2225 to 2235 refoam xmax question

[Loren42]

Disclaimer! This is not directed at anyone!!! OK;

I don't think people tinker (explore) enough, many read read read and puzzle but,,,

A 2225 is the 80's Chevy truck of the JBL line, I would be surprised if there is another 15 they made in larger quantities. I wouldn't discourage anyone from modifying either one.

So tune the box to 20Hz and see what you get, it won't be flat but what have you got to lose? Add some weight to the cone and see what you get. If you have the power and EQ what's the worst thing that could happen, pissed neighbors and or re-cone? 2225's are not in short supply.

Have some fun and see what you can learn.

Somewhere I once read, "Scientists are explorers, Engineers are pioneers, Philosophers are tourists."

You would probably not believe what I have run through a bandsaw over the years just to see whats inside. As a side note, every speaker that I have thrown in the trash since I was a kid got connected directly to the national power grid on the way out, I guess I will never grow up.

Have fun,
Barry.

I don't have a problem with experimentation, but you have to be careful with what you do.

For instance, tuning the box to 20 Hz is a bad idea. It is a common mistake for the uninitiated to do this, but it can have devastating consequences as shown below:



The red trace is the 2225H in the recommended vented enclosure (5 cubic foot) tuned to 30 Hz. The yellow trace is the same enclosure tuned to 20 Hz.

While it looks like you have extended the bass lower on the frequency plot, look at the cone displacement plot. Normally, xmax is reached at 26 Hz, but the cabinet tuned to 20 Hz exceeds xmax in two places, the first being at about 43 Hz, then again below 20 Hz. This is a good way to destroy a speaker that does not have a large xmax like the 2225.

The other experiment of adding mass to the cone is the green plot. I never really tried this except in theory, so I don't know what will really happen if you do this.

I added the 35 gram mass ring from the 2235H plus 15 grams of glue to approximate the mass of the 2235H cone. The result drops the Fs to about 26 Hz. Again the xmax is going to be close for this driver and you need a steep low cutoff at 26 Hz or higher to prevent over excursion.

Again, this is a theoretical experiment and the actual acoustic profile will most likely have gremlins that may not be desirable. The only way to determine that is to measure the speaker, which requires tools.

In reality, the xmax problems will not be a big problem unless you crank up the bass to compensate for the loss of bass. The yellow trace has a pretty steep rolloff and trying to compensate by assisting the rolloff with a bass boost goes badly wrong because the 2225H does not have sufficient xmax. You risk, at minimum, a lot of distortion, and at worst, voice coil damage.

[Artcore87]

Artcore,
Here are JBL's frequency response curves for the two woofers. As you can see, the 2225 is very predictable and has a rising response between 1k and 2k. the 2235 gets really hairy after 1k. I have never built with either of these woofers (Except the 2235 as subwoofers), so I will defer to the experts, but my opinion is that the 2235 will work better with a higher slope crossover at a lower frequency (600-800hz) so that the output is pretty low by the time those lumps hit. The 2225, however, is nice and predictable up high - more friendly with different crossovers. This is the reason you see 2225's used in DIY 2 ways with 2" horn drivers, while the 2235 is more often used in 4 ways or with mids that will go lower.

I think you can also see that inserting both drivers in your existing crossover will give very different performance at the crossover point to the midrange, hence the suggestion that you will need to modify the crossover.

