Hello,
I have now actually finished my "Tannoy 2528 Trapezoid" speakers that I had discussed with some of you some time ago here.
For the purpose of sharing my experience, I would like to post a build-thread here, but I've got a few questions...
I am not very experienced with taking measurements, and I am not exactly sure what people like to see measured. I have REW installed and a calibrated mic from minidsp on a tripod. I usually do sweeps from 20 to 16k, from the listening position, and then from 1× cone distance (not even sure why the latter...).
Unfortunately I cannot measure outside, too many neighbors. I could do a close measurement with the speaker in the center of the room maybe?
What else could I do to get some reasonable measurements without lots of room influences?
Any smoothing preferences? Or zoom preferences?
Waterfall?
Thanks...
I have now actually finished my "Tannoy 2528 Trapezoid" speakers that I had discussed with some of you some time ago here.
For the purpose of sharing my experience, I would like to post a build-thread here, but I've got a few questions...
I am not very experienced with taking measurements, and I am not exactly sure what people like to see measured. I have REW installed and a calibrated mic from minidsp on a tripod. I usually do sweeps from 20 to 16k, from the listening position, and then from 1× cone distance (not even sure why the latter...).
Unfortunately I cannot measure outside, too many neighbors. I could do a close measurement with the speaker in the center of the room maybe?
What else could I do to get some reasonable measurements without lots of room influences?
Any smoothing preferences? Or zoom preferences?
Waterfall?
Thanks...
(amazing specs on that driver - which crossover are you using?) you could do a "near cone" being careful not to poke the cone - that would tell the box tuning. Perhaps a near - port. A microphone plot on the floor might take out one dip.
What measurement software are you using? - more experience builders will have better advice than myself.
What measurement software are you using? - more experience builders will have better advice than myself.
@freddi: The drivers came with the original crossovers, and I figured I will use those for now.
I will post the technical data of the drivers. I don't know details about the crossover, although I believe the data includes the crossover point.
Posting the frequency response of cabinet with all drives is what I intended, but there are many ways to measure it, and also to plot that resulting graph (smoothing etc.). I was looking for suggestions in that area.
I will post the technical data of the drivers. I don't know details about the crossover, although I believe the data includes the crossover point.
Posting the frequency response of cabinet with all drives is what I intended, but there are many ways to measure it, and also to plot that resulting graph (smoothing etc.). I was looking for suggestions in that area.
A 1 m distance from baffle at maybe 2,83V; on axis with the driver centre, then 15,30 and 45 deg horizontally to the right, time window large enough to be able to see enough frequency resolution (about 100 ms), away from obstacles, elevated on a stand, diagonally positioned. Rotate the loudspeaker to arrive at the off-axis angles, the microphone stays foot, voltage too.
Thanks, I will have to do that. Not sure what I can use for stands though.
How do I arrive at 2.83 Volts?
8.23, 8.32, 3.28, etc... also valid...
Btw, room response measured with REW at listening position will definitely help to make a too long story short, showing how this animal might sound in your context. Though of course you will have to make plenty of different measurements to try understand how bad or good that room curve looks.
Set you mixer settings for good quality measurements, leave it at that, then send a 1 kHz sine with REW to the amp and with a multimeter measure the voltage at speaker out until it reads whatever voltage you're interested in. Before that turn off any EQ settings so it doesn't influence the measurement in any way. I assume you have already made sure that sound card measures flat. It's not mandatory to measure at exactly 2,83V. REW can probably measure input voltage too but be careful not to overload it. I think 1 volt is safe enough. If you can't find a stand for speaker and a mic, then don't. I have a pile of books when nothing else works.
Last edited:
First, let's be clear of one thing: the important measurement is how you like the sound.
But liking the sound isn't usefully analytic (or helpful in a thread).
I use almost nothing but a camera tripod with a calibrated* mic set where my head would be if I weren't at my laptop playing with REW. That provides a stable technique as you fiddle with your EQ, etc.
it just doesn't matter how "truthful" the REW curves are. You are testing so you can proceed in an orderly way to satisfy your ears. It is just as ill-informed to pontificate on the validity of "flat" as a smart goal as it is to have fantasies that you can (or even want) to replicate Carnegie Hall in your room.
I do each of my drivers, L and R, and everything together. With everything together, you might find some phase interference**, esp at higher freq. Phase matters very little for hearing (esp for authentic stereo recordings where the phases are often uncorrelated) and wanders about with each inch of movement of the mic. But good to note.
For purposes of sharing your experience in a thread, I'd do one channel FR, no EQ, at your head location, subjectively loud (say 90 dB), 1/12 smoothing, AND including distortion results. Yes, room modes get cooked in, but there really is no single defensible purist method for a home speaker.
Ben
*even a laptop mic is pretty good. And the individual calibrations of my three amateur-instrument mics is not really different than flat.
**where the sum is less than the parts
But liking the sound isn't usefully analytic (or helpful in a thread).
I use almost nothing but a camera tripod with a calibrated* mic set where my head would be if I weren't at my laptop playing with REW. That provides a stable technique as you fiddle with your EQ, etc.
it just doesn't matter how "truthful" the REW curves are. You are testing so you can proceed in an orderly way to satisfy your ears. It is just as ill-informed to pontificate on the validity of "flat" as a smart goal as it is to have fantasies that you can (or even want) to replicate Carnegie Hall in your room.
I do each of my drivers, L and R, and everything together. With everything together, you might find some phase interference**, esp at higher freq. Phase matters very little for hearing (esp for authentic stereo recordings where the phases are often uncorrelated) and wanders about with each inch of movement of the mic. But good to note.
For purposes of sharing your experience in a thread, I'd do one channel FR, no EQ, at your head location, subjectively loud (say 90 dB), 1/12 smoothing, AND including distortion results. Yes, room modes get cooked in, but there really is no single defensible purist method for a home speaker.
Ben
*even a laptop mic is pretty good. And the individual calibrations of my three amateur-instrument mics is not really different than flat.
**where the sum is less than the parts
Last edited:
Thank you for your input.
As a matter of fact, I have already made the measurements you suggested (from the listening position). Only hadn't looked at the distortion graphs yet, since I don't exactly know how to interpret them.
And I also agree very much that the most important thing is, how I like the sound.
Now I will just need to find some time to put this all into writing, resize the images etc... but I promise I will not forget to share this speaker-building experience, after all many of you helped me getting the plans for the cabinet structure right.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.