(Another) Respironics Encore Data Failure

I have been using a very careful protocol for the daily collection of data from my Respironics REMstar M-Series machine. After downloading my smartcard to EncoreViewer each day, I use the mycpap.org smartcard utility each day to reformat the card. When returning the smartcard to the slot in the blower, I listen for the "beep" and check that the card symbol appears on the LCD screen. Before going to bed, I check that the card symbol still appears on the LCD screen. Power to the machine is never turned off. When removing the card from the machine in the morning, I wait at least 15 minutes after the final blower-off, and listen for the "beep" as the card is removed.

I have always, until now, had full data access in the morning.

For last night's use, all of the above was carried out normally. I turned the machine on at 12:00 midnight. I arose three times, at 1:45, at 4:45, and at 7:00, each time manually turning off the blower by pressing the center blue-lit button. Up-times were each each less than 3 minutes, before restarting the blower. My final wake-up was at 9:00 am. This nightly pattern is in no way atypical for me.

EncoreViewer showed only the latter two sleep-segments, four+ hours of data from 4:45 am through 9:00 am. The initial two sleep-segments, almost five hours of data from midnight through 4:45, was not present. Three times I removed the smartcard from the reader, deleted the EncoreViewer data file from my computer, replaced the smartcard, and reloaded the EncoreViewer data. No change. The problem was clearly not with reading what was on the smartcard, but with the REMstar's writing of data to the smartcard. I have no way to determine whether the initial data was written and then deleted during my 4:45 am arising(though this strikes me as the more likely), or whether the machine inexplicably failed to write during the first part of the night.

Similar reports of total or partial night data loss has been reported here. Various solutions and suggestions have been offered. I post this incident report to document that very careful attention to procedure, and regular reformatting of the smartcard, do not prevent such problems. I have only recently switched from regular ResMed use (where I have never encountered data loss), and have so far been using the M-Series continuously for only three weeks. Twenty nights with no data problems, and then last night. So far, a 4.8% failure rate for the Respironics data management system. Not very impressive or encouraging. Velbor

I still have never experienced the missing data syndrome with my AFLEX machine that others have had. I don't do anything special or pay attention to anything before or after I remove my card for data download. I will post my serial number when I get home tonight.

I too have had nites when the card did NOT record until 3am - when I know I went to sleep long before that. Also have seen other nites when it recorded then stopped for unknown reason. I thought it might be oxidation on the gold contacts and lightly cleaned with a pencil eraser.

I wouldn't think erasing or reformatting the card will help a problem like this.

I have just experienced another partial-night data loss episode, similar to the one I described above. But this time, I think I know what happened.

My first awakening and arising last night occurred at 2:00am. As usual, I pushed the center button to turn the blower off. But I slipped, and hit the button twice, which turned the machine back on. A third button push was needed to turn it off once again. Up for a few minutes to do what needed doing, including documenting the incident, and then back to bed and machine. There were additional arisings later in the night, as is typical for me.

This morning, the EncoreViewer report reflected only data AFTER 2:00am. The initial 2.5 hours of data for the night was missing.

My supposition is that, for reasons unknown, the three rapid off/on/off button-pushes resulted in the preceding session data being erased from the smartcard. Similar events might possibly explain reports by others of random full-night data loss, if the final turning-off of the machine in the morning is not accomplished smoothly. Please be aware of this possibility. Velbor

I have only had software for about a month, but check my data daily. I have not had any issues at all.

I was reading somewhere that the machine may withhold data for an hour or two after the blower is cutoff so it can combine sessions. According to the site, the machine considers breaks in therapy of less than one hour part of the same session, so it probably keeps the data in some sort of RAM then commits it to the card after the 1 hr time period has expired.

I wake up at 7a every morning and make it a point to NOT remove the card until at least 10p. I have never reformatted the card and I haven't lost any data.

Kevin

klockemy wrote: I was reading somewhere that the machine may withhold data for an hour or two after the blower is cutoff so it can combine sessions.

