Fortius Functional Threshold Testing

Anything about Power (watts)

Moderators: mcorn, Tacx Moderator, Tacx Video Production, malfukt, Tacx support

User avatar
qcscmh
Posts: 777
Joined: Fri Dec 30, 2005 2:57 pm
Location: UK, Hampshire (The Tropical South)
Contact:

Post by qcscmh » Wed Mar 19, 2008 10:50 am

Forgetting the accuracy I find it strange that power is calculated diferrent ways in Catalyst ? surely it should be calculated the same way in all applications (VR, RLV, Catalyst ??) :? :?
Nothing interesting here > http://bikestuff.freehostia.com/

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Wed Mar 19, 2008 10:55 am

Phil wrote:Once the slope (real world averaged power from 0 up to your Vo2 max power) you should be cool. The % of error should be very low.
So for starters: While riding at Catalyst/RLV, whatever grade - stop applying torque to the pedals, just coast to provide cadence signal. You can just pedal backwards. Tell me what happens to speed/power readings. Explain why speed eventually won't go to zero :shock: Explain why power is registered for a long time even that there is no torgue on pedals.

Tell me how this is supposed to be realistic. How can you say that % of error is "low" then.

User avatar
Tacx Video Production
Posts: 3906
Joined: Sun May 09, 2004 11:49 pm
Location: Belgium
Contact:

Post by Tacx Video Production » Wed Mar 19, 2008 1:33 pm

Ive never seen this behavior with the Fortius? This is what you should have ...

http://www.tacx-video.com/Downloads/ima ... ortius.png This is the first 1hr of the Tour of Lombardy done on the Fortius.
Image

http://www.tacx-video.com/Downloads/ima ... SRMPro.png This is the SRM data from the same 1hr ride.
Image

Two very closly matching curves & matching times at the various power zones on the left side.

maybe you have something wrong with your unit? or there is a voltage issue like we have had with some of the other countrys in the past.

User avatar
Tacx Video Production
Posts: 3906
Joined: Sun May 09, 2004 11:49 pm
Location: Belgium
Contact:

Post by Tacx Video Production » Wed Mar 19, 2008 1:44 pm

qcscmh wrote:Forgetting the accuracy I find it strange that power is calculated diferrent ways in Catalyst ? surely it should be calculated the same way in all applications (VR, RLV, Catalyst ??) :? :?
Yes its a bit strange but it was a result of two seperate teams of programers doing the seperate parts. Its something Tacx is aware of & looking to fix this year.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Wed Mar 19, 2008 2:11 pm

Phil wrote:This is what you should have ...
Two very closly matching curves & matching times at the various power zones on the left side.
maybe you have something wrong with your unit? or there is a voltage issue like we have had with some of the other countrys in the past.
I've seen those pictures before and I believe it. Average power values could indeed be fairly close if you pay enough attention to thigs like calibrating the unit for the average power of your subseqent ride, if you keep riding steady enough so that brake temperature does not fluctuate so much etc.

But an overall accuracy of the system does not depend on right-prepared examples but on worst possible cases which could happen. I will provide my pictures later to show how the power reported by Fortius can be off. In Catalyst as you wish.

There is nothing wrong with my unit or my power. If you never seen Fortius reporting wrong power values when it should be ZERO than try it. You will see that this happens for every motorbrake unit in every country.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Wed Mar 19, 2008 2:18 pm

Phil wrote:
qcscmh wrote:Forgetting the accuracy I find it strange that power is calculated diferrent ways in Catalyst ? surely it should be calculated the same way in all applications (VR, RLV, Catalyst ??) :? :?
Yes its a bit strange but it was a result of two seperate teams of programers doing the seperate parts. Its something Tacx is aware of & looking to fix this year.
There is somrthing even more strange. If you claim that motorbrake measures power directly from an amount of energy sending back to the mains... how can it be different for Catalyst and VR? What you mean by "directly"? All that proves that in the process of measuring power there are lots of OTHER factors involved, not just simply measuring current you generate. There are lot of recalculations going on the way. This may introduce additional errors and you just proved that saying that in VR Fortius wrongly calculates power.

My point is that motorbrake is much less accurate when it comes to measuring power than strain gauge devices like SRM and PT. I remember that few years ago you claimed otherwise. You were wrong then.

mcorn
Moderator
Posts: 8205
Joined: Fri Nov 24, 2006 8:55 pm
Location: USA, Bellingham, Washington

Post by mcorn » Wed Mar 19, 2008 3:02 pm

Miguelon la bestia polaca wrote:If you claim that motorbrake measures power directly from an amount of energy sending back to the mains... how can it be different for Catalyst and VR? What you mean by "directly"? All that proves that in the process of measuring power there are lots of OTHER factors involved, not just simply measuring current you generate. There are lot of recalculations going on the way. This may introduce additional errors and you just proved that saying that in VR Fortius wrongly calculates power.

