John Galea's Blog

Just another WordPress.com weblog

Training with a heart rate monitor

I’m a bit of a weekend warrior, I like to mountain bike. And I have always looked for ways to ensure I am at least maintaining my cardio endurance and hopefully increasing it as the season goes on. For a long time I have used a heart rate monitor to help this process. There are lots of kinds of heart rate monitors that I have covered over time on this blog. In the end THE most accurate heart rate monitor is a chest strap. It also as it happens, tends to be the least comfortable. When you are pushing up a hill breathing heavily the strap around your chest restricts your breathing. And sometimes the chest strap can come loose and slide down.

At this point I have three heart rate chest straps I use:
1) Polar H7 chest strap that broadcasts only on Bluetooth low energy. This can talk to my Polar A300 for excellent data recording, or to my iPhone running whatever app you want. Polars own app called Polar Beat can be used to send the bluetooth data to both Polar Beat and the Polar A300 watch. It’s reasonably comfortable, and for the most part stays in place. It runs on a replaceable CR2032 and from within some apps you can see the battery level of the heart rate monitor (I haven’t had this long enough to comment on battery life). Accuracy is good as long as you properly wet the contacts. The electronics snap into two clips on the chest strap. They seem firm enough to hold it in place.
polar-h7-heart-rate-sensors-set-98309-en

2) Wahoo TICKR chest strap that broadcasts on both Bluetooth low energy and Ant+. I love this flexibility and it can broadcast to all of my Garmin devices as well as to my phone at the same time, or my Polar A300. Ant+ can broadcast simulataneously to as many devices as you wish. The Wahoo TICKR is oddly designed in that electronics clip into the middle of the strap meaning the electronics are being pulled by the strap. The TICKR is by far the most comfortable of the chest straps I’ve used and stays nicely in place. It run on replaceable CR2032 battery and you can tell the battery level from a number of apps on the phone (I haven’t had this long enough to comment on battery life). Via the Wahoo Fitness app you can even update firmware level. This is the first time I have ever seen this.
Wahoo-TickrX-2

Update 6/13/2017
My battery was down to 15% after a year! Impressive. Changing the battery takes a bit of patience and finesse. Whatever you do, don’t do it in a rush as I did. First off there is small rubber O ring that creates a moisture seal (for sweat and rain etc). Be careful with it. Second there is a small arrow on the back side, you have to align it with the unlock when putting the new battery back in. Then turn to the lock and your done.

3) Garmin Ant+ chest strap. Now that I am on an iPhone this is of little use (My Samsungs had the ability to receive Ant+). It’s always been accurate anytime I have run correlation runs. It’s also the least comfortable, but is well designed and rarely moves on the chest at all. It runs on a replaceable CR2032 battery and gets something like a year battery life. It interfaces best with Garmin devices, which as it happens are my favorite bike gadget.
5018-949_NOC02_view1_720x720

I also have a Scosche Rhythm+ heart rate monitor that goes on your arm and picks up the heart rate optically. This is by far the most comfortable of the heart rate monitors I use and my favorite. The Scosche broadcasts on both Ant+ and bluetooth low energy so is very flexible in terms of what it can talk to. It runs off a rechargeable battery and gets 6-8 hours of battery life.
rhythm_color_band_bk_1000

And now comes the hot topic of accuracy. What is accurate enough? What are you wanting to do with the heart rate data? These are personal choices. When I first started using a heart rate monitor I used it only to get a more accurate count of the calories I burned on a ride to ensure I was working on endurance. Calories burned is a simple calculation based on length of exercise and your average heart rate. If this is all you care about then you ought to go for the most comfortable heart rate monitor because as you will see in a bit accuracy at each and every data point is not all that important in that it does not dramatically effect the average heart rate, and so does not effect calorie counts. Each app calculates calories with their own magical formula. Comparing calorie counts between devices, or between apps is frivolous because you have no idea or control on what algorithm it uses. So choose something (an app, a device whatever) and just stick with it. One of the things I do like to have is a complete picture of my entire exercise in a week. So if you wear an activity tracker (I have a Fitbit Blaze and Charge), you will need to see what data inter operability options there are. Preferably automatic. I use Garmin connect, Fitbit and Endomondo. They all share data to one degree or another and ends up with a complete picture of my exercise in one place, well actually two Endomondo and Fitbit. Be careful if you use multiple devices/apps that your not double counting your exercise. Generally speaking these can be cleaned up manually if need be.

I started to have some suspicions that my Scosche was becoming inaccurate which frankly was what prompted me to look at this topic again. So I started out with an over 3 hours bike ride and used my Polar H7 logged by my Polar A300, as well as my Scosche logged by my Garmin Edge 305. Using dedicated devices to do the logging (instead of a phone app) gives you much better data for crunching numbers. Apps like Endomondo are less precise about how often data is logged. For example during a 3 hour ride I saw sampling rates averaging once per 3.8 seconds (Vs once per second like clockwork for the Garmin, and Polar A300) and at worst when the phone was busy doing god only know what of 17 seconds. Interestingly enough it makes little to no difference on the average heart rate, and thus no difference on the calorie count. If you were using an app to alert you on max heart rates then this might be something you want to worry about. I saw similar sampling rates for Endomondo running in the foreground and backup on iOS and on Android by the way.

scosche-polar
As you can see after an initial period the two tracked reasonably well. Crunching the data showed the two were within 10% of each other 97% of the time. That’s not bad correlation. And in the end it only effected the average heart rate by 2.4 beats per minute.

It’s worth noting that the Scosche Rhytm+ can not be used for HRV readings.

Calorie count is just one reason to wear a heart rate monitor. More serious athletes keep an eye not necessarily on the BPM but the zone your heart rate is in. Now being off by 10% at 180BPM doesn’t sound too bad but translate into BPM and thus into a heart rate zone and you have a bigger issue. Looking at the same data and changing the threshold to how often was the heart rate off by 10 BPM and you get a more troubling 10% of the time. And this is with the first 15 mins out of the data.

Another reason to keep an eye on heart rate is to watch to insure you are staying within a min/max. I use a heart rate alarm on my Edge 305 to insure I don’t spend too much time maxing out my heart rate. I either slow down, control my breathing or flat out stop. Being off by 10% again makes this difficult. Using the alerts on the 305 I have managed to lower my max heart rate by 20-30 BPM which has to be healthier. Similarly using the low alarm you can remind yourself when you are dogging it 🙂

So now let’s have a look at a comparison of my Polar A7 logged by the Polar A300 and my Wahoo TICKR logged by my Garmin Edge 305. Now this is impressive.
polar-wahoo
Now this is what I call correlation. In the whole 3 hour ride the two were only off by 10% a mere 9 data points (seconds) in over 3 hours. Impressive. And if I change the threshold to 10BPM (instead of %) there are only 11 data points where they differ. It’s worth noting that wearing two chest straps in a bumpy sport like mountain biking can lead to the two just bumping into each other which may explain even the minor differences.

So all in all, give some thought to what you want to do with the data, and use that to choose how accurate you need your heart rate monitor to be and how much discomfort your willing to endure to get that accuracy! I love crunching number and doing data analysis …

Advertisements

July 12, 2016 - Posted by | Activity Trackers, GPS Stuff, Uncategorized

No comments yet.

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: