Byrtt,
That looks really nice.
as you say, almost SH50 like. 
Can you also show your measured acoustic xo plot ? LF, HF, total?
You showed an earlier plot of response and phase that I got that looked pretty good. When was that ? I would like to go back to that - I can't seem to get my phase to look anything like that lately.
Maybe I should do a miniDSP reset and start with tabula rasa.
This has been a good discussion and glad we are looking into fine details. It seems that a DCR, although one may not actually use it to extract deep bass, is good to have from standpoint of providing wide bandwidth for use with a 1st order XO. Something for me to think about. Certainly can be modeled in Akabak and see how the phase does.
That looks really nice.
as you say, almost SH50 like. Can you also show your measured acoustic xo plot ? LF, HF, total?
You showed an earlier plot of response and phase that I got that looked pretty good. When was that ? I would like to go back to that - I can't seem to get my phase to look anything like that lately.
Maybe I should do a miniDSP reset and start with tabula rasa.
This has been a good discussion and glad we are looking into fine details. It seems that a DCR, although one may not actually use it to extract deep bass, is good to have from standpoint of providing wide bandwidth for use with a 1st order XO. Something for me to think about. Certainly can be modeled in Akabak and see how the phase does.
Outdoor Measurement
Speaker up on 6 ft ladder and driver/mic at 73in high, 1 meter away. The challenges of outdoor measurement include waiting for neighbor's AC unt to kick off, waiting for birds to stop chirping (have you seen a robin's voice spectrum?!), waiting for neighbor's children to stop yelling and talking, waiting for overhead airplanes to not fly by, waiting for a lull in the wind. It was tough - took a long time for all that to get 3 sweeps. Still has distant AC units in bass, and maybe a kid talking...
Acoustic XO ended up at 940Hz! Looks like the RS225 needs either PEQ to get it to behave more classically by pre-falttening, otherwise, too much output above 350Hz.
IR and SR:
Phase:
Speaker up on 6 ft ladder and driver/mic at 73in high, 1 meter away. The challenges of outdoor measurement include waiting for neighbor's AC unt to kick off, waiting for birds to stop chirping (have you seen a robin's voice spectrum?!), waiting for neighbor's children to stop yelling and talking, waiting for overhead airplanes to not fly by, waiting for a lull in the wind. It was tough - took a long time for all that to get 3 sweeps. Still has distant AC units in bass, and maybe a kid talking...
Acoustic XO ended up at 940Hz! Looks like the RS225 needs either PEQ to get it to behave more classically by pre-falttening, otherwise, too much output above 350Hz.
IR and SR:
Phase:
Attachments
Last edited:
If RS225-8 is so important to get filtered out as you and 5th element suggest then yes else some of the first setups that had acoustic XO around 730Hz had pretty good data measured FR/PHASE/IR/SR and sound clips sounded good especially a 60Mb FLAC he shared
It is rather important with that driver; most metal cone 8's really. Even with BW1 at 350 I have my doubts that the metal cone resonance is sufficiently down. The highest I think I've seen the RS225 successfully used is about 1200Hz with LR4 or LR8. BW1 @350 is -24db @5600 vs LR4@1200 is -48 @4800.
Plus, I thought you guys are trying to keep the mid range in the mid. 730Hz is well into the midrange.
A better driver in this price range for BW1 would be something like the Scanspeak 22W/8534G00. It's very smooth up top and can be used about as high as you want, beaming not withstanding.
On the otherhand, if it sounds good to you and you're happy with it then that's all that matters. You are of course lowering the magnitude of the breakup which will help prevent ringing, so that helps.
Speaker up on 6 ft ladder and driver/mic at 73in high, 1 meter away. The challenges of outdoor measurement include waiting for neighbor's AC unt to kick off, waiting for birds to stop chirping (have you seen a robin's voice spectrum?!), waiting for neighbor's children to stop yelling and talking, waiting for overhead airplanes to not fly by, waiting for a lull in the wind. It was tough - took a long time for all that to get 3 sweeps. Still has distant AC units in bass, and maybe a kid talking...
Yes, it sucks for me too. I live near downtown san antonio in a historic district right by a major highway. I've found Sunday mornings to be the best for me.
Acoustic XO ended up at 940Hz! Looks like the RS225 needs either PEQ to get it to behave more classically by pre-falttening, otherwise, too much output above 350Hz.
This is what I found too.
This is with peq on the breakup nodes only (two notches). Notice that it matches yours almost exactly except that it looks like I've applied a stronger flattening on the breakup. You appear to only be about 12-15db down @ 6K.
This shows the effects using the following settings.
Notice the low q cut centered @ 950Hz. This makes it fall into perfect BW1 @350 or so
If you look at IR, there is no reflection first 10ms which is why I showed the system response gated at 10ms in blue. I have limitations as to where I could put it - house is on a slope so no level spot for ladder farther away. House is good 10ft behind speaker. Wouldn't the well help actually to reduce back reflections?
