what is enough BSC

rjj45

As we sometimes say here in the South "How do it know?"
I have done a number of 2 ways and MTMs, and setting 3 dB of BSC is trivial, 
but for 3 ways, it's much tougher, because the Fc is in the middle of BSC territory.

Working on a 3 way with an RS270 woofer crossing to a dome mid around 800 Hz.
By ear, bass is good, but not great - think maybe I'll need to dial another 1 dB of BSC.
Here is close miked FR for the driver raw, and then with the proto crossover of
3.5mH and 660uF  2nd order.

What do you think (and what is your decision point) - is this about 3 dB of BSC?

dynamo

Unless I’m looking at your graph wrong, And if I’m correct in saying your blue line is the response after the crossover is applied; I’d personally want that blue line down flat on the 120db line to the start of the crossover slope. Right now I think you will be disappointed in the lower bass. 

By 200 hz, 100 for sure, all 6db of loss is complete, so if you want full 6db, go flat from 200. If you want 3db, use 200 as your reference point. You could consider working on the mid filter, get that dialed in where you need, then go from there on the woofer filter and mid padding, that might give you a better idea of where 6db up at 1k you are from 200. Then pull the mid down 3db at 1k (factoring the bandpass boost), for your 3db bsc. At least you will have a starting point. Not sure if that makes sense.

Honestly I’d take my measurements at 3x the baffle width, pull my woofer result into Response Modeler, enter the TS and box data, splice the simmed tail matched to the trace level at 200hz, and dial my filter to be flat from there. Then you aren’t guessing at a bsc number, you are simply making it whatever number it needs to be.

That said, I’d try to cross a touch lower if you can. I can only speak for the RS270P, but it does bass very well. It’s a nice woofer. I was able to cross the 270P to a coaxially loaded ND28 at 1k, but it was tricky to dial in and the sound was somewhat dark. Fortunately for me I like a darker meatier character to a speaker, but it’s not everybody’s cup of tea.

dcibel

Looks like 3dB of gain sort of, I assume you've adjusted the level in your before/after measurements to overlap them at 50Hz or so.

Remember that a nearfield measurement is without any baffle effects, the equivalent of a 2pi response for the bass region. Take your nearfield measurement and apply the baffle diffraction sim, then work from there.

FWIW I always go full 6dB, then EQ as needed for the specific room placement. Placement effects are a lot more complex than simply 3dB vs 6dB as I'm sure you can understand if you've ever taken in-room measurements. A couple bands of parametric EQ has proven to be far more effective than any amount of crossover fiddling.

One big thing to consider when designing a passive 3-way, is that a large series inductor will make a hump in the response around 100Hz or so which can throw the ultimate response off a bit further. If the design is ported, raising the tuning a bit to make a high Q response can "flatten the curve" lol, but at the end of the day the charts and graphs are just a tuning tool, it's your speaker so make sure it sounds good to you.

rjj45

Thanks Dynamo - I'll give that a try. For sure the bass will be better. 
I've got to cross the RS270 700-800Hz, because the dome mid would not be happy any lower. 
I hate the sound when a driver is stressed on transients.

PWRRYD

Is this the same $25 HiVi dome mid a bunch of us jumped on?  I've just now started laying out a plan for my pair.  I was shoot'n from the hip thinking a xo point right about where you are crossing based on my experience with my Gormacho project using the Morel CAM558 dome mid.

Not knowing a lot of details, 660 uF seems like a lot...

rjj45

No, it's a pair of Morel CAM558s I got last year from somebody. The funny thing for me I always forget about mid crossovers is that the LP components affect the response of the HP section.  So I have to tweak both.

rjj45

I'm doing a bit of a clone of Tony Gee's BackBox. He uses a Linkwitz transform on the RS270 with about 600uF caps. So far I haven't decided whether I like that better than normal RS270 bass. 

Wolf

An LT actually boosts voltage to extend the bass lower via more Xmax usage. The capacitively loaded Passive Assist (aka 3rd order sealed bass rolloff), makes the knee of the rolloff sharpen which lowers the F3.

I'm actually following your progress on this, as I have a quad of the 270 sitting here...

rjj45

Close miked RS270 without and with 600uF caps inline. Note 2dB per division. Box is about 26 liters.

ani_101

which is with the cap? the purple plot? Looks like the f3 is going to be extended a bit, but the black plot looks like it has greater thumping in the 80hz region, which translated to in your chest thump... maybe that's why you like it better?

Try playing something which goes lower - might change your opinion. James Blake limit to your love - smaller woofers capable of playing lower make a flapping sound and motion and sounds absolutely horrible. Larger woofer sound absolutely gorgeous. by smaller woofer i meant 8" and larger is 15".  5" and less do not event attempt the frequencies and everything just sounds peachy.... the 8" f3 is 35hz and the 15" f3 is 32hz. very similar, but the output and tactile feel of the 15 is very real. The 15" doesn't have much of a xmax either.... it's an actual woofer and not a sub.

ani_101

Do you have the calulcations for determining the cap? I have a sealed build somewhere on the way.... would like to try it.

rjj45

If I am able to attach a PDF - here is Tony Gee's writeup with calculations for the caps required. 
Yeah, I will need to listen to this a while. Like the thump in the 80 Hz region, but a bit greater extension
down low with the cap. Probably depends on the music also.

PDF won't attach. Here is the URL.
He posted it in the PE Gallery also.
http://projectgallery.parts-express.com/speaker-projects/black-box/

ani_101

Thanks Don. It says to keep QTC around 1. my enclosure is slightly oversized over 0.7 QTC to extend the low end.... so may not work. But i'll try anyways. How big is your enclosure?

rjj45

I'm pretty sure that my interior volume is about 26 liters, just a bit larger than Tony Gees which was 24 liters.
I modeled it also in 35 liters for a possible floorstander, but the lower F3 was not enough to sway me.

When I analyzed this project for future use about 6 months ago, I decided that empirical tests were the easiest method - dropped one of the RS270s into an existing test box and just measured bass with and without the caps.

ani_101

I just checked the model. It is 0.87 Q. I would also try the empirical method, though i am a bit ways off from trying it. The box is half done, and it is layered, so no changing that.... so a bit better than 0.7, but about 90% more volume than required for 1 QTC. but i gained about 5Hz in the f3 but going with the larger volume.... 

e6zion

DELETED wrong thread like an idiot