change tweeter db? need help

yes, you could just increase that series R before the network to match the sensitivity, and it might work well enough.

Ideally, though, you should rework the whole network. The T90 has a 5 db hump in its response abound 2KHz. The curves on the design you've linked to seem to have its high pass corner desiged to eliminate that hump. You may find that the T120's smoother response seems a bit low between 2-4 KHz with that filter. The T120 seems like a better choice for a ~2500 Hz XO, given the high resonance of the T90.

But, if you don't have the equipment and inclination to optimize the crossover I'd suggest just building the design as is with the T90. Lots of people seem to like the T90. I've never heard them, but I love the T120 in a similar system using the Focal 6W4254.
simon5 said:
Could you post the Xover you're currently using?

You'll probably need more than one resistor, at least two or the crossover frequency will change.

Because the resistor is places after the xover it doesnt change the impedance the filter "sees", thus no alteration of the frequency.

Single resistors dont work as well as Lpads but this is only of any concern if you cannot optomise the network. Both can work just as well as each other. Also an Lpad can be used to increase/lower the overall impedance section of the tweeter network.
In this case you'll only need one resistor, because it is outside the reactive part of the XO. I agree that simply adding a R in series with the tweeter and then adding the filter would affect response. If you want to pad the tweeter there, use an L pad of two resistors.

I'm using an active XO, 2400 Hz 4th order electrical LR, with 6 db of baffle step correction and a 7 db notch filter with a Q of 2.5 for the woofer breakup. I thought I might need a small notch on the tweeter, but I haven't bothered. The schematic is nothing special - I just used Linkwitz's building blocks. With the 6W4311 you'll need less/no notch filter and have higher max spl, but give up a little bass extension.

The electra like baffle ended up needing no phase correction. I rounded all but bottom edges to 1.25" radius. "The Edge" diffraction simulation program matches measured response when I assume the effective baffle width is 1" narrower than the box due to the roundover. (Plug/ thank you for that free program)


2005-04-09 10:53 pm
the same comments apply - the FR of the two tweeters is different, so the XO needs to be different to be optimal. The T120 is flatter, so to me it is easier to work with.

As for your latest design choices, I don't have time to crunch the numbers, and see what is done with them for actual crossover frequency and shape right now. While Zalytron's kit offerings are usually quite good, I think the T120 would be a better match for the 7W. I'd try to cross the T90 in at >3000 Hz, and the 7W will be beaming terribly by then. The 6W is still on the edge, but is a little beamy by 3k. The 7W/T120 combo would be a nice ~1 cf monitor that could go well into the low 40s and still play reasonably loudly. A 6W4254/T120 would have smoother dispersion characteristics and probably better imaging. It will want almost as much cabinet volume as the 7W4411. (can you tell I spent a lot of time evaluating these trade offs?)
5th element said:
Because the resistor is places after the xover it doesnt change the impedance the filter "sees", thus no alteration of the frequency.


It's only "after" the XO on one side of the leads. It's before the XO coming from the other side and it's all running through the same circuit.

Not sure what you are trying to say, but I don't read your statement as being correct.