What's new

So I got a HR monitor

I have a new Garmin HRM3 on the way, let's see how that does.

FileServlet
 
Its nothin to do with the computer side of things Owen.

The problem I have described may be a different issue to the one you have been experiencing and let me tell you, it definitely IS a computing issue; the constantly elevated rates I have seen in my ride data contain all the correct peaks and troughs one would expect from the terrain I was riding over, just the range maximum and minimum was a level too high; any software dealing with measuring a constant flow of data such as BPM will deal with it a number of stages. First, it will be wide band scanned for a general range, then fed through to the next narrow band detection unit. This is the same for all analysis systems, be they BPM, sound frequency, audio db level, whatever. The issue I have been experiencing is not strap related, but is actually a scanning error during the first wide band analysis, which throws the stream up into a range too high. Once it there, it seems to be being processed fine (hence the normal expected peaks and troughs shapes in the data). This is a common issue that has plagued all analysis software since processors got good enough to deal with highly detailed data streams.

What I would like to see on the Garmin, to overcome this, is a range switcher like in NI Traktor which can be used to knock the wide band data stream up or down some predefined ranges. To get really controlled data, adding some sort of feature rich Gate in there, to control the max and min of the HR info (and therefore controlling the window for error which is hard coded at the moment) would be amazing. In the audio world, stuff like the Sonalksis SV-719 is a standard tool. Some kind of cut down version with the parametres renamed to make the cycling / sports friendly would be ace; renaming Threshold to something like 'Sensitivity' or even 'Correction' would be just the ticket. Thinking about it, this could also be achieved with a simple quantizeation routine, which could have a user controlled threshold also http://en.wikipedia.org/wiki/Quantization_(signal_processing)

sonalksis_sv719.jpg


Which leads me onto the next point; yeah, the HR strap may be hooky on Garmins as well, causing spikes, and drop out (which is different to the main computing problem I have described, which I have been experiencing), but with a better computing system to deal with these peaks and drop outs (almost like an HR version of an audio compressor with a user controllable look-ahead) it would be a non-issue.

So yeah, I do agree that the strap is dodgy, but I don't agree that the errors have nothing to do with the computing side of the system. The fact that the Polar strap works better, but still has problems shows that the computing engine is not perfect.

So there you go!
 
...so this has piqued my interest, so I have given it a bit more thought.

For those people who have been getting elevated BPM rates, be they for an extended time, or just a spike here or there, would you be able to post the figures you have recorded?

An average of the last 30 seconds before the spike, and the spike itself would be fine. I have an idea that there is an error with the way the software, or the strap is dealing with the P wave, all parts of the Ventricular Activation Time and then the T wave; I suspect something in the signal path is interpreting either the P wave, the Q or S part of the VAT or the T wave as an actual R part, and therefore increasing the BPM. This would not affect the overall peak and trough trend over time, as any increasing or decreasing of real BPM would result in a real increase of decrease of the frequency of all parts of the wave, but would up the detected BPM to a point where it was either sent into a higher wide band processing range or would appear as a spike.

Graphically, I suspect that parts P and T, and maybe even Q and S are being counted by the software as a full R part, and therefore causing these spikes, or falsely sending the BPM into a higher wide band processing before quantisation (if Garmin actually quantises things at all :))

SinusRhythmLabels.svg
 
For those people who have been getting elevated BPM rates, be they for an extended time, or just a spike here or there, would you be able to post the figures you have recorded?
I have no idea what you're talking about, but if it makes you happy it can't be that bad :)

The bpms on this...

http://connect.garmin.com/activity/318667707

...should have been like those on this...

http://connect.garmin.com/activity/319227900

Same activity, same equipment, same user, same intensity, different day.
 
Half-Fast Mike -- are those activities perhaps private?

You do not have sufficient privileges to view the activity with id 318667707.
You do not have sufficient privileges to view the activity with id 319227900.
I was logged into my own Garmin account at the time.
 
Right, on the 'normal' one your max HR is 159BPM. On the 'wrong' one, your max HR is 211BPM. This is approximately 25% higher than it should be (if we just say that the two exercises were identical, for rough arguments sake). My idea that the P wave and T wave were being read as a full R wave pulse does not work here, as if this were true, the BPM would be at least 50% faster (if only the P wave, for example were being read on every beat as a full R wave). So, this kind of, maybe, rules out the theory that the strap itself is causing these reading spikes.

What it doesn't rule out, and what I honestly think is happening, is my theory that the information from the Garmin strap and transmitter are being sent to the main device OK, but the first stage wide band processor is throwing the signal up a level too high, for subsequent narrow band processing.

I will continue my investigations into this and post my incredibly dry findings, as they occur to me.
 
What it doesn't rule out, and what I honestly think is happening, is my theory that the information from the Garmin strap and transmitter are being sent to the main device OK, but the first stage wide band processor is throwing the signal up a level too high, for subsequent narrow band processing.

I will continue my investigations into this and post my incredibly dry findings, as they occur to me.

The Strap will definitely not send the sampled waveform to the head unit because of ANT protocol bandwidth and power consumption constrains. All base-band processing will happen in the belt and I assume it will be all analog (more power efficient). The Strap then process the resulting binary trigger signal and encodes it into the ANT+ protocol messages (timestamp + calculated HR + ...).
 