My point is that motorbrake is much less accurate when it comes to measuring power than strain gauge devices like SRM and PT. I remember that few years ago you claimed otherwise. You were wrong then.
I think that what Phil should have said is that the method chosen to measure power is based on current flowing back into the grid. However, as we all know, due to problems and bugs caused by the interactions between the regenerative (resistance) and motoring part of the system, at times this current (and therefore wattage) measurement is grossly inaccurate (particularly but not excclusively when going over 26 mph with 110 volt systems and over 32 mph with 230 volt systems).

The principle of measuring wattage based on current in a regenerative system is probably a sound one and probably would have worked fine without the integral motoring.

I have found that with all of our workarounds, our Fortius is within plus or minus 5% of our new PowerTap, but sometimes it is higher and sometimes it is lower, depending on the nature of the terrain in the course.

User avatar
Tacx Video Production
Posts: 3906
Joined: Sun May 09, 2004 11:49 pm
Location: Belgium
Contact:

Post by Tacx Video Production » Wed Mar 19, 2008 3:57 pm

Miguelon la bestia polaca wrote:
Phil wrote:
qcscmh wrote:Forgetting the accuracy I find it strange that power is calculated diferrent ways in Catalyst ? surely it should be calculated the same way in all applications (VR, RLV, Catalyst ??) :? :?
Yes its a bit strange but it was a result of two seperate teams of programers doing the seperate parts. Its something Tacx is aware of & looking to fix this year.
There is somrthing even more strange. If you claim that motorbrake measures power directly from an amount of energy sending back to the mains... how can it be different for Catalyst and VR? What you mean by "directly"? All that proves that in the process of measuring power there are lots of OTHER factors involved, not just simply measuring current you generate. There are lot of recalculations going on the way. This may introduce additional errors and you just proved that saying that in VR Fortius wrongly calculates power.

My point is that motorbrake is much less accurate when it comes to measuring power than strain gauge devices like SRM and PT. I remember that few years ago you claimed otherwise. You were wrong then.
Im not all that sure how the watts is calculated in the VR terrain. It has always evolved away from my eyes. At a guess they are using roller speed & break force or something like this. All I know is that the reported power has nothing in common with the watts you see in RLV/Catalyst or with a SRM, powertap etc. Im often bafled by it also.

Back in the early beta days of the fortius you had a seperate power out cable with a light. You could see the power produced by how bright the light was! It was simple case of your watts actually producing visible watts! Only friction loss had to be added to get very accurate power. Its a simple case of the energy you produce being converted into real watts. With a SRM you need to calibrate its zero setting & then you need to hang weights off the cranks & do some maths with crank lengths & pie to calculate a known torque. Its only accurate if you know what your doing. The Powertap is the same but with this you cant change its slope but just the zero setting. This can often change while out on the bike in changable weather.

Somehow they decided to drop the obviously cool light idea & go with the less obvious tech challenge of power back. in internals are the same though with the power produced being measured.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Wed Mar 19, 2008 5:08 pm

Phil wrote:Back in the early beta days of the fortius you had a seperate power out cable with a light. You could see the power produced by how bright the light was! It was simple case of your watts actually producing visible watts! Only friction loss had to be added to get very accurate power. Its a simple case of the energy you produce being converted into real watts.
Maybe it was simple and cool, maybe it was not.

The friction loss which wich should "only" be added actually makes all the difference. Calibration can measure it for one moment and after that it can change quite a bit.

Also, ever wondered why SRM or Saris produce their top ergotrainers still equipped with strain gauge powermeters? It's because ANY way of measuring (or more corretly - calculating) power from rear spinning wheel will be always worse solution than measuring torque on pedals.

Flying_Tort provided a link to SRM lab machine, and in it's technical data it is written:

The athlete’s power is measured with the SRM Powermeter Professional on the bottom bracket and with the SRM Powercontrol. The data gained with this kind of power measurement is not affected or distorted by the Eddy current brake nor by mechanical parts of the gearbox. The athlete’s power is measured directly at the point of output, on the cranks.

