Polar HR monitor problems

borisg

Just wondering if anyone has run into the same problems that I have with the monitor. Was very elated with I first attached the monitor to myself (after first wetting the contacts) and it seemed to be doing a good job at picking up my heart rate. However, after wearing the chest strap for most of the day today, the number of artifacts is incredible. Rewetting the contacts solves the problem, but as soon as the contacts dry, we're back to huge numbers of artifacts. The maddening thing about these artifacts is that the pulses are precisely 15 msec in duration and there's no intelligence in the Polar receiver which happily passes along a pulse that works out to a heart rate of 1000.

Presumably the monitors are setup for people who are exercising and thus sweating profusely. Aside from turning up the heat in my house, or going for a run every 30 minutes, I haven't been able to think of a way of dealing with this problem. I realize that not everyone is planning on having people wear the Polar HR monitor strap for a 24 hour period, but if you're planning to use it in your project, something to keep in mind.

EKG electrodes use an electrolyte and are waterproof enough so they don't dry out. Before I start adding various conductive gels to the electrodes, I'm curious if anyone has tried the Polar monitor for a period of hours when not exercising. Other option is to ditch the Polar monitor and start working on an EKG amplifier.

Ken Gracey

borisg,

I think I found a solution to the problem you are experiencing. On Friday I'm doing a presentation at the NextMed "Medicine Meets Virtual Reality" conference in San Diego. As part of my presentation I'm wearing a Polar Heart Rate Transmitter and displaying my heartbeat on a Propeller BOE while I talk. Initially I wet the contacts and this worked for a few minutes. Finally, I reached for the "Clorox Disinfecting Wipes" and cut a few 1" square pieces. I put these between the Polar electrodes (?) and my skin, and they've been working flawlessly for over an hour now. I'm not sure how long you need your system to run, but I imagine this setup will work for maybe an hour and a half.

In the picture below I took the strap off after an hour and switched to the Polar Heart Rate simulator, a little box with a dial that lets you simulate a heart rate [and test athletic equipment, I presume].

I've been reading your other posts with much interests, and neglect the fact that I haven't made time to reply. I'm interested in what you're doing and appreciate reading about it on the forums.

Ken Gracey

Beau Schwabe

While working in prosthetics, we always used an aloe vera based gel for MYO-electric sensors. You have to make sure that whatever you use doesn't have any corrosive effects on the electrode.... i.e. no cleansing agents. The byproduct of a corrosive reaction on the electrode can be harmful when in prolonged contact with your skin. That's why there is such emphasis on the type of electrode you use.


borisg,
Are you sure that your 15ms isn't closer to 16.67ms? Sounds like you need a 60Hz Notch filter on your electrode leads. With a filter you should /still/ be able to detect a Polar HR signal even with dry electrodes. (<based on prior knowledge of other types of EKG sensors)... <-- I don't know how sensitive the Polar HR monitor in the kit actually is so the mileage may vary.

borisg

Now back to my hardware hacking instead of meatware hacking for the next week. Thanks for the tips Beau. Just found another reason which might explain why I'm having problems. As I only have one Polar HR monitor, I soldered a socket for it on the Propeller Quickstart board that I built my preliminary project on. Today, when pondering what the problem was, noticed that the Polar HR receiver was wiggling in the female header on the board. Thus, I suspect that mechanical vibration likely plays a role as well as my stupid idea to flash an LED through a 1 Kohm resistor taken directly off the the Polar HR receiver output lead. I suspect this is transmitting noise from the Propeller board to the Polar receiver. The solution would be to solder the Polar HR receiver to the board but I've only got one unit and am reluctant to do so.

Beau, you're right about the width of the pulse - it comes back as either 15 or 16 msec using a 1 msec clock to measure the pulse and rarely I get a 17 msec duration pulse -- likely made the assumption that it was always 15 msec when I looked at the spec sheet and saw a series of 15 msec pulses.

Now the problem I have is that appears to be a worldwide shortage of the Polar HR receivers. What I'd like to do is to get another 3-4 of them so I don't have to worry about soldering one onto a board and potentially blowing the unit during testing/desoldering. (also I could build a few HeartLink monitor replacement units and start testing them on volunteers). I've soldered the 3 axis accelerometer that I'm using for movement quantification to the board as I have purchased lots of spares. Sparkfun no longer has their breakout board for the Polar HR receiver and just has a board which doesn't bring out the HR event pin. Have ordered some Olimex EKG boards which are setup as an Arduino or a Pinguino shield in case I blow my Polar HR receiver as the Olimex boards are easily available and save me the trouble of building an EKG amplifier (have done it in the past but now have more money than time and I do much better with digital electronics than analog).

So, if Parallax has some of the Polar HR receivers for sale, I'm very interested. They're not as good as an EKG signal, but the Polar HR chest strap is how I'm planning on monitoring respiration as soon as I figure out how to get a pressure or capacitative sensor as part of the mechanical system as the amount of tension produced by respiration should be easy to pick up with a sensor in series with the Polar chest strap.

Am assuming that will have to sample analog EKG data given the lack of Polar HR receivers as well as an EKG waveform is a lot more clinically useful as this allows on to get such parameters as PR interval, QRS width, QTc and ability to compare baselines of PR interval and QT interval. Will see if I can find a better socket for the Polar HR monitor but suspect that it will only work well in an ambulatory HR monitoring system if it's soldered to the board.

NWCCTV

@borisg, What program are you using on the prop for your data?