Telstar said:
Below the frequency the baffle supports the driver to, it will flap in the breeze. Best to filter it below that frequency.
dave
IMHyLPO (in my honest yet limited personal opinion), in most cases, a speaker system utilising a "full-range" (FR) driver will present most types of music more to my satisfaction if the lower frequencies are filtered away from the FR driver; larger, BLH loaded drivers may be the exception. I think this is especially true if one listens to a wide range of music and would prefer not to have that niggling thinking / feeling that at any moment things could fall apart... such a feeling does not lead to pleasant listening experiences (for me).
A PLLXO (1st or 2nd order high pass) would be my preference. I think I would use one even if I did not have any additional bass support, just so my amp could have it a little easier and the FR driver was a little better controlled (perhaps). I'm not as technically savvy as many here, but intuitively I'd not want my amp reproducing unnecessary frequencies (and sucking up headroom) and trying to control a driver flapping about unloaded. Seems uncool for both amp and the driver.
Cheers
Raymond
A PLLXO (1st or 2nd order high pass) would be my preference. I think I would use one even if I did not have any additional bass support, just so my amp could have it a little easier and the FR driver was a little better controlled (perhaps). I'm not as technically savvy as many here, but intuitively I'd not want my amp reproducing unnecessary frequencies (and sucking up headroom) and trying to control a driver flapping about unloaded. Seems uncool for both amp and the driver.
Cheers
Raymond
highpass on FR
It's not like I dont understand the problem (as it may seem from my annoyed reply, but Scott likes to repeat what he already told me). In the current setup, until I finish the baffles, add the woofers and complete the rest of the chain the way I want it, I wont use any filter to the FR, period.
Whether the distortion (going over xmax) point is at 90dB or lower i don't know that and I think that i cannot measure it, or maybe i can (there's a version of MLSSA for windows that i have in my bookmarks -we'll see).
Anyway, the point is that my ultimate goal is to go for a digital crossover before D/A conversion. And also DRC, in the digital domain. Until that I'm not wasting time trying caps & co.
What I can do to verify this distortion to my ear, is to use the highpass in the Reckhorn active XO and see if the FR will sound better or worse to my taste. This is assuming the S1 have a decent output impedance to match the F3.
To answer planet10, the baffle rollof for a 35cm wide baffle starts around 250hz i believe, that is the maximum XO point that I have considered, but i'm aiming lower 🙂
It's not like I dont understand the problem (as it may seem from my annoyed reply, but Scott likes to repeat what he already told me). In the current setup, until I finish the baffles, add the woofers and complete the rest of the chain the way I want it, I wont use any filter to the FR, period.
Whether the distortion (going over xmax) point is at 90dB or lower i don't know that and I think that i cannot measure it, or maybe i can (there's a version of MLSSA for windows that i have in my bookmarks -we'll see).
Anyway, the point is that my ultimate goal is to go for a digital crossover before D/A conversion. And also DRC, in the digital domain. Until that I'm not wasting time trying caps & co.
What I can do to verify this distortion to my ear, is to use the highpass in the Reckhorn active XO and see if the FR will sound better or worse to my taste. This is assuming the S1 have a decent output impedance to match the F3.
To answer planet10, the baffle rollof for a 35cm wide baffle starts around 250hz i believe, that is the maximum XO point that I have considered, but i'm aiming lower 🙂
Interesting post, Raymond, thanks.
A PLLXO (1st or 2nd order high pass) would be my preference.
Care to elaborate more on this PLLXO? You mean a digital XO using a PLL? That would require double AD/DA conversion and it's a total no-no in my book. So if an active XO has to be use, better it be electronic (if that's the correct word).
I think I would use one even if I did not have any additional bass support, just so my amp could have it a little easier and the FR driver was a little better controlled (perhaps).
