DIY speaker project

[zeeman]

Hi Guys,
I'm a long time reader of the Lansing Heritage site, but a new member. I'm in my 50's and have been a speaker nut since my teens. I built my first speakers ( all Electro-voice) when I was 19yrs old, almost 40 years ago.I'm now building a large speaker system and I could use your help modifying a crossover. The system is comprised of 2- JBL 2235H 15" woofers in parallel crossing to a TAD TD-4002 (16ohm)compression driver with a JBL 2380A horn and finally crossing to a JBL 2405 (8 or 16 ohm) ring radiator tweeter. I purchased 16 ohm versions of the 2405 diaphragm, but an 8ohm version is in them now. I will use whichever is better for the system.
I plan on building the charge coupled capacitor version of the L-300 crossover that I found on this forum. .
The original L-300 has a 1- 8 ohm woofer, a 16 ohm horn mid and a 16 ohm version of the 077 tweeter. My question is this: what mods would I need to make to the crossover because the woofer section is a 4 ohm load instead of 8 ohm? I'm using 2 woofers (93dB / watt each) so I won't have to turn down the mid horn ( 110dB / watt) too much, and for high efficiency. The tweeter is rated at 105 dB / watt. The woofers will be above and below the mid horn and tweeter in a tall (60") and relatively thin (17.5") cabinet that is 10 cubic feet net that the woofers share. There will be 4 - 4" ID ports in the back that will be 9.40" long. I used BassBox pro to get the box tuning (30Hz) and port length. I'm also kind of copying JBL older designs as both the L-300 and 4430 use a 5 cubic foot cabinet for a 15" woofer, so I figured I'd use a 10 cu.ft. cab for 2 of them. My speaker building experience has taught me that things generally go better when I copy what JBL did. Those JBL engineers knew a thing or 2 about speaker design. As the saying goes: "If you're going to copy something, copy the best".The cabinets are about 2/3 built and I have all the components. I'd like the crossover points to be around 800Hz and around 10kHz and I really like the charge coupling of the capacitors w/ a 9v battery

Do you think you guys could help me out with the crossover design? I wouldn't mind if you guys checked that the cabinet tuning and port length is correct as I don't trust some of the answers I get from Bassbox. You guys are the experts, not some $100 software. I will post pics when the speakers are farther along. Any help will be greatly appreciated.
Thanks,
Zeeman

[Horn Fanatic]

Hi Guys,
I'm a long time reader of the Lansing Heritage site, but a new member. I'm in my 50's and have been a speaker nut since my teens. I built my first speakers ( all Electro-voice) when I was 19yrs old, almost 40 years ago.I'm now building a large speaker system and I could use your help modifying a crossover. The system is comprised of 2- JBL 2235H 15" woofers in parallel crossing to a TAD TD-4002 (16ohm)compression driver with a JBL 2380A horn and finally crossing to a JBL 2405 (8 or 16 ohm) ring radiator tweeter. I purchased 16 ohm versions of the 2405 diaphragm, but an 8ohm version is in them now. I will use whichever is better for the system.
I plan on building the charge coupled capacitor version of the L-300 crossover that I found on this forum. .
The original L-300 has a 1- 8 ohm woofer, a 16 ohm horn mid and a 16 ohm version of the 077 tweeter. My question is this: what mods would I need to make to the crossover because the woofer section is a 4 ohm load instead of 8 ohm? I'm using 2 woofers (93dB / watt each) so I won't have to turn down the mid horn ( 110dB / watt) too much, and for high efficiency. The tweeter is rated at 105 dB / watt. The woofers will be above and below the mid horn and tweeter in a tall (60") and relatively thin (17.5") cabinet that is 10 cubic feet net that the woofers share. There will be 4 - 4" ID ports in the back that will be 9.40" long. I used BassBox pro to get the box tuning (30Hz) and port length. I'm also kind of copying JBL older designs as both the L-300 and 4430 use a 5 cubic foot cabinet for a 15" woofer, so I figured I'd use a 10 cu.ft. cab for 2 of them. My speaker building experience has taught me that things generally go better when I copy what JBL did. Those JBL engineers knew a thing or 2 about speaker design. As the saying goes: "If you're going to copy something, copy the best".The cabinets are about 2/3 built and I have all the components. I'd like the crossover points to be around 800Hz and around 10kHz and I real like the charge coupling of the capacitors w/ a 9v battery

