Pm3 Heartrate Average

[[old] ehagberg]

I'm looking at the heart rate data that the PM3 has logged in my logcard and notice that there are a number of places where it's logged numbers that I know I couldn't have had - even at rest - like 33 or 34 bpm.<br /><br />I know that my chest strap seems to lose contact a lot - maybe I don't sweat enough, or I'm so bony that the electrodes don't meet the skin all the time - but shouldn't those periods of lost contact (no HR) be getting thrown out when coming up with the average hr for a workout and its splits?

[[old] malor]

I always run the heart rate monitor's contacts under the faucet before putting it on to improve contact. You might try that and see if it helps. I'm not familiar with what the PM3 logs so I can't answer you question about that.

[[old] ehagberg]

<!--QuoteBegin-malor+Mar 6 2005, 01:08 AM--><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><div class='genmed'><b>QUOTE(malor @ Mar 6 2005, 01:08 AM)</b></div></td></tr><tr><td class='quote'><!--QuoteEBegin-->I always run the heart rate monitor's contacts under the faucet before putting it on to improve contact.  You might try that and see if it helps. <br /> </td></tr></table><br /><br />Yeah, I've done that too. I spit on the contacts as well. Both seem about equally effective (actually, the spit seems to work better, as it doesn't run off the contacts as easily). I still experience dropouts. The stiff plastic strap just doesn't bend around my chest in a way that allows it to keep contact over most of the electrodes - there's always significant space between some part of the strap and my chest.<br /><br />Maybe if I got one of the newer fabric polar chest straps I wouldn't experience the dropouts so often.<br /><br />Of course, the PM3 still shouldn't count the dropped periods at all, which is the problem that seems like it could be solved in the firmware.

[[old] psaint]

<br />[/quote]Of course, the PM3 still shouldn't count the dropped periods at all, which is the problem that seems like it could be solved in the firmware.<br />[/quote]<br /><br />When my HRM loses contact and the PM3 goes blank I get a zero at that spot in the log for the split. I can't say what your problem is but it doesn't seem to be the PM3.

[[old] ehagberg]

[quote=psaint,Mar 6 2005, 01:54 PM]<br />[/quote]Of course, the PM3 still shouldn't count the dropped periods at all, which is the problem that seems like it could be solved in the firmware.<br />[/quote]<br /><br />When my HRM loses contact and the PM3 goes blank I get a zero at that spot in the log for the split. I can't say what your problem is but it doesn't seem to be the PM3. <br />[/quote]<br /><br />I get a zero sometimes and other times I get those ridiculously low numbers. The low numbers never show up on the PM3, except in the data stored to the logcard, so I don't believe those low numbers are ever transmitted to the PM3.<br /><br />Which is why the PM3 must be doing some bogus math/interpolation somehow.

[[old] Porkchop]

I think that there are really two different problems here. <br /><br />First, the best way to make sure that the monitor contact is good is to use contact gel (and, of course, make sure the strap is quite snug). It can be found at any bicycle shop that sells heart rate monitors. It is similar in consistency to the gel that is used for ultrasound examinations of pregnant women. It works like a champ -- I very seldom lose contact.<br /><br />Second, as best I can tell, the "average" heart rate shown on a log card at the end a row is really the average of the heart rates shown at the end of each split. If your heart rate is significantly higher or lower during the split (i.e., between the times that the heart rate is recorded), then the "average" won't reflect reality. If you happen to lose contact at the end of a split (it's happened to me only once), the PM3 takes it as a zero, and you end up with an "average" heart rate that is proportionately lower than it should be. (The one time it happened to me, I "reverse engineered" the average, and came up with a much more appropriate, believable result.)<br /><br />Porkchop

[[old] ehagberg]

<!--QuoteBegin-Porkchop+Mar 6 2005, 05:21 PM--><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><div class='genmed'><b>QUOTE(Porkchop @ Mar 6 2005, 05:21 PM)</b></div></td></tr><tr><td class='quote'><!--QuoteEBegin-->Second, as best I can tell, the "average" heart rate shown on a log card at the end a row is really the average of the heart rates shown at the end of each split.  If your heart rate is significantly higher or lower during the split (i.e., between the times that the heart rate is recorded), then the "average" won't reflect reality.  If you happen to lose contact at the end of a split (it's happened to me only once), the PM3 takes it as a zero, and you end up with an "average" heart rate that is proportionately lower than it should be.  (The one time it happened to me, I "reverse engineered" the average, and came up with a much more appropriate, believable result.)<br /><br />Porkchop <br /> </td></tr></table><br /><br />That's not quite it.<br /><br />I'm talking about the split numbers being 33 or 34 - numbers that haven't ever shown up on the display. How did those come about?<br /><br />I think it would make more sense in any case if the splits and the total workout HR numbers that get logged were in fact the averages for each split and the overall average HR, rather than a number taken at the end of the split... with any zeroes thrown out when taking the averages.

