dc offset
Thanks GR and Ygg, here is my exact situation. I have built line array loudspeakers with 12-5" Fountek ribbon tweeters per side and 12 eton - 5" mid bass drivers in a ported encloseure which is 84"x13"w x 17"d heavily braced. The boxes are veneered MDF
at 1-1/4"thick with the tweeters in their own isolated chamber.
The crossover is a 24db per octave 4th order at 2600hz.
the tweeters are rated from 1200hz to 40Khz and the midbass are 50hz to 3500hz. With a line array you must minimize the seperation of the drivers to less than the wavelength of the highest frequency for good dispersion and prevent lobbing.
I want to bi amp using a digital crossover at 96db around 2500hz because the ribbons don't sound that good below this frequency but this pushes the eatons close to where they may start beaming.
I want to use my Aleph 1.2's for the tweeters but they have about 50mv of dc offset. The fountek spec sheet says that you must cap couple this driver for full power testing. Using your idea of a cap and sizing the cap for 120hz to cover any residual noise that may be coming from the PS==I get about 20uf. I have some 40uf PP left over from another project. I will use those and I am hoping with a crossover pt of 2500 hz/96db to the tweeters that they won't be noticed. The eatons will see my pass/tharagard A-75 and they are immune to the 20mv dc that the A75 has. The A75 is fused but the Aleph's are not. I can fuse them.
The fountek's are $100 a piece but I got them on sale from a supplier in Grand Rapids. I also got the Eatons on sale which provided the majority priority in their choice.
What do you think? I will be using the Deqx unit, I couldn't afford NP's but I still may make one. dave
Thanks GR and Ygg, here is my exact situation. I have built line array loudspeakers with 12-5" Fountek ribbon tweeters per side and 12 eton - 5" mid bass drivers in a ported encloseure which is 84"x13"w x 17"d heavily braced. The boxes are veneered MDF
at 1-1/4"thick with the tweeters in their own isolated chamber.
The crossover is a 24db per octave 4th order at 2600hz.
the tweeters are rated from 1200hz to 40Khz and the midbass are 50hz to 3500hz. With a line array you must minimize the seperation of the drivers to less than the wavelength of the highest frequency for good dispersion and prevent lobbing.
I want to bi amp using a digital crossover at 96db around 2500hz because the ribbons don't sound that good below this frequency but this pushes the eatons close to where they may start beaming.
I want to use my Aleph 1.2's for the tweeters but they have about 50mv of dc offset. The fountek spec sheet says that you must cap couple this driver for full power testing. Using your idea of a cap and sizing the cap for 120hz to cover any residual noise that may be coming from the PS==I get about 20uf. I have some 40uf PP left over from another project. I will use those and I am hoping with a crossover pt of 2500 hz/96db to the tweeters that they won't be noticed. The eatons will see my pass/tharagard A-75 and they are immune to the 20mv dc that the A75 has. The A75 is fused but the Aleph's are not. I can fuse them.
The fountek's are $100 a piece but I got them on sale from a supplier in Grand Rapids. I also got the Eatons on sale which provided the majority priority in their choice.
What do you think? I will be using the Deqx unit, I couldn't afford NP's but I still may make one. dave
CROSSOVER advises II
Thanks GRollins. Here is another question
If I use
- a woofer with frequencies - 3'500 hz
- medium with 500 hz - 10 hz
- tweeter 4000hz to 20khz
and I play with the crossover frequency between the woofer and the medium there are large / audible differences
For example if the woofer filter is set from - to 800 hz and the medium from 800 to 5000hz it sounds differently (more boxy, like the sound comes out of the box) that if i set the woofer from - to 1500hz and the medium from 1'500hz to 5khz. Why is that ? Does it mean the same frequency emited by the woofer will sound differently as the same emitted by the medium ?