Do you think you guys could help me out with the crossover design? I wouldn't mind if you guys checked that the cabinet tuning and port length is correct as I don't trust some of the answers I get from Bassbox. You guys are the experts, not some $100 software. I will post pics when the speakers are farther along. Any help will be greatly appreciated.
Thanks,
Zeeman

First, I have had very good results using Bassbox Pro. Two, with the component compliment you have in mind I would be building myself a pair of 4350 monitors. As for the venting, I favor large vents, and steer clear of multiple smaller vents. For a box the size of the 4350 I would use two six inch diameter ducts, and angle them, because they will be longer than nine inches. I find using large vents results in much tighter articulate bass at the tuning frequency. I also go for a symmetrical impedance on both sides of the lowest impedance. Unless you have the floor space to facilitate a sound stage, I would vent through the baffle so as to deploy the system against a wall.

Good luck, H.F.





























The dividing network design of course, will require consideration of the sensitivity of all the drivers. I myself do not use variable L-Pads in my networks, because a properly designed network doesn't require them.

[Allanvh5150]

Hi Zeeman,

The easiest way to run 2 woofers on that crossover is just to duplicate the 2235 section of the crossover. As for the 2405, they were all 8 ohms regardless of what was written on the foilcal.

Regards, Allan.

[macaroonie]

My experience with a similar setup ( 2235 / 2380 + B&C 85Ti / 2405 ) in 5 cu ft is that the 2235 can sound muddy up at the 800hz crossover.
I am running fully active so it was easy to gradually adjust. I have settled with a turnover to the horn at 650hz , this has removed a lot of the ' mud' in the midband. The 2235 has the mass ring as you know and while it is a super driver I really don't think it's up for playing cleanly at 800.
M

[zeeman]

Hi Zeeman,

The easiest way to run 2 woofers on that crossover is just to duplicate the 2235 section of the crossover. As for the 2405, they were all 8 ohms regardless of what was written on the foilcal.

Regards, Allan.

Hi Allen,
I'm not sure what you mean about duplicating the 2235 section of the crossover. Do you mean another woofer branch of the crossover? By that I mean another 2.2mH inductor, 4 more 40mF capacitors and 2 more 3 megohm and 1 more 10 ohm resistors arranged the same way and connected to the same main inputs. Would it be possible for you to make and post a sketch of the circuit design?
Best Regards,
Henry

[zeeman]

Hi Allen,
I'm not sure what you mean about duplicating the 2235 section of the crossover. Do you mean another woofer branch of the crossover? By that I mean another 2.2mH inductor, 4 more 40mF capacitors and 2 more 3 megohm and 1 more 10 ohm resistors arranged the same way and connected to the same main inputs. Would it be possible for you to make and post a sketch of the circuit design?
Best Regards,
Henry

I went and made a sketch of the circuit you proposed and posted a few pics of this project so far. Please let me know if this is what you meant about copying the bass section.

[Allanvh5150]

Hi Zeeman, that looks fine. It is certainly not a small box.

Allan.

[Doctor_Electron]

When cloning xover designs, if you feel the original does an excellent job then you should pay careful attention to the dc resistance of any and all inductors. Their DCR plays a role in slope shape as per attenuation interactions with the cabinet and other drivers. The overal interactions, attenuations, and slope shapes is are very complicated, and I have always felt that the designers knew what was up.

For example, I had a pair of Xovers from 4350's. After quite an effort in carefully melting and removing the (?) wax-sand ) potting in them and analyzing individual components with a General Radio Digibridge, some values of the inductors did not match theoretically calculated values. Those associated with the woofers had DCR's of 3.5 ohms, where a custom built coil might have almost no DCR at all. But one value is but a piece of the final "mix"

I'm sure that the electrical values of these coils were worked out for the best results possible as regarded the complex overall responses.

You could add a high quality non-inductive resistor in series with an available coil of low DCR to dial it in.

I think you mentioned 2235's. Is it not better when dual 2235's are used, to have a slight value of high-pass functionality in series with one of them to help them mesh?

Could be an important aspect.

From what experimenting and tuning of my 2235 based subs, they sounded much better with -3dB @ 125 hZ than my first stab of -3 dB @ 250 hZ. Drew Daniels and I did quite a bit of phone time as he helped me work things out (thanks Drew) and he concurred on that issue.

