Can cross a TMM this way?

[olepotato]

Hi all.
This is my first post, but I'm not new to this forum, most of the time just browsed and read.
Just like in real life, I'm also a quiet guy, most of the time listen and occasionally speak.

In this crossover, drivers frequency respond was traced from John Krutke's measurement in zaphaudio.com, crossover circuit created using Roman's 'software only' method as mention in rjbaudio.com.

A TMM system using SEAS's 27TDFC and Dayton's RS180.
The RS180 woofer acoustic center in the crossover is 0.9inch bellow surface, which was 'extracted' from Jon Marsh's Modula MT, assumed listening distance is 2m away at tweeter's lever.
9" width baffle, tweeter 1" off axis, upper woofer is 6" vertically bellow tweeter center, lower woofer 8" bellow upper woofer center.
The 180uH inductor is Jantzen 18awg air core, all other inductors are from Madisound.

In this crossover, the upper and lower woofers phase are matched together by crossover components, a vertical polar respond run should be able to show whether it works.
I'm using Speaker Workshop, so no polar respond available.

In the frequency respond curve, blue = lower woofer, green = upper woofer, red = tweeter.
This crossover circuit, can it works?

[cjd]

Can you? It seems so. It's two very different approaches to knocking down the top end of the RS180. Not entirely sure what you're asking though.

The approach and resulting difference in response between the two mid-woofers isn't putting this into 2.5-way territory, and I think you should do that for a TMM with these drivers.

While I've had success with the simulated modeling you're using, I've also learned it can lead you very very astray if you plan to build without live data. I still use it for sketching, and have found some of the pitfalls that have led to mistakes over the years, but I can't recommend designs put together this way any more.

C

[Johnloudb]

CJD, I have a question about 2.5 way speakers, if you don't mind. I've never been completely sure what that is. Is a 2.5 way where the lower woofer covers a more limited frequency band than the upper midwoofer? It would make sense I think since you don't want to cross the lower woofer with the tweeter cause of the distance between the two, but you still have the two woofers for more output in the lower frequencies. Wouldn't that also give about a 6dB boost in the frequencies where the woofer are working together? Anyway, not sure but if you could clear that up for me I'd be grateful.

[Jonasz]

CJD, I have a question about 2.5 way speakers, if you don't mind. I've never been completely sure what that is. Is a 2.5 way where the lower woofer covers a more limited frequency band than the upper midwoofer? It would make sense I think since you don't want to cross the lower woofer with the tweeter cause of the distance between the two, but you still have the two woofers for more output in the lower frequencies. Wouldn't that also give about a 6dB boost in the frequencies where the woofer are working together? Anyway, not sure but if you could clear that up for me I'd be grateful.

I believe you got that right!
The .5 woofer rolls in around basshelf territory.

[BobEllis]

But the simulated response does not show a woofer rolling off early - something's fishy, since UW has a healthy filter.

It also seems to me that the woofers are reversed. The one labeled upper woofer seems to be the .5 electrically.

[cjd]

But the simulated response does not show a woofer rolling off early - something's fishy, since UW has a healthy filter.

It also seems to me that the woofers are reversed. The one labeled upper woofer seems to be the .5 electrically.

You're only getting the first inductor and cap response included in the 2nd woofer. There are obviously many approaches you can take, but in this case yes - neither is a .5 behaviorally at all at this point. You simply have two techniques for knocking the response down, one per woofer.

Compare to the Ansonica response:



And the parallel crossover (this is a cascaded 2.5-way) to achieve this: (yellow is the parallel crossover in the response graph - you can see the series version too if you're curious.



A simple 2.5 way would be series inductor, parallel cap, woofer parallel; series inductor, .5 woofer parallel.

C

[olepotato]

Can you? It seems so. It's two very different approaches to knocking down the top end of the RS180. Not entirely sure what you're asking though.

The approach and resulting difference in response between the two mid-woofers isn't putting this into 2.5-way territory, and I think you should do that for a TMM with these drivers.

While I've had success with the simulated modeling you're using, I've also learned it can lead you very very astray if you plan to build without live data. I still use it for sketching, and have found some of the pitfalls that have led to mistakes over the years, but I can't recommend designs put together this way any more.

C

You're right, no way for reliable conclusion without live data.

This crossover created solely by chance.
I was a little tired after some fruitless attempt to create a working crossover, and connected the LW wrongly, but the resulting FR looks like usable.
After some adjustment to crossover components, i saw a nice flat system FR.
Just hopes that this can be a successful design, but apparently too many unknown factors.

I think i should go for my plan B, thank you for you comment, cjd.