I don't know what your other drivers or your crossover are, but I think you might find the 2225 blends better at the top of its range. Maybe keeping your 2225's and building a sub for the first octave is a consideration?
-John

HSCGuy, and Loren42, thank you very much for the hard data and insights, this is exactly the kind of helpful stuff I was looking for! I'm feeling better about the 2225's for a couple reasons, one being the better performance in it's upper ranges, since my midrange horn doesn't go all that low without sounding like a horn (below 1khz for sure). I'm also going to experiment with putting felt on the walls of the horn and modifying the mouth termination with sections of pvc covered in felt also to round off the termination, and also as a side benefit should get the end of the mouth out as far as the 1" wooden border around the baffle, to reduce it's diffraction effects I hope/think in theory. I may remove that border altogether... it just wouldn't look as pretty. In addition I will somehow damp the the horns internally (I've heard about some people using putty?) because there is some ringing during a sine sweep of moderate volume, which is not very noticeable with music, but obviously it's having some effect.
I will tune the box to 30hz as suggested, and it's gonna get some additional internal reinforcement (I believe there's enough displacement that lowering it might actually help... I have yet to measure to be sure though.)

Would anyone recommend or NOT recommend putting a variable l-pad (forgive me if they are ALL variable...) on the midrange? Does this degrade the signal significantly?

I only hope that I find the 2225's to be loud enough down to 30hz (not really worried about frequencies below that, no pipe organ music here... though I do listen to some electronic music with very low frequencies). A 30hz F3 would be awesome, 35hz would be good enough.

[Loren42]

Would anyone recommend or NOT recommend putting a variable l-pad (forgive me if they are ALL variable...) on the midrange? Does this degrade the signal significantly?

I only hope that I find the 2225's to be loud enough down to 30hz (not really worried about frequencies below that, no pipe organ music here... though I do listen to some electronic music with very low frequencies). A 30hz F3 would be awesome, 35hz would be good enough.

I use an L-Pad to set the level, then replace the L-Pad with resistors once you figure out the actual level you need.

Looking at the red trace you are 9 dB down at 30 Hz. In reality you will be another 3 dB (or more) down due to diffraction losses on the front baffle.

Looking at the numerical data on the plots I gave you shows F3 for the red trace at 97.7 Hz. That isn't anywhere near where you want it and the only way you can get close is with a 2235H.

I have a plot that shows an F3 of 36 Hz with a 2235H in a 5 cubic foot box tuned to 30 Hz. The 2235H likes more volume and my box has 6.7 cubic feet with an F3 of 32 Hz when tuned to 30 Hz.

No way will the 2225 get you what you want without a sub.

[Artcore87]

I use an L-Pad to set the level, then replace the L-Pad with resistors once you figure out the actual level you need.

Looking at the red trace you are 9 dB down at 30 Hz. In reality you will be another 3 dB (or more) down due to diffraction losses on the front baffle.

Looking at the numerical data on the plots I gave you shows F3 for the red trace at 97.7 Hz. That isn't anywhere near where you want it and the only way you can get close is with a 2235H.

I have a plot that shows an F3 of 36 Hz with a 2235H in a 5 cubic foot box tuned to 30 Hz. The 2235H likes more volume and my box has 6.7 cubic feet with an F3 of 32 Hz when tuned to 30 Hz.

No way will the 2225 get you what you want without a sub.

Well, I'm interested in that xmax graph because I only expect it to extend that low after I EQ it... I'm sorry I forgot to mention that and make it clear. I would like an F3 of 30-35hz after equalization, and I will drop the 16hz band (and maybe one above it) to prevent over-excursion below that... and when they are operating in that respect the only question then is how loud they'll be able to do that, and I'm hoping that it is loud enough. If not, 2235h's will be in my future, or a proper reconing. It seems odd to think that these 400w woofers cannot produce low bass as loudly as the cheap RTR 15"s that are in there now. Even they seem to have more xmax than the 2225... but maybe not, perhaps when I push them hard they're simply operating out of (semi) linear xmax range between that and xmech... but still they can go loud and clean, and at least have low distortion with single note bass notes and (with) or bass drum hits. If they sound that clean out of xmax then maybe pushing them won't sound too bad... they have the same xmech as the 2235h's. The only reason I would do that is for the those occasions when I simply want a healthy dose of bass, sometimes I am a bass fiend. Normally I'd keep them well within very-low-distortion limits for 90% of listening.


