high power output over long durations only if I could lose enough heat and that is pretty difficult on a turbo. Not many turbos are set up for a 40+km wind flow and low-ish humidity.

zigzag · « **Reply #5 on:** 26 October, 2015, 10:39:03 pm »

the real power meter is much more precise to measure your output/intensity and can be a very useful tool in the actual tt race when you need to pace yourself and not burn out at the start or on the inclines. i've used both hrm and pm alongside and hr is nowhere near as useful due to the lag and other factors affecting it (but definitely better than nothing).

simonp · « **Reply #6 on:** 27 October, 2015, 12:15:52 am »

I thought some research had found that HR produced better results than power.

Here you go:

http://www.ncbi.nlm.nih.gov/pubmed/19204572

Big problems with these kinds of studies: they’re just too small.

rob · « **Reply #7 on:** 27 October, 2015, 08:40:06 am »

Thanks for pointers to traineroad. I had looked at it but we only have one lap-top in the house and it isn't mine. I'm actually not too bothered about getting accurate power data as I race on HR and RPE, anyway. What I was looking for was a metric that allowed me to pace my sessions better.

I re-ran my 1hr tempo test last night and managed to get the effort a bit smoother. There was still a drop off in cadence as I started at circa 90rpm and was doing 70rpm by the end. My average cadence was 74rpm, so I will try the same session next week and ride the whole hour at 75rpm and see how HR/RPE responds.

I'll do something similar with my threshold efforts, with a view to steadily increasing the cadence - in truth 70rpm is a bit slow for me, so I'm hoping to put some improvements in to get this into the 80s fairly quickly. I have a couple of bigger chainrings that I can swap over as fitness builds.

Ningishzidda · « **Reply #8 on:** 28 October, 2015, 09:22:10 pm »

Quote from: Chris S on 26 October, 2015, 03:12:30 pm

How about hooking up your turbo and gubbins to something like TrainerRoad? I've seen comparison graphs elsewhere that show a perfectly acceptable comparison between a power meter and virtual power.

https://www.trainerroad.com/virtual-power

I notice they are using a Kurt Kinetic Road Machine, which has a repeatable power curve. Better than most. KK freely publish the curve in MS Excel, and a 4th order Polynomial trendline will give you the coefficients to plot power for any speed.
You can either perform 60 minute FTPs or CT20 tests with a bike computer on the driven wheel.

A useful addition to the KK road machine is the Pro Flywheel which goes most of the way to simulate the Standing start of a TT.

If you already have a turbo, the 'test' part of the FTP protocol will be the only part that will be variable. The 'warm up' part will always be the same for consistency's sake. The 'preconditioning' cycle of a motor vehicle emissions test ( as Volkswagen know too well ) is consistent. The 'preconditioning' cycle of a cyclist's FTP test should also be consistent.

rob · « **Reply #9 on:** 29 October, 2015, 10:32:36 am »

2 * 20 @ 101% threshold HR last night. At the end of the second interval I was going slower than at the end of a tempo session. I think all this info discovery is depressing me and whilst last years training seemed to make a huge difference I was still doing it wrong.

I think as I use a warm-up to get my HR up towards target I think I'm pushing too hard and compromising the interval and the rest of the session. I'm going to pace some sessions by cadence/speed next week and see how HR responds rather than the other way round. It's 6 months until I race again so plenty of time to experiment.

Pedal Castro · « **Reply #10 on:** 29 October, 2015, 03:28:47 pm »

My understanding of sweet spot training is that you should be doing it at 90-95% of THR. I certainly couldn't do 2*20 at anything above my THR. To improve your threshold you could do 2*20 at 90-95% and if you manage that OK add another 20 interval and see if you can complete 3*20 at that HR range. As you continue to do these then either your THR increases or you get faster for the same HR, or both.

I have just started using TrainerRoad which will complement my commute based training this winter. I will train with virtual power on the turbo but HR on the road, the relationship between the two is very interesting.

What I found last winter was that it was very difficult to ride at the same HR into the wind than with it, usually a 10bpm difference at least. Possibly psychological but more likely due to lower cadence.

rob · « **Reply #11 on:** 29 October, 2015, 05:07:01 pm »

By last Summer I was able to do 2hrs at Sweet Spot - which I paced by using a Heart rate right at the top of my measured zone 3 (162-163bpm). 163bpm is almost exactly 95% of my threshold HR of 172bpm. The last 30 mins were a but gnarly but I could complete the session. You do need a good dvd to keep you distracted and not clock watching.

