Allright:
I think I know what I am going to do-
Fountek jp2 (seems almost as good as the AC for 1/4 the price)
------
Audax 8" Aerogel sheielded AM210Z2
------
Sonicraft Sc12Nrt (again almost as good as the aura for 1/3 the price)
These woofers both seem to have a nice balance between stiffness of cone and breakup modes.
What I want to do is build a passive 24dB/oct HP filter for the tweeter to protect it- My thought is to put this at the lowest possible freq- about 2k and then use the active filter on a 6dB or so slope to smoothly XO to the woofer- I will start by with no LP filter on the woofer at all- and maybe use a 6dB active (or some notches) to control the very minor break up modes that do exist.
I heard a system with the 6.5' audax aerogel running with no crossover at all- The mids were startaling and the imaging was fantastic- the only problem with this system is that there was group delay on the other drivers (wich were crossed over) so there was some comb filtering- I will be able to adjust this in the active XO so this should be no prob here.
So can I use this calculator to figure my passive HP filter values?
http://www.apicsllc.com/apics/Misc/filter2.html
Just wanted to make sure that in my setup the lack of a woofer will not effect anything. It looks like it won't because when I change imedance values for the woofer the L and C values remain the same for the tweeter.
So then I will HP filter the audax at around 100hz or so with the softest slopes that will allow it to attain higher max spl's without compression and to decrease the slow rise in distortion from the Sc12nrt woofer above 100hz.
The boxes will be 1.5" thick MDF walls and baffles with extinsive shelf bracing and rounded edges all covered in some sort of acoustic damping matierel. The Audax will se about a 15L chamber and the Scnrt about 50L both slightly smaller than .707 Qtc to yeild a <1dB boost at the low end (for a little more punch)
I want to use sealed cabs throughout- I know this will not lead to the highest possible SPL's but the system should be plenty loud if I can limit over excursion of the 8" audax. Plus I am not looking for true subwoofer bass in this system and I like the tightness of sealed cabs.
I am thinking about a hafler p3000 for the audax's and a hafler p1500 for the tweets, and then my bryston 4b st for the woofers. This way at least the most critical frequencies (100-20000hz) will have very similer amp's.
One question- I do not want to spend more than $100 on the pair of passive HP filters- What caps and chokes would be recomended for this application?? i know this is subjective again but I would appriciate any opinion's- thanks again,
Ryan
I think I know what I am going to do-
Fountek jp2 (seems almost as good as the AC for 1/4 the price)
------
Audax 8" Aerogel sheielded AM210Z2
------
Sonicraft Sc12Nrt (again almost as good as the aura for 1/3 the price)
These woofers both seem to have a nice balance between stiffness of cone and breakup modes.
What I want to do is build a passive 24dB/oct HP filter for the tweeter to protect it- My thought is to put this at the lowest possible freq- about 2k and then use the active filter on a 6dB or so slope to smoothly XO to the woofer- I will start by with no LP filter on the woofer at all- and maybe use a 6dB active (or some notches) to control the very minor break up modes that do exist.
I heard a system with the 6.5' audax aerogel running with no crossover at all- The mids were startaling and the imaging was fantastic- the only problem with this system is that there was group delay on the other drivers (wich were crossed over) so there was some comb filtering- I will be able to adjust this in the active XO so this should be no prob here.
So can I use this calculator to figure my passive HP filter values?
http://www.apicsllc.com/apics/Misc/filter2.html
Just wanted to make sure that in my setup the lack of a woofer will not effect anything. It looks like it won't because when I change imedance values for the woofer the L and C values remain the same for the tweeter.
So then I will HP filter the audax at around 100hz or so with the softest slopes that will allow it to attain higher max spl's without compression and to decrease the slow rise in distortion from the Sc12nrt woofer above 100hz.
The boxes will be 1.5" thick MDF walls and baffles with extinsive shelf bracing and rounded edges all covered in some sort of acoustic damping matierel. The Audax will se about a 15L chamber and the Scnrt about 50L both slightly smaller than .707 Qtc to yeild a <1dB boost at the low end (for a little more punch)
I want to use sealed cabs throughout- I know this will not lead to the highest possible SPL's but the system should be plenty loud if I can limit over excursion of the 8" audax. Plus I am not looking for true subwoofer bass in this system and I like the tightness of sealed cabs.
I am thinking about a hafler p3000 for the audax's and a hafler p1500 for the tweets, and then my bryston 4b st for the woofers. This way at least the most critical frequencies (100-20000hz) will have very similer amp's.