The "cool" solution of lighting bulb is still driven by those eddy currents. Anyway, motorbrake is much more than just a bulb so all this is not actually that important ;-)

User avatar
Tacx Video Production
Posts: 3906
Joined: Sun May 09, 2004 11:49 pm
Location: Belgium
Contact:

Post by Tacx Video Production » Wed Mar 19, 2008 6:18 pm

Miguelon la bestia polaca wrote:
Phil wrote:Back in the early beta days of the fortius you had a seperate power out cable with a light. You could see the power produced by how bright the light was! It was simple case of your watts actually producing visible watts! Only friction loss had to be added to get very accurate power. Its a simple case of the energy you produce being converted into real watts.
Maybe it was simple and cool, maybe it was not.

The friction loss which wich should "only" be added actually makes all the difference. Calibration can measure it for one moment and after that it can change quite a bit.

Also, ever wondered why SRM or Saris produce their top ergotrainers still equipped with strain gauge powermeters? It's because ANY way of measuring (or more corretly - calculating) power from rear spinning wheel will be always worse solution than measuring torque on pedals.

Flying_Tort provided a link to SRM lab machine, and in it's technical data it is written:

The athlete’s power is measured with the SRM Powermeter Professional on the bottom bracket and with the SRM Powercontrol. The data gained with this kind of power measurement is not affected or distorted by the Eddy current brake nor by mechanical parts of the gearbox. The athlete’s power is measured directly at the point of output, on the cranks.

The "cool" solution of lighting bulb is still driven by those eddy currents. Anyway, motorbrake is much more than just a bulb so all this is not actually that important ;-)
So this is good, we are now talking about just the acuracy of calculating friction/wheel drag/drive chain loss from the spin down test. We have shown as in the charts Ive posted show that the Fortius can be very close to a 'calibrated' SRM Pro. Yes I will admit that if you can calibrate a SRM Pro well enough the ideal place to measure power is the crank.


User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Wed Mar 19, 2008 6:33 pm

Phil wrote:So this is good, we are now talking about just the acuracy of calculating friction/wheel drag/drive chain loss from the spin down test.
So to be more precise, I'm talking about stability of the system. Friction can change since calibration, but the most important is thermal instability.

The heat-induced drift that occurs in eddy-current braking systems is well-documented, and has absolutely nothing to do with how power is measured, or even if it is measured at all.

I think this is quote from Andrew Coggan when he was discussing Computrainer on wattage group. It is different than Fortius, but both use eddy curents in braking.

BTW: Don't go to far, drive chain loss cannot be calculated from spin down test. Not at all.
Last edited by MirekD on Thu Mar 20, 2008 8:06 am, edited 1 time in total.

User avatar
qcscmh
Posts: 777
Joined: Fri Dec 30, 2005 2:57 pm
Location: UK, Hampshire (The Tropical South)
Contact:

Post by qcscmh » Wed Mar 19, 2008 9:08 pm

Phil, do you have any thoughts on why a diferrent power supply and grossly effect the speed on VR ?

http://www.tacxvr.com/phpbb2/viewtopic. ... highlight=

I am out of my depth on this conversation.

But dont forget I have an imagic system not Fortius.

Is the method of power calculation the same for imagic and Fortius ?
Nothing interesting here > http://bikestuff.freehostia.com/

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Thu Mar 20, 2008 12:28 pm

Phil wrote:Wohoo your talking about the VR world here??? The power in this is wrong & I cant deny this. Its much more like the old i-magic software. Anyone with a power meter will see this very quickly. Tacx is working on a solution to this.

Im talking about Catalyst which is what Real life & Ergo Video is built on. This is the part of the software used for testing. The way it calculates power is very accurate. Once the slope (real world averaged power from 0 up to your Vo2 max power) you should be cool. The % of error should be very low. We want it so you can adjust the slope while riding to make it even simpler.

The VR worlds uses a compleatly different formula for calculating power to Catalyst & RLV.
I have no reason to doubt what you wrote, but OTOH no reasons to believe you cos in my rides I see no difference between power registered in VR and RLV.

I checked many of my recent rides, both VR and RLV and they are all pretty the same. The power profile/curve derived from both softwares are nearly identical. Any differences are statistically non-important.

Like this:
http://www.virtualcycling.org/Applicati ... px?id=4501
and
http://www.virtualcycling.org/Applicati ... px?id=4637

RLV ride: 40km, 59 min, av power 393W, av HR 163
VR ride: 52km, 1h 12 min, av power 387W, av HR 166

Both rides done at FTP-like effort, less than 2 weeks between them. I have many more examples proving, that (at least for me) there is no difference between power in VR and RLV.