After I have compared my 3-way with a FR without bass support, trust me that you would want all of the lower notes that they can play 😉
Before listening, i also thought that i may go fine with 150ish+ FR. Niet, not possible. Forget about it 🙂
I'm not as technically savvy as many here, but intuitively I'd not want my amp reproducing unnecessary frequencies (and sucking up headroom) and trying to control a driver flapping about unloaded. Seems uncool for both amp and the driver.
This is absolutely correct, on a theoretical point of view. But adding something in the signal path can create colorations, filters, and other distortions that may or not may be worse than the ones produced by the FR itself when pushing too hard.
See my previous post to understand what I'm heading to.
A PLLXO (1st or 2nd order high pass) would be my preference.
Care to elaborate more on this PLLXO? You mean a digital XO using a PLL? That would require double AD/DA conversion and it's a total no-no in my book. So if an active XO has to be use, better it be electronic (if that's the correct word).
I think I would use one even if I did not have any additional bass support, just so my amp could have it a little easier and the FR driver was a little better controlled (perhaps).
After I have compared my 3-way with a FR without bass support, trust me that you would want all of the lower notes that they can play 😉
Before listening, i also thought that i may go fine with 150ish+ FR. Niet, not possible. Forget about it 🙂
I'm not as technically savvy as many here, but intuitively I'd not want my amp reproducing unnecessary frequencies (and sucking up headroom) and trying to control a driver flapping about unloaded. Seems uncool for both amp and the driver.
This is absolutely correct, on a theoretical point of view. But adding something in the signal path can create colorations, filters, and other distortions that may or not may be worse than the ones produced by the FR itself when pushing too hard.
See my previous post to understand what I'm heading to.
Telstar,
PLLXO is for "Passive Line Level CrossOver", strictly in the analog domain.
You find more about it at Planet10s site.
PLLXO is for "Passive Line Level CrossOver", strictly in the analog domain.
You find more about it at Planet10s site.
Re: highpass on FR
Which gives the constant impression that you don't actually understand the problem, hense the reason I and several other people here keep repeating the fact that you need to protect the wide-band unit from the bass if you want to keep distortion down to acceptable levels. Do you honestly think we're doing it for the good of our health? The rest of your remarks indicate some questionable thinking in areas, which I dare say other people here will be kind enough to point out, but personally, I'm not going to waste any more of my time. Brick walls, and beating heads spring to mind.
Telstar said:
Which gives the constant impression that you don't actually understand the problem, hense the reason I and several other people here keep repeating the fact that you need to protect the wide-band unit from the bass if you want to keep distortion down to acceptable levels. Do you honestly think we're doing it for the good of our health? The rest of your remarks indicate some questionable thinking in areas, which I dare say other people here will be kind enough to point out, but personally, I'm not going to waste any more of my time. Brick walls, and beating heads spring to mind.
Day 6
Day 6
Because of being very busy, I was able to finish the test baffles only this mornimg.
Quick comments: two layers of pressed cardboard are very strong, easy to cut, but quite flexible (not sure if it is good or bad for the sound).
Sound (vs drivers on the floor, inclined): Better in everything. Especially bass extension. The speakers disappear, which i cannot say about the boxed ones due to placement.
Note that the drivers placement in the test baffles gives emphasis on the mid-midrange, as you can see in this Edge graph:
It wont be the definitive baffles, they'll be taller and a bit narrower with drivers centered at 85cm from the floor, instead of 65cm. But the cardboard baffles could not sustain the drivers weight in that position. You can see the drawn circle in the right baffle.
This placement rolls off the high-high freqs, that helps with theharshness that i couldnt stand during the first days of burn-in.
Now burn-in restarted and I wont post new comments before the 100 hours mark.
Lots of pictures follows. Here and there, there are wrong holes drilled, and pencil marks. But hey, it was the very first thing that i built and basically tool-less. I could have damaged the speakers, but i didnt.
Overall i'm very pleased.
Two pressed cardboard sheets: 70x100x0.3cm. A third sheet will be used to test wings, when i'll add the woofers:
Ready to work on the first cardboard sheet:
Left speaker, back. Note the two glued sheets (30 hours of vinilic glue dry under a ton of books. but i dont have the picture of that). Note also the rings to protect the cardboards.