Buuut I think you're right... eventually I'm going to need 2235's for the real performance I want. This will just have to do for now! Thanks again.

[grumpy]

Last on this from me...

I'd recommend you cut wood and get a feel for how the 2225 will work
in your system (I'd further recommend a higher box tuning... more like
40Hz... should be nice and punchy, but do as you will).

Room gain will help -some- on the low end, but don't expect miracles.

Also, don't be surprised if the smoothness shown on old sales sheets
is not really comparable between decades of product... (2225 vs 2235...
it -appears that the 2225 curve is more a plot of the spec, -derived- from
data, where the 2235 plot actually represents a measured driver... having
measured one or two myself).

Get your system out there, show what you're assembling, ... have some
fun.

Later.

[Loren42]

Last on this from me...

I'd recommend you cut wood and get a feel for how the 2225 will work
in your system (I'd further recommend a higher box tuning... more like
40Hz... should be nice and punchy, but do as you will).

Room gain will help -some- on the low end, but don't expect miracles.

Also, don't be surprised if the smoothness shown on old sales sheets
is not really comparable between decades of product... (2225 vs 2235...
it -appears that the 2225 curve is more a plot of the spec, -derived- from
data, where the 2235 plot actually represents a measured driver... having
measured one or two myself).

Get your system out there, show what you're assembling, ... have some
fun.

Later.

I agree! Just keep expectations realistic and have some fun learning.

As Grumpy said, retune your cabinet to 40 Hz. I would also bump the internal volume up to 6 cubic feet (you can always make it smaller and it will allow an upgrade path for the 2235H).

Here is a comparison with the old 30 Hz tuned box 5 cubic feet (red trace) and a 6 cubic foot box tuned to 40 Hz (yellow trace).

You will have a much better sounding bottom end with the 40 Hz tuning.

[Artcore87]

I agree! Just keep expectations realistic and have some fun learning.

As Grumpy said, retune your cabinet to 40 Hz. I would also bump the internal volume up to 6 cubic feet (you can always make it smaller and it will allow an upgrade path for the 2235H).

Here is a comparison with the old 30 Hz tuned box 5 cubic feet (red trace) and a 6 cubic foot box tuned to 40 Hz (yellow trace).

You will have a much better sounding bottom end with the 40 Hz tuning.

Hmm, you're right. The yellow trace has better performance to 35hz, which is my goal... I'll then use EQ to cut the ultra low frequencies to stay within xmax. Thanks. I'll have pictures and more info as I gather it.

[Artcore87]

I agree! Just keep expectations realistic and have some fun learning.

As Grumpy said, retune your cabinet to 40 Hz. I would also bump the internal volume up to 6 cubic feet (you can always make it smaller and it will allow an upgrade path for the 2235H).

Here is a comparison with the old 30 Hz tuned box 5 cubic feet (red trace) and a 6 cubic foot box tuned to 40 Hz (yellow trace).

You will have a much better sounding bottom end with the 40 Hz tuning.

Loren42 would you be so kind as to run that program with a box volume of 4.6 cubic feet? It turns out this is the volume of my cabinets... and minus the horns internal volume and the drivers own displacement it's even somewhat smaller than that, you may be better able to approximate the difference... i would say no less than 4 cu. ft, maybe up to 4.2 cu. ft.

Currently it has one port that is 4.5" x 6.5", so please model that as well.

And, again, if you would so generously be able to find the optimum tuning for a box of this size (and optimum port dimensions), I would be most grateful.

Also, the port is in the rear... how would this act differently than if the port was in front? They'll see time both corner loaded, and simply back-loaded (one wall), in not-so-large rooms.

[Loren42]

Loren42 would you be so kind as to run that program with a box volume of 4.6 cubic feet? It turns out this is the volume of my cabinets... and minus the horns internal volume and the drivers own displacement it's even somewhat smaller than that, you may be better able to approximate the difference... i would say no less than 4 cu. ft, maybe up to 4.2 cu. ft.

