ResMed S9’s Exhale Whistle

You’re very excited because you just received your ResMed S9 Autoset or Elite. You get it home, plug all the parts together, put on your mask and start drifting off to sleep. Suddenly, you can’t help but notice that every time you exhale there is a very very faint whistle coming from the S9’s rear.

It’s not that noticeable, yet for some of us, it’s a bit like the Gilligan’s Island theme song or the SpongeBob SquarePants theme song. You just can’t get it out of your head.

The sound can be slightly different from machine to machine, sometimes higher in pitch, sometimes lower, depending on year and origin of manufacture. Some machines apparently show no signs of  having the whistle. It may also be possible that it exists in all machines, but that many of us who get one are hard of hearing from years of snoring and ill sleep. Other circumstances may also take a hand in whistle brightness/loudness, for example the patient’s general state of flow limitation.

For purposes of documentation, the below two videos are a comparison between the M-Series Pro and the S9. In both instances, the microphone was set one foot away from the blower, level with the blower on a bed stand in a typical bedroom situation.

[youtube_sc url=”” title=”Respironics%20M%20Series%20Pro%20Audio%20Sample” autohide=”1″]

[youtube_sc url=”” title=”ResMed%20S9%20Autoset%20Audio%20Sample” autohide=”1″]

I’ve included a visual comparison from Audacity to show how much larger–hence louder–the waveform is for the M-Series Pro, an older model from Respironics that’s no longer made. Both waves are shown at the same magnification. (If you listen starting around second 35 on the M-Series example, you can hear my heartbeat in the open air column patency feedback in the sound of the machine. Pretty interesting stuff!)

Upon comparison, I find the noise to be barely noticeable with the S9 below the level of my mattress on a lower shelf on the nightstand, and certainly less noisy than the M-Series Pro, even with the whistle. If you can get used to traffic or trains going by your house, you can get used to this much more easily. There may be the one or two odd duck machines out there with truly loud whistles that are just menacing enough  to a reasonable night’s sleep to you or your sleep partner that sleeping with it on is unbearable. In this case, it really would pay ResMed to get your DME to switch out the machine for a new one and return your offending blower to ResMed for study. Approach the DME respectfully and ask for a contact within ResMed so you can transact a full swap without missing a night on CPAP. Make a recording of the machine’s behavior as I have, and play it for your DME and for ResMed so they can decide what to do.

Is there any way to fix this thing?

Sorry, you really can’t fix it yourself. It’s internal to the blower design. There is some constriction in the air column between the blower and the hose that, due to the way air pressure is regulated, acts like a one of the pipes from a pipe organ in your local place of worship, except not nearly as loud. That constriction is likely integral to the pressure regulation method or it would have been designed out in some version of the S9.

I can’t dictate ResMed policy, but enough people have complained about it on enough forums that I suspect it’s a priority for them to get it fixed in the next generation of blowers.

I’m not sure if there are any similar blower noise concerns in the Phillips Respironics System One series. If you know of any, I’d like to hear from you!

If you have an S9 that whistles loudly, I’d love to hear it. To emulate my method, place the S9 on a night stand on the right side of the bed, about a foot below the mattress. Place your mic about a foot from the right side of the machine, level with it and send me a recording. (Also do a recording at a foot and a half from the right of the S9.) Please use a ruler and make sure the mic is not hand-held, but rather sitting stationary on a tripod or a cardboard box.

Leave a Reply

Your email address will not be published. Required fields are marked *