Right speaker, back:
Left speaker, front. The black allen screws looks very good, but they were a bitch to screw.
Right speaker, front:
Closeup of the feet. I used angled metal hangers for shelves, plus some of the cardboard that was holding the speakers before. Perfect fit.
Completed speakers (playing music):
The very, very basic tools used:
Day 6
Because of being very busy, I was able to finish the test baffles only this mornimg.
Quick comments: two layers of pressed cardboard are very strong, easy to cut, but quite flexible (not sure if it is good or bad for the sound).
Sound (vs drivers on the floor, inclined): Better in everything. Especially bass extension. The speakers disappear, which i cannot say about the boxed ones due to placement.
Note that the drivers placement in the test baffles gives emphasis on the mid-midrange, as you can see in this Edge graph:
An externally hosted image should be here but it was not working when we last tested it.
It wont be the definitive baffles, they'll be taller and a bit narrower with drivers centered at 85cm from the floor, instead of 65cm. But the cardboard baffles could not sustain the drivers weight in that position. You can see the drawn circle in the right baffle.
This placement rolls off the high-high freqs, that helps with theharshness that i couldnt stand during the first days of burn-in.
Now burn-in restarted and I wont post new comments before the 100 hours mark.
Lots of pictures follows. Here and there, there are wrong holes drilled, and pencil marks. But hey, it was the very first thing that i built and basically tool-less. I could have damaged the speakers, but i didnt.
Overall i'm very pleased.
Two pressed cardboard sheets: 70x100x0.3cm. A third sheet will be used to test wings, when i'll add the woofers:
An externally hosted image should be here but it was not working when we last tested it.
Ready to work on the first cardboard sheet:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker, back. Note the two glued sheets (30 hours of vinilic glue dry under a ton of books. but i dont have the picture of that). Note also the rings to protect the cardboards.
An externally hosted image should be here but it was not working when we last tested it.
Right speaker, back:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker, front. The black allen screws looks very good, but they were a bitch to screw.
An externally hosted image should be here but it was not working when we last tested it.
Right speaker, front:
An externally hosted image should be here but it was not working when we last tested it.
Closeup of the feet. I used angled metal hangers for shelves, plus some of the cardboard that was holding the speakers before. Perfect fit.
An externally hosted image should be here but it was not working when we last tested it.
Completed speakers (playing music):
An externally hosted image should be here but it was not working when we last tested it.
The very, very basic tools used:
An externally hosted image should be here but it was not working when we last tested it.
Sweet -- will the wings be asymmetrical and if so, how will you determine the relative dimensions? I.e. will the width / height ratio be based on a classical ratio like phi? .6:1:1.6 or something like that?
I''m building something very similar with the Visaton B200's. If the wings don't work out, I have an active crossover and will add helper woofers in the manner of "early" MJK.
I''m building something very similar with the Visaton B200's. If the wings don't work out, I have an active crossover and will add helper woofers in the manner of "early" MJK.
rjbond3rd said:
I actually plan to try Derek's VPL wings. They are longer on top and bottom and shorter in the middle. He's helping me with the dimensions for a single 12" woofer per side. Woofer that will be positioned pretty close to the floor to use that for augie.
As long as the average wings length does NOT surpass the width of the baffle, one should be free of their resonance.
The wings usually work out 😉
But I'll be using an active XO as well (dunno yet if included or not in the bass amp).
Day 12
Here I am with a due update.
Burn in continues and it's taking long: 90h and counting. In the meanwhile, this morning i added some felts to dampen the backwave, since the speakers are and will remain pretty close to the back walls (around 60cm)
You cant see from the pictures, but the felt layers have been cut with growing size holes. There are 7 layers of felt and only the back of the magnet is open.