My experience is that full data is available 15 minutes after blower shutoff, and possibly sooner. There is no "combining" as such; the data for a session is treated as a unit, until a break exceeds the criteria for establishing a separate "session":

klockemy wrote: According to the site, the machine considers breaks in therapy of less than one hour part of the same session,

That is my understanding as well. My experience is that breaks of 45 minutes or less are treated as a single session; breaks over 1:15 are treated as separate sessions. Presumably the cutoff point is one hour.

klockemy wrote: so it probably keeps the data in some sort of RAM then commits it to the card after the 1 hr time period has expired.

Rather, it appears that data is written to the card in "real time". If the card is not in place, the data for that period is irretrievably lost. ResMed machines store night data in memory, and download to a smartcard which is inserted AFTER the session. Respironics appears to keep only minimal data in the machine itself, for use in its LED display, and all data is written to the smartcard "on the fly".

Good to know that some users have not encountered data loss. Perhaps it only us folks with shaky fingers on the buttons ?!

Velbor wrote:Good to know that some users have not encountered data loss. Perhaps it only us folks with shaky fingers on the buttons ?!

The occasional times when I lost data seemed to occur on weekends .... when I slept over 9 hours. I wondered if that longer session messed up the data but can't be sure of that.

Mindy

I loose data on my "Classic Tank", everytime I let it go past 7 days or forget to put the card back in the machine, Why! Jim

Hi,

I use Encore Pro and in that program I can set what time the computer program uses to indicate the start of the treatment day. This "Start of day" can shift parts your "Segmented" sleep data from one day to another. Mine is set for 12 noon. This means that anytime that any data that ends (read this you push the off button) from noon one day to noon the next day will be part of the collected data and statistics for that day.

I also believe that "DSM" has indicicated that Respironics machines also tend to slowly lose time on their internal clock (like minutes a month.) So, the older the machine the "more off the time." Only Respironics seem to be able to reset the internal clocks in their units--Oh boy a "repair trip to the DME"--no thanks!!!!

Finally, the Blower unit's clock is set to GMT and the Encore software uses your computer clock's time zone to adjust the readings to your "local" time.

So, if your computer clock is off or the clock in the Respironics unit has shifted or you have the "Start of day" set to occur during your sleep period that could cause data to "show up" on the wrong day (like the day before?) Have you checked to see if this is what is going on?? Have you checked the Encore Viewer software to see "at what time it starts counting a new day?"

Just some thoughts.

Hi Builta,

Actually someone else reported that lost time observation (may have been Wulfman). But I do recall there being some recent threads that again commented on the slow time loss.

The description here is not one I have ever experienced but I have never used an M-Series just the 'tank' Remstar & 'tank' Bipaps. I have had an odd problem where the data in the charts showed with stipes through it (as if a block of data failed to get written).

My best guess from the symptoms described is a software issue to do with how the machines triggers the writing of the data to the smartcard.

DSM

The standard for removing a device from electronic equipment is to
1. power off the equipment.
2. remove the device.
3. reinsert the device.
4. power the equipment back on.

That being said, I believe the card can be removed/reinserted safely while powered on (hot swap).
However, the time syncronization may get confused, and you'll lose data until it re-synchs.

So, If you do do a hot swap, I would recommend powering the equipment off(at the plug), then on, to do a reset.

If that doesn't work, as the Wizard from the comic strip Shoe would recommend:
Sacrafice a goat and leave the entrails over the keyboard(control panel).

JayR_1945 wrote:So, If you do do a hot swap, I would recommend powering the equipment off(at the plug), then on, to do a reset.

If that doesn't work, as the Wizard from the comic strip Shoe would recommend:
Sacrafice a goat and leave the entrails over the keyboard(control panel).

Sorry, JayR -- you're going to need to sacrifice that goat! I always powered off before removing the card and it still lost data occasionally. Firmware.

