(10-23-2013 09:00 PM)Peter_C Wrote: My machine (Resmed "VPAP Auto 25') Came out around 2007-8 (remember, please, I am really bad on dates!!) and I got it around that time. ...
I wonder how to learn the expected leak rate of a given mask, and, how does one tell the machine that XX leak is to be expected?
My machine reports leaks in L/s. My range lately has been 0.34 to 0.64 L/s or, 20.4 to 38.4L/m which seems to be high enough to **possibly** create/cause errors/issues and/or data errors to occur - so I need to work on this.
Remed draws the Large Leak line at 24 L/min or 0.4 L/sec in the ResScan software. Resmed has long maintained that their machines can compensate for leaks up to 24 L/min without any real problems. But once leaks get to that level their machines start having problems maintaining therapeutic pressure and recording accurate data. Short lived large leaks are usually not that big of a problem in terms of quality of therapy or data, but the longer the large leak lasts, the more likely it is to adversely affect the quality of the therapy and the data.
On the S9 machines (which are the current generation of Resmed machines), there is a mask fit feature on the machine's LCD. Mr. Green Smiley face shows up indicating that the Large Leaks are short enough to NOT affect therapy whenever the 70% leak rate is less than 24 L/min. Mr. Red Frowny face shows up when the 70% leak rate is greater than or equal to 24 L/min and Mr Red Frowny indicates that the leaks are both large enough and long enough to be a serious problem. In other words, the Resmed engineers are not particularly worried about Large Leaks above 24 L/min unless they are lasting at least 30% of the night on the newer machines.
But I don't think the machines are as old as yours have a feature similar to the Mr. Green Smiley/Mr Red Frowny twins on the LCD's sleep quality report.
The leak numbers on your machine's LCD are 95% leak numbers. In other words, the leak rate was AT or BELOW the reported number for 95% of the night and it was AT or ABOVE the reported number for 5% of the night. Hence if the leak rate on your LCD is strictly less than 0.4 L/sec, then you know that your leak rate was in LARGE LEAK territory (AT or ABOVE 0.4 L/sec) for no more than 5% of the night. And so on any night that your reported leak rate is STRICTLY less than 0.4 L/sec, you can conclude that Large Leaks probably were not too much of an issue.
But on any night where your machine's leak rate is AT or ABOVE 0.4 L/sec, you really have no idea just how bad the leaks really were based only on the machine's LCD data. It could be that your leak rate was AT or ABOVE 0.4 L/sec for exactly 5% of the night (and leaks really aren't that big of a deal) OR it could be that your leak rate was AT or ABOVE 0.4 L/sec for 90% or more of the night (and leaks were a VERY big deal. You just don't have enough data to figure it out.
To make it easier to understand what I'm saying, let's look at three very simplified data sets each containing 20 data points and pretend that they're the whole leak data for the entire night.
Data set 1:
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.2, 0.2, 0.2, 0.3
Data set 2:
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1, 0.2, 0.2, 0.3, 0.4
Data set 3:
0.0, 0.0, 0.0, 0.1, 0.1, 0.2, 0.2, 0.3, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.4
Note that in data set 1 the 95% is the the RED 0.3 number, which is the 19th number on the list. The only number on the list that can exceed this number is the 20th number on the list. If the 20th number on the list is 0.4, then we hit Large Leak territory for 5% of the night. If that 20th number is another 0.3, we don't.
In both data sets 2 and 3, the 95% value is the RED 0.4 number. It is the 19th number on both of these lists. 95% of the data points in each of these sets have values that are AT or BELOW that number. (And 5% of the data points---i.e. the 20th number in the data set--have a value that lies AT or ABOVE the 95% level.)
But if these data sets represented leak numbers, in data set 2, the large leaks last only 10% of the night, whereas in data set 3, the large leaks last 60% of the night. Big difference.