I understand the utility of <1/4 wavelength driver spacing. I'm also aware of rising response of woofers and the need to cross well before cone breakup. Unfortunately, there are WAY more things that I don't know than those that I do know!
I'm thinking of making a WAW with an 8" full range and an 8" woofer. I know this is the multiway forum, but I respect all opinions, including those of the multiway way guys and I think the question is more appropriate here. Mods please feel free to move if you want.
Anyway, If I put the 8" full range on a baffle at about 90cm height and the 8" woofer centered at about 20cm, they'd have say roughly 70cm separation -- dictating about a 100 Hz cross with the 1/4 wavelength rule.
Partial motivation would be to get the floor reinforcement with a sealed cabinet. My question is, with a 6dB/octave, 1st order filter slope and the sealed box, would the falloff be fast enough that I don't hear 2 sources at, for example, 300 Hz?
I guess the jewels of the question are: does filter order greatly affect acceptable driver spacing?
Muchas gracias. Hope everyone is safe from the coronavirus.
I'm thinking of making a WAW with an 8" full range and an 8" woofer. I know this is the multiway forum, but I respect all opinions, including those of the multiway way guys and I think the question is more appropriate here. Mods please feel free to move if you want.
Anyway, If I put the 8" full range on a baffle at about 90cm height and the 8" woofer centered at about 20cm, they'd have say roughly 70cm separation -- dictating about a 100 Hz cross with the 1/4 wavelength rule.
Partial motivation would be to get the floor reinforcement with a sealed cabinet. My question is, with a 6dB/octave, 1st order filter slope and the sealed box, would the falloff be fast enough that I don't hear 2 sources at, for example, 300 Hz?
I guess the jewels of the question are: does filter order greatly affect acceptable driver spacing?
Muchas gracias. Hope everyone is safe from the coronavirus.
So the 1st order alignment will be more than fine at 100hz with only 70cm
Not sure what your asking, but I’ll try so excuse me if I am not saying correct
BW6 filters when aligned are 90deg from each other. That one fact is the only real issue as far as the filter goes (not reflections or baffle) that will cause 2 sources to emerge, however that’s not how your going to hear things unless there headphones ;-)
The two signals will sum just like two speakers wired in series sum. You’ll hear the sum of the two.
My 1st advice is to use a LR2 filter, the alignment is way easier and there is none of the phase issues. The LR2 has fast rise time and is plenty transient. If it’s a passive filter the BW6 I can understand a little more, if it’s a active filter why go through the headache for such a minimal amount of transient response when a LR2 will be a ton easier to get right.
Not sure if you know, an LR2 the phase rotation on the low pass and high pass go the same direction when one driver (the low pass) polarity is flipped. That’s perfect phase coherence.
Why wouldn’t anyone want that, it’s the next best thing to linear phase.
If your bound to the BW6 what I personally would do is measure the drivers in there baffle and see what is actually going on acoustically. You can do the a*b=c thing or just compare them and overlay them. Than measure them together and look at how they sum and each driver individually and see the acoustic behavior.
In my experience with crossovers I can fiddle with crossovers all day and sim a driver and crossover but it’s always the diffrent story once installed in its home. You can get close with sims but acoustic slopes are usually much farther out on most frequencies above 100 ish
Does filter order greatly affect driver spacing? Well yes , if your mid is crossed at 1k with a BW6 and a tweet or horn crossed let’s say an oactave up at 2k BW6 that’s 1000hz or 1.128 feet you can move the tweeter away from the mid and still get them to blend right with eq and time alignment signal delays. I know poor example and who would do that except maybe a car audio install or something but yes, that same senerio with a LR4 you have to be crossed at 6db down so .25feet is now the norm at 1khz (1:4wave) so yes a shallow slope does allow you to move the driver farther apart ,
At 70cm roughly 2 feet you would need to cross lower than 500hz with a 4th order 1000hz 2nd order and 2000hz BW6
So if your crossing at 100hz where you need to be lower than 2000hz I would say you can go with a higher order filter. And use LR!