User avatar
Tacx Video Production
Posts: 3906
Joined: Sun May 09, 2004 11:49 pm
Location: Belgium
Contact:

Post by Tacx Video Production » Thu Mar 20, 2008 1:47 pm

hmm send me the files, there will be differences unless they changed things already. Guess Im going to have to find space to set up the trainer again. (Im living out of boxes just now with a dam slow drawn out house move).

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Thu Mar 20, 2008 2:34 pm

All my rides (hundreds of them) are available for download from VCF site, those particular two are here:
http://www.virtualcycling.org/Applicati ... px?id=4637
http://www.virtualcycling.org/Applicati ... px?id=4501

But I'm not sure if those can help you in any way. What differences will you be looking for?

All I say is that if ride for a given period of time with a given level of effort I see no difference in power registered in VR and RLV software. The above example is for about 1h ride treshold effort. If I do 20 min or 5 min or 1 min max effort I get the same wattage values too. There is no repeatable pattern to be spotted like say x % difference between RLV and VR.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Post by MirekD » Thu Mar 20, 2008 8:27 pm

Phil wrote:If your the Polish 30+ National champion then how can you say the power is completely wrong with the Fortius? Looking at your data in the wattage thread its spot on what you would expect from such a rider no matter the weight for those durations. I was guessing your power was off or you were semi Pro or a very good amateur.
Phil, I never said that "power is completely wrong with the Fortius". All I say is that it can be off for 5-8% for majority of riders and even more for those who deliver more power overheating motorbrake. My real world FTP is 365W, PowerTap measured (not corrected for chaindrive losses).

Those two rides are about the difference between VR and RLV when it comes to (average) power registered. I just see almost no difference.

Also, I'm not complaining about inacurracy of motorbrake. It's not about Tacx, nothing personal too. It's just the thing that all electromagnetic trainers suffer from power drift and in some cases, like FTP testing (yes, I'm referring to the topic of this thread) it can lead to errors few times greater that errors usually introduced by strain gauge devices.

If you combine this with the great second post from markmergler1970 - there are two good reasons which should be considered before testing on Fortius.

Flying_Tort
Posts: 21
Joined: Sun Mar 11, 2007 4:21 am
Location: Canada, Ontario

Five Second Power

Post by Flying_Tort » Sat Mar 22, 2008 7:16 am

I tried to build a couple of testing protocol on Catalyst including a 5 second test and one minute power in order to determine power profiles in reference to Hunter and Coggan's charts.

It seems that the smallest period of time the Catalyst and Analyser allow for is 10 seconds.

Is there something that I am missing, or does the software not allow for smaller increments than 10 seconds when creating a Catalyst program or trying to review a brief section on Analyzer.

Does anyone have any thoughts?

Has anyone tried doing these tests? and if so, what were your results as far as watts/kg?

rocketrod
Posts: 44
Joined: Sun Jul 01, 2007 10:39 am

problems with doing FTP testing

Post by rocketrod » Sat Mar 22, 2008 8:08 am

Here is the main problem with timed 1 hour testing. Most people
can't tell the diffence between their 1 hour speed, their 2 hour speed, or
their 10 minute speed. So the most common mistake is they start out
at their 2 hour speed and do this for 50 minutes, then they realize there is 10 minutes left so they go at 10 minutes pace for the last 10 minutes.

Or they might start out at their 10 minute pace, and then after 7 minutes they start to get tired so they slow down to their 2 hour pace, then when they get close to 1 hour, they speed up againto 10 minute pace.

That is why the stepped heart rate is the best. If you go to the www.Saris.com website Allen Lim and Robby Ventura give an example on how to determine FTP. They use the RPE and ventilary threshold method, just like Lance did. You gradual increase pace slowly like increase every five minutes, and just when you start to have trouble speaking due to involutary breathing that means you are approaching anaerobic threshold. But if you are just gasping for air and your eyes are going cross eyed and everything is going dark, that means you are way beyond anaerobic threshold.

Here is an example way. Suppose you start out at 150 watts for 5 minutes, increase to 170 watts for 5 minutes, then say you reach
240 watts and you just start to breath for five minutes, then you try 260 watts and all of a sudden you start gasping and your legs feel like bricks.
That means your threshold is closer to 240 watts. After a while you can tell when you are bellow threshold and when you are above threshold.
Monitoring heart rate is a more sensitive test than looking soley at power for threshold since heart rate seems to be more related to the levels of lactic acid in the body while power can be affected by your glycogen, ATP short term systems. For example it is possible to put out a lot of power say 500 watts for a short time but the lactic acid levels don't reach threshold, and to go above lactic threshold levels, and start to pedal with low power but the heart rate and lactic acid levels still are high. So power in itself is not a good indicator of lactic acid levels, but heart rate is.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Re: problems with doing FTP testing

Post by MirekD » Sat Mar 22, 2008 9:22 am

rocketrod wrote:Here is the main problem with timed 1 hour testing. Most people can't tell the diffence between their 1 hour speed, their 2 hour speed, or their 10 minute speed. So the most common mistake is they start out at their 2 hour speed and do this for 50 minutes, then they realize there is 10 minutes left so they go at 10 minutes pace for the last 10 minutes.

