Hi - asking here because I'm hoping someone will have a better memory (or Google-foo) than I do.
I'm interested in a technique for building a 3-way active system with two amp channels where on channel drives the mid and the other drives the bass and tweeter.
I had in mind (and this may be from an old SpeakerBuilder article, but I don't have the old magazines any more) that this was patented.
Has anyone tried this - specifically with an analogue series crossover to protect the tweeter and divide the A and C signals?
Any links to prior art on this (including any such patent etc) would be great. I goofed when I wired my home - just put two lots of coax in to the speaker locations and didn't run a cat5 at the same time.
Thanks
James
I'm interested in a technique for building a 3-way active system with two amp channels where on channel drives the mid and the other drives the bass and tweeter.
I had in mind (and this may be from an old SpeakerBuilder article, but I don't have the old magazines any more) that this was patented.
Has anyone tried this - specifically with an analogue series crossover to protect the tweeter and divide the A and C signals?
Any links to prior art on this (including any such patent etc) would be great. I goofed when I wired my home - just put two lots of coax in to the speaker locations and didn't run a cat5 at the same time.
Thanks
James
Sorry for my ignorance but glancing at the patent it appears to successfully combine all the disadvantages of passive xovers with all the disadvantages of active ones.
What am I missing?
What am I missing?
Well, if you only have (or want to use) two channels of active amplification, its handy. And the high and low pass filters in the split aren't there to perform the normal crossover function - they are there to prevent crosstalk between the two ranges, so the passive low pass for the woofer can be well above its cross to the midrange, and the high pass for the tweeter can be well below its cross to the midrange.
Stopping the high freq get to the woofer stops it exciting cone breakup modes (though, I suspect with the relative sensitivities of the drivers I'm expecting to use its not such a big deal) - but I definitely want to prevent the bass signal hammering my (ribbon) tweeter. At least, that's my understanding.
Stopping the high freq get to the woofer stops it exciting cone breakup modes (though, I suspect with the relative sensitivities of the drivers I'm expecting to use its not such a big deal) - but I definitely want to prevent the bass signal hammering my (ribbon) tweeter. At least, that's my understanding.
I don't really see the point in "saving" one channel of amplification. If you already have two channels of amplification present, you already have a power supply for that. Increasing the PS capacity by 50% and adding on another amplifier probably costs less money than creating the passive filters after the W+T amp as shown in the patent. WHen this was written, amplification was likely much more expensive than today. Now there are many inexpensive options in many amplification classes (e.g. AB, T, D).
Since there is (as shown in the patent) already an active 3-way crossover before the amps, putting passive elements after the amps sort of removes one of the main advantages of active circuitry - the direct connection between amp and driver.
-Charlie
Since there is (as shown in the patent) already an active 3-way crossover before the amps, putting passive elements after the amps sort of removes one of the main advantages of active circuitry - the direct connection between amp and driver.
-Charlie
It is using the Patent Office to keep people from doing stupid things. Anyone (with a clue) is going to look at the diagram in the patent and think "I can avoid that simply by dropping the passive stuff and adding a third amp".
And they'll get a better speaker for it . . .
Sure, but in my case:
- I have a pair of stereo Cyrus power amps that have a form factor reasonable to site under the speakers
- I have 2 off coax from where I'd like to put the controller to each speaker site
- I'm happy to build the 3-way active crossover part in a PC
So it means I can have fun playing with digital crossovers with the kit and wiring I have in place.
So no, I don't think its a stupid thing. Sorry dewardh - sure you want to accuse people of having no clue just like that?
Now, if I could use the coax to put 4off S/PDIFs down the lines then maybe I'd get some cheap DACs and a couple of T-amps to run the tweeters, but running multiple digital consumer signals seems to be bizarrely tricky and I'm not sure I want to pay for an AES16 and level/protocol converters yet.
Right now I'm keen to have fun with digital crossovers and speaker stuff, not building amps and opamp active stuff. Though miniDAC is certainly tempting.
- I have a pair of stereo Cyrus power amps that have a form factor reasonable to site under the speakers
- I have 2 off coax from where I'd like to put the controller to each speaker site
- I'm happy to build the 3-way active crossover part in a PC
So it means I can have fun playing with digital crossovers with the kit and wiring I have in place.
So no, I don't think its a stupid thing. Sorry dewardh - sure you want to accuse people of having no clue just like that?
Now, if I could use the coax to put 4off S/PDIFs down the lines then maybe I'd get some cheap DACs and a couple of T-amps to run the tweeters, but running multiple digital consumer signals seems to be bizarrely tricky and I'm not sure I want to pay for an AES16 and level/protocol converters yet.
Right now I'm keen to have fun with digital crossovers and speaker stuff, not building amps and opamp active stuff. Though miniDAC is certainly tempting.