At the same time, I moved farther the Linns, leaving the test baffles to touch only a corner (for safety) and almost parallel. This way the backwave does not reflect on the other speakers and thanks to the deep felts, reflects also less onto the back wall.
Results?
The sound is more pleasant, i'd like to raise the volume up, rather than put it down. It's obvious that a 20% of the backwave was bouncing on the Linns, with obvious problems. The felts are useful, on the definitive baffles I will order the wool ones, as recommended by John K. For me they are worth the money.
Other changes after about 90h of burn-in: Some more bass came out, HF are less fatiguing. The shouting remains, especially with some recordings. It has to be noted that these fullrange without XO driven by my F3 are absolutely merciless of poor recordings. This can only improve with the addition of a preamp of the same level of transparency. Dunno instead when will be ready the new source (read a very, very special DAC).
Picture of the 7 layers of felt, from above:
Left speaker with felt:
Right speaker with felt:
Next steps:
1) More burn-in, at least another week, with the current setup, then measurements
2) Second version of test baffle, in mdf this time, with hole for the woofer, and more measurements
3) Adding woofers (trying also side placement and U, VPL shapes) and wool felts, more measurements
4) Definitive baffles made of at least 2 layers of birch plywood and one of mdf.
About woofers, i'm choosing between the Beyma SM100N and the Peerless SLS-315. A behringer dcx2496 or another one that is not available yet will be used to cross and equalize, and a dedicated amp of at least 100W will power them.
Here I am with a due update.
Burn in continues and it's taking long: 90h and counting. In the meanwhile, this morning i added some felts to dampen the backwave, since the speakers are and will remain pretty close to the back walls (around 60cm)
You cant see from the pictures, but the felt layers have been cut with growing size holes. There are 7 layers of felt and only the back of the magnet is open.
At the same time, I moved farther the Linns, leaving the test baffles to touch only a corner (for safety) and almost parallel. This way the backwave does not reflect on the other speakers and thanks to the deep felts, reflects also less onto the back wall.
Results?
The sound is more pleasant, i'd like to raise the volume up, rather than put it down. It's obvious that a 20% of the backwave was bouncing on the Linns, with obvious problems. The felts are useful, on the definitive baffles I will order the wool ones, as recommended by John K. For me they are worth the money.
Other changes after about 90h of burn-in: Some more bass came out, HF are less fatiguing. The shouting remains, especially with some recordings. It has to be noted that these fullrange without XO driven by my F3 are absolutely merciless of poor recordings. This can only improve with the addition of a preamp of the same level of transparency. Dunno instead when will be ready the new source (read a very, very special DAC).
Picture of the 7 layers of felt, from above:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker with felt:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker with felt:
An externally hosted image should be here but it was not working when we last tested it.
Next steps:
1) More burn-in, at least another week, with the current setup, then measurements
2) Second version of test baffle, in mdf this time, with hole for the woofer, and more measurements
3) Adding woofers (trying also side placement and U, VPL shapes) and wool felts, more measurements
4) Definitive baffles made of at least 2 layers of birch plywood and one of mdf.
About woofers, i'm choosing between the Beyma SM100N and the Peerless SLS-315. A behringer dcx2496 or another one that is not available yet will be used to cross and equalize, and a dedicated amp of at least 100W will power them.
little update
I consider burn-in done.
The shout remained, so it will need to be filtered with a notch.
The next steps will be:
1) Measurements. I made myself a gift and got an used Earthworks M30 mic. I'm going to take drivers, speakers and room measurements very seriously.
2) Correction and XO though a Behringer dcx2496 (modded)
3) Woofers (Beyma SM112N) and other test baffles
I'm going for a vacation now and I wont be thinking about my speakers everyday. But I'll be badder and better when i come back 😉
I consider burn-in done.
The shout remained, so it will need to be filtered with a notch.
The next steps will be:
1) Measurements. I made myself a gift and got an used Earthworks M30 mic. I'm going to take drivers, speakers and room measurements very seriously.
