S9 flow skew bug... Anyone else seeing this?

General Discussion on any topic relating to CPAP and/or Sleep Apnea.
User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Mon Mar 29, 2010 9:04 am

Nord wrote:The skew has arrived... but only for the start and end of Flow. All my events are in the correct position for Flow... odd. My starts and finishes are out by about 10-20 seconds depending. Going to do some more investigatin'

Nord
That's precisely how mine started. Maybe we can track this down. How many nights has it been since you started using the machine? I'll compare with the same number of nights on my machine to see if there is an equivalent change around the same time.

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF

User avatar
Nord
Posts: 565
Joined: Sun Mar 07, 2010 4:30 pm
Location: GTA Canada

Re: S9 flow skew bug... Anyone else seeing this?

Post by Nord » Mon Mar 29, 2010 3:57 pm

Hi DD
This is my second S9 - I had a humidifier heater problem with the original and my DME switched it out on March 23. So I have 6 full sessions of Detailed data up to/ including today before noon. Since the skewing has arrived on my machine... I am not sure how much of the data has changed from the original DL.

I have taken a close analysis but there are very few patterns to see.
As of today... this is what I see:

1. a pattern seems that 95% of the time - the Flow Data begins on the same second that it ends - suggesting that there may be some rounding off occurring or that the Hi-Rez data at 25 Hz cannot perfectly sync with the Low Rez 2 Hz Data

2. Flow is separate from each other data posting

3. Leak; Pressure; Minute Ventilation and Flow Limitation are all in sync and begin and end exactly the same times

4. Snore and AHI are Indexes and carry through all 24 hrs if there is a report

5. Events, of course, occur at separate times (hopefully at the correct times)

March 23
3 separate sleeps where Flow lost 2 seconds at each of the 3 sleeps (skewed)
OA's (2) reported correctly with Flow Data or perhaps 1 second early (may be rounding)

March 24
4 separate sleeps where Flow Data lost 3 seconds; 2 seconds; 3 seconds and 2 seconds each time (skewed)
OA's (3) reported correctly with Flow Data or perhaps 1 second early
H (1) reported correctly with Flow Data

March 25
2 separate sleeps where Flow Data lost 50 seconds and 2 seconds (skewed)
OA's (3) reported correctly with Flow Data
H's (7) reported correctly with Flow Data 1 time; out 1 second 5 times and out 2 seconds once - all were late

March 26
3 separate sleeps where Flow Data lost 73 seconds; 1 second and 2 seconds (skewed)
OA's (2) 1 reported correctly with Flow Data and 1 I could not find where it should be pointed (may have been lost in the 73 seconds as it was close to the end of the sleep...

March 27
2 separate sleeps where Flow Data lost 62 seconds and 2 seconds (skewed)
OA (1) reported with Flow Data out by 1 second late
H (2) 1 reported correctly with Flow Data and 1 reported out by 3 seconds

March 28
2 separate sleeps where Flow Data lost 3 seconds and 2 seconds (skewed)
OA's (2) 1 reported correctly with Flow Data and 1 was 1 second early
CA (1) reported 2 seconds late

I don't see any Trends to note and it doesn't seem that the errors are additive - they appear to be independent from each session to another. Perhaps the errors and skewed effect is more based on more events.

Since switching the S9's and the ID in ResScan, I have been getting less events so I have less evidence. I have not looked at my previous 10 days with the S9 yet. I will start looking for patterns there but since the only shared piece of evidence is ResScan 3.10.

If there is something I see... I will let everybody know about it.

Nord

User avatar
dave21
Posts: 720
Joined: Tue Mar 09, 2010 1:05 am
Location: United Kingdom

Re: S9 flow skew bug... Anyone else seeing this?

Post by dave21 » Tue Mar 30, 2010 1:10 am

I power cycled my S9 and wiped my SD card (formatted in Windows as FAT32) and last nights data is now all completely lined up. One thing I have noticed is that I tend to backup the date files to a folder on my PC every day for the previous days/nights data. One thing I have found is that some of the summary files although say 5kb one day might be 3kb a week later.

