Using PowerStroke with DFPM as cadence sensor

Post Reply
Erik W
Posts: 2
Joined: Tue Jun 23, 2015 3:20 am

Using PowerStroke with DFPM as cadence sensor

Post by Erik W »

I have a racing bike with a Newton 5 and a power2max DFPM mounted on it.
There is no traditional cadence sensor with a magnet on a crank arm.
The Newton 5 in "normal" modes seems to read and display the cadence data from the DFPM just fine.

My Newton 5 has the PowerStroke option enabled, but so far I have not used this function.

I was now about to start out investigating the PowerStroke possibilities, but came to think of one thing right away:
As I understand it PowerStroke needs to know the crank arm angular position to do its analysis, hence also the need to define the position of a "normal" cadence sensor.
How will that work with a DFPM (if it at all works?) which is "undefined" in terms of crank arm absolute position.
I'm not even sure how the cadence data is transmitted from a "cadence" ANT+ sensor, is it a "beep" every time the magnet passes the sensor, or is the angular velocity calculated and transmitted?
Velocomp
Velocomp CEO
Posts: 7793
Joined: Fri Jan 11, 2008 8:43 am

Re: Using PowerStroke with DFPM as cadence sensor

Post by Velocomp »

If you plan to use PowerStroke with your DFPM, just start your ride with the crank in the same position. There is likely a magnet that is used for cadence in the P2M, but I am not sure.
John Hamann
Pete
Posts: 110
Joined: Wed Mar 05, 2014 8:09 am
Location: Melburn, AUS

Re: Using PowerStroke with DFPM as cadence sensor

Post by Pete »

Unlike most crank-based power meters on the market today the Power2Max does not require a cadence magnet. Instead, the unit includes an accelerometer to automatically determine your cadence. In doing so, it’ll then transmit that cadence information to your head unit, which will display it in the appropriate data fields (typically cadence). As well as record it for later reference:
Velocomp
Velocomp CEO
Posts: 7793
Joined: Fri Jan 11, 2008 8:43 am

Re: Using PowerStroke with DFPM as cadence sensor

Post by Velocomp »

Since they're using an accelerometer to measure cadence, then presumably they have some trigger point for starting a new cadence "revolution".

I would look around on a P2M forum (assuming there is one) to try to find out how their cadence algorithm works.
John Hamann
EHB
Posts: 129
Joined: Sun Oct 26, 2014 5:52 pm

Re: Using PowerStroke with DFPM as cadence sensor

Post by EHB »

I joined the ANT developer network. It allows you to read each of the protocols. I won't bore you with why.

Essentially, whatever it the protocol is, be it HR, Power, cadence etc., there are eight chunks of data. In the cadence protocol timing data is sent. However, in the Power protocol the space is used up for other stuff and it doesn't send this information. It just sends a cadence number (or PowerTaps send speed).

I suspect Powerstroke needs that info. I bet the Newton could guess where the cranks are using its accelerometers, but as the Isaac software requires to know where the magnet is I doubt it works that way. Velocomp will be able to confirm.

I too have a Power2Max and haven't ever tried to use it for cadence for Powerstroke. Let me know if you get to work as I can get rid of a sensor. The Wahoo and Garmin magnet less sensors both use accelerometers, but transmit using the more detailed cadence protocol. I'm going to play with the Wahoo one tomorrow and see if it works. I know when in the rotation they trigger as the guys there sent me a photo.
Erik W
Posts: 2
Joined: Tue Jun 23, 2015 3:20 am

Re: Using PowerStroke with DFPM as cadence sensor

Post by Erik W »

EHB wrote:I joined the ANT developer network. It allows you to read each of the protocols. I won't bore you with why.

Essentially, whatever it the protocol is, be it HR, Power, cadence etc., there are eight chunks of data. In the cadence protocol timing data is sent. However, in the Power protocol the space is used up for other stuff and it doesn't send this information. It just sends a cadence number (or PowerTaps send speed).

I suspect Powerstroke needs that info. I bet the Newton could guess where the cranks are using its accelerometers, but as the Isaac software requires to know where the magnet is I doubt it works that way. Velocomp will be able to confirm.

I too have a Power2Max and haven't ever tried to use it for cadence for Powerstroke. Let me know if you get to work as I can get rid of a sensor. The Wahoo and Garmin magnet less sensors both use accelerometers, but transmit using the more detailed cadence protocol. I'm going to play with the Wahoo one tomorrow and see if it works. I know when in the rotation they trigger as the guys there sent me a photo.
I work as an electronics developer so I might also be inclined to join the ANT developer network - in my case it would be just because I'm curious as to how these devices communicate.

Interesting information about the ANT+ protocol. I assume then that the power2max DFPM sends cadence data as part of the Power data packets then?
I kinda assumed that the P2M would send two packet types , one for Power and the other for Cadence, but I have not studied the ANT+ protocol i detail so I guess I was wrong about that.