As long as the speaker to house and back to mic distance is longer than speaker to ground to mic then you should be good. They may be becoming omni directional, but the waves still have to travel around the baffle, to the house and back to the mic before the ground reflection gets to the mic to be a problem
Ground plane would give you a more accurate measurement of the bass in room, but before room gain.
Think it maybe was the first attempt to BW1 and the special and beautiful was it only had electric XO, delay and Linkwitz transform. I plague you for a mdat file to help in design simple passive to replicate the responses and from that can attach what i think nice data reminded no EQ at all at that time. Together with the live set up show same slopes run in electric loopback to use as sort of textbook belonging to 730Hz BW1. Think electric XO was set 350Hz, delay for 10F in 0,2mSec area, and padding for 10F in area -5-6dB. 182,5Hz three octaves below acoustic 730Hz XO point 10F is still there and starts its error fall off at 176Hz. RS225-8 is not following three octaves up to 2920Hz so maybe 10F is fire ring too late which can make the low Q boost as shown picture two with HP driver a 0,03mSec error too late. Think if 10F is steeped a little bit forward system phase and minimum phase will begin be coherent and SR fall a little in amplitude but be more strait from 0-2mSec as is from 2-4mSec, but that RS225-8 is not holding its slope there will probably lead to a fair low Q dip then.
5th element and satx warns about using RS225-8 up high, last plot is group delay RS225-8 and the more than 10mSec spike above 3kHz i can see is not going to blend friendly with 10F's timing there.
Agree good discussions and looking into box environment for 10F to extend low end getting acoustic slope in 600-500Hz area would probably balance last bits if possible to modify
.Attachments
Last edited:
Byrtt - thanks for all the help with the XO sims!
SATX - thanks for the tips and fine points of improving this xo. It is trickier than I thought. Imagine if this was done passively how much of a pain that would have been. Fortunately, no caps or coils were harmed in the making of this speaker.
I think this work over past few days will really help to polish this already very nice sounding speaker to be exceptional.
A lot of fun here.
Getting back to extending 10F bass response for better low order XO: would a simple bass reflex tuned for 80Hz help? That would get bandwidth available to go 2 octaves below current 350Hz f3 of the Dagger which is natural 1st order already but needs cone control and bass extension. Wouldn't this mess with the GD? Maybe not as Byrtt is showing how the 10F and TC9 in a DCR tuned for 60Hz seems to do well.
SATX - thanks for the tips and fine points of improving this xo. It is trickier than I thought. Imagine if this was done passively how much of a pain that would have been. Fortunately, no caps or coils were harmed in the making of this speaker.
I think this work over past few days will really help to polish this already very nice sounding speaker to be exceptional.
A lot of fun here.
Getting back to extending 10F bass response for better low order XO: would a simple bass reflex tuned for 80Hz help? That would get bandwidth available to go 2 octaves below current 350Hz f3 of the Dagger which is natural 1st order already but needs cone control and bass extension. Wouldn't this mess with the GD? Maybe not as Byrtt is showing how the 10F and TC9 in a DCR tuned for 60Hz seems to do well.
Last edited:
This looks good. So you just added the 3PEQ's to the RS225? What electronic XO freq is this on RS225 and 10F?
Yes three peq's, dual notches on the breakup and the 950 cut. Small eq on 10f too. This is with xo set at 225 for the Rs225. No xo on the 10f which gives a perfect BW1 response @ 350 in 1L sealed, but no protection for it. Acoustical system xo about 325-350. I don't know if this would provide you with sufficient baffle step or not though, since you seem to have dialed in about 7db before.
Here is the system with phase. Gray is total phase, orange and blue are driver phase. about 60 degree phase difference, but apparently BW1 won't actually align? I would think this would be really phasy, but that could be part of the 1st order magic?
To make this work, I had to apply a large amount of additional delay to the woofer (don't know if that's ok or not, never done it) Notice that there is almost no cancellation anywhere.
Here it is with no delay on the RS225. The gray line now represents reversed polarity. Kind of looks similar to the phase problems you've been having.
And here are the settings for the 10f
Byrtt - thanks for all the help with the XO sims!
SATX - thanks for the tips and fine points of improving this xo. It is trickier than I thought. Imagine if this was done passively how much of a pain that would have been. Fortunately, no caps or coils were harmed in the making of this speaker.
I think this work over past few days will really help to polish this already very nice sounding speaker to be exceptional.
A lot of fun here.
Getting back to extending 10F bass response for better low order XO: would a simple bass reflex tuned for 80Hz help? That would get bandwidth available to go 2 octaves below current 350Hz f3 of the Dagger which is natural 1st order already but needs cone control and bass extension. Wouldn't this mess with the GD? Maybe not as Byrtt is showing how the 10F and TC9 in a DCR tuned for 60Hz seems to do well.