I could also do 2*20 mins at 101% of threshold HR (175bpm) which is the session I did last night. If your threshold/FTP is the max you can produce for an hour then 2*20mins should be do-able with the right level of focus.

Where I'm getting to know is that I warmed up so entered the interval/session at the target HR and what this seems to mean is that I worked too hard in the warm-up and in the first few minutes of the session and this compromised output at the end. I need to start pacing intervals so that I enter then at a lower HR but can maintain the effort until the end. From there I can start to steadily increase the effort/duration and see how HR reacts (hopefully falls) as there are fitness gains.

Pedal Castro · « **Reply #12 on:** 29 October, 2015, 07:48:09 pm »

THR and FTP are not the same thing though. FTP is the maximum average power you can sustain for one hour but THR is most commonly defined as OBLA (onset of blood lactate accumulation). My understanding of OBLA is just under and an hour is possible but just over and after 5 minutes or so it becomes quite a struggle to keep up a constant output. I know that in my case as long as HR remains 158 or below I can carry on for hours, 160 and above and I know I can't keep it up for more than a few minutes without blowing. This figure matches very closely to an Conconi type test I did lat year to estimate my OBLA HR.

Ningishzidda · « **Reply #13 on:** 30 October, 2015, 07:12:57 am »

Keep the resistance of the turbo on the same curve, which is what the Kurt Kinetic Road Machine does. Fit a computer to the driven wheel and watch avg Speed through a 10 mile session. Don’t worry about speed or power or HR. Just make sure the ‘warm up’ is the same every session.

IF the time it takes to complete the 10 miles reduces and the avg speed increases, you’re getting better.

With the avg speed noted, refer to Kurt Kinetic’s curve to get an estimate of the Power.

For a CT20, ride for 20 minutes instead of 10 miles. Eventually, they should agree.

rob · « **Reply #14 on:** 30 October, 2015, 08:53:31 am »

I have done a couple of ramp tests in the past and have a good handle on my zones and where my lactate threshold (OBLA) sit. From these I have been able to design sessions that push threshold from below or pull from above. Adam Topham's TT book has been quite useful although I don't do the pyramid intervals which look horrific. Given I am training for longer TTs (50m and up) I do most work sub threshold/tempo/sweet spot and one session a week above threshold.

I tried to do one of my favourite sessions this morning which was 5min on/off under over threshold, but tried to use cadence rather than HR. Again I found it wasn't hard enough and ended up overcooking the early part and ruining the last 15 mins of the session. So I'm now concluding that a) I have been warming up too hard and b) My HR has a real lag. I need to stop worrying about not working hard enough early in a session in order to push hard to the end. Will try the same set of sessions next week and review.

Pedal Castro · « **Reply #15 on:** 30 October, 2015, 02:59:18 pm »

Going back to your original question I would say focus on a constant cadence not HR. If you look at myTrainerRoad FTP test trace you can see that although the oxygen deficit induced HR "lag" is pronounced for up to 2 minutes my HR doesn't really stabilise until the second 10 minutes. However in the 3rd and 4th quarters the average HR and average power are constant.

simonp · « **Reply #16 on:** 30 October, 2015, 03:03:49 pm »

In my last 20 minute test my HR climbed throughout.

https://www.trainerroad.com/cycling/rides/2179426-20-Minute-Test

Pedal Castro · « **Reply #17 on:** 30 October, 2015, 03:57:15 pm »

Quote from: simonp on 30 October, 2015, 03:03:49 pm

In my last 20 minute test my HR climbed throughout.

https://www.trainerroad.com/cycling/rides/2179426-20-Minute-Test

But so did your power after the first quarter. Also it would not be implausible to expect HR to increase with temperature so if there is inadequate cooling HR will increase even at a constant power output, this can be a marked effect on turbos, particularly if used indoors.

rob · « **Reply #18 on:** 30 October, 2015, 04:29:03 pm »

Quote from: Pedal Castro on 30 October, 2015, 02:59:18 pm

Going back to your original question I would say focus on a constant cadence not HR. If you look at myTrainerRoad FTP test trace you can see that although the oxygen deficit induced HR "lag" is pronounced for up to 2 minutes my HR doesn't really stabilise until the second 10 minutes. However in the 3rd and 4th quarters the average HR and average power are constant.

Thanks. I'm with you, but it takes a fair bit of will power not to push too early in a session as I have a limited amount of time to train so want to make every session count. Roughly I estimate my upper tempo cadence to be 75rpm and my threshold to be 80rpm. Using the Kurt Kinetic graph this roughly equates to 175w and 200w, respectively. This feels a bit low given the times I did last year, but 1) it's an estimate only and 2) I'm a bit out of shape. Extrapolating this tells me that 90rpm would be, again roughly, 250w so my target is to be able to do this for an hour within the next 6 months.

