Garage Speakers using JBL 2204

[glenn]

First off thanks to everyone who contributes to this site. Ive learned much and am grateful for the information
I was going through my stuff and found a pair of 2204h woofers and some 104h2 midrange drivers. Im thinking of making a pair of speakers so I have music outside in my garage. I ran some winisd simulations and a 3.1 cubic foot box tuned to about 45 hz looks like a good choice. The crossover point between the 2204 and the 104h2 would be 800 hz. Does this seem reasonable so far? I havent picked a tweeter yet (none in stock here) so if anyone has a recommendation I would appreciate it. I do have a three way active crossover and three spare crown amps so I can go that wa but I would rather keep it simple with a passive crossover and one amplifier .
Let me know what you think good or bad. I realize that the 2204 doesnt have the most low end in the world but 2203 are hard to find. In this application I think that these will sound punchy which is a good thing.
Any advice is much appreciated.
Regards,
Glenn

[RMC]

Hi Glenn,

Your project does make some sense, though I don't have much tech info about the 104h2 (5" cone). It will also depend on your choice of tweeter... I do have some observations that I share with you below.

In the 2204H data sheet (10/85), JBL did give an example of response curve for a vented box of 2 ft^3 with tuning (Fb) at 50 hz. The response is pretty flat, as in one of my simulations, but with an F3 of about 55 hz.

I did run a few other quick simulations in Winspeakerz speaker design software, with your proposed box of 3.1 cu. ft, and tunings at 45, 50 hz and in between these two to see effect on bass response. Since that driver showed a little rising response of a good 1 db (above bass output) from about 200 hz and up, your 45 hz tuning (F3 about 42 hz) showed a little weaker bass level than the mid-bass output. Not exactly as punchy sound as you expected. Maybe a little more mid-bassy sound...

"In this application I think that these will sound punchy which is a good thing." To about match bass/mid-bass levels and to get a little punchy sound from that box/driver combination, I would rather look at a 50 hz box tuning (F3 about 45 hz). Not much difference (loss) between F3 of 42 and 45 hz re low- bass, but at least your bass level in the 55-65 hz range, where a small bump would be, will then be about equal to mid-bass level, for a more balanced sound and a bit punchy. Thiele/Small analysis assumes half-space box placement, meaning your speakers should be at or close to one reflective surface (e.g. floor or wall).

The sensitivity rating of the 2204H is given by JBL as 95 db, 1 W, 1 M (100-500 hz) and that of the 104h2 MAY be 91 db, so some level matching issues could be on the way for you... The Tech Manual of the JBL L100T which uses the 104h2 says the loaded box sensitivity is 91 db (same for midrange driver or padded down?) and the x-over used is at 800 hz to mid (good for you) at 12 db/oct. JBL tech sheet on 2204H says the highest recommended x-over is 1,200 hz, if need be. Search to find the exact 104h2 sensitivity rating to help.

Matching drivers' sensitivities here may not be as critical as it seems. Since these are garage speakers, not studio monitors, and considering you already have the most expensive drivers needed, thereby saving money, you may still want to try the two drivers (bass and mid) together anyway in view of the Fletcher and Munson curve which showed the human ear is more sensitive to midrange sounds in the 1,500-6,000 hz or so range, than to lows and highs. It is possible you perceive after all, that even though there might be a 4 db level difference between low and mid drivers, the result remains quite acceptable to your ears because of their naturally higher sensitivity to midrange sounds... Maybe some mid-bass EQ on 2204H to reduce that level a little?

I don't know how powerful of a Crown amp you're planning to use, and how loud you plan to listen, but the software tells me that in the BASS range you can expect to reach Xmax at 40 hz with 300 W and a SPL of 120 db at 60 hz. It is good practice with vented boxes to high-pass filter the low range, just below box tuning frequency (e. g. 40-45 hz if Fb at 50 hz), with a steep filter.

The driver displaced volume in the box for 2204H is given as 4 L (0.15 cu. ft.) by JBL, plus add that of other components and good bracing for your gross box volume calculation. My software recommended a minimum vent area of 23 sq. in. and JBL used a 25 sq. in. vent in their example above, most probably because of the 2204H power input rating of 350 W. Two 4" dia. vents would meet that requirement. If you don't ever plan to use that much power at any time, then you may get away with a smaller vent area.

BTW There can be small differences in results from one software to another when making simulations. Regards,

Richard

[RMC]

Hi Glenn,

Sorry, I can't help you on fixing your damaged driver since I'm not a speaker repair expert. But you may try submitting your driver repair question to Ken who is usually very helpful here. He posts under the name "EDGEWOUND" here in the forums. His web site is uplandloudspeaker.com He's a JBL repair expert...