(All my numbers were off the cufff I didn’t calculate anything, was just for demonstrating a idea. Plz calculate the numbers for accuracy)
Hope that helps.
Not sure what your asking, but I’ll try so excuse me if I am not saying correct
BW6 filters when aligned are 90deg from each other. That one fact is the only real issue as far as the filter goes (not reflections or baffle) that will cause 2 sources to emerge, however that’s not how your going to hear things unless there headphones ;-)
The two signals will sum just like two speakers wired in series sum. You’ll hear the sum of the two.
My 1st advice is to use a LR2 filter, the alignment is way easier and there is none of the phase issues. The LR2 has fast rise time and is plenty transient. If it’s a passive filter the BW6 I can understand a little more, if it’s a active filter why go through the headache for such a minimal amount of transient response when a LR2 will be a ton easier to get right.
Not sure if you know, an LR2 the phase rotation on the low pass and high pass go the same direction when one driver (the low pass) polarity is flipped. That’s perfect phase coherence.
Why wouldn’t anyone want that, it’s the next best thing to linear phase.
If your bound to the BW6 what I personally would do is measure the drivers in there baffle and see what is actually going on acoustically. You can do the a*b=c thing or just compare them and overlay them. Than measure them together and look at how they sum and each driver individually and see the acoustic behavior.
In my experience with crossovers I can fiddle with crossovers all day and sim a driver and crossover but it’s always the diffrent story once installed in its home. You can get close with sims but acoustic slopes are usually much farther out on most frequencies above 100 ish
Does filter order greatly affect driver spacing? Well yes , if your mid is crossed at 1k with a BW6 and a tweet or horn crossed let’s say an oactave up at 2k BW6 that’s 1000hz or 1.128 feet you can move the tweeter away from the mid and still get them to blend right with eq and time alignment signal delays. I know poor example and who would do that except maybe a car audio install or something but yes, that same senerio with a LR4 you have to be crossed at 6db down so .25feet is now the norm at 1khz (1:4wave) so yes a shallow slope does allow you to move the driver farther apart ,
At 70cm roughly 2 feet you would need to cross lower than 500hz with a 4th order 1000hz 2nd order and 2000hz BW6
So if your crossing at 100hz where you need to be lower than 2000hz I would say you can go with a higher order filter. And use LR!
(All my numbers were off the cufff I didn’t calculate anything, was just for demonstrating a idea. Plz calculate the numbers for accuracy)
Hope that helps.
Last edited:
In what way?
Consider this, you listen at 3m from the speaker, one driver at 90cm, the other at 20cm, and you listen at a height of 90cm? In this case the lower driver is 8cm further away, a quarter wavelength at more than 1kHz.
Sage advice!
Also, I've read the 1/4 wavelength center to center driver spacing, and for the life of me it does not make sense at the listening location, and as I understand it fails to take into account head and ear related issues. I have repeatedly violated that standard with no ill effect. Honestly, not sure how you keep to this with most tweeters and their big circular mounting plates anyway.
I'm sure this is the point in time when some one points me to a book on speaker design I need to read.
I would suggest a smaller midTweeter. A big FR usually brings bass to the party, but you pay further up. And with helper woofer the extra bass brings you nothing.
In a WAW any midWoofer breakup is usually high enuff not to worry much about it.
It was Danely who showed that once you get to 1/4 wavelength centre-to-cente at the XO frequency, the wavelengths become so long that the drivers are essentially co-incident. Also, XO on a WAW is where the drivers are fairly well behaved, so phase coherance with 1st order XO is fairly easy to achieve.
That is a low XO for a WAW, what you are describing is a sub/satelittle system. subtely different than a WAW.
dave
Thanks dude. I'm sure your rationale for suggesting that is solid and I appreciate your input.
I've (probably poorly) implemented a few of this sort of design with 3", 4", and 7" (12pw) mid-Tweets. In my head and with my implementations, the 6.5" was totally more satisfying than the smaller drivers. I'm sure there were weaknesses including radiation patterns and rear wave reflections, but there you have it.
