Replacing my tweeters a good idea?

While i hope that nobody takes this personally, but "just inserting" a resistor into a speaker circuit is NOT a good idea. Outside of lowering the output level, it will alter the crossover frequency, change damping and tonal characteristics, etc... The ENGINEERS at Infinity didn't just throw parts into a box and hope that it worked out okay. Nonetheless, there is always room for improvement in any design though and this is especially true of mass market brands and speakers in specific. Your best bet is to look elsewhere in the chain to see what may be aggravating the problem. As someone mentioned, the use of low grade crossover components INSIDE the speaker may also be causing problems. You really have to look at the system AS A SYSTEM and NOT cast the blame on just one component. Unless you have changed things around enough and substituted other components in their places, you can't be sure exactly what is causing the problem. You might be surprised at what can cause "less than stellar" performance out of a system. My suggestion is to look at the source components and then work your way back. Is the problem only with one source ( cd's ) or when listening to any source ? This will give you an idea as to whether the problem is SYSTEM based or COMPONENT based. From there, the path to take will be much clearer. Sean >

Sean, you are partially correct about what happens when a resistor is inserted into the speaker circuit. Output is lowered, tonal characteristics are changed. THAT IS THE POINT!!! Crossover frequency is left UNAFFECTED by resistors. Resistors merely attenuate the output. Here we want to tame a bright speaker. MANY, if not most speaker companies wire a 1 to 3 ohm speaker in series with the tweeter, on purpose. They want the speaker to be 3 db down at 15K. Why? Because most people complain the same speaker without that resistor is too bright. Crossover frequency IS determined by capacitors AND inductors. A capacitor in series limits a lower frequency being passed to a driver. And inductor in series limits a higher frequency being passed to a driver. A first order crossover has a capacitor on the positive side on the tweeter leg, and an inductor on the positive side of the woofer leg. If you had a midrange, it would have both a capacitor and an inductor in series(again, on the positive side) with that driver. Rolloff would be 6 db/octave, depending on the component values used. Adding an inductor in parallel to the tweeter, and a capacitor in parallel to the woofer(and both in parallel to the midrange) is a second order crossover(12 db/octave), and so on. Alternating these components in series and in parallel(maybe confusing, but no need to explain in detail here) takes us up to 4th order(normally). Resistors don't factor into this crossover equation AT ALL. Their purpose. To attenuate drivers. Say a tweeter with 91 db/2.83 V sensitivity is used in a speaker with an 89 db/2.83 V sensitive woofer. The tweeter needs to be quieted down a bit(we can't increase the woofer output), for the speaker to be balanced(I still listen before inserting the resistor - just to get a baseline feel). We may also need other things in the circuit(won't discuss them here), but they come later.

Audiopath..... you think I should try your stuff first? I am not sure about how to do room treatments, how do i treat the first and upper corner reflections?

also, the freq response is quoted as going to very high, like 35Khz or 45Khz, i forget... is it worth me using a high pass to cut it off at 20khz??

Trelja...I couldn't have said it better. Bryhifi...I spent two grand on power tools, bought 3 or 4 20 foot sections of PVC water pipe, and built a pretty cool cylindrical 3-way using SEAS 5 mid and 0.75 plastic dome, with Peerless 8 down-firing woofer. I had built a similar cylindrical speaker as a kid using 3 foot clay sewer pipes, mounting a coaxial Jensen or Utah (remember them?) driver upward-firing into a plaster-filled kitchen funnel for 360 dispersion. What a hoot! So I decided to revisit that idea with modern materials. Well, no-one seemed excited about a cylindrical shape, so I never built the second side. Sound was pretty good, although woofer/mid crossover was problematic. Got good at voicing crossovers, and noticed that I could hear 1/3 dB changes over 1.5 octaves in the upper crossover region. After seeing that raw driver sensitivity tolerances are typically +/- 5-10 times that level I quickly realized that it would be impossible to make speaker systems consistently without customizing each crossover, or spending a fortune to get tight-tolerance drivers. I built with others 20 pair of a nice 6.5/1 two-way a few years back that ALMOST got into production as a reinforcement system for a digital piano mnaufacturer. A local high-end dealer liked them A LOT, but we couldn't get the cabinet maker to take a repeat order at a reasonable price. Voicing a matched-pair of speakers is a labor of love, and I have great appreciation for those who do it well. I preferred these two-ways to many commercial designs for a long time, but after I heard what Bruno Pelchat et al at Verity Audio accomplished with their Parsifal Encores, in my living room, my jaw dropped! Had to have 'em! So I'm done...for now. (PS. Want a good deal on Seas 5" 'bullet' dustcap, or those tweeters? How about the Peerless CC line 8"? A nice router...or maybe a belt-sander? Oh well, it was fun. It's all about the process, anyway, right?) Shabbado...you mean "low-pass", and no, that's NOT what you want to do. Room treatment, upstream components, maybe 1-2 ohms on the tweeter: try 'em all, and have fun!