In the near future I will have another important suggestion for you regarding the matching of driver sensitivities and diffraction losses in the low frequencies.

According to John L. Murphy, "Loudspeaker designers have traditionally compensated for the diffraction loss by reducing the level of the tweeter and making other adjustments in the crossover." Which you may have to do for mid and high frequencies, depending on speaker placement and also on number of mids used. Regards,

Richard

[glenn]

Richard,
Thank you for taking the time to look at my project. Im going to do a comparison between Winisd and Eminence speaker design software to see what you brought up about the differences in models. I also did some looking on the 104h2 and the sensitivity difference between it and the 2204. Im looking into if the 104h2 is a good driver to use on the edgar midrange horn. If it is a workable combination that would boost the sensitivity of the midrange driver over that of the 2204. I like to build things so that would make a nice project . I dont have any compression drivers in stock but maybe someone can suggest on that will do the job and not be overly expensive. Im not anchored to using the 104h2
On the tweeter front Im wondering if you or anyone has an opinion on the Selenium/JBL tweeters that are out there? They look like possible candidates but I have no idea as to their sonic qualities. 2402 bullets would also work well but I would have to track a pair of those down.
The crown amps I have are the DC300a Series 2, D150a Series 2, and some D75's. Though I work in broadcasting (High power shortwave transmitters) and understand the "basic" design of crossover networks, the subtleties of crosssover design is beyond me. Thats the difference between a good crossover and a bad one. I need to find some good information on the subject. I could take the easy way out and triamp but thats alot of gear in the garage!
Thank you for your guidance,

Glenn

[glenn]

Richard and Ivica thank you for you suggestions. Richard I made note of all your suggestions and will start building in the fall when its not so darn hot and humid here in NC. Ill post a question about welding that frame up on another forum. I may just contact edgewound directly and see what he says. I hate to just toss it.
Thanks again for your time and your hard work,
Glenn

[DualTriode]

I now have something similar going on.

I have a new garage and need some tunes. First I will try the tri-amp arrangement to get it going. Then I will cobble together a passive crossover when I settle on mid drivers.

I like the 2204 H better than the 2206 H. The 2204 H has an additional half octave bass extension.

On the shelf I have a pair of both 2123H’s & 2118H’s and cross them over at 400hz. The 2123H’s will get the first tryout. I have the impression that the 104’s will struggle to keep up (as in less efficient and less power handling ability). Often you will see 2123H’s on ebay. Currently there are several pairs of 2118’s listed.

For tweeters I plan to use a pair of 2425H’s compression drivers crossed over at 2Khz, also on the garage shelf.

Thanks DT

[DualTriode]

I have been playing with these drivers.

The mid is now a JBL 2123H crossed over at 300hz and 2000hz.

At the 2k cross over the Directivity Index matches the JBK 2425H/Eighteensound XT1086 horn/waveguide.

The 2204 2123 2425 threesome sounds really good without a lot of high frequencies bouncing around the room.

Thanks DT

[RMC]

Hi Glenn,

Your project seems to be taking a different turn with your reply on edgar horn, compression driver and 2402 tweeter (40° conical dispersion). Work well in a garage ? Long-throw stuff in a usually short-throw environment ! It doesn't sound serious to me, unless you have a really HUGE garage...

I did have a look recently at a JBL/Selenium tweeter model ST 350 which seemed interesting in their 2014 Black Catalog with 120° H X 40° V dispersion. However, in the 11/05 spec sheet for same tweeter on JBL's site the dispersion pattern is given as 50° X 50° with no explanation as to why the difference and no indications that the tweeter was changed or modified... They have to get their act straight it seems.

The advantage of having many amps as you do is bi or tri-amping with adjustable gain controls on each amp to match the sound levels, i. e. adjust gain for the differences in drivers' various sensitivities. By bi-amping only you may save some gear: amp for bass and amp for mid/tweeter, with passive X-over for mid/high only and possibly some padded down on tweeter for level matching with the rest.