Thanks GRollins. Here is another question
If I use
- a woofer with frequencies - 3'500 hz
- medium with 500 hz - 10 hz
- tweeter 4000hz to 20khz
and I play with the crossover frequency between the woofer and the medium there are large / audible differences
For example if the woofer filter is set from - to 800 hz and the medium from 800 to 5000hz it sounds differently (more boxy, like the sound comes out of the box) that if i set the woofer from - to 1500hz and the medium from 1'500hz to 5khz. Why is that ? Does it mean the same frequency emited by the woofer will sound differently as the same emitted by the medium ?
Daly21k,
I confess that I'm not clear as to what the question is/was. Just me being short on sleep, probably.
Nicola,
Absolutely, positively, without a doubt the same frequency will sound different through different drivers. This is (part of) why I was making skeptical noises in an earlier post about a woofer that claims to go from 30Hz to something in the mid-thousands of Hertz. It's just not going to happen. Not gracefully, anyway.
Why does it sound different?
Let me count the ways...
Actually, I can't at the moment. I'm way short on time.
The short and dirty version is that:
A) The woofer cone will start beaming.
B) The woofer will start breaking up. Note that this doesn't mean that the driver is dying--it's a term meaning that the cone ceases to behave like a piston. Resonances develop within the cone itself. This technically counts as distortion.
C) You could conceivably overdrive the midrange (exceed X-max) such that it becomes non-linear. This counts as distortion, too.
D) Etc.
With luck there will be a reasonable compromise between the two drivers. When you get down to choices like this--meaning the difference between the same music played with two different crossover points--it's called voicing the speaker.
Grey
I confess that I'm not clear as to what the question is/was. Just me being short on sleep, probably.
Nicola,
Absolutely, positively, without a doubt the same frequency will sound different through different drivers. This is (part of) why I was making skeptical noises in an earlier post about a woofer that claims to go from 30Hz to something in the mid-thousands of Hertz. It's just not going to happen. Not gracefully, anyway.
Why does it sound different?
Let me count the ways...
Actually, I can't at the moment. I'm way short on time.
The short and dirty version is that:
A) The woofer cone will start beaming.
B) The woofer will start breaking up. Note that this doesn't mean that the driver is dying--it's a term meaning that the cone ceases to behave like a piston. Resonances develop within the cone itself. This technically counts as distortion.
C) You could conceivably overdrive the midrange (exceed X-max) such that it becomes non-linear. This counts as distortion, too.
D) Etc.
With luck there will be a reasonable compromise between the two drivers. When you get down to choices like this--meaning the difference between the same music played with two different crossover points--it's called voicing the speaker.
Grey
CROSSOVER advises II
Thanks GRolins,
I'm fairly new in this area; it is almost discovery. The active cross over let me explore thing I would never have had the opportunity to explore with the passive crossover since I'm not able to build one.
It is interesting for me to use a maesurment software like SmartLive with a pink noise, have a fairly +-1bd linear overall speaker response from 50hz to 18khz but a different sound when it comes to play classical instruments (piano is the most apparent) depending on where the bass / mid loudspeaker crossover frequency is set...
Thanks for your time and answer, that helps me a lot
Thanks GRolins,
I'm fairly new in this area; it is almost discovery. The active cross over let me explore thing I would never have had the opportunity to explore with the passive crossover since I'm not able to build one.
It is interesting for me to use a maesurment software like SmartLive with a pink noise, have a fairly +-1bd linear overall speaker response from 50hz to 18khz but a different sound when it comes to play classical instruments (piano is the most apparent) depending on where the bass / mid loudspeaker crossover frequency is set...
Thanks for your time and answer, that helps me a lot
All this just makes me want to ask the question I asked once before and never got a strait answer... And, although I've seen refrence to it, there are countless variations to most things Nelson does... What ever happened to the future project, on the PassDIY site, known as, the High Low Pass??? Has it come to Pass, and I missed it??? 
