3 way crossover help

[omni]

I am doing a 3 way design using the Peerless sls 830669 12 inch woofer in sealed 2 cubic ft box for Qtc .9, along with Seas CA15RLY midbass, and Seas 27TDFC tweet. I have no measuring equipment. Can anyone give me some crossover suggestions, or help me model one? It was reccommended a 250 Hz and 2500 Hz crossover frequency set up, with 12, 12, 18db/oct slope, for woofer, midbass, tweeter respectively. I am not familiar with this.....Any suggestions or help would be greatly appreciated..........

[Jonathan Bright]

Hi omni, what you say is fine. The cross over points and slopes are okay. I like 12db (second order) slopes as they are more toleraant of variations in parameters than 3rd order so they are a good place to start if you haven't done this sort of thing before.. If you can get hold of a basic text book that would be a good place to start. They will show you tables of the appropriate values for these frequencies. You will find that your speaker impedence will need to be known (8 or 4 ohms). Have you got a local library nearby where you can borrow a text? Or buy one. If the speakers have different efficiencies you can lower the volume of the relevant units with a simple resistor network. Again a basic text will explain this. (But it won't be helpful if the bass unit is louder than the mid/high units. ) On the other hand you are obviously on the Net so Google a few terms and see what come up. There are some very technical sites that are quite sophisticated with immpossible math and there are very good basic sites that you will appreciate more easily. If your not formally trained then you just have to check around and pick things up as you go along. After looking at few sites or books you'll be asking more specific questions and getting a better response from this site. At the moment your initial post is a bit too general for people to respond to okay? Good luck and stick with it. The other thing worth saying is that what ever you end up with it will sound really good because you did it yourself.

[omni]

Thank you, Mr. Bright. The drivers I mentioned are all nominal 8 ohms, however, my understanding is that in the crossover, they may not actually be 8 ohms, and that simply plugging 8 ohms into crossover design formulas, may do me a disservice. I have Vance Dickasons cookbook and have read it extensively, however, more contemporary research has appeared to show that there are other issues to be concerned with. Example: Zobel networks: 2 schools of thought apparently exist. 1- it is absolutely needed to maintain woofer impedence in order to make a crossover work properly., and 2- that Zobels can actually be harmful to the amplifier because impedance maintenance in the woofer can draw too much power, therefore, creating a strain on the amp.,being that the woofer draws 75% of the amps power, and the midbass and tweeter draw less power. What I seek is to get a crossover design which best utilizes the drivers I mentioned in my original post. I will be using air core, inductors, good quality capacitors and resistors, looking for a transparent, open sound with tight, transient bass response, airy mids, and crystal clear, crisp, smooth highs, with the drivers, I believe to be worth it. I am not a wealthy person, but I think this is attainable with these particular drivers. Any help from the experienced audiophiles on this site will be appreciated. Maybe some of you have worked with these drivers?

[tinitus]

I have looked at driver specs and curves, and it looks like you have made wise choices.....it should be possible to get get a good result

Because the drivers are sound and without peaks I would adwise you to make your filters differrent than you suggest

woofer 12db - mid 6db/6db - tweeter 12db

And crossover points I would try 300/3000 hz

It is possible to tune this into very good result

But why you have chosen enclosure with high Q=0.9 I dont know

You have a woofer with medium high Qt and 90db...combined with midrange with lower sensivity....so you should carefull not to get a muddy and boomy bassystem or you could end with the feel of lack of midrange

[tinitus]

And You really should make big structured "matrix" for bass enclosure ...and finer "matrix" for midrange

[sreten]

From

http://www.speakerbuilding.com/content/1030/index.php

Is as good a place to start as any I've seen.

If the impedance of your drivers is different, scale the c/o values, its
unclear whether the c/o is for the 4 or 8 ohm version of the bass unit.

/sreten.

[omni]

I am greatly appreciating the replies to my request for help on this 3 way... Tinitus, my theory on a .9 Qtc for my enclosure is dictated by a few factors. 1- my boxes are already built from back in the 70s, as I followed recommendations from my research, back in the day, and 2- I prefer a little more damping factor with a smaller box. I feel this will provide me a more solid sounding, tighter, more transient bass, as well as allowing the midbass to cover a wider bandwidth. In addition, information I have found on the Peerless woofer recommendations suggests a smaller box, sealed is what this woofer is designed for. Do you think .9Qtc will contribute to boominess? Understanding the different sensitivities, I was planning on using 14 guage air core inductors throughout the entire crossover circuit, whereby the dcr for the inductors in the midbass and tweeter sections will maximize their particular sensitivities, as well as perhaps lower the sensitivity in the woofer section, thereby bringing the whole system into balance, sensitivity wise. Am I mistaken in this theory? Tinitus, is your 6db/6db recommendation on the band pass in order to allow more midrange sound ? Also could you clarify what you mean by structured Matrix for the enclosures. Sreten, I have seen this diagram, before, in my research, and ironically have printed it out, and have it in my diy files. Its amazing, the coincidents we diyers experience, thank you very much...... Would Lpads introduce distortion into the system or not? Also, as far as mounting drivers, I have used silicone...........Anything wrong with this ? As some of you may have figured out by now, I am a protege of the old Speakerlab era, some ideas of theirs which I subscribe to, and others, I am not so sure....... This is a reason I am here seeking help, and appreciate any input.......Respectfully..........Omni.

[tinitus]

reply to your questions

Qt=0.9 is high and could give a lose and "slappy" bass, but there are other factors that determine how it turns out in reality....and if you use a midrange with lower sensivity it may be trouble

But how do you know it will be Q=0.9

I recommended "matrix"....it is made by deviding the interior of the enclosure into smaller "bits"...is said to lower Q or have a similar effect ...among other things...but you did not inform that you had enclosure allready

6db/6db on midrange can be made to give you close to 12db/12db on acoustic side....and with good driver thats all needed....but there should be RC also to get it work properly

Schematic suggested by "sreten" looks OK and a good place to start...but you dont need L-pad on midrange

[tinitus]

must comment further.....2.6mH/1.2ohm in paralel with midrange....leave that out too

There are other adjustments possible, but will only comment on that if you choose to do this crossover

[Jonathan Bright]

Hi omni, Sounds like you knew more about the subject than you were initially displying. When I first read you initial post I also thought that a Q of 0.9 was too high. I may have missed understood you subsequent post but a BIGGER box will give a lower Q qand not a higher one. Many people find that a Q of 0.7 (ie maximum flatness) will sound a bit boomy or overpowering in the bass. If you can move up to a bigger box you may enjoy the results more. You won't get much attenuation of the signal with increased resistence in the inductor to the bass unit and if is getting up towards one ohm that will worsen the damping factor and make the bass more boomy again. If you can afford to you might consider bi-amping!!!!.