working with RS150&SEAS27TBFC/G

[nicholas_23]

I'm working with this combo and came up with a summed Xover point of 2447 using a 2nd order butterworth on both drivers. In order to keep things flat, there is also a fixed Lpad to bring the tweeter down roughly 4.7 db. The RS150 xover point is set to 1500 and the SEAS at 3000. Now even though this sums a flat response between the two drivers, is there an issue with the gap in between (1500-3000)? Should one try to get both drivers closer to each other? Just curious if it will be missing anything because of the gap from 1500 to 3000. The summed Xover point for the two drivers is about 6.7 DB's down using this topology.

Hope I described this correctly and makes some sense.

TIA
Nick

[cjd]

Got a network you can share?

I'm really unclear how you're finding one crossed at 1500Hz and one at 3000Hz and having them sum flat, unless you're talking where it *starts* to roll off and you have some really low slope crossover (which, by the way, the RS drivers are probably not suited to due to their harsh breakup).

But seeing what you're looking at would definitely help. I've worked with this combination (MTM) and can model your stuff to see what's going on and probably some more useful feedback.

Right now, I'm just left confused.

C

[nicholas_23]

Thanks for the info CJD , yes it is the roll off for both. I will go back an rethink.
Thanks
Nick
> Got a network you can share?

> I'm really unclear how you're finding one
> crossed at 1500Hz and one at 3000Hz and
> having them sum flat, unless you're talking
> where it *starts* to roll off and you have
> some really low slope crossover (which, by
> the way, the RS drivers are probably not
> suited to due to their harsh breakup).

> But seeing what you're looking at would
> definitely help. I've worked with this
> combination (MTM) and can model your stuff
> to see what's going on and probably some
> more useful feedback.

> Right now, I'm just left confused.

> C

[cjd]

> Thanks for the info CJD , yes it is the roll
> off for both. I will go back an rethink.
> Thanks
> Nick

I wasn't necessarily suggesting a rethink unless you went 1st order at 1500Hz on the RS150. :) Not entirely sure how your ears would like it.

It really helps to at least give us an idea what your target slopes are and the like, because otherwise there is just little to go on. Most of my post was trying to guess what was happening based on what you did say. :)

C

[unknownuser]

> I wasn't necessarily suggesting a rethink
> unless you went 1st order at 1500Hz on the
> RS150. :) Not entirely sure how your ears
> would like it.

> It really helps to at least give us an idea
> what your target slopes are and the like,
> because otherwise there is just little to go
> on. Most of my post was trying to guess what
> was happening based on what you did say. :)

> C

If you look at his "crossover frequencies", you can see that they are 1 octave from each other. They are also 6.7dB down. The "crossover frequency" is probably the knee of the slope. From his description, I guess that he probably has quasi-2nd order L-R slope on the woofer with actual cross over frequency of ~2.5KHz.

To the original poster: The answer is yes, if it sums flat, then that's what you will hear; the output of the system is the sum of the output from each drivers. But have you taken the driver offset into consideration? To get an accurate simulation, you'll have to take the location of the drivers into consideration...unless you have coaxial driver like KEF's Uni-Q.

Try posting the picture of the crossover (frequency response part, not the schematic). More people can help you then. Verbal descriptions are always somewhat vague.


(Originally posted by: kyle)

[scottsehlin]

I am working on a small two way using those drivers - I have been listening to them a lot over the last month with a variety of crossover approaches (intentional and otherwise). I would be happy to take a look at your design if you can share the specific component values for your crossover. My drivers still are connected to my crossover by wires coming out the port hole, so if I happen to have the right components around, I might be able to mock your crossover up and listen...

[nicholas_23]

Thanks Scott , I may take you up on that as I am still waiting for the H1212 to show up.
Also , and I know this is taboo but I am using the manufactures measured response for the tweeter .
I will try to get the information to you later this afternoon.

P.S. Whats your impression of the design you came up with? The good the bads the.......

Thanks
Nick

> I am working on a small two way using those
> drivers - I have been listening to them a
> lot over the last month with a variety of
> crossover approaches (intentional and
> otherwise). I would be happy to take a look
> at your design if you can share the specific
> component values for your crossover. My
> drivers still are connected to my crossover
> by wires coming out the port hole, so if I
> happen to have the right components around,
> I might be able to mock your crossover up
> and listen...

[nicholas_23]

Thanks guys , I will try to get more information out later this afternoon. Will posted back when it is available.
> If you look at his "crossover
> frequencies", you can see that they are
> 1 octave from each other. They are also
> 6.7dB down. The "crossover
> frequency" is probably the knee of the
> slope. From his description, I guess that he
> probably has quasi-2nd order L-R slope on
> the woofer with actual cross over frequency
> of ~2.5KHz.

> To the original poster: The answer is yes,
> if it sums flat, then that's what you will
> hear; the output of the system is the sum of
> the output from each drivers. But have you
> taken the driver offset into consideration?
> To get an accurate simulation, you'll have
> to take the location of the drivers into
> consideration...unless you have coaxial
> driver like KEF's Uni-Q.

> Try posting the picture of the crossover
> (frequency response part, not the
> schematic). More people can help you then.
> Verbal descriptions are always somewhat
> vague.

[scottsehlin]

I am pretty happy with my design at this point - but of course I am biased. I should get some less biased opinions at Chicago Audiofest in a couple weeks.

I use a pair of Ellis 1801's as a reference. The RS150-TBFC combo has many similarities - great resolution, low distortion. The biggest weakness for this type of speaker is that it is somewhat unforgiving of poor quality source gear or recordings. Although they have less bass extension than the 1801's, it is surprising how well they do for such a small speaker/box. If you really want full bass extension and SPL, you will need a sub or a larger speaker though. I found the TBFC a little tricky to work with, but really rewarding once dialed in. I guess my overall impression is very positive - shocking level of sound quality for about $120 worth of drivers.

Everyone has an opinion on using manufacturer's response curves - I do it sometimes too. I really think the biggest advantage of measuring yourself is you can measure all of your drivers under the same conditions. If the manufacturers sheet tells you what baffle the driver was mounted on, how many volts were used to drive it, and what distance away the microphone was placed, then it is very possible to deconstruct it and predict quite accurately what will happen in another design. Roman Bednarek does a good job of describing how to do this on his website. The other important consideration when designing this way is to make a good estimate of the relative acoustic centers of the drivers - which helps determine their phase alignment. This is roughly the difference between the depth of the woofer vs. tweeter voice coil.