Home » Audio » Speaker » Altec 902 / Edgarhorns: compensation experiments
Altec 902 / Edgarhorns: compensation experiments [message #17671] Thu, 24 March 2005 17:13 Go to next message
Floyd Andrews is currently offline  Floyd Andrews
Messages: 31
Registered: May 2009
Baron
I've just gotten a pair of Edgarhorn 650 hz round horns to use with Altec 902 drivers. These have replaced a pair of JBL 2345 horns with Selenium 205 1" drivers. The new setup sounds fabulous; I couldn't be more pleased. After waiting about a week to get used to the new gear, I started playing with compensation components. My system has an active crossover, so the 902's are driven directly from a small tube amp. I wasn't sure that I even needed more high end, but I wanted to try it out. 50 ohms in parallel with .47 mfd was way too hot. So was 25 ohms and .47 mfd. Right now I'm using 8 ohms and .47, but I think that this still may be too much. Does anyone out there have any experience with this driver/horn combination? What is the effect if I go to a smaller capacitor, say .33 or .22? All suggestions appreciated. thanks.

Re: Altec 902 / Edgarhorns: compensation experiments [message #17673 is a reply to message #17671] Thu, 24 March 2005 21:17 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18786
Registered: January 2001
Illuminati (33rd Degree)

My experience with the round 650Hz tractrix horns Edgar sells is that they need 8dB compensation above 8kHz with an Eminence PSD2002 and 12dB compensation above 4kHz with a JBL 2426.

JBL 2426 on Edgar round 650Hz tractrix horn

Start EQ around 4kHz at 6dB/octave so that there's some boost at 8kHz and full augmentation at 16kHz. The collapsing DI of the tractrix horn provides some EQ, but not enough to make response flat above 8kHz with either the JBL 2426 or the Eminence PSD2002.


Thanks, Wayne! [message #17674 is a reply to message #17673] Fri, 25 March 2005 09:53 Go to previous messageGo to next message
Floyd Andrews is currently offline  Floyd Andrews
Messages: 31
Registered: May 2009
Baron
Thanks for your response. Will continue the experiments this weekend.

Series and Shunt Resistance [message #17675 is a reply to message #17674] Fri, 25 March 2005 12:38 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18786
Registered: January 2001
Illuminati (33rd Degree)

Just a suggestion - You might want to use a resistor in shunt along with a resistor in series. Bypass the series resistor with capacitance for high frequency augmentation. If you just have a series resistance, it will raise the amount of midrange as well as the top octave or two. The highest frequencies are augmented by the rising impedance caused by voice coil inductance. The midrange peak is caused by various resonances, such as diaphragm mechanical resonance, horn flare 1/4 wave resonances and interaction of crossover capacitors and voice coil inductance. If you're familiar with Spice, you can use it to calculate a network that will work for you. There are some illustrations of this in the documents at the links below.
Re: Series and Shunt Resistance [message #17681 is a reply to message #17675] Sun, 27 March 2005 11:07 Go to previous messageGo to next message
spkrman57
Messages: 522
Registered: May 2009
Illuminati (1st Degree)
I would use the 14 ohm shunt with a 30 ohm series(paralleled with .33 ufd) like the 4 pi crossover.

By the way, I am getting ready to try a comparison between the JBL 2426 and the Altec 902 on the 650hz round tractix(Edgar) and the 650hz wooden tractix(Martinelli). Two different drivers and 2 different horns to mix and match and see which I like best.

By the way, I still prefer passive crossover on my JBL/Altec systems. Makes life easier.
Ron

Uniform DI verses Collapsing DI [message #17682 is a reply to message #17681] Sun, 27 March 2005 15:34 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18786
Registered: January 2001
Illuminati (33rd Degree)

I would suggest that you compare a system with uniform or matched directionality to another with axisymmetric horns having collapsing DI. More specifically, you would be comparing a loudspeaker system that uses tractrix round horns with another that uses 90x40 horns. That involves using matched mids and bass systems and, in the case of the uniform DI speaker, you'll also want a crossover that minimizes comb filtering between adjacent subsystems. In my opinion, the type of system you are making determines the horn you should use.