[[old] Porkchop]

<!--QuoteBegin-ehagberg+Mar 6 2005, 04:28 PM--><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><div class='genmed'><b>QUOTE(ehagberg @ Mar 6 2005, 04:28 PM)</b></div></td></tr><tr><td class='quote'><!--QuoteEBegin--><br />*    *    *<br /><br />I'm talking about the split numbers being 33 or 34 - numbers that haven't ever shown up on the display. How did those come about?<br /><br /><br />I think it would make more sense in any case if the splits and the total workout HR numbers that get logged were in fact the averages for each split and the overall average HR, rather than a number taken at the end of the split... with any zeroes thrown out when taking the averages. <br /> </td></tr></table><br /><br />I can't figure out the reason for your first point, either. It sounds like a software error to me.<br /><br />I agree entirely with your second point. As I understand it, the more sophisticated and expensive Polar models do something like that, but I suspect that the PM3 simply doesn't have that capability. I wonder whether it has the processing capacity to add the feature without affecting other features. I suspect that the operating assumption when writing the PM3 software was that most longer rows would be more or less steady-state efforts. I frequently row using protocols that include either fartlek-type intervals or sprint intervals, so my "average" is probably quite inaccurate on a lot of occasions, depending on where in any particular split my maximum effort happens to fall. <br /><br />Maybe someone from C2 could clarify????<br /><br />On the other hand, I'm not sure that it really matters very much. Personally, I don't actually do much sophisticated analysis with my heart rate numbers, anyway. I suppose a "true" average would provide a basis to compare similar workouts to determine whether there has been an improvement in "true" cardiovascular fitness over time or to compare dissimilar workouts to determine which was more "challenging." It might also provide some information about recovery times. For most of us, though, I would guess that is much more information than we actually need or would use.<br /><br />Part of your observation is simply a matter of rising expectations from technology. In my youth, my cross-country coach simply told us: "Today, I want you to go run 6 miles. I'll see you in the locker room after you get back." My team (no thanks to me) did pretty well even without knowing our heart rates. Later, people were spending a lot of time taking carotid pulses during and after exercise and trying to calculate percentages of 220-age in their heads while walking around panting and trying to count their pulses at the same time. Now, we have instant access to our heart rates, and still we want more -- contemporanous information is not enough. We want instantaneous analysis as well.<br /><br />Me -- while I occasionally get frustrated, I'm still in awe that I can get any information at all. (And Mrs. Porkchop still programs the VCR for me when I want to record something.) <br /><br /><br />Porkchop

[[old] BigFatFishy]

I haven't used the PM3 HR monitor attachment, but I'm a big HR monitor user and here's what I have observed using one on the Erg.<br /><br />I row on a Model C with a PM2 so I use my own Polar model 410 with the thick plastic chest strap. Since it's a pain to try to look at your wrist while you're rowing I tried takine the watch off and attaching to the handle. That didn't work so well, so I took the watch off the handle and put it on the Erg itself, between my feet. It did not read my heart rate accurately at all this way. Watching it as I rowed, I noticed that the HR would drop as I pushed away on a stroke, and speed up as I returned. I think it was missing the beats while I was farthest away, even though it was only the length of my legs.<br /><br />So I think the issue may be that the chest straps don't transmit very far, even to the point where if the watch is any farther away from the chest strap than your wrist is then you won't get an accurate reading. This could affect the receiver you hook into the PM3 in the same way.

[[old] ehagberg]

I'm not getting slower readings on the monitor.<br /><br />I either get what appear to be reasonable numbers, or the signal drops out.<br /><br />I'm just asking that the times when the signal is dropped shouldn't be counted toward the average HR numbers. The "zeroes" in the data must be getting used, as I'm seeing numbers like "34" that never have appeared on my monitor - the only way that number could be arrived at would seem to be by averaging-in the zeroes as well as all the valid HR datapoints.<br /><br />It would help if someone from C2 could comment on this thread.