But no matter what it's a fun and challenging pursuit, and when they are done right and posiitioned right in the listening room you will reap the harvest. Those subs with the 4410's as MB/Mid/High above 125 hZ were very clean and dynamic. On a great recording you could hear the detail and timbre of every bass note played... which is tough result to achieve.

If I could just get a hold of two now for short of a King's ransom I would do it again.

They were truly awesome.

[zeeman]

I know it's been quite a while since last posted, but the project has progressed. The cabinets are built, veneered, stained and varnished. The crossovers are built and tested. Since the crossovers have large components and that I wanted to mount them on the back of the cabinets to keep them away from the driver magnets, I split the crossover into 2 sections - the bass section and the mid/high section
I remember one of you guys suggested that perhaps both woofers shouldn't maybe cover the exact same range, so I started looking at other JBL models that use 2-15" woofers and ran across the JBL 4435 which uses 2 very similar woofers to the 2235H's I'm using. Their spec sheet (www.jblpro.com/pub/obsolete/443035.pdf) states that one of the woofers operates below 100 Hz, which sounded like a good idea, so I went looking for the schematic for the 4435 crossover (www.jblproservice.com/pdf/studio%20monitor%20series/4435lr.pdf) found that it's fairly easy to do. Here's a sketch of the schematic using the 4435's circuit for the low range woofer and sharing the negative terminal of the upper range woofer, which I kept the same as I posted before( the charge coupled L-300 crossover with the extra woofer)
Does this seem like a good idea to you guys? I'm concerned that maybe the lower woofer sharing the negative terminal with the upper woofer would maybe screw up the crossover for the other woofer or something like that. if so, I could completely copy the 4435's bass section crossover, but I'd prefer not to cross at 1KHz.

What do you guys think? Thank you in advance for your responses.
-H

[srm51555]

The negative terminal on the 2nd woofer in the 4435 looks like it's connected the same as what you've shown in you schematic, but my crossover experience is limited. Maybe someone will confirm. Cab's look good though.

[Ian Mackenzie]

I have only just seen this thread.

For clarity the 4435 network was customised for the specific drivers and bi radial horn.

The way the 4435 functioned as that the woofers summed to maximally flat response with a common tuning and relied on the specific T/L parameters of a pair of 2234 drivers.

The 2334 is a 2235 with the mass control ring removed. The mass of the ring is considerable in relation to the mms of the cone.

If you are intent on the 4435 configuration I recommend you see a JBL dealer for removal of the mass rings.

The 2234 performs significantly better above 300 hertz.

The 4435 network attenuated the slave woofer above 100 hertz so only one woofer is operating at the crossover network of 1000 hertz.

The alternative is to look at this as a D'Appolito configuration but the crossover design would be complex and require a software optimiser.

As for the 2380 horn the L300 network will be sub optimal.

The Tad 4002 on the 2380 is a totally different beast.

The 2380 horn is a bi radial CD. Horn is requires CD Equalisation to realise is true potential.

Have a look at my comments in the 30-50 ohm resister thread or consider an active network with generic CD compensation.

Hope this steers you in the. Direction of a desireable outcome.

[zeeman]

I have only just seen this thread.

For clarity the 4435 network was customised for the specific drivers and bi radial horn.

The way the 4435 functioned as that the woofers summed to maximally flat response with a common tuning and relied on the specific T/L parameters of a pair of 2234 drivers.

The 2334 is a 2235 with the mass control ring removed. The mass of the ring is considerable in relation to the mms of the cone.

If you are intent on the 4435 configuration I recommend you see a JBL dealer for removal of the mass rings.

The 2234 performs significantly better above 300 hertz.

The 4435 network attenuated the slave woofer above 100 hertz so only one woofer is operating at the crossover network of 1000 hertz.

The alternative is to look at this as a D'Appolito configuration but the crossover design would be complex and require a software optimiser.

As for the 2380 horn the L300 network will be sub optimal.

The Tad 4002 on the 2380 is a totally different beast.

The 2380 horn is a bi radial CD. Horn is requires CD Equalisation to realise is true potential.

Have a look at my comments in the 30-50 ohm resister thread or consider an active network with generic CD compensation.