It's not just a matter of comparing one particular horn with another, that's kind of like comparing a car's tires without discussing suspension. Unless they are of the same general type, a comparison isn't really appropriate. Some things can be quantified but overall performance cannot. Some tires are more suited for one kind of vehicle or another, like tall, bouncy slicks for a drag car or short, rigid tires for a road race. Both are designed to maximize traction, but one is optimized for straight-line launch and the other for lateral hold. Either way, the tuner has to optimize properly or the car isn't setup right. Put a drag slick on an indy car and you can out-corner it in a mini-cooper. Put a Formula I tire on a top fueler and you'll melt it to the rim without even leaving the lights.

So the long story short is that I don't think it is going to really work comparing an individual round horn to a 90x40. The system optimizations are different. One is optimized when power response is uniform, the other when on-axis response is uniform. A nice experiment might be to compare a loudspeaker system with all round horns to another with all 90x40 radial horns. If you like the tractrix curve, you could use round tractrix horns for one system and for the other, use radial horns having a tractrix expansion in the vertical plane and uniform 90° in the horizontal plane. But whatever the flare profile, I suggest the 90x40 horn be used in a system that has a DI matched midwoofer or a cornerhorn that uses all 90° flares. I'd use the round horn in a system that has all axisymmetric horns. Then you can compare systems, because I think that makes a lot more sense.


Re: Series and Shunt Resistance [message #17685 is a reply to message #17681] Mon, 28 March 2005 19:19 Go to previous messageGo to next message
Floyd Andrews is currently offline  Floyd Andrews
Messages: 31
Registered: May 2009
Baron
thanks, spkrman! I will try those values next. I'm very interested in hearing the results of your comparison of the two horns. Martinelli was on my short list of horns before I got the Edgars.

Re: Series and Shunt Resistance [message #17686 is a reply to message #17675] Mon, 28 March 2005 19:25 Go to previous messageGo to next message
Floyd Andrews is currently offline  Floyd Andrews
Messages: 31
Registered: May 2009
Baron
Thanks, Wayne! Right after I got your message I went home and put a 25 ohm resistor in shunt before the series resistor. That value was used just because I had it on hand. You were right (as usual) Sounded much better. Now to get a few more resistors to try for an optimum combination. I'm going to try some of the values that you use in the Pi crossovers next.

Re: Series and Shunt Resistance [message #17689 is a reply to message #17686] Mon, 28 March 2005 20:41 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18786
Registered: January 2001
Illuminati (33rd Degree)

Glad that helped, Floyd. Do plug the numbers into Spice or measure the system, whichever is best for you. That will make sure you've chosen values that give you good clean response. Depending on the reactive values surrounding the circuit, you may want the shunt resistance after the series resistance instead of before it. It will also help you select the values that work best.


Wish i understood everthing you said... [message #17691 is a reply to message #17682] Tue, 29 March 2005 12:32 Go to previous messageGo to previous message
spkrman57
Messages: 522
Registered: May 2009
Illuminati (1st Degree)
Wayne,
The "DI" factor loses me. I know I just tried out my 1st JBL compression driver(2426J). I used the 4 Pi-pro figures(15 ohm shunt and 33 ohm series/.33 ufd cap). I also used .68 ufd bypass caps on the PE 1.6 khz hi-pass crossover.

So far I think the caps are still new and not smoothed out(less than 2 hrs time on them). I find the JBL to be rather bright in the upper midrange/lower treble area used on the Martinelli horn. I know that the 2426J is a 16 ohm driver, but I have used Altec 8 ohm and 16 ohm drivers without changing up the crossover with little difference of sound.

I think the JBL with titanium diaphram is more brittle sounding than the aluminum of the Altec drivers I have used. I set up the crossover this time with easy "connect/disconnect" terminals to swap out the horns quickly.

I have my Altec 902B on my 650hz Edgar horn and will be trying it out tonight. I think the JBL needs less HF compensation than the Altec 1" drivers.

Ron

Previous Topic: PHY-HP - Know anything about these?
Next Topic: Does anybody here know how to add and subtract?
Goto Forum:
  


Current Time: Tue Nov 26 14:10:25 CST 2024

Sponsoring Organizations

DIY Audio Projects
DIY Audio Projects
OddWatt Audio
OddWatt Audio
Pi Speakers
Pi Speakers
Prosound Shootout
Prosound Shootout
Miller Audio
Miller Audio
Tubes For Amps
TubesForAmps.com

Lone Star Audiofest