I am 1/2 even contemplating going to a 12" + 12" just to see if it floats my boat!
As always, the diyaudio community is awesome.
ff85wk, 126en, Alpair7.3, CHR70, Alpair10.3. I have no doubt that my WAW bass augmentation executions were less than ideal. The 126en and A7.3 were in your FH3, the A10.3 in Scott's Poplar, ff85wk and CHR70 in simple reflex. Most xovers between 150 Hz and 250Hz.
I’d rate teh A10.3 ahead of teh A12p stock (unless you like the vinatge top of teh A12p (then you should look at A10p). The others all have some significant issues, the FF85wKeN once tarted up is one of my favourites, but if you want a really “big” sound probably lacks some cone area. CHR was never a favorite, and FE126 really needs some work.
dave
dave
This could be a consideration through your crossover if things change much, but there are ways to manage.
Sorry, not sure what you are saying. Do you just mean using different filters to tidy up directivity aberrations?
Just like the larger driver at high frequencies, the off axis levels can fall and this is normally ok, but with a crossover it happens in the middle of the spectrum so the effect comes in and goes back out again. Sometimes this doesn't happen, don't try to fix it if it isn't a problem but if it is, it's about driver spacing and floor spacing first, then maybe the crossover.
Sage advice!
Also, I've read the 1/4 wavelength center to center driver spacing, and for the life of me it does not make sense at the listening location, and as I understand it fails to take into account head and ear related issues. I have repeatedly violated that standard with no ill effect. Honestly, not sure how you keep to this with most tweeters and their big circular mounting plates anyway.
I'm sure this is the point in time when some one points me to a book on speaker design I need to read.
At the listening location isn’t where the problems originate, by that time they should have summed and several propagation and modulation cycles.
I think it’s more about getting them to sum properly, of off axis energy can interfere and if it’s a 1/4wave away or more it will start to combfilter and cancel just like a reflection from a wall (kinda sorta) but will definitely comb filter and cause small notches and ripples in response and lead to non minimum phase behavior.
They would have to be about 28ms apart to hear two sources (about 30 feet) otherwise they’ll sum and it your ears will hear one sound source. It’s whats it’s composed of is what matters
It’s actually pretty important. I’m sure your music still sounds good breaking the norms, however the level of detail in the presentation is diminished and stage placement is detrimentally affected
Read Floyd Toole (or Tooyle) book. It’s very good
Also read here
Last edited:
At the frequencies involved to get a woofer & a midTweeter to be within a 1/4 wavelength both drivers are essentially omnidirectional. One of th ekeys to getting a good blend between the drivers requires a fairly extended response for the bass driver so the top of the woofer is unlikely to be an issue.
You may run into a problem if the woofer does not have th erequire extention.
dave
Yeah there’s compromises, but a well thought out system with average gear can be stunning if you can get things to work together properly.
If the time domain and frequency domain are addressed weather by dsp , placement, quality gear, and a sober approach one should net a system that should be mostly free of any compromises that severely hinder performance minus the sheer addiction factor for wanting to try something new lol
I did make a WAW, a bit like you want with an Alpair 10M gen 3 and a Scanspeak 26W/8534G00, that is a 5.25" fullrange driver and a 10 woofer. But i put the woofer on top of the box, not near the floor (that is too far from the fullrange) and crossover at 250Hz in first order. I still got the bass i need with the right crossover. First i made everything ported, but i moved to sealed boxes, and both work very well (sealed is better than ported in this case).
I also tried a 8" seas CA22 with that alpair, and that did not integrate well. If you use a 8" woofer, i would even use a smaller FR, like a Alpair 7 (4") or something similar. I don't know if there is a scientific base for this, but it seems that smaller woofer like smaller FR drivers more than larger.
I also tried a 8" seas CA22 with that alpair, and that did not integrate well. If you use a 8" woofer, i would even use a smaller FR, like a Alpair 7 (4") or something similar. I don't know if there is a scientific base for this, but it seems that smaller woofer like smaller FR drivers more than larger.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.