You could get to Tamagawa by whatever's your preferred route, and then go upstream. (I'd recommend crossing it and then going upstream, crossing back somewhere around Fuchuu. The Kanagawa section of this is mostly along the "cycle road", but partly along an actual road, one that's decent enough.) Going all the way to the end of the "cycle road" and back would take you well over 60km, but if you ever do want to do it then (unsurprisingly) it's flat. Or after going some way upstream you can head westwards, and encounter some real hills. It does all require a bit of planning and memorizing (I'm poor at both). The less-thinking, more sociable alternative is to join a "Half Fast" ride; in the past, news of these has been posted here on the preceding Thursday or thereabouts.

If you're not familiar with "cycle roads", just remember that they are designed not for ease of cycling but instead as good places to practise your reaction times, etc. I very nearly pranged a kamikaze baseball kid two days ago; I said nothing (he "came out of nowhere" and I didn't even have time to shout or ring my bell) but his dad (or some adult) apologized to me.

I'm gonna go look for the fabled "cycle road" along the Tamagawa early tomorrow morning. The flat part appeals to me right now. Weather looking decent though. Thanks for the advice.
 
wexford, here is my regular training route along the Tamagawa. I join it near Komae, getting there via Setagaya dori (Route 3). The upper 11 km is the best part. From Meguro you can either head towards the river via Komazawa dori (Route 416) and then towards Komae, or head up towards R3 and follow that to Komae. There's not much of a bike path between Futakotamagawa and Komae. Between Komae and the Tsurukawa Highway bridge (R19) either stay on the road or follow the bike path once the road meets it but in that case take it easy because it's usually crowded there. Above the Tsurukawa Highway bridge is where you can go faster, but you always have to watch out for pedestrians, runners, other cyclists, people with pets, etc. There are just fewer of them than downstream and the path is wider than near Futakotamagawa. I think the busiest time is before sunset.
 
Thank Joewein. I found the cycle road this morning. Success at last. It was a tad confusing from time to time as it switched sides of the river but it got me a 60km ride today. Yeah! The way back it all made sense as I'd been down it once. Places were quite crowded but overall it was better than stopping at lights and fighting cars all the time. Great ride and I could see mountains hiding in the background so very promising.

3rd day back on the bike. Took it easy mostly. Heart rate was mostly in the 140s again. It's kinda annoying me that its not a tad lower. When I got home I had 120, sat down and it went down to about 100. Decided to take my pulse and it was indeed quite fast. I counted 95, the 500 read 96 so that looks accurate. Took me resting pulse the other day when I woke up and it was 59. Anyway. Gonna just see what happens over the next few weeks as I hopefully get fitter. Btw mostly riding a cadence of around 90 in lo ish gears. Not so fast.

Btw. Hundreds on cyclists out today.
 
Resting pulse of 59:eek: Yikes! That is high... isn't it?
 
Blob--wait till you're 61. Just checked and I'm at 67... Once upon a time (when 30) I was playing daily racquetball, it would sometimes be under 40.

on edit: Still just sitting, but now at 75bpm, and 127/84.
 
Resting pulse of 59:eek: Yikes! That is high... isn't it?
Well they say the average resting heart rate is around 70 so wouldn't say 59 is high at all. If it was 99 then yeah.

140 on a training ride is just about perfect really. I tend to try to keep mine around 120 to 140 when on training rides and 138 to 150 on training runs. I have noticed a significant jump in bmp recently now that the weather is getting hotter. Your heart rate will be lower in the winter when it is cooler.
 
Your the last of the Niigata hard men though JDD. You'll be cranking out your schedule till your 80 ill bet.



Blob--wait till you're 61. Just checked and I'm at 67... Once upon a time (when 30) I was playing daily racquetball, it would sometimes be under 40.

on edit: Still just sitting, but now at 75bpm, and 127/84.
 
My Garmin HRM went mental on Saturday. I have the 2nd generation soft strap, with the hook rather than the 1st generation which uses the transmitter to fasten the belt closed. It has been generally good, but a particular combination of circumstances seems to affect it badly.
On downhills, if I am wearing a windproof, particularly if I am only wearing a single layer under the windproof, the HR readout starts going higher and higher; almost proportional to speed. If I slow right down for a few seconds, say for a short uphill, it corrects itself, then as my speed rises again the HR rises too.
I think maybe the flapping of the windproof or the passage of the (essentially, plastic) windproof through the air might be generating a static charge which might in turn affect the readout.
Hmmmmmm.
 
My Garmin HRM went mental on Saturday. I have the 2nd generation soft strap, with the hook rather than the 1st generation which uses the transmitter to fasten the belt closed. It has been generally good, but a particular combination of circumstances seems to affect it badly.
On downhills, if I am wearing a windproof, particularly if I am only wearing a single layer under the windproof, the HR readout starts going higher and higher; almost proportional to speed. If I slow right down for a few seconds, say for a short uphill, it corrects itself, then as my speed rises again the HR rises too.
I think maybe the flapping of the windproof or the passage of the (essentially, plastic) windproof through the air might be generating a static charge which might in turn affect the readout.
Hmmmmmm.
You are right, Alan. It's a pretty common occurrence. These HR monitors are temperamental things.
I wore my modified polar wearlink strap with my garmin edge 510 on Saturday's ride and the strap worked flawlessly for all 10 hours (the edge unit lost all ANT+ signals twice but that wasn't the strap's fault).
 
I noticed that in winter as well, If i would unzip my jacket and let the wind in a bit, Suddenly the thing would go mental.
 
Back
Top Bottom