Currently it has one port that is 4.5" x 6.5", so please model that as well.

And, again, if you would so generously be able to find the optimum tuning for a box of this size (and optimum port dimensions), I would be most grateful.

Also, the port is in the rear... how would this act differently than if the port was in front? They'll see time both corner loaded, and simply back-loaded (one wall), in not-so-large rooms.

This is what I get for 4.2 cubic feet. I think you would be happier with the results if you could get the internal volume up to 6 cubic feet, but if you must use this volume, here are the details, which is not that bad:



The rectangular vent is fine. The maximum air speed is 12 m/s, which is quite acceptable. For comparison and contrast, making the box 6 cubic feet yields the following:



The yellow trace is the 2225H. The orange trace is a 2235H. All you need to do in this case is change port length to retune the box lower and drop in the 2235Hs. Notice the F3 on the 2235H. Six cubic feet gives you the option to change drivers without making major changes to the box. All you need to do is change the port length, which in this design is two 4" diameter vents.

What I have given you is only the first cornerstone of the design. There is a lot more work to do to complete a design. Vance Dickason's The Loudspeaker Design Cookbook would go a long way toward completing the design.

As far as a rear firing port goes, I don't think it makes a big difference, but perhaps someone else here might speak up about that?

[Artcore87]

This is what I get for 4.2 cubic feet. I think you would be happier with the results if you could get the internal volume up to 6 cubic feet, but if you must use this volume, here are the details, which is not that bad:



The rectangular vent is fine. The maximum air speed is 12 m/s, which is quite acceptable. For comparison and contrast, making the box 6 cubic feet yields the following:



The yellow trace is the 2225H. The orange trace is a 2235H. All you need to do in this case is change port length to retune the box lower and drop in the 2235Hs. Notice the F3 on the 2235H. Six cubic feet gives you the option to change drivers without making major changes to the box. All you need to do is change the port length, which in this design is two 4" diameter vents.

What I have given you is only the first cornerstone of the design. There is a lot more work to do to complete a design. Vance Dickason's The Loudspeaker Design Cookbook would go a long way toward completing the design.

As far as a rear firing port goes, I don't think it makes a big difference, but perhaps someone else here might speak up about that?

Thank you, but I have to apologize I was not clear about the port... it is 4.5" diameter, 6.5" long.... any chance you could re-model with that spec? And I will read the suggested material.

[1audiohack]

I have years ago beat the 2225 thing to death in my home trying to make them something they are not. I hope you don't think I would intentionally lead anyone astray, sometimes it seems you've just got to hear the differences yourself before you concede the fact that forcing a driver into a duty it was not designed for is folly.

Anyway you said the RTR's play clean and loud right? I would continue the path your now on with the 2225's and make a box for the RTR's and use them for subs.

It's very gracious of Loren42 to run sims for you!

All the best.

[Loren42]

Thank you, but I have to apologize I was not clear about the port... it is 4.5" diameter, 6.5" long.... any chance you could re-model with that spec? And I will read the suggested material.

If you look at the gray box on the left of my plots, under Box Properties you will see a term Lv = 6.174". That is the length of the port. The heigh and width are 4.5" by 6.5".

Your 4.5" diameter round port is too small. I would suggest two 4" ports or go with one 6" inside diameter port.

If you want a flush mount tubular vent, just calculate the length as follows:

Lv = [ (1.463 * 10^7 * R^2) / (Fb^2 * Vb) ] - 1.463 * R

Lv = port length in inches
fb = Desired tuning frequency
Vb = box volume in cubic inches
R = radius of port in inches

The minimum port area in your case works out to a minimum of 25 square inches. Two 4" diameter ports have an inside area of 12.57" each. Just double that number and calculate the value of R using PI*R^2 = A (it works out to a diameter of about 5.768". Calculate R by 5.768"/2 = 2.884"). Use that calculated value of R (2.884") for calculating the length of two 4" ports. Make sense?