Sorry, just had to ask
Sorry, just had to ask
GRollins said:
maybe for garden party,but not for living room.....
that's only in case if man choose this path ;
me not-I'm more one way guy ,or one and a half ,or one augmented from both ends.......
equalizing drivers is not exactly my preferred game, just because I have opportunity to choose pretty adequate drivers to my ears (intentionally didn't sayed linear) .
but-when you stash few dozen' kilodollars in active system, choosing to not equalize drivers completely -in most efficient way -is total waste of muuuni.
digital processor with xover functionality is different story.....
shelvings,phases,decays,etc........all stuff included and that anywhere in spectrum -repeatedly
maybe I'm not informed,but seems to me that XVR1 is not equipped with that sorts of gizmos
dc offset affect on drivers
GR, I have been travelling on business so this is late response.
My original question was if 50mv of dc offset from my Aleph 1.2 power amps would hurt my ribbon tweeters which in a bi-amp mode are not protected by the caps in a passive crossover. The active crossover will be in front of the amps. Your suggestion of putting a cap in series with the output of the tweeter amp was a good one. I have using the standard formulas come up with 40uf pp cap. The formulas actually say 20uf but I have these leftover.
Since I am crossing over at 2500hz at 96db per octave, I am hoping that the caps will not affect the sound. Theoretically they should not add any phase shift at these frequencies. I am using 12 tweeters perside so they would only see 4-5mv each anyway but since they are expensive I want to make sure I don't dynamite them. I never see this subject come up in discussions about bi-amping maybe I am the only one in the world that worries about such a thing. Tube amps and Mac amps don't put out any dc offset due to their xformers. Protection circuits on high powered amps usually activate at 500-700mv so I am not sure that they are the answer. thanks for the help, I got the answers that I needed. dave
GR, I have been travelling on business so this is late response.
My original question was if 50mv of dc offset from my Aleph 1.2 power amps would hurt my ribbon tweeters which in a bi-amp mode are not protected by the caps in a passive crossover. The active crossover will be in front of the amps. Your suggestion of putting a cap in series with the output of the tweeter amp was a good one. I have using the standard formulas come up with 40uf pp cap. The formulas actually say 20uf but I have these leftover.
Since I am crossing over at 2500hz at 96db per octave, I am hoping that the caps will not affect the sound. Theoretically they should not add any phase shift at these frequencies. I am using 12 tweeters perside so they would only see 4-5mv each anyway but since they are expensive I want to make sure I don't dynamite them. I never see this subject come up in discussions about bi-amping maybe I am the only one in the world that worries about such a thing. Tube amps and Mac amps don't put out any dc offset due to their xformers. Protection circuits on high powered amps usually activate at 500-700mv so I am not sure that they are the answer. thanks for the help, I got the answers that I needed. dave
I wouldn't worry about a few mV across a tweeter. Zero would be better, obviously, but in the real world a small amount of DC will not hurt. All bets are off if the amp's output fails and dumps the rail through the drivers...
The turn-on/turn-off thump (if any) is more problematical.
I'm not a big fan of digital crossovers. I don't feel that they sound as good as they should. Yes, you can dial-in some pretty extreme parameters and eq the drivers and all that, but you can also push pottery clay through a window screen and have it come out the other side looking somewhat like it did when it went in. It's the 'somewhat' that bothers me. All those little criss-cross cuts kinda take away from the original in my opinion.
What is cool is to take one of those self-eq do-dads and let it do its thing, then use that as a rough template for analog eq. Note that I said rough, not final. It works better in the low bass than higher up, and I wouldn't trust it at all above 200-300Hz.
Grey
The turn-on/turn-off thump (if any) is more problematical.
I'm not a big fan of digital crossovers. I don't feel that they sound as good as they should. Yes, you can dial-in some pretty extreme parameters and eq the drivers and all that, but you can also push pottery clay through a window screen and have it come out the other side looking somewhat like it did when it went in. It's the 'somewhat' that bothers me. All those little criss-cross cuts kinda take away from the original in my opinion.
