If I understand what's been said here and above, adding the 0.5 additional woofer would cause close to a 6dB increase in SPL, 3 dB because of the additional driver, and another 3dB because paralleled they now present a 4 ohm load to the poweramp instead of 8 ohms (?). I'm guessing that by "0.5" you mean the second woofer is rolled off lower than the first, to create the anti-bafflestep attenuation curve(?). Since there would be a phase differential between the 2 woofers, due to the different inductors (rolloff frequencies), I wonder if they would add elegantly acoustically?
Baffle step rolloff depends on size of wavelengths, size of front surface of the baffle, and proximity to walls or anything that is strong and/or dense enough to reflect the lower frequencies. All published frequency response graphs are done with the driver on a large wall, and facing into an anechoic chamber. That's the standard anyway. Not necessarily real-world. If you need to do a lot of correction for baffle step, doing it passively at a passive speaker crossover you will likely lose a large amount of system efficiency in the mid and high frequencies. You might effectively turn a 100watt poweramp into a 10 watt amp (for about 10dB of total correction - which is probably close to realistic). I prefer to do all EQ and crossovers actively for this reason and many others. A decent Baxandall tone control might be all you need, depending on the variables.
Last edited:
Hi,
In the passband they are in phase. In the stopband of the 0.5 way
section the phase difference gradually increases towards 90 degrees.
So basically the drivers never cancel each other in the area of interest.
And if the cabinet needs full BSC then the output of a 0.5 way pair
will add to flat, and by definition being flat, add to give linear phase.
All done correctly of of course, and it is never simple to get flat.
rgds, sreten.
In the passband they are in phase. In the stopband of the 0.5 way
section the phase difference gradually increases towards 90 degrees.
So basically the drivers never cancel each other in the area of interest.
And if the cabinet needs full BSC then the output of a 0.5 way pair
will add to flat, and by definition being flat, add to give linear phase.
All done correctly of of course, and it is never simple to get flat.
rgds, sreten.
Just to complicate things, I have never had a textbook BSC as the smoothest solution. Sometimes a little more, sometimes a little less. So, what is the right answer? As always, measure in the intended location in a prototype cabinet.
About half the time, I can get my compensation by upping the LP inductor. Other times, I have to add a second coil and resistor. I am not a fan of .5 config as I do not like low impedance speakers.
About half the time, I can get my compensation by upping the LP inductor. Other times, I have to add a second coil and resistor. I am not a fan of .5 config as I do not like low impedance speakers.
what I was wondering - if one FR driver has nothing in the signal path and the other has an inductor, is there a phase issue there?
EDIT just read sretens post ..
Last edited:
instead of using a resistor, what if the inductored driver was place at the rear (bipole?)
Last edited:
The amusing answer is active.
The theoretical example above is my own setup. Drivers are 8Ω driven in pairs from a 4Ω amp. Do I see that hypothetical perfect 3dB gain, nope. Unless you have an amp that can run the numbers linear current and linear voltage flat you will never see that 3dB, maybe 2dB, even less once DCR from passive parts is added on. So the full compensation never occurs. Baffle step is normally much higher today than in the past due to the extreme narrowness of most designs. This wasn't always the case, if the enclosure is wider then BSC doesn't cause issue until much lower. We like to toss around names for everything, defining function. This is to be expected and wanted, but what if our sub/woofer/midbass is all in the same driver? For example a sealed (AS) pair combined with an MLTL pair per speaker that defines the MMMM section. The net result combines multisub + control of the low midband from a distributed approach. An MTMMMMM hardly defines what it does.Good example of are the GR Research M130X and M165X.
Haven't seen this done either, would work, but the shift in level changes the BSC point acoustically. Perhaps not enough to affect SQ
The amusing answer is active.
Most decent amplifiers do reasonably well at providing what's necessary in getting you that full 3dB when going from a 8 ohm single, to 4 ohm pair of drivers.
All amplifiers will have no trouble delivering twice the current unless they are driven into clipping or their protection circuitry is activated. What determines the voltage loss, when going from an 8 to 4 ohm load is usually the size of the RE resistors present within the output stage of the amplifier. If appropriately sized these don't cause a tremendous amount of loss, we're talking 0.25-0.5dB tops. You also need to make sure that the DCR of your loudspeaker cables is appreciably low too and then of course the DCR of the series inductor. All of these things can be low enough to cause only a small amount of loss, but if you're getting any more than 1dB, then you're doing it wrong.
