Outstanding!
Thank you for taking the time to do all of this. I will be ordering the books you recommended today. I used to have the Loudspeaker Design Cookbook, but it appears to be a casualty of my move to Florida.
Needless to say I will need some time to digest all of this and doubtless I will have more questions, if I may trouble you with a few I have now.
First, the wiring diagram for the crossover illustrates the L-Pads as a single fixed resistor symbol. Typically, a variable resistor is illustrated as a 3-terminal device (i.e., the symbol is represented as a combination of a fixed resistor and an arrow pointing to the center of the symbol's squiggle, which represents the wiper of the pot). So, I am uncertain as to how that is wired in your illustration.
Second, both my girlfriend and I were a little taken back by the driver selection you recommended. No doubt that the Dayton speaker you recommended will perform as advertised, but we both are left to wonder what are the advantages to these higher end drivers we see. Perhaps this is a case of just trying to find something to fit the current design and I realize that my choices are going to be limited.
Again, that you for taking the time to educate me on this subject. Also, your listening room/studio is one of the most impressive rooms I have seen. Bravo! Do you also play a musical instrument?
Regards,
Loren
First the L-pad. These are fixed and not variable. The value of the resistors are given. Variable resistors upset the crossover and also lead to crackles. The speaker balances should not be used as tone controls. The values should be set for good after full evaluation. From the modeling these values look correct. We might have to bypass Rp1 on the tweeter circuit with a 2mfd cap, to pull up the response droop above 10 KHz. I doubt you will though. I don't usually correct HF roll off above 10KHz. The tweeter had a nasty peak, which I got rid off, but it has left a slight roll off above 10 KHz. I would not correct it off the bat, I would listen first.
Now to driver selection: we have few to choose from. If I might say so, the concept of your speaker is very retro. In the fifties sixties and seventies there were lots of speakers on this model, virtually none now. Most of them were "raucous squawkers". Now narrow fronted cabinets have become popular because of reflection issues from the large front baffle. The most popular models are either bookshelves, or the two and a half way towers, where there are two bass mids, a tweeter and the lower bass mid shelved in to compensate for the diffraction loss of the narrow cabinet. This has another advantage of avoiding two crossovers in the sensitive mid range area. It also saves a lot of dollars on crossover components.
These speakers, especially for HT, have the bass extended by subs operating below either 60 or 80 Hz.
Now sub woofers are big, heavy and of lower sensitivity. They will not be suitable for your application, as they lack sensitivity and wide enough frequency response.
Now there an inverse correlation between magnetic flux, and therefore sensitivity and bass extension.
There is an inclination to believe that a 15" driver must have deep bass. Nothing could be further from the truth.
Most large cone, non sub woofers are basically for high spl professional application. Bass extension for pro work and PA is not a prime consideration and often a detriment. What is required is a robust driver of high sensitivity, that will handle lots of power and generate high spls. The fact they are expensive and sound awful is beside the point to the pro. They want a rough tough speaker that will work in a smallish cabinet, or be able to use multiples without enclosure size getting out of hand, that can belt it out and cross to a compression HF horn with active crossover between 500 and 1200Hz.
So these drivers tend to be stiff to make them robust, have light paper type cones and huge flux density. This tends to highish Fs, low Qts and lowish VAS. In essence belters, but with poor bass. Cut off usually between 45 and 60 Hz, usually nearer the higher number. Your current driver fits that description.
So we need a driver of intermediate sensitivity, low Fs, reasonably smooth response to 400 Hz, and an out of band response, that can at least be tamed. It needs to work in a ported enclosure around 6 cu.ft.
The driver I selected for you fits the bill. It has an Fs around 20 Hz. Remember a speaker will not perform below Fs and usually F3 is about 7 to 10 Hz above Fs. Most 15" non sub drivers have an Fs between 30 and 40 Hz. The Qts and VAS of your driver also made it look promising for your application. It has a lossy surround and a stiffened cone, to help smooth the mid range response. All stiffened cones break up and when they do you get a sharp peak. This one is no exception, and has a doozie at 1500 Hz. I have tamed it in the crossover. Remember you always have to consider the in band and out of band responses of all drivers. This is because crossovers are not brick wall and there is driver overlap. This is especially problematic in three ways as there is overlap at both ends of the bandpass. This leads to band pass gain, which if not properly handled leads to the "Raucous Squawker". I suspect your current speaker fits that description to a degree and may be totally. The band pass gain is minimized by spreading the crossover points as far as you can. We have yours spread about three octaves. The rest of the bandpass gain is taken care of in the crossover design.
So those are the reasons I chose the driver I did. The selection of 15" drivers for your application is extremely limited. That driver is cost effective. This is important, as when you add up the cost of building those crossovers you will have sticker shock.
Use wire wound air core chokes. You may use an iron cored choke for Lm. Use Polypropylene caps. You don't need exotic caps. I use the French Solen line. Do not use non polarizing electrolytic caps. You may have to parallel some caps to get the right values.
I would build the low pass, band pass and high pass filters on separate boards. Keep the inductors as far apart as possible, and if one is in the horizontal plane, then put the next vertical, to minimize mutual inductance.
Check your work very carefully for errors, shorts and poor solder joints.
I'm busy right now as I'm getting ready to fly to England on Thursday, for a three week stay. I will be on GMT then, six hours ahead of of CST which I'm on here. Also on Sunday we are traveling to our families second home on a piece of National Trust land just north of
Southwold, Suffolk.
Notice the Canon. Governments never change, no matter what Senator Obama thinks. These canon were put there after the battle of Sole Bay against the Dutch in 1672. His Majesties government provided and installed the Canon at great expense to the exchequer. However it was up to the Bergers of Southwold to vote the money for the powder and shot. The wise Bergers knew that no warship could come in range of the guns because of the shallows, so they never voted the money for the powder and shot.
The first and only time the guns were fired was as part of the Queen Mother's 100th birthday celebration a few years ago!
At Sole View Eastern Bavance, there is no phone service, internet or cell phone service. I will be at Sole View for five days, then I should have Internet most of the rest of the time.
Anyhow I fly back Oct 9, so if I'm slow to respond to your questions, you know why.
Good luck with your project.