One question- I do not want to spend more than $100 on the pair of passive HP filters- What caps and chokes would be recomended for this application?? i know this is subjective again but I would appriciate any opinion's- thanks again,
Ryan
Ok so that link above won't work for a passive 24dB because it is taking the woofer into effect-
This one looks good because it allows me to enter 0 ohm for the woofer (wich as there is no woofer it should have no imedance or resistance at any freq)-
http://www.carstereo.com/help/Articles.cfm?id=12
Also i noticed that it is difficult - if not impossible- to get the exact values for these components- how close do they need to be? Is it adviseable to use a bunch of low value components in series?
Also Ceramicman- I presume your name implies that you are a fan of the Accuton drivers? I am also thinking about the C²23-6 for the tweeter-
Do you think If I went with this driver that I would still require the passive HP filter? It would still be actively x'd over I just wonder if an amp transient might damage it (the driver) thanks again,
Ryan
This one looks good because it allows me to enter 0 ohm for the woofer (wich as there is no woofer it should have no imedance or resistance at any freq)-
http://www.carstereo.com/help/Articles.cfm?id=12
Also i noticed that it is difficult - if not impossible- to get the exact values for these components- how close do they need to be? Is it adviseable to use a bunch of low value components in series?
Also Ceramicman- I presume your name implies that you are a fan of the Accuton drivers? I am also thinking about the C²23-6 for the tweeter-
Do you think If I went with this driver that I would still require the passive HP filter? It would still be actively x'd over I just wonder if an amp transient might damage it (the driver) thanks again,
Ryan
Ok now youve settled on the drivers (which look like nice choices).
Unfortunatley xovers arent that easy, ideally you need some modelling software. Without that you can get pretty good results by trial and error.
When trying to match drivers you have to take the acoustical roll-off & add your xover electrical roll-off to sum correctly.
First off look at the woofers low-pass rollof, in this case (for the Audax am210z2) its about 1st order from 2k.
The tweeter (jp2) is about 3rd order at 2k, so it would only require a first order (single cap) to match the woofer.
However I can not recommend a single cap on the tweeter as it modfifys the response. At minimum I would go 2nd order high pass, which would mean increasing the woofers slope to 4th order or pick a higher frequency (which you cant really do as you are at the woofers upper limit).
Anyway with a bit of tweaking the combination may work.
Unfortunatley xovers arent that easy, ideally you need some modelling software. Without that you can get pretty good results by trial and error.
When trying to match drivers you have to take the acoustical roll-off & add your xover electrical roll-off to sum correctly.
First off look at the woofers low-pass rollof, in this case (for the Audax am210z2) its about 1st order from 2k.
The tweeter (jp2) is about 3rd order at 2k, so it would only require a first order (single cap) to match the woofer.
However I can not recommend a single cap on the tweeter as it modfifys the response. At minimum I would go 2nd order high pass, which would mean increasing the woofers slope to 4th order or pick a higher frequency (which you cant really do as you are at the woofers upper limit).
Anyway with a bit of tweaking the combination may work.
IMO, the passive crossover should be kept to a minimum, eg: just a 1st or 2nd order protective HP for the tweeter, and do most of the filtering before the power-amp. A second order passive filter should be more than enough to protect a ribbon tweeter against things like ground hum from the amp. Straight DC is not a problem for a ribbon tweeter due to its transformer.
I've got a passive-active system with active filters for the sub, and passive for a pair of 2-way monitors. I recently upgraded the caps in the passive section from polyester to polypropylene, and to my own surprise it made a big improvement. It really toned down the treble, and removed layers of distortion. I can foresee another similar improvement when I finally get round to upgrading to fully active.
The problem with passive filters is not just the size of the components, but also the magnitude of the signal going through them. One big advantage of active filtering is the tiny ripple across the capacitors that are used. A passive cap may have 10V or even over 20V peak of high-frequency ripple, but even with very low ESR and ESL (equivalent series reactance) it may ruin the sound due to hysteretic effects from dielectric absorption and dielectric saturation. OTOH, an active Sallen-Key filter produces less than a few mV of ripple across its caps, so even lesser quality caps with relatively bad characteristics should have better results.
This is a good active filter design tool, albeit a rather slow Java app. (Analog Devices, under "interactive design tools".) Here you can eg: design a 3rd or 5th order active filter, and get a good C value for the passive RC section by substituting the approximate nominal resistance of the tweeter. You then end up with a 2nd or 4th order active filter with a passive filter that sums to a 3rd or 5th order one.
Re: Audax AM210Z2, have you checked on their website www.audax.fr to see how they compare with others? IMO Seas Excels are better, although there aren't that many options with magnetic shielding.