Or they might start out at their 10 minute pace, and then after 7 minutes they start to get tired so they slow down to their 2 hour pace, then when they get close to 1 hour, they speed up againto 10 minute pace.
It is a problem indeed to maintain relatively steady power output during 1h testing especially if this supposed to be your max level. But even if you will swing - there is a concept of normalized versus average power which could help you to determine FTP. But still you should really do your best to exhaust yourself completetly during that 1h.

It is all the same as 1h flat TT strategy. Dont go all out at the very beginning. Try to find your max steady power output after few minutes and hold on to it.
rocketrod wrote:That is why the stepped heart rate is the best. If you go to the www.Saris.com website Allen Lim and Robby Ventura give an example on how to determine FTP...
This is tricky. They talk about "treshold power", "Steady State Power Fitness Test", "maximum steady state" different "lactate tresholds" based on lactate concentrations levels etc.

I can't find the term FTP there. However "functional treshold power" and "treshold power" sound similar they are not necessarily the same thing. I already written that FTP is the term given by Andrew Coggan and he gave it the precise definition.

Allen Lim is Ph.D. physiologist too, but they use different approaches to training with power. I'm not saying which is better, who am I anyway, but it's very risky to mix their concepts. Like deriving FTP concept from Coggan, going for testing to Lim and going back to Coggan's power profile tables. The best thing is to choose one ot them and stick to one concept only. Especialy if you want to go further like TSS concept, cyclingpeaks software etc.
rocketrod wrote:So power in itself is not a good indicator of lactic acid levels, but heart rate is.
Well, this is something A. Coggan would argue with ;-) In his general approach to training with power - only power matters actually. HR and lactate levels are dependant variables. They only show your physiological response to training load. Power measures the training load itself therefore is independant variable.

These kinds of duscussions are endless on wattage google group. If you want it more - go there and check it out.

User avatar
MirekD
Posts: 202
Joined: Thu Dec 30, 2004 8:03 pm
Location: Poland

Re: Five Second Power

Post by MirekD » Sat Mar 22, 2008 10:08 am

Flying_Tort wrote:I tried to build a couple of testing protocol on Catalyst including a 5 second test and one minute power in order to determine power profiles in reference to Hunter and Coggan's charts.

It seems that the smallest period of time the Catalyst and Analyser allow for is 10 seconds.

Is there something that I am missing, or does the software not allow for smaller increments than 10 seconds when creating a Catalyst program or trying to review a brief section on Analyzer.

Does anyone have any thoughts?

Has anyone tried doing these tests? and if so, what were your results as far as watts/kg?
You're right about 10 sec in Catalyst.

And I tried many different tests in Catalyst, both for imagic and fortius.

When it comes to Coggan power profiles - they are cool, right? ;-) I was excited about W/kg too and especially about those categories. Am I good enough or not? ;-)

But the most important thing is not about W/kg and categories, what really matters is the SHAPE OF THE PROFILE cos it tells you lot about you, what kind of rider you are, what are your strenghts and weeknesses etc.

5 sec. power is important as it determines your sprinting abilities.

Unfortunately, 5 sec. max power and sprinting in general is something to hard for electromagnetic trainers like fortius. They are not like Wingate machines.

You can try to test it without bothering about 10 sec. Catalyst segment limit - even with 0 slope. Try to accelerate slowly to the point which would be your "flying start". Prepare the right gearing as your cadence will incerase. Then go all out! But not for 5 sec, try at least 15-20s. You will accelerate during first few seconds, then you will reach max speed. Hold it until you fade out.

Save your workout, open it in Analyzer and there you are. Find the 5 sec period where your power was at max.

But the 5 sec max power I ever reached on Fortius was 25-30% lower than real max power measured outside. This is due to the lag in measuring (Phil already admitted that) and the simple thing that sprinting indoors will never be the same as outdoors.

Now, 25-30% error is HUGE and when it comes to real Coggan's power profile table it can ruin it completely. So I see no point of performing test like this on Fortius.

Post Reply

Return to “Wattage”

Who is online

Users browsing this forum: No registered users and 2 guests