tvrgeek, amplifiers have quite a major impact , audio connoisseurs have selected passive speakers for many awards. Try this soon. http://www.diyaudio.com/forums/solid-state/212521-amplifier-tasting-group.html
OnAudio,
There is no relationship between Le (inherent to driver) and the amount of BSC required (dependant on baffle size).
What I said, it is never easy, and rules of thumb, well, I have learned by now where you might as well stick it (the thumb, not the rules). Like Pano said many posts ago and I am not going to look it up, so in my own words: it all depends. And the only way to find out is to measure raw drivers, model, calculate, build, measure, think, refine model, calculate, modify and measure again, etc.
vac
There is no relationship between Le (inherent to driver) and the amount of BSC required (dependant on baffle size).
What I said, it is never easy, and rules of thumb, well, I have learned by now where you might as well stick it (the thumb, not the rules). Like Pano said many posts ago and I am not going to look it up, so in my own words: it all depends. And the only way to find out is to measure raw drivers, model, calculate, build, measure, think, refine model, calculate, modify and measure again, etc.
vac
Passive crossovers allow us to have plug and play and they also condition the sound coming from the amplifier.
For electronic crossovers, amplifier needs to be selected carefully to sound right. It is for this reason others have used a passive crossover to feed the amplifier.
For electronic crossovers, amplifier needs to be selected carefully to sound right. It is for this reason others have used a passive crossover to feed the amplifier.
haha! Ok maybe +2 then, OT but my first childhood injury involved a 70s racing Tuscan that was just lying around in the garden...
I disagree.
I frequently swap amps in my active set up (between MC2 Audio and QUAD for example; one's a rather good class AB bipolar, the other a current dumper) and the impact is minimal.
IME if an amp is ok to drive passive speakers it will work fine as part of an active set up.
vac: The all pass (time aligning) network was a pure marketing excercise, mostly for the North American market as time-alignment was THE buzzword back then. Even Tannoy engineers have admitted as much and that the all pass actually degenerates the sound quality (not surprised as it places more coils in the signal path).
The timing difference between tweeter and woofer is about 6 microsec, smaller than audible to human perception.
I simply use a parametric eq to recreate the notch filter and the 6dB/oct boost most constant directivity horns require for a flat output.
What does have an audible impact on the sound is the phase between pass bands at crossover. Luckily my xovers allow for continuous adjustment through 360deg and the best sounding settings did happen to agree with what I calculated (I first set them by ear and then did the calculations only to find that by ear I set everything where it should be).
Charles, didn't know the background of why Tannoy was doing this, but it makes sense. Urei had this big thing about being able to reproduce (sort of) square waves. First Tannoy I converted to active was the Little Red Monitor in the middle of the eighties, and it really was an improvement even then with LF356.
OnAudio, passive filters are a much more difficult load for amplifiers to drive. Further, conditioning sound is not something I would like my amps to do; that's where effect boxes come in (DSP, analogue filtering). Good amplifiers don't add their own sound.
vac
OnAudio, passive filters are a much more difficult load for amplifiers to drive. Further, conditioning sound is not something I would like my amps to do; that's where effect boxes come in (DSP, analogue filtering). Good amplifiers don't add their own sound.
vac
I personally have heard very good active setups as well as horrible ones. The same for passives. Charles I couldn't fail to notice your selection of amplifiers 😉 Also check post #4 .
Vac as an active supporter what is your 'selection' of amps 😉 .
Vac as an active supporter what is your 'selection' of amps 😉 .
You know, it's just that I hear this dragged out all too often. I could just as well say that active gives good results to those who don't know what the worthwhile goals are.
Here I've taken a plain 2nd order passive electrical with tweeter L-pad and woofer rise compensation and added a single RLC conjugate (yellow), vs the woofer as the amp sees it on its own.
Attachments
OnAudio
McIntosh 2205, Quad II, 2 Luxman A/AB (high end kit only sold in Japan at the time), some Chinese tube amps, assortment of class D amps, in between, many came and went.
All good amplifiers sound pretty much the same i.m.o. For example, the Luxman can be switched between 2*150 W AB or 2*25 W class A. Everyone I tried, including myself, could not distinguish between the two modes in blind tests. Did a lot of comparisons. Usually when there are audible differences, it comes down to the interaction between amp and speaker.
Big exception was a (Chinese) SET with directly heated cathodes, which sounded very pleasing, but measured like sht. Enough heat to grill a chicken on. A couple of years ago I have imported on a very small scale a very wide variety of tube amps from China, just because I wanted to try them out before selling them on. Some of them where actually quite good, but no keepers.
vac
vac
I am usually listening to prototypes I am working on, and for the last year or so that is bi-amped with 41Hz Tripath.
McIntosh 2205, Quad II, 2 Luxman A/AB (high end kit only sold in Japan at the time), some Chinese tube amps, assortment of class D amps, in between, many came and went.
All good amplifiers sound pretty much the same i.m.o. For example, the Luxman can be switched between 2*150 W AB or 2*25 W class A. Everyone I tried, including myself, could not distinguish between the two modes in blind tests. Did a lot of comparisons. Usually when there are audible differences, it comes down to the interaction between amp and speaker.
Big exception was a (Chinese) SET with directly heated cathodes, which sounded very pleasing, but measured like sht. Enough heat to grill a chicken on. A couple of years ago I have imported on a very small scale a very wide variety of tube amps from China, just because I wanted to try them out before selling them on. Some of them where actually quite good, but no keepers.
vac
vac
I am usually listening to prototypes I am working on, and for the last year or so that is bi-amped with 41Hz Tripath.