Re: Accuton C23, I hadn't thought about those mainly because of your high power requirements. But now that I think about it, I've actually found my ones to be quite grunty.
CM
I've got a passive-active system with active filters for the sub, and passive for a pair of 2-way monitors. I recently upgraded the caps in the passive section from polyester to polypropylene, and to my own surprise it made a big improvement. It really toned down the treble, and removed layers of distortion. I can foresee another similar improvement when I finally get round to upgrading to fully active.
The problem with passive filters is not just the size of the components, but also the magnitude of the signal going through them. One big advantage of active filtering is the tiny ripple across the capacitors that are used. A passive cap may have 10V or even over 20V peak of high-frequency ripple, but even with very low ESR and ESL (equivalent series reactance) it may ruin the sound due to hysteretic effects from dielectric absorption and dielectric saturation. OTOH, an active Sallen-Key filter produces less than a few mV of ripple across its caps, so even lesser quality caps with relatively bad characteristics should have better results.
This is a good active filter design tool, albeit a rather slow Java app. (Analog Devices, under "interactive design tools".) Here you can eg: design a 3rd or 5th order active filter, and get a good C value for the passive RC section by substituting the approximate nominal resistance of the tweeter. You then end up with a 2nd or 4th order active filter with a passive filter that sums to a 3rd or 5th order one.
Re: Audax AM210Z2, have you checked on their website www.audax.fr to see how they compare with others? IMO Seas Excels are better, although there aren't that many options with magnetic shielding.
Re: Accuton C23, I hadn't thought about those mainly because of your high power requirements. But now that I think about it, I've actually found my ones to be quite grunty.
CM
Thanks again CM-
How low do you think I can push the XO point for the passive filter? Madisound suggusts XO freq at 2k but they dont say what slope that is. . . My thought here is that I want to use the softest slopes possible so that the transition from a cardioid dispertion pattern to the line source is very smooth.
One question- the audax 8" is about 6-8dB down at 4k- So if what I am wondering is if I will need to (activly) cross over the tweeter so that it is 6dB down at the same point? Or will they need to combine at 3dB down in order to sum to a flat response?
AP:
This system is all active except the tweeter will need some type of HPF to avoid damage from say an amp's power on transient, etc.
CM:
I am going to use that behringer XO- So I only need to calculate the values for the passive HP filter-
Also The audax looks to be really nice on their website- It has a very tight waterfall, and a very smooth HF rolloff- If I were going for a dome tweet I would think a little more about the seas drivers- but I feel like I need to choose a mid woofer with a smooth rooloff so that I don't have a hard time mating with the different propogation patterns of the ribbon. Do you know where you can see the waterfall plots from the seas drivers?
Thanks again guys,
Ryan
How low do you think I can push the XO point for the passive filter? Madisound suggusts XO freq at 2k but they dont say what slope that is. . . My thought here is that I want to use the softest slopes possible so that the transition from a cardioid dispertion pattern to the line source is very smooth.
One question- the audax 8" is about 6-8dB down at 4k- So if what I am wondering is if I will need to (activly) cross over the tweeter so that it is 6dB down at the same point? Or will they need to combine at 3dB down in order to sum to a flat response?
AP:
This system is all active except the tweeter will need some type of HPF to avoid damage from say an amp's power on transient, etc.
CM:
I am going to use that behringer XO- So I only need to calculate the values for the passive HP filter-
Also The audax looks to be really nice on their website- It has a very tight waterfall, and a very smooth HF rolloff- If I were going for a dome tweet I would think a little more about the seas drivers- but I feel like I need to choose a mid woofer with a smooth rooloff so that I don't have a hard time mating with the different propogation patterns of the ribbon. Do you know where you can see the waterfall plots from the seas drivers?
Thanks again guys,
Ryan
RyanC said:
Dunno, the nearest I can point you to is www.seas.no for the frequency response graphs. I always find it a pain to compare plots between different companies. For example the Audax AM210Z2 looks like it's measured on a large wide baffle (half-field), while the Seas datasheet says that the W22EX001 is measured in free-field in a 20L box. This automatically gives the Audax speakers a 6dB sensitivity advantage at low frequencies. The Audax graph uses 0.1-octave smoothing, while the Seas one appears not to.
With a 100Hz crossover, a 6.5" speaker may be able to do the job. Depending on what woofer you end up using, you may be able to make the crossover higher, eg: 125Hz or 150Hz. Then the power requirements for the midrange, are definitely only due to heat rather than Xmax (for 6.5").
Re: ribbon directivity. I don't think it should be much of a problem at low frequencies like 1 to 3kHz. It becomes more and more directional at higher frequencies.
CM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.