Issues with CMS50E pulse oximeter - new and different issues

Hi all,

I just got a CMS50E pulse oximeter because a few folks here seemed happy with it.

It works great, but I'm having some trouble with the software.

I get the drivers installed and it does show up in the devices in Windows' control panel and is on COM1. When I launch the SpO2 Manager it just doesn't see it. When I try SpO2 Review, it hangs on creating a new session. I know it's seeing something because when I tried to change the COM port # I got an error saying it was already open and couldn't be changed. Also when I was connecting using a plain old USB cable, it did not hang... using the bundled cable it definitely sees something but fails.

I can read data off of it on my Mac if I install the serial chip manufacturers driver, so I know it's not defective... at least in as much as you can get data off it. I can't make heads nor tails of it. I've been trying to reverse engineer the data stream for a couple days now. I figured it was time to ask the Windows question b/c I'm not really a windows guy so I figure it's some obvious thing I haven't tried.

I read the couple other troubleshooting threads here and didn't see anything like my problem... may have missed it, though. Granted, that would be what clued me into the "wrong cable" problem.

Which OS are you using? I'm ok using Win XP SP3. It doesn't work on my other machine with Win 2k.

Are you using the green icon = SPO2 review? This one works. The red icon is the SPO2 manager - it does not receive data. Look for the green icon, it works. You have to click on the magnifying glass - it will say waiting for data. Then connect the oximeter and select upload data - turn this to on. Should work just fine.

Green icon, it's the one that hangs. The red icon just does nothing.

It stops hanging and says no device available as soon as I unplug the USB cord... though not the oximeter from the cord. Makes sense, tho, that cord has a chip in it that does the serial to USB conversion.

It's Windows 7, 64 bit. I dug around the history and saw some other folks who said they had it working on Win 7 64 with the older version of the software. Tried both with similar results. Even updated the driver to the latest from Silicon Labs. Still no go.

Don't ask me how I forgot to include IT troubleshooting 101 info.

I'm going to try to fire up a Win XP VM I have around and see if that works. I'm not optimistic, though.

The red icon is for live SPO2 recording while the pulse-ox is attached to the computer via usb.

If you have the 50E plugged in to USB cable, start the red icon, and then turn on your pulse-ox and put you finger in. The readings will show on your screen. Does this work for you?

Have you tried uninstalling everything including drivers, rebooting, and then trying to do a fresh install again?
(Sounds like you have, but just checking.)

I can confirm that Version 0.982 rc1 found here viewtopic/t60488/viewtopic.php?f=1&t=42197&p=557523worked for me on Windows 7 x64 Home Premium.

---------

It could be something else that is on your COM1 port and is interfering. When in the 50E software version 0.982, I can not change the COM port under Connection. It is fixed. I can change the CMS Contec device in the drop-down menu though.

If that is the case then it seems to me that to change the COM port you have to:
Go to Control Panel-> Device+Printers-->
Right click on Silicon Labs CP210x USB to UARTCM Bridge and Select Properties
Click on Hardware, then click Properties
Click on Port Settings, then click Advanced
Change the COM port Number from the drop down menu.
Click OK OK OK.

Start the Green SPO2 Review Icon, and see if you can start a new session (also that the com port has changed), and upload the data recorded on the pulse-ox.


Hope that helps.

Yep, did that, nope it doesn't work.

Haven't reinstalled the OS, but I'm not gonna go that far. It's honestly less effort to write my own software. I'm guessing their software hunts down all available COM ports and gets hung up on one of the other's that's available. I was hoping that making the pulse ox COM1 would get it to stop looking once it was found.. obviously no such luck.

I've just gotten a python script that can read it live and am about 15 minutes from one that can parse the recorded data. It only took me three days of obsessing over it.

I lied.

I was only five minutes from being able to parse the recorded data.

I've got the timecode header figured out, though there are three extra bytes I'm not sure about, yet. I bet they're something to do with the length of the recording.

Also need to figure out the right frequency for the data points, but that'll be easy with some trial and error.

I'll throw it up for folks to grab if anyone wants.

If this helps at all, I believe the data is recorded once per second.

It does, very much, in fact.

I had it on all night last night, 20363 seconds works out to 5 hours 39 minutes 23 seconds, so about right for what I slept last night.

I was going to guess one entry per second and run some varying length recordings to confirm it. I'll still probably do some just to be absolutely sure and partly to see how inaccurate it's timing circuit might be.

The other thing that you need to look for is that it takes 4 - 6 seconds before it records the first value. There is some over sampling and averaging being carried out and I don't know the specifics of all of that. This delay may throw your values off a little. If there is some difficulty in detecting a good signal, the delay increases.

Are you able to capture the (live) waveform data, or just data points?

Crusher, what about this simpler O2 meter for $99:

http://www.go2nonin.com/

You put it on once or twice a night at different times.

Avi, Nonin makes very good quality oximeters.

The main problem with spot checking is that the odds of monitoring at the same time you are experiencing a desaturation during sleep is not very good. Spot checks give you values closer to average values.

For example, my average saturation before xPAP was 96+%. However, I had desaturations down to below 80%. After starting xPAP therapy my average is still 96+%, but my desaturations don't fall below 90%.

That unit looks awkward to wear while sleeping. I've had pretty good luck with this one. I had a problem with downloading not working several months after purchase. Formosa Medical replaced it for free. I reviewed it.
http://www.amazon.com/Wrist-Pulse-Oxime ... B001O4P2IQ

My only gripe is that my ASV clock is slow by 21 minutes. So, it takes a bit of calculation to see if a particular event in my ASV report is related to a signal in pulse or oxygen saturation.

coolguyclay wrote:Are you able to capture the (live) waveform data, or just data points?

I can read waveform data while it's in USB mode, but in record mode it only records data points.

I could probably hack something up to read the serial port and leave it in USB mode all night. Not sure it's relevant, though.. what would you use a historical HR waveform for, that you couldn't get from 1 Hz HR signal?