I'm not sure if this might be causing the effect (and I hadn't been write protecting my SD card) so I don't know if the summary data is being skewed along with the detailed/high resolution data too and whether it's the S9 or ResScan doing it.

Next time (hopefully there won't be a next time) it happens to me I'm going to just power cycle the S9 and see if that makes a difference and then compare the files (beyond compare) in windows and see what differences there are in the raw files.

Thanks
Dave

_________________
Machine
Additional Comments: Running AirSense 10 AutoSet CPAP, previously S9 AutoSet and S8 AutoSet Spirit
Image

fiberfan
Posts: 262
Joined: Sat Feb 13, 2010 2:50 pm
Location: UT

Re: S9 flow skew bug... Anyone else seeing this?

Post by fiberfan » Tue Mar 30, 2010 12:25 pm

I had a flow skew today. I forgot to lock the card before downloading data in ResScan. When I saw the flow skew I removed the card from the card reader and closed ResScan without saving. I locked the card and repeated the download and the flow data now lines up correctly.

_________________
Humidifier: S9™ Series H5i™ Heated Humidifier with Climate Control
Additional Comments: ResScan 3.14 and SleepyHead software.
So many ideas, so much fiber, so little time - http://fiberfan.blogspot.com/

User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Tue Mar 30, 2010 12:47 pm

fiberfan wrote:I had a flow skew today. I forgot to lock the card before downloading data in ResScan. When I saw the flow skew I removed the card from the card reader and closed ResScan without saving. I locked the card and repeated the download and the flow data now lines up correctly.
Sweet! Excellent debugging logic. Have to remember that one.

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF

User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Tue Mar 30, 2010 3:02 pm

dave21 wrote:Next time (hopefully there won't be a next time) it happens to me I'm going to just power cycle the S9 and see if that makes a difference and then compare the files (beyond compare) in windows and see what differences there are in the raw files.

Thanks
Dave
I wonder if the power-cycle part of this is unique from the skew bug or related. I'm looking forward to your results.

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF

User avatar
Nord
Posts: 565
Joined: Sun Mar 07, 2010 4:30 pm
Location: GTA Canada

Re: S9 flow skew bug... Anyone else seeing this?

Post by Nord » Tue Mar 30, 2010 3:50 pm

DreamDiver wrote:
dave21 wrote:Next time (hopefully there won't be a next time) it happens to me I'm going to just power cycle the S9 and see if that makes a difference and then compare the files (beyond compare) in windows and see what differences there are in the raw files.

Thanks
Dave
I wonder if the power-cycle part of this is unique from the skew bug or related. I'm looking forward to your results.
Hi DD

I really believe that the "bug" is more related to memory within the S9. The skew can be corrected after we go back to the S9 and re-set it... causing the machine to unload its' stored memory. I believe it is either not enough memory designated or it is "slower" than required for the Hi-Rez, which is I guess, is causing some problems that can be "cured" by unloading (shutting down) and then reinstalling on the card.

I am now shutting down the machine and format the SD card ( I use 3 SD cards) after each session and restarting at sleep time, although I haven't verified the results yet... Now, I also back up each session on my hard drive.

I think it may be the best answer for the moment.

Has anyone got skewed results after leaving the SD card in the S9 for 7 days before DL'ing the data. That would be easier that doing each day

BTW: just in case someone hasn't discovered... I don't unplug the unit to "power down", I hold the power button down for 3 seconds until the S9 shuts down. There was only brief mention of this in the manual and didn't actually say thats what it did.

Nord

User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Tue Mar 30, 2010 4:04 pm

Nord wrote:...
I really believe that the "bug" is more related to memory within the S9. The skew can be corrected after we go back to the S9 and re-set it... causing the machine to unload its' stored memory. I believe it is either not enough memory designated or it is "slower" than required for the Hi-Rez, which is I guess, is causing some problems that can be "cured" by unloading (shutting down) and then reinstalling on the card.

I am now shutting down the machine and format the SD card ( I use 3 SD cards) after each session and restarting at sleep time, although I haven't verified the results yet... Now, I also back up each session on my hard drive.