Going back to your points on HR there's a good article on riding a 12hr here :-

http://www.abcc.co.uk/how-to-ride-a-12-hour-time-trial/

This was written prior to power meters being too common but the HRs in the table actually work really well for me so my HR goes :-

10m - 180+
25m - 175+
50m - 170
100m - 160
12hr - 150

As you say with the 150bpm I can hold it for 12hrs, but the last 2-3 are a bit tough and I dropped a few beats at the end of the 12 I did last year. My measured threshold HR is 172 so being able to run a bit over for a 25 feels about right.

Rob

Pedal Castro · « **Reply #19 on:** 30 October, 2015, 07:36:44 pm »

I was aware of that article by Malcolm (I was an ABCC coach myself in the 80s/early 90s) but the bit on HRs is not significant in my opinion. It did prompt me to look at all my old HR data I recorded for TTs in 89/90. Interestingly I see a very similar initial rise in HR and then very stable HR for the majority of the ride but with a slight uplift at the end or a slight decrease which I suspect was due to having a bit left in the last mile or the opposite! Most interesting is that of the two 25s I have data for the much better performance was 5bpm lower that slower one.

My conclusions from then and what I can see now is that using HR to guide race effort should be treated with caution as so many things can elevate or depress it. However as a training aid to keep you in the right zone I find it really useful.

As I said upthread I find it much harder to keep my HR high when riding into the wind, either training or racing. Having thought about that and considered what you have said Rob, I now suspect that it could be a cadence issue, I.e. I struggle to keep up my cadence into the wind and end up with a lower HR for the same? perceived effort.

Personally if I were you I would go with a lower gear and higher cadence on the turbo.

rob · « **Reply #20 on:** 31 October, 2015, 09:33:31 am »

All useful points, thanks.

On racing by HR, I do find it quite variable. I did 3 25s last year and HR varied by 5-10bpm. On my PB ride (1:00:17) I was still lifting towards the end. Feeling I had a bit left I went harder at the start of the next 2 rides and then struggled later on. In retrospect I think I paced the first one right, but I do curse those 17 seconds.

Commonly accepted theory is that, for a fairly aero position 230-240 watts will get you under the hour. My position isn't the best but I estimate my FTP was around that number last year.

The weakness I noticed last year, racing on fixed, was pushing headwinds and riding up steady drags. I did some high intensity low cadence intervals last year and I hated them. By mid Summer I was really spinning in order to get to the effort I required on the turbo. In order to fix both issues I have upped the gear on the turbo from 49*19 to 48*17. All efforts will now be at a lower cadence, but I agree I may have overlooked it. I will give it a few weeks and see how it goes and if cadence increases as fitness starts to rebuild. The other issue is that I have 49 and 50 tooth chainrings but nothing smaller.

I will end up buying a power meter I'm sure but I am trying to put it off for a while. This project has already been pretty costly.

rob · « **Reply #21 on:** 03 November, 2015, 02:06:06 pm »

65 mins at Tempo effort last night. Went a bit too hard for the first 30-odd minutes and backed off slightly for the last half an hour. Ave RPM was very similar to previous workouts without the huge drop-off in output.

Heart Rate took 30mins to get into the range I would expect for an effort like this.

L CC · « **Reply #22 on:** 03 November, 2015, 02:51:38 pm »

I've come to find I take a ridiculous amount of time to warm up properly. 30km or so at audax pace. It's slow enough that I'm concerned about finding a different warm up strategy when I start racing.

Ningishzidda · « **Reply #23 on:** 03 November, 2015, 02:55:26 pm »

Quote from: simonp on 30 October, 2015, 03:03:49 pm

In my last 20 minute test my HR climbed throughout.

https://www.trainerroad.com/cycling/rides/2179426-20-Minute-Test

Please tell us there was a five minute nothingness between the warm-up and the test.

Five minutes to go to the lavvy and do some stretches.

simonp · « **Reply #24 on:** 03 November, 2015, 03:29:14 pm »

Quote from: Ningishzidda on 03 November, 2015, 02:55:26 pm

Quote from: simonp on 30 October, 2015, 03:03:49 pm
In my last 20 minute test my HR climbed throughout.

https://www.trainerroad.com/cycling/rides/2179426-20-Minute-Test

Please tell us there was a five minute nothingness between the warm-up and the test.

Five minutes to go to the lavvy and do some stretches.

Sorry, can't help you there. It was 5 minutes at ~50% FTP.