BR tuned low may be ok. I think I would try aperiodic first as that should give extension between sealed and ported, but will damp the resonance of the 10f, so will help control cone and distortion there. Aperiodic will also not need to be any larger that what you currently have. BR will likely take up more internal volume. With the cone shape you have now, that should be a fairly ideal shape for aperiodic. The formula is 10:1 large side to small and then you lightly stuff the small opening while measuring impedance until the fs peak is sufficiently damped. This along with a slightly higher xo might get you where you need to be.
BR would probably give more extension, but the impedance peaks around tuning might cause strange things to happen to the phase in the stop band
So,
This looks fantastic. Can't wait to try it out as soon as I have time. The miniDSP and amps are going into a real box with binding posts and RCA inputs right now. Figured it was time to get over the "lab" version of bare boards glued onto a sheet of foam core
Still not a metal box but hey, we all have our limitations.
Given how good the phase looks in the SATX sims - I am anticipating that the SR is going to look even better. Will try to see if I can get some cone control on the 10F. Although for 95dB Max SPL indoors as near fields I will be just fine with 10F flying loose and full range. Would like to avoid LF inter modulation distortion though more than the xmax getting bottomed out.
This looks fantastic. Can't wait to try it out as soon as I have time. The miniDSP and amps are going into a real box with binding posts and RCA inputs right now. Figured it was time to get over the "lab" version of bare boards glued onto a sheet of foam core
Still not a metal box but hey, we all have our limitations.
Given how good the phase looks in the SATX sims - I am anticipating that the SR is going to look even better. Will try to see if I can get some cone control on the 10F. Although for 95dB Max SPL indoors as near fields I will be just fine with 10F flying loose and full range. Would like to avoid LF inter modulation distortion though more than the xmax getting bottomed out.
Thanks. To get the phase looking like this took -0.13 msec of additional delay on the woofer, no delay on the 10f. This is quite a bit, without it you see the last graph which looks something like what you were getting; cancellation below crossover with polarity one way and above with it reversed.
It should be pretty accurate, but will probably need some fine tuning as far as phase is concerned. I estimated the acoustic center of the woofer to be 18mm behind the mid driver which should be in the ballpark. I would adjust the delay and measure until you aren't getting cancellation around 1k and if you're getting cancellation below the xo then flip polarity before setting delay.
The graph I posted that looked almost exactly like yours where the rs225 was crossing at 900 or something should point to the sim being rather accurate so I think you should be able to recreate what I have with similar eq settings.
Here's what I get with the 10f ported in 1.5L tuned to 80. You can see the box splice at 300 on the 10f. It'll give you an idea, but the acuracy is always questionable when you splice a box response to frequency response around 300Hz. I was able to hit BW1 target at 350 with BW1 electrical xo at 200Hz. Phase looks ok I guess. This is with no delay on either driver.
I didn't do this in a box modeling program. I used response modeler so no port, you just set tuning. I can do it in unibox later if you can't.
No delay would certainly be easier passive. In this case the xo on the 10f causes the necessary phase shift. I don't even know how you would do it passive if it needed delay unless you did a stepped baffle or asymmetric xo or relaxed slopes etc.
No delay would certainly be easier passive. In this case the xo on the 10f causes the necessary phase shift. I don't even know how you would do it passive if it needed delay unless you did a stepped baffle or asymmetric xo or relaxed slopes etc.
miniDSP and amps boxed up
My system finally looks tidy and is portable. This box has a magnetic snap lid and is 15in wide x 10in deep x 1in high. 4 sets of 5-way binding posts for output, 2 x RCA inputs, miniDSP volume knob, LM317 linear power supply for miniDSP when in standalone. Flip top open and connect USB cable to program. Amps are dual Ybdz TPA3116D2's with oscon 330uF SEPF cap mod and bootstrap snubber mod on treble channels. Good for 40watts/ch with under 0.1%THD. Uses dual 19v 4.65amp smps bricks.
It's a chocolate box in case you are wondering.
Anyhow, I am back up and running again with sound. I can even take these speakers with me and setup anywhere I want good sound with the nifty amp box now done.
My system finally looks tidy and is portable. This box has a magnetic snap lid and is 15in wide x 10in deep x 1in high. 4 sets of 5-way binding posts for output, 2 x RCA inputs, miniDSP volume knob, LM317 linear power supply for miniDSP when in standalone. Flip top open and connect USB cable to program. Amps are dual Ybdz TPA3116D2's with oscon 330uF SEPF cap mod and bootstrap snubber mod on treble channels. Good for 40watts/ch with under 0.1%THD. Uses dual 19v 4.65amp smps bricks.
It's a chocolate box in case you are wondering.
Anyhow, I am back up and running again with sound. I can even take these speakers with me and setup anywhere I want good sound with the nifty amp box now done.
Attachments
Last edited:
