SPL Logger difficulties

[LordBobbin]

I am using a MiniDSP UMIK-1 with calibration file set, to take multiple measurements of sound exposure over time. I'm having two problems.

First issue: when I open a saved file into the SPL Logger after recording, it no longer shows the curves for any of the LAeq fields, and that's the only data I need! During the initial recording all data shows great. But none of my measurement files will trigger that curve to generate upon reopening.

Second issue: possibly not REW's fault, when doing an SPL recording, input will randomly stop for a few seconds. There's no regular duration between the dropouts, and no specific time length. Generally the dropouts are between 2-10 seconds. It drops down to about +22 when this happens. I tried recording audio into Audacity, and there are no dropouts at all. Another curiosity is that the drop seems to not happen if audio input is quickly varying.

Many thanks in advance for any advice anyone might have!

 

[John Mulcahy]

Can you attach an example saved file? Which REW version are you running?

 

[LordBobbin]

Hi John, thanks for your reply. I'm using 5.30.8. I'll attach two files, a recent one that has the issue, and an old test I was doing (using C weighted instead of A) which DOES display all the averaging curves.

I have just opened the text file to explore, and see that the LAeq column shows "infinity" instead of a number. I'll assume that's causing the problem in part - but there IS data in the LAeq1m and 10m columns. Any ideas?

 

[LordBobbin]

Update - I tried restarting everything and recording again. While recording, the LAeq graph wasn't appearing, and I hovered over the line - REW says "graph is off the top of the chart" (or something similar), so that explains why all of the data is recorded as infinity, for that column. I've just reverted back to 5.20.13 which I was using previously and for the recordings that did work. Lo and behold, LAeq is working properly. Checked the text files of both, and it's consistent - 5.30 doesn't log LAeq for me, while 5.20 does.

I'm running MacOS 10.15, so perhaps I didn't read the release notes for supported systems before updating REW.

This is separate from my "dropout" issue, but I did record a good example of it which I'll attach. Ambient noise in the house is at about 39dBA, but you can see at the end of the log that LASMin drops to 22 and flat lines for a few seconds before "waking up". Any ideas there?

 

[John Mulcahy]

Thanks. The LAeq problem happens when not using rolling Leq, I'll fix it for the next build. The dropouts look like a mic issue, best check with miniDSP.

 

[LordBobbin]

One more update on the missing curves problem: I just edited the text file, and replaced every ∞ with 100. NOW all of the curves show. The LAE, LZPeak, and of course LAeq curves are messed up. So I opened just the recorded data set into OpenOffice (using space as cell delimeter), copied the measurements from LAeq10 to the missing column, exported as a CSV text file using spaces instead of cells as the delimeter, copied that new data set back into the original file, and BOOM it all works now. I've got my measurements and can create screen shots of the graphs!
So in case anyone else has a botched data set... this is an option for repair.

 

[LordBobbin]

Thanks. The LAeq problem happens when not using rolling Leq, I'll fix it for the next build. The dropouts look like a mic issue, best check with miniDSP.

Thank you John! This software is fantastic. I really appreciate your help on this issue - wouldn't have found a solution for my little problem without you suggesting to look at the text file. I'm very happy now as I thought my 3 hours of SPL measurement were lost. Much appreciated!