I can't think of any amp (off the top of me head) that can do 100 into 16Ω, 200 into 8Ω and 400 into 4Ω loads and 800 into 2Ω. Normal is more like 100w into 8Ω with 150-160w into 4Ω as long as we don't enter into current clipping. This is where cheaper designs fall short of even this potential.
Wires are excluded from this. I would assume this is considered from the onset.
Isn't interesting to note that amp performance affects BSC?
If we used bel canto C5i for $1800, sure, Oh wait spied one on eBay for $1200. Doesn't make much sense on a pair of $500ea speakers where it would still require a second amp as a min. Personally would flip those numbers around and spend more on better drivers than ridiculous priced electronics to ensure near double power rating (120w
, ensuring proper BSC.The amusing answer is active.
Hi,
Your still talking utter nonsense, active makes little difference to the response.
That an amplifier can't provide double the power into half the impedance
is irrelevant below full power and nothing to do with frequency response.
They all can easily at lower power double power into any half impedance
load unless your pedantically talking tosh about amplifiers with significant
output impedance, hardly the typical case, with complex interactions.
An active 0.5 way will give fulll 6dB BSC by definition, and your musings
are clueless, or are correct but you've got your your physics very wrong.
Your active setup its amusingly wrong, and no answer. A pair of 0.5
way drivers, 0.5 wayed with a further pair in parallel is utterly clueless.
rgds, sreten.
Last edited:
You're confusing two different things with one another. The maximum output power is not the same as the voltage loss created by the instrinsic losses within the amplifier and loudspeaker as a system.
You could have an amplifier that would deliver 50 watts into 8 ohms and 50 watts into 4 ohms, as limited by the output stage, but providing the instrinsic losses are small you will still see close to the 3dB rise from the doubling of the output current. Yes, you wont see this gain if you're running the amplifier at its current limits - ie right up until clipping - but we don't typically listen like this.
Sorry was thinking in terms of current drive, instead of voltage drive.
sreten, love how you fell all over yourself (twice). If you knew the answer stating voltage vs current drive would have been sufficient to wake me up.
My 4way active is not a comparison of a 2.5 way and should not be considered as such. What is different is how the low end is handled where a MLTL's handles the bottom end along with an AS pair. The MLTL is crossed lower (<140Hz) than the AS (~250Hz) of the bass section. The upper MTM's are crossed at ~100Hz to reduce doppler effect. Top end is ~2.2k and up. In the region where these three sections overlap is handled with negative eq/shelving filters. Result equals flat on axis with 8 multipoint sub loading and midbass/low midrange compensation in a stereo pair. BSC is around 200Hz for it's 23" elliptic width.
Naw don't have a clue, just misplace them
Please read the post again, this time starting with the first sentence. YES
Now if it's some unwritten requirement or a sticky I've missed about flailing of arms and groveling in forgiveness, FORGETABOUTIT
Listen to effecy of BSC before and after
This may be of interest to the thread topic of is it important. You can listen to the effect for yourself - I turned on BSC halfway through.
http://www.diyaudio.com/forums/full-range/252627-viva-la-vifa-curvy-cabinet-dcr-tc9fd-2.html#post3850829
http://www.diyaudio.com/forums/attachments/full-range/405088d1394537733-viva-la-vifa-curvy-cabinet-dcr-tc9fd-dcr-vifa-clip-bsc-compare-02.zip
This may be of interest to the thread topic of is it important. You can listen to the effect for yourself - I turned on BSC halfway through.
http://www.diyaudio.com/forums/full-range/252627-viva-la-vifa-curvy-cabinet-dcr-tc9fd-2.html#post3850829
http://www.diyaudio.com/forums/attachments/full-range/405088d1394537733-viva-la-vifa-curvy-cabinet-dcr-tc9fd-dcr-vifa-clip-bsc-compare-02.zip
But does flatter response actually mean better sounding? It didn't seem like it when I tried it....just saying.This may be of interest to the thread topic of is it important. You can listen to the effect for yourself - I turned on BSC halfway through.
http://www.diyaudio.com/forums/full...fa-curvy-cabinet-dcr-tc9fd-2.html#post3850829
http://www.diyaudio.com/forums/atta...et-dcr-tc9fd-dcr-vifa-clip-bsc-compare-02.zip
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.