I don't think I'd want to drive my bass drivers with the a 12W SET amp, nor would I want to drive my ribbon tweeter with a 200W SS PA amp.
horses for courses 😉
horses for courses 😉
If one were restricted to two amps I would always run the bass active and split mids and tops passively.
Would seem the most sensible way to me.
Would seem the most sensible way to me.
That was my first thought - but then I'd be having to build a passive crossover. And I'm a code monkey, so that's no fun.
While there are passive components in this unusual split, my recollection from the original article (which wasn't the patent page, but I might get further using the names on it I guess) and a rough sketch suggests that it might be a lot easier to build a splitter that's not actually in the crossover bands. I'm going to work through some numbers assuming a series crossover.
While there are passive components in this unusual split, my recollection from the original article (which wasn't the patent page, but I might get further using the names on it I guess) and a rough sketch suggests that it might be a lot easier to build a splitter that's not actually in the crossover bands. I'm going to work through some numbers assuming a series crossover.
Well, all this '+1' business is just dull. Its potentially 'the obvious thing to do', but I don't see the point of saying '+1' unless there is some justification. Personally, I find it attractive to do 3 off digital filters rather than do the crossover optimisation thing - and for potential mids like (say) the FR125S there's quite a big efficiency disparity that will just leave the tweeter padded down a long way.
If you think there are overriding technical reasons then pray do let on. (Yes, I know the component values are easier: but also their accuracy is much more critical if they are actually doing response shaping rather than blocking unwanted noise).
I'm not an experienced crossover engineer, and I don't have a parts bin to play with, so a coarse approach out of the critical regions seems to give me a lot more freedom to play with biquads and try things I couldn't try with passive crossover. Not to mention delays, and trying an FFT approach too.
This seems sensible to me: what's the objection? You think it won't work, or just playing safe? Where's the fun in that?
If you think there are overriding technical reasons then pray do let on. (Yes, I know the component values are easier: but also their accuracy is much more critical if they are actually doing response shaping rather than blocking unwanted noise).
I'm not an experienced crossover engineer, and I don't have a parts bin to play with, so a coarse approach out of the critical regions seems to give me a lot more freedom to play with biquads and try things I couldn't try with passive crossover. Not to mention delays, and trying an FFT approach too.
This seems sensible to me: what's the objection? You think it won't work, or just playing safe? Where's the fun in that?
It's a +1 because the topology as outlined in that patent is something I doubt anybody has ever built.
It uses more stages in the active section, an additional bandpass plus a mixer.
The passive section includes a low pass for the woofer that requires far bigger component values, a wacking great coil for a start than a 2 way passive for the mid/top would require so it will cost more and with the coil in series with the woofer you're losing one of the advantages an active system has when driving a woofer which is to direct couple to it without loses caused by DCR.
It uses more stages in the active section, an additional bandpass plus a mixer.
The passive section includes a low pass for the woofer that requires far bigger component values, a wacking great coil for a start than a 2 way passive for the mid/top would require so it will cost more and with the coil in series with the woofer you're losing one of the advantages an active system has when driving a woofer which is to direct couple to it without loses caused by DCR.
There's nothing wrong with the digital active crossover approach . . . what's ****** is to put passive filters on the woofers and tweeters instead of simply using a third amp and making the whole thing fully active.
I don't like that topology either.
But I am also convinced that it was developed for an aplication where it had its advantages (and if it was only to save some cents in a mass-produced product probably). Large companies like this one don't spend lots of money for patents that don't make sense to them IMHO.
Regards
Charles
But I am also convinced that it was developed for an aplication where it had its advantages (and if it was only to save some cents in a mass-produced product probably). Large companies like this one don't spend lots of money for patents that don't make sense to them IMHO.
Regards
Charles
Woofers on the whole tend to be even less efficient than mids, so the tweeter would have to be padded down even more wasting power a long the way. More power is wasted at the necessary series inductor and before you know it your bass amp clips thus frying the tweeters.
While when you run the bass active you can use a cheap brute of an amp for that and get the full advantage the amplifiers control over the connected driver where it matters most, if the bass amp clips it'll sound bad but it won't fry the tweeter.
For the mids and tops you can use a more refined amp that still has sufficient headroom.
All this of course if you really are restricted to two amps. 3way active would be the easiest and best solution.
While when you run the bass active you can use a cheap brute of an amp for that and get the full advantage the amplifiers control over the connected driver where it matters most, if the bass amp clips it'll sound bad but it won't fry the tweeter.
For the mids and tops you can use a more refined amp that still has sufficient headroom.
All this of course if you really are restricted to two amps. 3way active would be the easiest and best solution.
- Status
- Not open for further replies.