Hope this steers you in the. Direction of a desireable outcome.

Hi Ian,
Thanks for your help and suggestions. I will employ the 4435's lower range woofer's crossover circuit and will have the mass rings removed from all four 2235's, no problem. I'm friends with the owners of a JBL service center. I'll drop them off tomorrow.
The other thing I gathered was that you really don't like the 2380A horn for this application. I didn't think it would make that much difference, but I'm not married to that horn. I just thought it looked good. It seems the constant directivity aspect of the horn is the issue - okay for pro sound use where the CD equalization is expected to be used, but apparently not so good for this application.
I just started looking for a different style horn and found the elliptical tractrix wood horn, which is about the right size to replace the 2380, looks pretty cool, and is available in a kit form that's reasonably priced. Here's the website with some specs (http://www.fastlaneaudio.com/eliptrac.html)
What do you think of this horn? Is it much better than the JBL 2380A for this application w/ the TAD 4002? I also have a pair of JBL 2446H drivers. Is this driver better than the TAD for this application? The TADs have the original beryllium diaphragms, the 2446's have titanium diaphragms from Orange county speaker.
Thanks again for your help and guidance. I look forward to your response
-H

[ivica]

Hi Ian,
Thanks for your help and suggestions. I will employ the 4435's lower range woofer's crossover circuit and will have the mass rings removed from all four 2235's, no problem. I'm friends with the owners of a JBL service center. I'll drop them off tomorrow.
The other thing I gathered was that you really don't like the 2380A horn for this application. I didn't think it would make that much difference, but I'm not married to that horn. I just thought it looked good. It seems the constant directivity aspect of the horn is the issue - okay for pro sound use where the CD equalization is expected to be used, but apparently not so good for this application.
I just started looking for a different style horn and found the elliptical tractrix wood horn, which is about the right size to replace the 2380, looks pretty cool, and is available in a kit form that's reasonably priced. Here's the website with some specs (http://www.fastlaneaudio.com/eliptrac.html)
What do you think of this horn? Is it much better than the JBL 2380A for this application w/ the TAD 4002? I also have a pair of JBL 2446H drivers. Is this driver better than the TAD for this application? The TADs have the original beryllium diaphragms, the 2446's have titanium diaphragms from Orange county speaker.
Thanks again for your help and guidance. I look forward to your response
-H

Hi zeeman,

Such horn shape

http://nebula.wsimg.com/962165466544...&alloworigin=1
is very unusual for me to be called 'tractrix' shape, but every day we can learn something new.
May be 'conical' would be better name.
Up to now something like this has been called 'tractrix'
http://www.audioheritage.org/vbullet...1&d=1161638799
http://www.audioheritage.org/vbullet...1&d=1162896873



regards
ivica

[Ian Mackenzie]

For home use yes.

CD has its pros and cons.

But I would contact them and explain your intention before going ahead.

Make sure the throat matches you driver

They will probably be to recommend a crossover so ask about that.

[zeeman]

I'm back again. I have some news and it's mostly all good. As you can see from the photos, I've only gotten 2 of the 2235's without the mass rings back from the shop as there were some issues. Someone had done a bad job refoaming them and after I got them back they started rubbing. I'm glad I wasn't playing loud and noticed after about 20 minutes. When the mass ring were removed, Mark used the proper shims to center the voicecoil, but with an improperly applied (not using shims) surround foam pulled the voice coil slightly off center. Bottom line, I had to get all 4 of them refoamed. That's why I've only gotten 2 of them back. I was lucky that I had 2 Electrovoice EVM-15B's I wasn't using to fill in for the bottom woofers while they're being repaired. The lower woofer is the one that only goes up to 100 Hz.Now for the good news. THEY SOUND GREAT! The blend between the upper woofer and the horn is seamless. The blend from the horn to the tweeters also seamless. That TAD driver really is something special. You may notice I sanded and varnished the 2405 tweeters' housing and phase plug. I just thought it looked good. The only variation from the previously posted crossover is that I didn't employ the resistor for the tweeter. It seem to reduce the tweeter output too much. I didn't remove the resistors. They're still in the cabinets in case I change my mind, but they're not in the circuit.

Sincere Thanks to Allanvh5150, Doctor Electron, and especially Ian Mackenzie. Thank you so much for your advice and guidance. I couldn't have done this without you guys.