I think it may be the best answer for the moment.

Has anyone got skewed results after leaving the SD card in the S9 for 7 days before DL'ing the data. That would be easier that doing each day

BTW: just in case someone hasn't discovered... I don't unplug the unit to "power down", I hold the power button down for 3 seconds until the S9 shuts down. There was only brief mention of this in the manual and didn't actually say thats what it did.

Nord
That's a neat trick with the 3-second power button shut down.
I'm a little slow today, so I hope you'll bear with me.
By holding down the power button for three seconds, this resets a memory cache for flow data in the S9, and flow data should sync back up with the other data?

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF

User avatar
Nord
Posts: 565
Joined: Sun Mar 07, 2010 4:30 pm
Location: GTA Canada

Re: S9 flow skew bug... Anyone else seeing this?

Post by Nord » Tue Mar 30, 2010 4:15 pm

DreamDiver wrote:That's a neat trick with the 3-second power button shut down.
I'm a little slow today, so I hope you'll bear with me.
By holding down the power button for three seconds, this resets a memory cache for flow data in the S9, and flow data should sync back up with the other data?
Yup... Page 16 S9 Clinical Manual points to the button and says "power save mode". When I tried it... the machine completely shuts down w/o removing the plug. Near as I can tell, there is no mention of it anywhere else. It would be just like any computer and will empty the cache... so far it has worked for me.

Nord

User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Tue Mar 30, 2010 5:10 pm

Nord wrote:Yup... Page 16 S9 Clinical Manual points to the button and says "power save mode". When I tried it... the machine completely shuts down w/o removing the plug. Near as I can tell, there is no mention of it anywhere else. It would be just like any computer and will empty the cache... so far it has worked for me.

Nord
So then... by powering down this way after every night, are you saying this alleviates the need to lock the card at all for transfer to ResScan 3.10?

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF

User avatar
coreyg
Posts: 53
Joined: Wed Jan 13, 2010 4:29 pm
Location: AZ

Re: S9 flow skew bug... Anyone else seeing this?

Post by coreyg » Tue Mar 30, 2010 5:43 pm

I've been powering the S9 down like this since day 1 each morning. So far, out of 17 days, I have not seen the time-skew issue.

I power up the system before I put the SD card back in, let it read it and then I power it back down. A couple times I've put the card in without powering it up, but I wondered if the card was in correctly or not and whether it'd read it correctly if it didn't know it'd been removed.

EDIT: I have started locking the SD card when it's plugged into the computer just in case to avoid any issues from ResScan or Windows in general.

-Corey

_________________
Mask: Mirage Quattro™ Full Face CPAP Mask with Headgear
Humidifier: S9™ Series H5i™ Heated Humidifier with Climate Control
Additional Comments: It's actually ResScan 3.10

User avatar
Nord
Posts: 565
Joined: Sun Mar 07, 2010 4:30 pm
Location: GTA Canada

Re: S9 flow skew bug... Anyone else seeing this?

Post by Nord » Tue Mar 30, 2010 6:50 pm

DreamDiver wrote: So then... by powering down this way after every night, are you saying this alleviates the need to lock the card at all for transfer to ResScan 3.10?

I just started using this feature (power down) yesterday... but that makes sense.

I have locked the card for the past several days as well but... the Software has never warned me about it not being able to write to the card. I believe the only corruptions people are getting, come from ejecting the card early or improperly.

Today I took a blank formatted SD card and installed it in the S9 before a nap. It wrote detailed and Hi-Rez data for that nap only... not for any other session that we believe is stored. It also wrote all the Summary data for each session since I started using this S9. When I re-read the ResMed "Data Management Guide" again. It does NOT say that the Detailed or Hi-Rez data is stored on the S9... it just says that up to 365 days Summary; 30 days Detailed and 7 days Hi-Rez can be transferred to ResScan by SD card.