2) Correction and XO though a Behringer dcx2496 (modded)
3) Woofers (Beyma SM112N) and other test baffles
I'm going for a vacation now and I wont be thinking about my speakers everyday. But I'll be badder and better when i come back 😉
Do I read it correctly that your "burn-in" was 90hrs? If so, then it might be premature to tailor any correction filters to what you're currently hearing or measuring.
I'd be inclined to ride them hard for at least another 100hrs - maybe you can arrange someone to babysit them during your vacation?
Further, you might want to consider some degree of physical treatment to the drivers to mitigate the "shout" issues acoustically, rather than by either passive or active filters/ EQ.
for example, phase plugs, whizzer damping and EnABL cone treatment can transform such a driver dramatically for the better
I'd be inclined to ride them hard for at least another 100hrs - maybe you can arrange someone to babysit them during your vacation?
Further, you might want to consider some degree of physical treatment to the drivers to mitigate the "shout" issues acoustically, rather than by either passive or active filters/ EQ.
for example, phase plugs, whizzer damping and EnABL cone treatment can transform such a driver dramatically for the better
chrisb said:
No, they are about 150. I will play more when i come back from vacation but not 8h/day marathons.
In short no, I dont want to physically mod the drivers (except maybe dammar). After i filter them digitally, i'll decide if to keep or sell them and try more linear drivers. I have three FRs on top of my list. I started with cheap ones to see if the whole OB/FR concept was suited to my room and my listening habits and the answer to that is "yes".
update
Back from short holidays and before going soon to a business trip, it's time for an update.
1) The shout has been tamed with digital EQ, of about 6db. I'm not completely satisfied with the sound, but I'm still on paper baffles (grr me wants to kill the carpenter) and yet the overall sound became more pleasant than my Linns, which are going for sale.
I have been suggested an analog notch filter by Bert, which translates to about -3db at 3k. i'm using about 9db of digital attenuation. I will try it later maybe. At the same time, i also put a low shelf at 100hz. This improved the clarity of the sound and i'm sure it reduced distortion. The poor Hemps with only 1,9mm Xmax* distorts a LOT in the lower regions, which compromises all response. Icing on the cake I boosted from 100 to 200hz to compensate the baffle rolloff.
2) The woofers, a pair of Beyma SM112N have been ordered yesterday and will be shipped tomorrow.
3) The enclosures are MIA due to busy carpenter. I'll see what I can do to burn the woofers in.
4) The XTZ Sub Amp 1 has also been ordered from the very nice italian distributor. Thanks to the help from Rikard of XTZ, the bass boost frequency (7,5db) will be changed to 30hz, replacing a resistance and a cap. This should suit most of my needs.
5) A MSB multichannel volume control will be used as master volume control and my laziness will be preserved thanks to its remote.
[* Yes, yes I have been told that several times, but I replied that i would have tried that, just not passively, which is what i have done]
Back from short holidays and before going soon to a business trip, it's time for an update.
1) The shout has been tamed with digital EQ, of about 6db. I'm not completely satisfied with the sound, but I'm still on paper baffles (grr me wants to kill the carpenter) and yet the overall sound became more pleasant than my Linns, which are going for sale.
I have been suggested an analog notch filter by Bert, which translates to about -3db at 3k. i'm using about 9db of digital attenuation. I will try it later maybe. At the same time, i also put a low shelf at 100hz. This improved the clarity of the sound and i'm sure it reduced distortion. The poor Hemps with only 1,9mm Xmax* distorts a LOT in the lower regions, which compromises all response. Icing on the cake I boosted from 100 to 200hz to compensate the baffle rolloff.
2) The woofers, a pair of Beyma SM112N have been ordered yesterday and will be shipped tomorrow.
3) The enclosures are MIA due to busy carpenter. I'll see what I can do to burn the woofers in.
4) The XTZ Sub Amp 1 has also been ordered from the very nice italian distributor. Thanks to the help from Rikard of XTZ, the bass boost frequency (7,5db) will be changed to 30hz, replacing a resistance and a cap. This should suit most of my needs.