What is cool is to take one of those self-eq do-dads and let it do its thing, then use that as a rough template for analog eq. Note that I said rough, not final. It works better in the low bass than higher up, and I wouldn't trust it at all above 200-300Hz.
Grey
crossover analog vs digital
GR--I intend to try all three types, passive, active digital, active analog. In reality when you have to tame a peaky driver with notch filters and attenuation then an active analog system can be difficult unless you have some form of an equalizer or use passive elements. I like the idea of the analog filter since you can still play vinyl records without digitizing anything. My reason for the digital crossover was the room correction since to modify my basement listening room would cost a whole lot more than the room correction device. I bought the book Speaker building 201 and its crossover section is excellent and gives you some ideas where to get free software. I bought it from www.meniscusaudio.com I have considered the www.borbelyaudio.com active crossover but I am waiting for NP's penutltimate crossover to be revealed. dave
GR--I intend to try all three types, passive, active digital, active analog. In reality when you have to tame a peaky driver with notch filters and attenuation then an active analog system can be difficult unless you have some form of an equalizer or use passive elements. I like the idea of the analog filter since you can still play vinyl records without digitizing anything. My reason for the digital crossover was the room correction since to modify my basement listening room would cost a whole lot more than the room correction device. I bought the book Speaker building 201 and its crossover section is excellent and gives you some ideas where to get free software. I bought it from www.meniscusaudio.com I have considered the www.borbelyaudio.com active crossover but I am waiting for NP's penutltimate crossover to be revealed. dave
Re: Re: CROSSOVER advises II
Could you be more precise ? Do you mean some passive element should be inserted between amps and drivers to compensate some of the driver behaviour ? That is what you mean with EQ ? Is my understanding correct ? Recently I had a discussion with a friend who uses a 4 way amps with digital filter (a japanese one something like accuphase) and what he call a digital voicing to adapt the loudspeaker response to the room (if I get it correctly) He explained only the digital filter was able to handle phase, time delay and other compensation compared to an analog system. I'm not quite sure I got everything - I have to visit him - but from what I understood this is way way way above my financial capacities anyway. Is it something like that you mean with EQ ?
Regards
Zen Mod said:
just for record
don't be sure that you are free of any passive element between amps and drivers.....
you can't find many drivers that don't need any sort of EQ or ,at least ,Boucherot comp
Could you be more precise ? Do you mean some passive element should be inserted between amps and drivers to compensate some of the driver behaviour ? That is what you mean with EQ ? Is my understanding correct ? Recently I had a discussion with a friend who uses a 4 way amps with digital filter (a japanese one something like accuphase) and what he call a digital voicing to adapt the loudspeaker response to the room (if I get it correctly) He explained only the digital filter was able to handle phase, time delay and other compensation compared to an analog system. I'm not quite sure I got everything - I have to visit him - but from what I understood this is way way way above my financial capacities anyway. Is it something like that you mean with EQ ?
Regards
Re: Re: Re: CROSSOVER advises II
what I meant is
impedance corrector near driver's Fs (known as Boucherot or smthng)
dip corrector filter
peak corrector filter
shelving filters (for up and down)
just look for xover schematics of good spk boxes,and you'll see that many of them use linearizing chains in front of drivers
I know for some cases where LS3/5 were used in active ,but that doesn't mean that you cut each passive element from xover.......
name of the game is to linearize each driver even before putting them together in combination;
Nicola said:
what I meant is
impedance corrector near driver's Fs (known as Boucherot or smthng)
dip corrector filter
peak corrector filter
shelving filters (for up and down)
just look for xover schematics of good spk boxes,and you'll see that many of them use linearizing chains in front of drivers
I know for some cases where LS3/5 were used in active ,but that doesn't mean that you cut each passive element from xover.......
name of the game is to linearize each driver even before putting them together in combination;
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.