In other words, ResMed may expect that all of us leave the SD card in the S9 for up to 1 week if we expect Hi Rez data to be stored; up to 30 days if we expect all the Detailed data; and up to 365 days if we expect all the Summary data... to then be transferred to ResScan. When we take a nap or even remove the card from the S9... it probably over-writes all the info that it has. It does not over-write the Hi Rez except for the same session date because it does not store it. Since the S9 tries to write Detailed and Summary data... it may skew the info based on just trying to write data that it doesn't have when it writes the Summary data for each date.

In the ResScan 3.10 Clinician Manual on page 16-17 shows a difference between S8 and S9. It says that Detailed data is available on the S8 directly but through the S9 only through SD card... not thru ResLink or Flow Generator. I'm guessing that's a back-handed way of saying that Summary data is only stored in the S9 and the S9 must collect Detailed data (beyond 1 session) by SD card or lose it. It only stores in memory for last session but it probably is trying to over-write data for other dates because it has Summary data for those dates.

I don't believe that it matters about locking the card in the reader because ResScan (I don't think) is causing any problems with skewed data.

I think it was Bill Bolton that stated the Detailed data was stored in S9 memory for up to 30 days and Hi Rez for 7 days based on his experience with the S8... but I'm not finding that. When I take a blank SD card and place in the S9... I only get Summary plus the current session for Detailed/ Hi Rez.

I'm still playing and experimenting to find solutions to Data hygiene.

Nord

User avatar
KatieW
Posts: 1672
Joined: Sun Sep 27, 2009 11:07 am
Location: southern AZ

Re: S9 flow skew bug... Anyone else seeing this?

Post by KatieW » Tue Mar 30, 2010 10:33 pm

Nord wrote: I believe the only corruptions people are getting, come from ejecting the card early or improperly.
Nord, by early or improperly, do you mean before closing down ResScan? Or do you mean not using Windows Explorer to "eject", as AirBreather explains in this thread:

viewtopic.php?f=1&t=50788&p=467268#p467268

I have to tell you, I really appreciate the work that you, DreamDriver, Dave, and others are doing. It's beyond my capability, and I would be lost without it.

_________________
Mask: Pico Nasal CPAP Mask with Headgear
Additional Comments: Resmed AirCurve 10 ASV and Humidifier, Oscar for Mac
KatieW

User avatar
dave21
Posts: 720
Joined: Tue Mar 09, 2010 1:05 am
Location: United Kingdom

Re: S9 flow skew bug... Anyone else seeing this?

Post by dave21 » Wed Mar 31, 2010 12:20 am

I would still recommend locking the SD card if possible. The main reason is that our PC's have lots of nice and wonderful interesting software installed, anything to antivirus software or antispyware software to in some cases viruses and spyware software. These love to check removable media and we don't know what they might be doing. Some packages open the files in WRITE mode even though they're only going to be reading files which is potentially dangerous.

So from now on I'm going to be locking my SD card when I take it out of the S9 just to try and make sure I don't see any skewing problems occurring. If I do get skewed data again I'm going to then start power cycling the S9 on a daily basis for a week and see if that then stops the skew.

My gut feeling similar to Nord's is that it's most probably the S9 that's skewing the data for some reason and not ResScan. If you write protect an SD card and read it in ResScan it doesn't complain about not being able to write to the files on the SD card, so I don't think ResScan is writing anything to these files but other applications might be.

_________________
Machine
Additional Comments: Running AirSense 10 AutoSet CPAP, previously S9 AutoSet and S8 AutoSet Spirit
Image

User avatar
DreamDiver
Posts: 3082
Joined: Thu Oct 04, 2007 11:19 am

Re: S9 flow skew bug... Anyone else seeing this?

Post by DreamDiver » Wed Mar 31, 2010 2:45 am

Locking SD cards has at least drawback. It causes more wear and tear on the card than just inserting and removing the card from a reader/recorder. I think I need to find a cheap source for 1GB SD cards for backup. Anyone got a cheap source? Is the included card class 2, 4 or 6? I'm guessing 2.

_________________
Mask: ResMed AirFit™ F20 Mask with Headgear + 2 Replacement Cushions
Additional Comments: Pressure: APAP 10.4 | 11.8 | Also Quattro FX FF, Simplus FF