MJL21193 said:
But any significant cabinet coloration is going to be down in level 10-40 dB from the level shown in the FR so won't show up there. The impulse has some significant differences, but it is too hard to interpret what they mean. So we know there are differences, the CSD is perhaps the tool that will let us interpret them.
dave
Member
Joined 2002
Is there at any time on this form that people stop and say hey I'M BEING A CHILD i should stop bickering. This is frustrating reading you guys go back and forth bickering about who's right and who's wrong. GIVE IT UP!
MY GOD!!
MY GOD!!
jleaman said:
Here's some friendly advise: If you don't like our discussion here, you are free not to read it. The only comments in this thread thus far that resemble childishness are yours.😉
maxro said:
Panel noise and driver output were measured at the same power level and settings, to demonstrate the relative noise level.
Impulse test were taken at a higher level to get a good reading.
planet10 said:
I'm confused now Dave. Are you saying 10-40 dB down from the drivers FR or the measure panel FR? If it's the first, that's covered in the panel FR, if it's the second, the sound level is already high, well within audible range, and clearly shown at a maximum -18 dB from the driver level.😕
I may go to the trouble of installing Arta for the CSD plots.
MJL21193 said:
Didn't you say that you didn't/couldn't block the sound of the driver? Unless you window really small this will contaminate the panel measurements?
So i'm assumming the primary level is that from the driver and not from the panel.
dave
Okay, a fence sitting post to encourage you both. 😉
I'll go back to what I stated much earlier in the thread. I believe there may be differences, but ones that only come into play on larger, unbraced panels, such as used on the horns and TLs that Dave loves so. I understand that MJL's test boxes are relatively small, and as such, may not show these differences, no matter what test regime is used, (though CSD plots would be nice!), and as such he has proven to my satisfaction that for small bookshelf type speakers there is no problem with using MDF over ply. Now, if he has the time, it would be nice to see a similar comparison with a typical HL cabinet. 😉
I'll go back to what I stated much earlier in the thread. I believe there may be differences, but ones that only come into play on larger, unbraced panels, such as used on the horns and TLs that Dave loves so. I understand that MJL's test boxes are relatively small, and as such, may not show these differences, no matter what test regime is used, (though CSD plots would be nice!), and as such he has proven to my satisfaction that for small bookshelf type speakers there is no problem with using MDF over ply. Now, if he has the time, it would be nice to see a similar comparison with a typical HL cabinet. 😉
planet10 said:
I attempted to block the output using a towel draped over it. It really cut down the output (by at least 60-70% approx) but made next to no difference on the panel FR results. They still hit the -18 dB mark. For this reason, I didn't think the drivers output had a large effect on the panel test. Also, with the towel, I didn't want the added variable of it's random movement against the speaker cone to muddy the results.
Only practical way to completely take forward cone output out of the equation, is to build a large, open ended, heavily stuffed box and mate it to the test box to dissapate to output. For my homebrew tests, I thought this was overkill.
MJL21193 said:
18 dB down is still probably mostly the speaker. A panel resonance that high up would be seriously bad. A towel will have little or no effect except at frequencies where things are already pretty directional.
dave
Hi John,
The testing you are performing is something that no one has been able to point to current results. I don't think there is overkill.
Before you install ARTA, lets see if it will do what we want. See post 101. http://www.diyaudio.com/forums/showthread.php?postid=1216198#post1216198
What we need is the period based CSD. I have just installed the latest version, yet it is not obvious how to do that test. I'm still looking, or overlooking something!
Cheers,
Geoff.
Edit: BAD grammer
The testing you are performing is something that no one has been able to point to current results. I don't think there is overkill.
Before you install ARTA, lets see if it will do what we want. See post 101. http://www.diyaudio.com/forums/showthread.php?postid=1216198#post1216198
What we need is the period based CSD. I have just installed the latest version, yet it is not obvious how to do that test. I'm still looking, or overlooking something!
Cheers,
Geoff.
Edit: BAD grammer
Geoff H said:
Geoff, I'll keep plugging away here with what I have (satisfying Dave is challenge enough) if you could keep me posted on Arta's capabilities.
At times it felt as though my brain was literally cooking inside my head, trying to get Speaker Workshop running right. Therefore I'm hesitant to switch and go through the same thing again. With that said, if Arta has more to offer, I'll give it a go.
planet10 said:
I'll do as posted above and fire the test boxes into that sub. It's big enough to get a lot of fiberglass into to eat the drivers output.
Any doubts Dave?
MJL21193 said:
It could work... you'll only know after you try & it is certainly worth a try ... you'll want to gasket the front of the box....
dave
Moderator Note: the discussion starting here and for about the next 4 pages was split from this thread http://www.diyaudio.com/forums/showthread.php?s=&threadid=103023 on speaker basket damping. dave 
I guess I'll have to (as usual) settle for really good. They say the devil is in the details, with forward progress after the smallest adjustment. Endless tinkering sometimes nets negitive results.
Dave, you will damp your driver basket, but you will not damp the panels it's attached to. Maybe bigger gains in noise reduction could be had if you concentrated on the real source of the noise.
planet10 said:
I guess I'll have to (as usual) settle for really good. They say the devil is in the details, with forward progress after the smallest adjustment. Endless tinkering sometimes nets negitive results.
Dave, you will damp your driver basket, but you will not damp the panels it's attached to. Maybe bigger gains in noise reduction could be had if you concentrated on the real source of the noise.
MJL21193 said:
Tried that. Not as effective as proper bracing. Panel damping just pushes the panels in the direction opposite of what i want so is counter-productive
dave
planet10 said:
Dave, it seems you are saying that if you damp the panels, they won't be braced. Braceing and damping the panels is what I do. The objective is to reduce audible vibrations, right?
You go to a lot of trouble to critically damp your driver basket, then ignore the box. The noise potential is greater from the box panels than from the drivers frame.
Bracing the panels drive the resonance up, damping drives the resonance down but also decreases the decay time, therefore you fight from two fronts rather than one. Makes sense to me.
MJL21193 said:
Yes. Adding damping to the panels means the need for even more bracing... once i have reached a threshold there is little need to do more, Adding unneeded panel damping just complicates the build (have to add damping & more bracing) and increases energy storage.
Panel damping is something i've abandoned as counter-productive.
dave