5) A MSB multichannel volume control will be used as master volume control and my laziness will be preserved thanks to its remote.
[* Yes, yes I have been told that several times, but I replied that i would have tried that, just not passively, which is what i have done]
Hemp Acoustics FR8 drivers...
don't sound good until at least 200 hours are on them . In the right baffle they can be very good.No bass impact, but usable bass down to the low 40's (regardless of what others may say, they played at least as low as a powered Polk subwoofer, reputedly good for high 30 Hz bass extension).
don't sound good until at least 200 hours are on them . In the right baffle they can be very good.No bass impact, but usable bass down to the low 40's (regardless of what others may say, they played at least as low as a powered Polk subwoofer, reputedly good for high 30 Hz bass extension).
Re: Hemp Acoustics FR8 drivers...
200 eh? Well, as long as I'm home, I'm still let them play... I think I'm close to 150 now.
Probably by then the 6db cut at 3k will reduce to 3db.
You are right on the bass, it's there, but its very much attenuated and distorted at high SPL. I'm pretty sure i'll highpass them. Not usable with the size of my baffles under 200hz - you probably had them in a JE. Cant make wider in my room. But the Beymas are coming 🙂
Thanks to chime in 🙂
Nanook said:
200 eh? Well, as long as I'm home, I'm still let them play... I think I'm close to 150 now.
Probably by then the 6db cut at 3k will reduce to 3db.
You are right on the bass, it's there, but its very much attenuated and distorted at high SPL. I'm pretty sure i'll highpass them. Not usable with the size of my baffles under 200hz - you probably had them in a JE. Cant make wider in my room. But the Beymas are coming 🙂
Thanks to chime in 🙂
low frequency distortion
hadn't experienced any, but haven't listened to them in a long time.. They do love the juice though. What amp are you driving them with?
hadn't experienced any, but haven't listened to them in a long time.. They do love the juice though. What amp are you driving them with?
Re: low frequency distortion
First Watt F3. And it's not clipping.
I heard LF distortion when bumping the LF with digital EQ (again without clipping, I let it clip and I hear the difference).
I think that either the lower registers of the speakers are not burned in or, more likely, when boosting the LF too much (over 10db) they simply cant take that much with 1.9mm Xmax.
Nanook said:
First Watt F3. And it's not clipping.
I heard LF distortion when bumping the LF with digital EQ (again without clipping, I let it clip and I hear the difference).
I think that either the lower registers of the speakers are not burned in or, more likely, when boosting the LF too much (over 10db) they simply cant take that much with 1.9mm Xmax.
Re: Re: low frequency distortion
remember the efficiency is said to be -3dB than commonly advertised. Adding the felt can create an near aperiodic behaviour, and perhaps reduce efficiency in the bass further.
A 10dB boost in bass can cause distortion, even if the amp is not clipping. I've found that using a full width top plate between the uprights to help create a significant change in bass response on the JE open baffle. I also use a full width upright "stabilizer" (a top plate mounted at the bottom rear of the stabilizers).
Regardless of drivers being used, everyone who has tried a full width top plate and a lower upright stabilizer has reported good results and an improvement in the bass performance of their OB speakers (regardless if "JE type" or others...)
Telstar said:
remember the efficiency is said to be -3dB than commonly advertised. Adding the felt can create an near aperiodic behaviour, and perhaps reduce efficiency in the bass further.
A 10dB boost in bass can cause distortion, even if the amp is not clipping. I've found that using a full width top plate between the uprights to help create a significant change in bass response on the JE open baffle. I also use a full width upright "stabilizer" (a top plate mounted at the bottom rear of the stabilizers).
Regardless of drivers being used, everyone who has tried a full width top plate and a lower upright stabilizer has reported good results and an improvement in the bass performance of their OB speakers (regardless if "JE type" or others...)
Attachments
- Status
- Not open for further replies.