Jbl 2267h

[zimpah]

With the explanation given the cubic foot mentioned should be about right for all that's in the box.

Btw you stated the cab volume but not the tuning frequency i think, these go hand in hand to produce a LF alignment. ISD should have given you Fb (port tuning freq).

RE I want to input the net amount of liters? Yes net air volume is what the woofer works with.

RE why does WinISD ask for detailed woofer dimensional measurements? It seems you are wrong in the presumption. I'll take for granted here that you're not talking about parameter Sd for example which is about cone area, but rather you have in mind things such as driver frame diameter, depth, hole cutout, etc. If so these have nothing to do with software box modeling which uses only T/S parameters. Instead its more like "driver statistics" kept in the database for speaker building phase, a place where they're easily found if needed.

I use both Win ISD and Winspeakerz. ISD has a poor driver database vs 1500+ in the latter, its a good start. The reason i mention this is the vent tube diameter. The ISD 4" vent is totally unacceptable for that 15" at the drive levels you refer to. The 6" is a minimum. Winspeakerz takes into account driver input power to recommend a vent dia. or multiple vents with size and length. If i had modeled your cab would probably get over 6". At high drive vents eventually "choke" if not large enough, unable to pass the required amount of air. Cab then reacts more like a sealed box.

You complain about lack of bass and indicate the Win ISD box is about the same as the one you made. So, looking at other items:

Could there be in your system a LF high-pass filter you might have left activated?

We've seen before a member having overpowered his system and a number of crossover components where pretty "cooked".

I note your description didn't mention any bracing, this is surprising for cabs of this power, even with good wood used, you talked about 1500 watts of power that's a fair amount. Most cabs do require panel bracing.

RE Anyways, let me know your thoughts on the above

Its possible cab weakness might be bracing and vent size in view of power levels used. Put your hand on box at different places during high drive, feel vibrations? Might indicate a lack of bracing. Also, at low drive this time, say 50-100W, is the bass sound better (e.g. deeper) than at high drive? If so could be an indication of port choking at high level. I hope the above helps you.

Richard

Hi Richard,

Thank you. Great info and very helpful for me.

I'm surprised that WinISD asks for all the measurements of the driver but doesn't take them into account when calculating volume of air space, etc... It seems like an obvious thing to do. But now that I know for sure it doesn't automatically do that, I can make more informed decisions about the information i'm getting out of the program.

As far as port tuning frequency, I modeled the cabinets with a number of different frequencies. It seemed to me that tuning to 30hz gave the best response curve. I tried 20, 25, 30, 35, 40, and 30 had the flattest and largest response curve and the best at low frequencies. Is this the appropriate way to decide this issue? Again, i've never really understood tuning a cabinet or port to a frequency. I don't understand the purpose of doing it. My guess is that it's done because it affects the overall frequency response. So tuning to one frequency will give a better overall frequency response than a different tuned frequency. Is that correct?

As far as other concerns like filters/crossovers/etc... I don't think that's the problem. The whole setup is run from my PC into a MiniDSP active crossover, then output to the amplifiers. Each driver has it's own channel of amplification (except the woofers which share one channel). Each driver has specific crossover settings and PEQ filters applied based on measuring the driver with a mic and using REW to generate filters to flatten the frequency response of each. I have opted to simply turn off any DSP or crossover built into my amplifiers.

As far as cabinet bracing. You may be right. The cabinets have minimal bracing. I simply built them as carefully as possibly, glued them, nailed them, and silicone sealed them, then put 10 plus coatings of poly on the outside. The only bracing I used was two triangular wedges for every 90 degree corner. Basically, whenever two sides meet at 90 degrees, I added two 4 inch wooden triangles that fit into the 90 degree corner and glued and nailed them into both of the sides. This is probably not the clearest explanation. But I think you probably understand what I'm saying. Here's a picture of what i'm saying:


This isn't my cabinet, but an example of something similar. Just basic corner wedge braces.

I'll do as you suggested and see if they're vibrating or flexing while in use.

One other thought, I installed steel ball bearing roller feet on both to make the cabs easier to move around. They are probably acting like isolation spikes since only the smallest portion of the round steel balls are actually in contact with the floor. Could decoupling the cabs from the floor be part of the problem? I imagine that free-floating, anti-gravity speakers would be less impactful since they wouldn't be anchored to anything. I'll try putting a soft mat underneath both cabs as a quick test. I don't know if this is part of the problem, but it seems relevant to me. I'm just not sure exactly how or why.

Again, thank you for your assistance. It's definitely instructive.

EZ

[RMC]

Winspeakerz too asks a ton of dimensional info, you're not required to fill all of them, in fact the software asks for more than i have data to enter. When my patience is depleted i enter the important ones only, e.g. Mfr, model, impedance, power capacity, size, depth. The TS parameters represent pretty much what the software needs to give you a picture of driver performance, maybe not dispersion pattern. The TS should all be entered for better/detailed results.

D.B. Keele has shown, with his approx vented box method, that only three parameters are required to model a box (Fs, Qts and Vas) in order to get Vb, Fb and F3. Naturally, you don't get all the nice extras such as excursion curve, max power curve, etc. In the old days this pocket calculator method is what audiophiles used, including me.

With regards to box tuning, you decide which response curve you prefer. Would be interesting if you post it. Personally i model for flat response or so, and don't lose sleep for a one db ripple. Yes, i think you got the trick, normally one tries to go for flat response and wide bandwidth. There's always exceptions, for a dance club box tuning for a bass bump might appeal to some DJs, less EQ required.

Oversimplified, think of box volume as determining driver LF capability and box tuning as a refinement adjustment for flat, bumpy, etc. frequency response. That's about as basic as i can explain it. Your woofer is a 40hz one and since it has a Qts of .42 in principle would lead to a fourth-order C4 low-frequency alignment. Not a "strong" C4 (like Qts .5 or .6) because its at the beginning of the range, but such alignment usually has some response ripple and the box is tuned somewhat lower than driver Fs. Your Fb is lower than driver Fs. You seem to be on the right track. I can take my retirement now!

What you show on pic is definitely not bracing, rather glue blocks or cleats. These reinforce the cab joints only. They don't reinforce the box panels to prevent them from vibrating. With the power levels you have in mind its important. Plus that might mean some more box volume addition for space taken in the cab by bracing.

On the other hand, since you're after the fact here (boxes already made), one way to try to "save the furniture" is to use cross shape bracing (+), front to back and left to right panels, midway. Used alone its not ideal in the present case but will improve the matter up to a point. Trying to "salvage" the situation to avoid new boxes and minimize space taken in the box by bracing (e.g. using pieces of 2"X2" glued and screwed).

RE Could decoupling the cabs from the floor be part of the problem?

YES. Most speaker softwares assume 2 Pi box placement. That means DIRECTLY ON THE FLOOR (or in a large wall). The ISD response you see on computer screen is a 2 Pi response. When cabs are elevated off the floor, low frequency diffraction loss sets in, reducing the amount of LF. So another item to take care of...

Richard

[zimpah]

Winspeakerz too asks a ton of dimensional info, you're not required to fill all of them, in fact the software asks for more than i have data to enter. When my patience is depleted i enter the important ones only, e.g. Mfr, model, impedance, power capacity, size, depth. The TS parameters represent pretty much what the software needs to give you a picture of driver performance, maybe not dispersion pattern. The TS should all be entered for better/detailed results.

D.B. Keele has shown, with his approx vented box method, that only three parameters are required to model a box (Fs, Qts and Vas) in order to get Vb, Fb and F3. Naturally, you don't get all the nice extras such as excursion curve, max power curve, etc. In the old days this pocket calculator method is what audiophiles used, including me.

With regards to box tuning, you decide which response curve you prefer. Would be interesting if you post it. Personally i model for flat response or so, and don't lose sleep for a one db ripple. Yes, i think you got the trick, normally one tries to go for flat response and wide bandwidth. There's always exceptions, for a dance club box tuning for a bass bump might appeal to some DJs, less EQ required.

Oversimplified, think of box volume as determining driver LF capability and box tuning as a refinement adjustment for flat, bumpy, etc. frequency response. That's about as basic as i can explain it. Your woofer is a 40hz one and since it has a Qts of .42 in principle would lead to a fourth-order C4 low-frequency alignment. Not a "strong" C4 (like Qts .5 or .6) because its at the beginning of the range, but such alignment usually has some response ripple and the box is tuned somewhat lower than driver Fs. Your Fb is lower than driver Fs. You seem to be on the right track. I can take my retirement now!

What you show on pic is definitely not bracing, rather glue blocks or cleats. These reinforce the cab joints only. They don't reinforce the box panels to prevent them from vibrating. With the power levels you have in mind its important. Plus that might mean some more box volume addition for space taken in the cab by bracing.

On the other hand, since you're after the fact here (boxes already made), one way to try to "save the furniture" is to use cross shape bracing (+), front to back and left to right panels, midway. Used alone its not ideal in the present case but will improve the matter up to a point. Trying to "salvage" the situation to avoid new boxes and minimize space taken in the box by bracing (e.g. using pieces of 2"X2" glued and screwed).

RE Could decoupling the cabs from the floor be part of the problem?



YES. Most speaker softwares assume 2 Pi box placement. That means DIRECTLY ON THE FLOOR (or in a large wall). The ISD response you see on computer screen is a 2 Pi response. When cabs are elevated off the floor, low frequency diffraction loss sets in, reducing the amount of LF. So another item to take care of...

Richard

Hi Richard,

You're the best! Thanks for your time and education. Your input has filled in a lot of gaps in my knowledge, and i'm confident I can do what needs to be done to get the best out of these speakers.

I'll repost after I have a bit of a chance to work some modification magic and see what my results are.

Cheers,
EZ

[short_circutz2]

The woofers are wired in parallel to one amplifier channel in order to get the maximum power out of the amplifier into each woofer.
EZ

600 watts at 8 ohms in stereo to each speaker, or 1200 watts from one channel into a 4 ohm load of 2 speakers in parallel resulting in...guess...600 watts to each speaker.

There is no logical reason to have things hooked up the way you do. Either way, the power to each woofer will be the same...

[Mr. Widget]

To answer you, I built a pair of speakers, each has one 2267H woofer, one 2069 midrange, and one D2 tweeter. The woofers are wired in parallel to one amplifier channel in order to get the maximum power out of the amplifier into each woofer.

If I understand your post, you have a stereo pair of speakers, but you are running the left and right woofers in mono and in parallel? If that's the case, you should try it wired conventionally and enjoy true stereo.

600 watts at 8 ohms in stereo to each speaker, or 1200 watts from one channel into a 4 ohm load of 2 speakers in parallel resulting in...guess...600 watts to each speaker.

There is no logical reason to have things hooked up the way you do. Either way, the power to each woofer will be the same...

... and then there is that.


Widget