[Artcore87]

If you look at the gray box on the left of my plots, under Box Properties you will see a term Lv = 6.174". That is the length of the port. The heigh and width are 4.5" by 6.5".

Your 4.5" diameter round port is too small. I would suggest two 4" ports or go with one 6" inside diameter port.

If you want a flush mount tubular vent, just calculate the length as follows:

Lv = [ (1.463 * 10^7 * R^2) / (Fb^2 * Vb) ] - 1.463 * R

Lv = port length in inches
fb = Desired tuning frequency
Vb = box volume in cubic inches
R = radius of port in inches

The minimum port area in your case works out to a minimum of 25 square inches. Two 4" diameter ports have an inside area of 12.57" each. Just double that number and calculate the value of R using PI*R^2 = A (it works out to a diameter of about 5.768". Calculate R by 5.768"/2 = 2.884"). Use that calculated value of R (2.884") for calculating the length of two 4" ports. Make sense?

It does make sense, thank you very much. Since the cabinets already have the one 4.5" circular hole cut in them I will simply go with a single 5 7/8" port.

And 1audiohack, I can't BELIEVE I didn't think of that myself!!! I was planning on selling them and some other drivers I have to a friend as a first speaker build project on a budget. The RTR drivers don't have the upper bass/low midrange detail that I want that the 2225's should provide, but as subs for under 40-50hz they should perform just fine... with 2 15"s at normal listening levels they shouldn't have to move too much and should thus be very accurate. I'll power them with another 45w (thats a CLEAN <.01% thd 45 watts) tk2050 tripath amp like I have driving the mains. These woofers have much tighter bass with this amp than any other I've tried (several decent mid-fi receivers and a vintage onkyo). If I get a behringer electronic crossover I will buy 2 lower power tripath amps for the tweeters and horns, for a total of four amps.

If I decide to go the subwoofer route... would it be better to tune the 2225's higher than originally planned (40hz) for optimum performance in their range? And secondly what should that range be... I could x-over as low as 40hz to the subs... but would 50hz be better, with a higher main cabinet tuning? I'd like to keep the accuracy and detail of the 2225's as low in the range as I can. Also, in this case should I high-pass the 2225's? Or would adding additional x-over components compromise sound quality too much? If I leave the mains passively crossed, how should I cross between them and the subs? Another passive with a high pass and low pass between 40-50hz, 12db/oct... steeper? Or would I simply split the full signal from the point of my graphic EQ and send the full signal to the mains and the subs, with a passive x-over in the subs? Or would this screw up impedance matching at the amplifier input?

Unfortunately there is no info online (that I've been able to find) about these RTR drivers, so I don't know where to start as far as building a cabinet... except for guessing of course. I don't have the necessary equipment to figure out their specs. They were always sold in vented enclosures, so I'll assume it's best I go with vented and that they would not work well sealed... good assumption? I'll give them each their own box and run stereo... use them as speaker stands (and flip my speakers vertically so the horns/tweeters stay at ear level, with the 2225's on top). I'd like to tune them to 25hz. I wouldn't tune them lower than that because I doubt these woofers would really operate well THAT low, since they're not subs or extreme woofers in the least. I mean they're not bad, decent sized magnet assembly... a lot bigger than the magnets on my KLH 15"s which I also still have... bigger spider as well, and it looks like a slightly larger voice coil. How would I approximate a good box volume without knowing the driver specs?

[Loren42]

How would I approximate a good box volume without knowing the driver specs?

Get Vance's book. He explains exactly how to determine the driver's Theil/Small constants empirically so you can design the best enclosure. You don't need as much equipment as you think.

[clubman]

I have a pair of 2225's in a large box(cant remember cubic feet) tuned to 40hz and love the sound they have. They are used as HT mains