[[old] BobD]

My PM3 (Mod C) apparently loses the signal. The HR goes blank or only shows a heart symbol. My Polar watch (M52) contiues to receive the signal and tracks my HR. I am 185cm (6ft, 1in) and have long arms. Is my PM3 sensor too far to the rear perhaps?

[[old] c2jonw]

Currently, if the PM3 senses an unusually rapid change in heartrate value it should blank out on the PM3 display. If you happen to be at the end of a split or interval when the hr is blank on the PM3, then a blank hr will be logged on your card. Note that the PM3 is logging the hr value at the end of the split or interval, not the average hr for that split or interval.<br /><br />When the workout is complete, the hr values at the end of each split/interval are summed and averaged. The blank values are added into this, and this is probably not the way it should be done as the resulting average hr for the entire workout will be misleading. This will probably be changed in the next revision. <br /><br />I am not sure why an abnormally high or low split/interval hr value would show up on the logcard but not on the monitor itself. Wireless heartrate transmission can be fouled up in many ways, particularly as we get more cell phones, wireless speakers, computers, garage door openers, etc, etc thrown into the mix. Heartrate problems are high on our priority list and we are looking into them. JonW

[[old] John Rupp]

<!--QuoteBegin-BigFatFishy+Mar 11 2005, 12:33 PM--><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><div class='genmed'><b>QUOTE(BigFatFishy @ Mar 11 2005, 12:33 PM)</b></div></td></tr><tr><td class='quote'><!--QuoteEBegin-->I tried takine the watch off and attaching to the handle.  That didn't work so well </td></tr></table><br />I taped a 4 inch piece of 1/2" pvc to the top front of the handle. This works very well as it is easy to loop the watch over for rowing. I can take the watch off very easily for running, then slip it back on just as easily to row again.<br /><br />See post #9 on this thread:<br /><a href='http://concept2.ipbhost.com/index.php?s ... 7&hl=heart' target='_blank'>http://concept2.ipbhost.com/index.php?s ... art</a><br /><br />As to gels, I mixed my own from similar ingredients to the commercial ones, but that didn't work any better then just splashing water in my chest and the monitor.

[[old] GeorgeD]

I agree with John and just splash some water on my chest and the back of the strap and have had no problem. I have one of the 'soft belts' and it is quite firm and no lost contacts.<br /><br />I wear my watch on my wrist and again have no problem catching a look at it on the recovery as the wrist is flat anyway. A glance is all it takes as I dont need to be staring at my HR all the time and once every 5 mins is enough.<br /><br />I think there can be a danger constantly focusing on it as there is no doubt in my mind that anxiety over what is happening will cause an elevated HR .... catch a glance and then get on with what your about rowing.<br /><br />George

[[old] ehagberg]

<!--QuoteBegin-c2jonw+Mar 14 2005, 06:15 PM--><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><div class='genmed'><b>QUOTE(c2jonw @ Mar 14 2005, 06:15 PM)</b></div></td></tr><tr><td class='quote'><!--QuoteEBegin-->I am not sure why an abnormally high or low split/interval hr value would show up on the logcard but not on the monitor itself. <br /> </td></tr></table><br /><br />how 'bout this for a theory... If I'm doing a long "just row" session (as I often do), the final splits that are recorded are done for each 20 minutes, but the splits recorded along the way are done every 5 minutes. Now, I'm just guessing, but maybe the 20-minute split HR numbers are arrived at by averaging the four 5-minute split numbers (which are recorded at each 5-minute mark), and if there are some zeros (drop-outs) at one or more of the individual 5-minute points, that would make the 20-minute split look ridiculous (I have a couple in today's workout that show my HR at 68 and 100bpm, which didn't show on the monitor at the end of any 5-minute split).<br /><br />Maybe that explains the weird numbers?<br /><br />In any case, I still think all the numbers should be averages rather than the number at the end of the split. Is there any reason not to do it this way, or at least to make the average number at the end of a session _really_ work out the the running average for the workout and not the average from the split points? The split-point average seems artificial to me - often those are exactly the points at which I'd expect to see abnormal numbers, for example if doing intervals, you'd be going harder (and your HR going higher) than at any other point - making the "average" much higher than the real average HR.