Anyhow, from what I've read an accelerometer based cadence sensor should be able to sense crank position so depending on how the data is sent through ANT+ the PowerStroke analysis could work.
Yesterday I did a 12 km TT which was recorded on Newton on my bike with a P2M crank and a "speed only" magnet based sensor.
I downloaded the ride on ISAAC and clicked the PowerStroke icon, and it completes the analysis with what seems to be meaningful output.
I'm not sure whether Newton/ISAAC is able to detect that the cadence data came from the P2M, or if it is able to detect that there is only RPM data without "timing" or what ... ?
I also need to study more the details of the Powerstroke analysis itself to fully understand the conclusion.
ronpei
Posts: 120
Joined: Sat Jun 08, 2013 7:47 am

Re: Using PowerStroke with DFPM as cadence sensor

Post by ronpei »

I assume the Garmin magnetless (is that a word?) cadense sensor is probably an accelerometer as well. They told me that the revolution starts from where ever the sensor is when it wakes up. So for it to work perfectly with Powerstroke you would have to let it go to sleep with it aligned with the non drive side chain stay. For the P2M I would say the same approach would work since it is likely the same idea, the RPM cycle starts where ever the pod is when it wakes up.
EHB
Posts: 129
Joined: Sun Oct 26, 2014 5:52 pm

Re: Using PowerStroke with DFPM as cadence sensor

Post by EHB »

I was going to copy and paste from the power profile, but the thisisant.com website is down. This is quite irritating as I can't check what I was saying and prove it. I have however looked into it previously and decided it wasn't possible.

In essence the power meter transmits the power profile. This can include a number for cadence and speed, but does not include a timing number.

A separate magnet less garmin cadence sensor will transmit using the cadence profile, which does include the timing number.

Even if a power2max does count cadence from the same point as a garmin magnet less sensor it doesn't send out the timing signal. Incidentally the Wahoo cadence accelerometer sends out the timing from when the word "wahoo" is upright, so in essence most likely when the non-drive crank points directly forward. The Wahoo people helped me out by telling me that, although I haven't tested it yet. They have sent a photo, which I would post but the forum does not allow. Keeping the cranks in the correct position till starting cycling to use the garmin sensor is comparatively a pain. It also does Bluetooth so it can be used in the gym with most phones. I've already tested it in a 70.3 and it worked well. Last year I did a full Ironman with a standard S&C sensor, but had no bike computer as the magnet got banged out of place during transition....

What exactly a Newton does if there is no timing signal I don't know. That the software wants to know whether the sensor is above or below the Chainstay suggests to me it is important. Velocomp may be able to help us there…

The Ant developer network is free (when it works) and allows you to download the developer tools. If you've got an ANT USB stick you can record the raw data from Ant devices. I think it is eight simultaneously with the newer USB sticks/chips. It is not very well designed for tinkerers though. Bit disappointing how many things transmit at 250ms intervals, but exactly the same data for four cycles. I used it to prove a garmin 910XT wasn't receiving data that I could easily record cycling around with a laptop on my back. The software is PC only but I just created a windows virtual machine on a Mac and it worked fine.

FWIW, my setup, which appeared to work great until I crashed, for the 70.3 Triathlon was
1) Newton+
2) MIO Fuse wrist Heart rate Monitor (recorded HR during the swim!)
3) Power2Max DFPM - no magnet to knock
4) Wahoo cadence sensor - no magnet to knock
5) Garmin magnet less speed sensor - no magnet to knock

When I was running with my bike during transition I just hit the Newton's centre button before I got to the mounting line. It did everything else.

The one thing that didn't work was that my running watch didn't lock onto GPS in the transition tent and wanted me to stop running outside, although it eventually gave up pestering me. Not at all related to the Newton that though.
EHB
Posts: 129
Joined: Sun Oct 26, 2014 5:52 pm

Re: Using PowerStroke with DFPM as cadence sensor

Post by EHB »

The thisisant.com website is back up and running again.

The standard power profile 'page' includes 'instantaneous cadence'. There is no timing data sent. Having read some of the alternative profiles it does appear that that a power meter transmitting torque rather than power can do so at a timing event, so I could see how you could use when the data is sent to time when the cadence timing event happens. Would have to be really quick though and the Power2max is always 2-3 seconds behind.

In contrast the standard cadence 'page' includes 'Bike Cadence Event Time' as two Bytes. The same data is sent from combined Speed and Cadence sensors.

I'm using the word page as that is what their documents call it. Essentially there is a standard page and then extra or alternative versions (eg. One every now and again showing manufacturer info of battery status etc.) each of which is also called a page.
james f
Posts: 3
Joined: Tue Mar 01, 2016 6:46 am

Re: Using PowerStroke with DFPM as cadence sensor

Post by james f »

Got various new concept about cadence sensor.
Post Reply