Mindy

mindy wrote:

Velbor wrote:Good to know that some users have not encountered data loss. Perhaps it only us folks with shaky fingers on the buttons ?!

The occasional times when I lost data seemed to occur on weekends .... when I slept over 9 hours. I wondered if that longer session messed up the data but can't be sure of that.
Mindy

I too have seen data loss when sleeping over 9hrs for 2 times in the 7 session period. That usu, happens on weekends and quite honestly, I don't think is a lot of time to use a device designed to enhance sleep. I have seen my sessions going for 9.1hrs result in data loss.

I too have done the double stroke on the off button. Could be on to something there, Velbor. It was my feeling that if a session terminated at 59 minutes then it prob had to wait to terminate under normal circumstances but that could be off base, too.

I wish I knew 'exactly how' the data is written and stored on the card. I'm sure that would answer a lot of questions.

GumbyCT wrote:I too have seen data loss when sleeping over 9hrs for 2 times in the 7 session period. That usu, happens on weekends and quite honestly, I don't think is a lot of time to use a device designed to enhance sleep. I have seen my sessions going for 9.1hrs result in data loss.

I too have done the double stroke on the off button. Could be on to something there, Velbor. It was my feeling that if a session terminated at 59 minutes then it prob had to wait to terminate under normal circumstances but that could be off base, too.

I wish I knew 'exactly how' the data is written and stored on the card. I'm sure that would answer a lot of questions.

I have had sessions up to 9:37 duration with no data loss, so if there is a time limit, it seems to be well over 9 hours. (ResMed, by the way, has a 10 hour limit for data display. They insist that after that, no data is lost, and that correct statistical handling of all data continues, but that graphic display ceases.)

I have been able to effectively access the data used by EncoreViewer. The .xml files which are retained in C:/Neo/SCDecodedFiles can be loaded into Excel and manipulated fairly well. This is not truly "raw" data, in that it is not a direct representation of the information stored on the card, but rather is produced in the process of report generation by the Encore software. Still, it offers a great deal of insight into how Respironics data is managed. (Let me know if you'd like more detail on this process.)

Data is sampled every 30 seconds. Pressure and leak data are recorded every 30 seconds from 0:00:00 session-start time. If there is a short blower-off break in the session, there are "disconnect" and "reconnect" entries, and recording resumes on the new schedule, still based on the original start time (though with the 30-seconds sampling now at new "odd" times, e.g. :27 and :57 rather than at :00 and :30). After a long break, a new session, with a new 0:00:00 start time, is initiated. Within each session, OA, NR, VS, FL and H events are recorded simply as time stamps: time is identical to the 30-second pressure and leak times, but no duration information is recorded. (Variable Breathing, whatever that is, is similarly recorded, but apparently not used by EncoreViewer.)

builta, thank you for your comments about the Respironics internal clock. This is another area of concern, with issues possibly depending on differences between local time and machine clock time. I can confirm the Respironics clock being set to GMT, and Encore software providing local times based on computer settings. I've presented information about this in another post about another issue I've had with Respironics data, at:
viewtopic.php?f=1&t=42182
My machine is well over a year old and sat in its box for most of that time (while I used ResMed), and as indicated in the above thread, the clock is still surprisingly close to correct. There is no evidence of my data loss problem being due to clock/day issues.

Jay R, although I've not been able to locate "specific" statements from Respironics that the Encore card system is intended for "hot" insertions and removals, all evidence points in that direction, and that process seems to be implied. There is no power on/off switch. The machine provides and indicator "beep" when the card is "hot" inserted or removed. There is also an LCD screen indicator which indicates correct card presence, which works "hot". (ResMed, by the way, expects a "hot" card insertion in the morning, at which time night detail data is downloaded from machine internal memory onto the smartcard.) My partial-night data losses, rather that the full-night losses described by others, would not seem to be explainable by this concern.

The mystery continues. Thanks for all of your offerings of clues. Blessings, Velbor