I've put together a measuring setup to measure drivers. We're using Arta, our home made 'measuring device', a 80 x 80 x 80 cm box and Mikrotech Gefell mics. Now that everything seems to work as it should, some questions arise regarding various measurements and procedures.
1. Cumulative Spectral Decay vs. Burst Decay
What's the difference between Cumulative Spectral Decay (CSD) and Burst Decay (BD)? They both look like a waterfall plot (which is what I'm after) only the BD looks much cleaner and is much easier to produce. When using CSD you have to specify Window Rise Time whereas BD seems to determine this by itself. I'm just not sure what the difference is between these measurements?
2. Mic Distance
What would be a sensible mic distance for a waterfall plot? The measurements above was made from a 6,5 driver, nearfield with the mic just a few mm from the dustcap.
As a result, the measurement is pretty 'clean', meaning that it clearly shows the driver resonance at 10k without a lot of 'noise'. Off course the downside is that the frequency response is way off due to the minimal mic distance and I'm guessing that I am missing out on some speaker nonlinearities as well.
So what makes sense in terms of mic distance vs. cone diameter for this type of measurement?
1. Cumulative Spectral Decay vs. Burst Decay
What's the difference between Cumulative Spectral Decay (CSD) and Burst Decay (BD)? They both look like a waterfall plot (which is what I'm after) only the BD looks much cleaner and is much easier to produce. When using CSD you have to specify Window Rise Time whereas BD seems to determine this by itself. I'm just not sure what the difference is between these measurements?
2. Mic Distance
What would be a sensible mic distance for a waterfall plot? The measurements above was made from a 6,5 driver, nearfield with the mic just a few mm from the dustcap.
As a result, the measurement is pretty 'clean', meaning that it clearly shows the driver resonance at 10k without a lot of 'noise'. Off course the downside is that the frequency response is way off due to the minimal mic distance and I'm guessing that I am missing out on some speaker nonlinearities as well.
So what makes sense in terms of mic distance vs. cone diameter for this type of measurement?
Audiomatica's Knowledge Base resource page has lots of useful information.
Also,read all the linked to articles by Joe d'Appolito .
Also,read all the linked to articles by Joe d'Appolito .
Audiomatica's Knowledge Base resource page has lots of useful information.
Also,read all the linked to articles by Joe d'Appolito .
I know this response is late (as usual) but nevertheless: Thank you!
ihad to dig through some of the stuff by Joe D'Appolito and found the answer I was looking for. Off course it has been staring at me since the beginning
With CSD the plot is generated in the time domain (ms) whereas the Burst Decay plot is show in Periods (Cycles). And while both methods have their pro's & cons (or limitations) I would say that plotting in Periods makes much more sense to determine the decay of a loudspeaker of a wide bandwith.
Also I took up on your advice and read "Testing Loudspeakers" by Joseph D'Appolito which I can highly reccomend.
Still a bit unsure about the mic distance but I'd guess that the calculation (or rule of thumb) for the far field (referenced to the cone diameter) goes for this case as well. I'll do some tests and see if this is true...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.