Richard

[glenn]

Richard,
After looking at things it may be best to use something other than the 104h2 for the midrange. Ive got 9 of them here and just wanted to use stuff on hand. Might be best to sell them and buy something more in line with the 2204. In the house I use a pair of 4408a with a home built sub that uses a special crossover much like Rod Elliots elf sub. I cross it over with a modified Behringer crossover and a DC 300A for the sub in bridge mono and a pair of D75A for the two 4408a. Works very well. I thought about taking it outside and building something for in the house but Kim is unhappy with big boxes in the living room these days.
Ill look int the ST350 . It maybe a good driver for me. Ive heard little pro or con on Selenium compression drivers so used JBL may also be the best solution in the end.
Thanks again for your input.
Regards,
Glenn

[RMC]

Hi Glenn,

I'm trying to help you save your original project since there's some good things in it. Plus putting your 104h2 up for sale sounds like the waste of good midrange drivers, since in my recent search for specs for these I noticed there's a ton of them up for sale on the Web, which means you may not get a good price for them.

With so many 104h2 you have in stock, here is a worthwhile idea to consider seriously in my view and would mostly settle the sensitivity issue with 2204H vs 104h2. Use double 104h2 per box, wired in parralell and in a one above the other configuration. You will gain 3 db sensitivity from the pair for a more acceptable 94 db compared to 2204H 95 db.

JBL uses the 104h2 in L100T and L100T3 enclosures which both have rated sensitivities of 91 db. They are used from 800 hz to 4,500hz (high enough for a good choice of tweeters).

I'll be back tomorrow with more on pros and cons of using two of these, I'm in a rush now.

Richard

[RMC]

RE: "The 2204 2123 2425 threesome sounds really good without a lot of high frequencies bouncing around the room.


Hi DT,

But that doesn't mean or guarantee that the lows and/or low-mid won't be bouncing around in the garage? Regards,

Richard

[DualTriode]

Hello Richard,

Yes there are low frequencies bouncing around in an Modal way and there are direct and reflected high frequencies, in between the low and high frequency regions there is a Schroder frequency.

https://www.soundandvision.com/conte...nd-tell-part-1

Thanks DT

[RMC]

Hi Glenn,

After some more search, I found the 104h2 is also used in the 4410 and 4412 studio monitors. In the spec sheets of these two monitors more interesting info is available about the 104h2 compared to L100T and L100T3 spec sheets. For both of these 44 series monitors it says " The basic sensitivity of the midrange transducer is 94 db, one watt at one meter, ...". Only 1 db difference with 2204H. Great news for your sensitivity issues! That makes your pile of 104h2 shine even more...

Both of these boxes use crossovers at 800 hz and 4.5 khz as in the L100T and L100T3 mentioned previously. Moreover, the impedance curves for these 44 series boxes show an impedance of about 10 ohms at 800 hz crossover frequency for 4410/4412 and going down to about 7-8 ohms at 3.5 khz, then rising again. This 10 ohms impedance number represents valuable info for your crossover design purposes, supposing it has not been "doctored" by JBL's crossovers in those boxes.

In order to get some type of very rough idea, I would take DC resistance measurements on your many 104h2 using the resistance measurement (ohm) function of a digital multi-meter. See the results you get.

Many speaker design softwares also include a crossover design application, Win ISD Pro for example has a small basic one, but you can easily find many better ones on the Web.

Just in case you would still want to pursue the double 104h2 idea re increased output: I mentioned in my last post a 3 db sensitivity gain for two close 104h2, wired in parallel, one above the other and both with same power input. But these two can also accept twice the input power compared to a single one. Hence another possible 3 db gain in output (SPL), for a total of 6 db greater output ability with two units compared to a single one. I say close for coupling effect and one above the other to preserve horizontal dispersion.

There's a side-effect to mutual coupling of drivers in a box (e.g. 2 X 5") , you should be aware of about the drivers' sound dispersion: BEAMING. This is a directional property occuring when multiple cone drivers are close to each other. " The assumption is made that the drivers are arrayed in a vertical line, so that maximum coverage will be obtained horizontally." (John M. Eargle (JBL), Loudspeaker Handbook, Chapman & Hall, 1997, P. 80). Vertical driver alignment leads to increased vertical driver beaming ("focus") and maximum available horizontal coverage. Horizontal driver alignment leads to increased horizontal driver beaming and maximum available vertical coverage!

Taken from a 1996 Electro-Voice data sheet "Cone loudspeakers may be stacked for greater acoustic ouput and a narrower beamwidth. (...) This principle is already employed in the dual-woofer TL770D, and is responsible for the higher sensitivity and narrower vertical beamwidth (with the system long axis vertical) relative to similar single-woofer systems.

