Noise level comparison - Old Remstar Auto vs. New M series

dsm wrote:As mentioned before, more like a batch problem than a wide variation in construction.

Actually, those batch problems are driven by variation in construction (either physical composition or assembly). And that is the point of imposing statistical controls in the manufacturing process. If I get some time in the next day or two, I may come back to this thread and touch a bit more on how acoustical edges distributed throughout a CPAP system, and how sympathetic or natural mechanical frequencies distributed throughout a CPAP system can throw manufacturing and design engineers quite a surprise from time to time. These distributed areas of acoustical concern are by no means limited to the blower motor, blades, or bearings.

I don't think the thread was pointless, I just think it pointed out more than ever that this is an impossible trait to quantify.

Actually, I have to very respectfully disagree with just the laaaaast part of that statement, guest. These acoustical properties are very measurable and they really don't come close to outstripping the capabilities of statistical controls. .

-SWS wrote:

I don't think the thread was pointless, I just think it pointed out more than ever that this is an impossible trait to quantify.

Actually, I have to very respectfully disagree with just the laaaaast part of that statement, guest. These acoustical properties are very measurable and they really don't come close to outstripping the capabilities of statistical controls. .

You still think so? What about the guy earlier who mentioned that he thought his machine was very quiet, but his wife said she could hear it two rooms away? What about that person next to you at the street light with the ultra loud car stereo? I think some of them are stupidly loud. Obviously the occupants don't think so.

It still just seems to me that trying to quantify "Too Loud" is an endeavor that is doomed to failure since it's a subjective concept.

But hey, I'm wrong on a daily basis about all kinds of other stuff so who knows eh?

Anonymous wrote:It still just seems to me that trying to quantify "Too Loud" is an endeavor that is doomed to failure since it's a subjective concept.

I agree it is impossible to try to quantify "too loud" because that is a subjective assessment. It is, however, possible to quantify "louder than specs indicate". What seems to be the issue here is the unexpected variation between 2 identical models. One would assume they would be equally loud with an identical pitch, yet many users have found this is not the case. It comes down to, as -SWS so succinctly put it, variations in construction (either physical composition or assembly). With that in mind, the solution would be to impose statistical controls in the manufacturing process to address this problem.

-SWS wrote:Someone on TAS once posted a link to an internal Resmed memo that conveyed the results of their own investigation into field complaints about noisey machines. I kind of vaguely think that memo may have concluded their own third-party component supplier sent out a batch of blower motors that just so happened to mechanically resonate at the wrong pitch. Reading that study I got the impression that statistical quality controls were not routinely employed by Resmed regarding acoustics in particular.

Obviously the manufacturers are aware of this problem, but perhaps they're having a hard time justifying the expense required to resolve it.

Using just the example of fan blades.....when you take into consideration the volume of them that are made (in molds).....it isn't hard to imagine the small imperfections in their shape, balance or whatever that would cause them to create variations in pitch as they're spinning. Add to that some imperfections in some motors and other moving parts and statistically you're going to have machines that sound different than others. Some VERY quiet and some with annoying characteristics.
Like I said before, they're just going to take care of the ones that people complain about.....it's much more cost effective (to them) that way.

Den.

Anonymous wrote:

Anonymous wrote:It still just seems to me that trying to quantify "Too Loud" is an endeavor that is doomed to failure since it's a subjective concept.

I agree it is impossible to try to quantify "too loud" because that is a subjective assessment. It is, however, possible to quantify "louder than specs indicate". What seems to be the issue here is the unexpected variation between 2 identical models. One would assume they would be equally loud with an identical pitch, yet many users have found this is not the case. It comes down to, as -SWS so succinctly put it, variations in construction (either physical composition or assembly). With that in mind, the solution would be to impose statistical controls in the manufacturing process to address this problem.

Well I do totally agree with that. However, isn't someone ultimately STILL having to arbitrarily (Through statistical analysis) determining a point that is considered "too loud"?
We know any 2 units are going to sound slightly different. There's no way around that. So someone has to decide a point at which they are "too different". Isn't that still an arbitrary point which is still going to leave someone unhappy? Isn't there always going to be someone for whom it will be "too different"?

Anonymous wrote: isn't someone ultimately STILL having to arbitrarily (Through statistical analysis) determining a point that is considered "too loud"?

The manufacturers establishes acceptable parameters of deviation (quality control). Respironics indicates their M Series Bipap Auto with Biflex has a sound pressure level of 30 dB(A), but they don't indicate the control accuracy (+/- %?). Do they have an established variance? I don't know. They should, and they wouldn't want it to be too large (and certainly not a larger variance than their competitor's) because most people request the quietest machine they can get their hands on. How do they determine what is considered an acceptable variance? I don't know. I imagine it depends not only on the cost of imposing statistical controls on the individual components, but also on what the competition is doing. At this point, it doesn't seem any of the manufacturers are as concerned about it as the users are, which is unfortunate. It would be a boon to us if they tried to out-quiet each other with the slightest deviation possible. But I'm not holding my breath.

The real losers are those new to CPAP who end up with a loud machine and think that's as good as it gets, then give up. Wulfman is right, it seems the manufacturers are content to deal with returned machines and not interested in addressing the issue at the factory to prevent it from occurring in the first place. I'd be curious to know how prevalent the problem is, but don't think we'll be getting those statistics from the manufacturers anytime soon!

Pointless wrote:My GOD what a pointless thread....

LOL. Spoken like a true troll.

The difference in sound level seems to be common for the same machine, not only in CPAP machines but with other "blower" mechanisms. Take for example, the heppa air filter purifiers at work, we have two in the office and as a side by side comparison today someone said theirs's was loud...itt actually was much louder than the other identical model a few cubicles down.