By analogy, a similar situation of beaming also happens with mid/high frequency horns combined. In his "Handbook of Sound System Design", ELAR, 1989, in section "Directional properties of combined radiators", sub-section "Techniques for narrowing coverage", John Eargle of JBL states "In-line vertical stacks of horns have long been used for producing narrower vertical coverage that can be provided by a single horn. (...) Note that the horizontal pattern of these arrays is the same as for a single horn. If horns are placed side by side, horizontal coverage will be narrowed, again at the expense of considerable lobing. (...) Note here that the vertical pattern of the horizontal array is the same as that of a single horn." (P. 90-91). Read as if horns = Midrange drivers for better understanding.

If double midranges are a must, then have the box/midrange drivers vertically aligned for improved midrange horizontal dispersion up to tweeter, and improved stereo effect. Depending on tweeter choice, somewhat matching tweeter horizontal dispersion pattern, as the case may be.

For direct sound to the listener, the beaming issue isn't a big thing if you are mostly on-axis and not too close. If you move around or are located off-axis, it may be noticeable. As for reflected sound in a room, it will depend if none/small/large imbalances exist with coverage patterns from each driver used.

A "non-conventional" suggestion for a tweeter could be: Scan-speak Discovery H2606/9200 Horn Loaded 1" Textile Dome Tweeter, from Denmark at Madisoundspeakerstore.com in the USA ($36.80US). This is a 1" dome tweeter in a small horn for added sensitivity (95 db). This unit replaces the Vifa H26TG05-06, which replaced the Vifa H25TG35-06 that I have a few of. Nice sounding sound reinforcement tweeter at a great price/performance ratio. Don't be mistaken on the specs page, the response graph scale at bottom right IS 100,000 hz not the usual 20 khz... Max input 100-200W with 12 db/oct. high-pass filter at 2.5 khz depending on noise or music.

Richard

[glenn]

Richard,
Thanks for all the wonderful information. Being that I have nine 104h2 driver Ill put two in each box. That will give me a little more head room and allow adjustment of the level between the woofer and the midrange. I work in shortwave broadcast and we use what is called a curtain array antenna. It is nothing more than a stack of dipole radiators in the vertical and horizontal planes. The number of dipoles in the vertical stack determines the radiation angle just like in the speaker array and the number of stacks in the horizontal plane determines the beam width. A wave is a wave if its audio or electromagnetic. Im guessing I need to study up on loudspeaker theory! I do remember JBL made couple of line arrays years ago using four LE5's (I think) and another using a pair of Le5 and a slant plate lens along with if my memory serves me right four D208. Im guessing that whole idea was pattern control.
Im going to take your advice and start looking at getting those tweeters and doing some modeling using the two different programs I have.
Can you suggest any books that deal with the subject of crossover networks? I dont like using computer programs unless I have a thorough understanding of the subject. Garbage in garbage out. Im good at designing filters in the HF range but there are some things in crossovers that near as I can tell arent that straight forward.
I appreciate your time and effort helping me down the road of fewest bumps. When I get it done Ill send you a picture or two
Regards,
Glenn

[ivica]

Richard,
Thanks for all the wonderful information. Being that I have nine 104h2 driver Ill put two in each box. That will give me a little more head room and allow adjustment of the level between the woofer and the midrange. I work in shortwave broadcast and we use what is called a curtain array antenna. It is nothing more than a stack of dipole radiators in the vertical and horizontal planes. The number of dipoles in the vertical stack determines the radiation angle just like in the speaker array and the number of stacks in the horizontal plane determines the beam width. A wave is a wave if its audio or electromagnetic. Im guessing I need to study up on loudspeaker theory! I do remember JBL made couple of line arrays years ago using four LE5's (I think) and another using a pair of Le5 and a slant plate lens along with if my memory serves me right four D208. Im guessing that whole idea was pattern control.
Im going to take your advice and start looking at getting those tweeters and doing some modeling using the two different programs I have.
Can you suggest any books that deal with the subject of crossover networks? I dont like using computer programs unless I have a thorough understanding of the subject. Garbage in garbage out. Im good at designing filters in the HF range but there are some things in crossovers that near as I can tell arent that straight forward.
I appreciate your time and effort helping me down the road of fewest bumps. When I get it done Ill send you a picture or two
Regards,
Glenn

Hi Glenn,

If talking here about UHF section of the speaker I want to pay your attention to the SELENIUM D220Ti driver and Selenium HM 17-25 horn,

https://www.parts-express.com/seleni...-8-18--264-270
https://www.parts-express.com/seleni...8-tpi--264-308

I have very good experience with them, and the price is acceptable, I think
and You will be remained "living in the JBL yard"


regards
ivica