+10
There are many negatives when adding multiple amplifiers in parallel feeding from a single source, some may or may not be affected by the tonal balance changes..
Never heard a Tannoy or urei in a studio setting that was worth mentioning, well apart from party levels .... 🙂
Domain Game
Your program source is digital whether you like it or not.
Do your signal processing (DSP) there as well to channelize and tailor the drive signals.
Here you have additional degrees of design freedom not available to you in the analog domain.
Note that driver offset can be easily accommodated here; so, there is no longer an excuse for ugly cabinetry.
Focus your efforts on tweaking the power amplifiers to match the requirements of the drivers you are using.
In this setting, use of a DC blocking, series capacitor in midrange and tweeter drive circuits remains a requirement.
Regards,
WHG
Your program source is digital whether you like it or not.
Do your signal processing (DSP) there as well to channelize and tailor the drive signals.
Here you have additional degrees of design freedom not available to you in the analog domain.
Note that driver offset can be easily accommodated here; so, there is no longer an excuse for ugly cabinetry.
Focus your efforts on tweaking the power amplifiers to match the requirements of the drivers you are using.
In this setting, use of a DC blocking, series capacitor in midrange and tweeter drive circuits remains a requirement.
Regards,
WHG
Last edited:
Yeah, I'm not buying the "passive networks are hard to drive" bit either.
I've read a lot of research on the speaker as a load and there are only a few things that matter:
1) The average impedance. The lower the average impedance level the higher the current requirement and the higher the general heating effect. Remember that most of the long term average power of music is in the midrange so that is where the impedance matters most of heating effects.
2) High phase angle combined with low impedance. If the real part drops low at the same time that the phase angle is high then this leads to high peak current and may trigger current protection circuits (in solid state amps). Ham-fisted passive networks with crossover points near woofer resonance (3 and 4 way) may lead to this but it is easy to design around if you know what to look for.
3) Low impedance at high frequencies. This tends to push amplifier instability issues. This was always an electrostatic speaker potential problem. Ironically, adding Zobel networks to tweeters is going in the wrong direction since amplifiers to to be very happy driving the rising impedance of an inductive load. Dropping it down to resistive doesn't do the amp any favors (although most amps are okay with it).
A passive network (compared to an individual driver) will have a more complex impedance curve (more "bumps"), but if you stay away from the above conditions the amplifier will be just as happy driving it.
David S.
I've read a lot of research on the speaker as a load and there are only a few things that matter:
1) The average impedance. The lower the average impedance level the higher the current requirement and the higher the general heating effect. Remember that most of the long term average power of music is in the midrange so that is where the impedance matters most of heating effects.
2) High phase angle combined with low impedance. If the real part drops low at the same time that the phase angle is high then this leads to high peak current and may trigger current protection circuits (in solid state amps). Ham-fisted passive networks with crossover points near woofer resonance (3 and 4 way) may lead to this but it is easy to design around if you know what to look for.
3) Low impedance at high frequencies. This tends to push amplifier instability issues. This was always an electrostatic speaker potential problem. Ironically, adding Zobel networks to tweeters is going in the wrong direction since amplifiers to to be very happy driving the rising impedance of an inductive load. Dropping it down to resistive doesn't do the amp any favors (although most amps are okay with it).
A passive network (compared to an individual driver) will have a more complex impedance curve (more "bumps"), but if you stay away from the above conditions the amplifier will be just as happy driving it.
David S.
On op-amps:
I use two BSS FDS360 switched to mono as crossovers. Bought both s/h and could never get the stereo balance right between them, even when I measured and adjusted to be as near perfectly level as possible.
Then I opened them up and found one stuffed with TL072s and the other with LF353s.
Got the second biggest improvement after going active by swapping all those with OPA2134s.
The drivers I use are the same as the ones in LittleReds but mine came out of DC4000s and had previously been re-coiled by Roger at Lockwood which increased power handling to somewhere around 300Wrms.
Still got another minty pair of LittleReds with all the factory accessories including Allen keys, tech manual and leads to bypass the original passive xover.
Allen,
Also @ David
My statement was a bit blanket and based on commercial stuff I measured. What I mean is speakers that dip below 2 Ohm and rise to 15 somewhere. But of course it can be done right to as you showed. Some raw drivers are not an easy load either, and the specimen you have might be one of them; eyeballing the Le it would be around 1,5 mH and the drivers I typically use are about a third of that.
This is a measurement of raw driver (impedance cyan, elect. phase red) and the same driver in a 2nd order 2way with a Zobel (impedance orange, phase blue). So, not that much difference when done right with drivers that fall in place.
Actually, my concern with passive filters is more in the other direction; the fact that around xover your effective damping factor gets really low. I am a control freak and I want my amplifiers to control my drivers.
@ a.wayne
Tannoy is not that bad, but have been long surpassed by better speakers. Concentric is a billiant idea that has always failed. Mine are just taking up space with their surrounds rotting away at the moment. Wish I were in Japan, the audiophile scene there loves them and pays top Yen.
vac
I can testify to that. Sounds real "zingy".
i dont see the amp issue with passives, if your f1 Z peak, or twin peaks, of woofer resonance is usually fairly large, and if this/these are larger that any other peaking, then imho, then which is worse? Really? I can empathise with a general 'constant z' approach, thus why i see the merits of aperiodic TLs and the like. At the end of the day with protection caps only, or without, active still has to deal with impedance at fs and the general rising characteristic. In my limited experience the fs peak is much larger than any in the x over.
I would desire a smaller impedance peak at Fx than Fs, and i generally avoid whacky Z at cross over. With the right tweeter and done well, the peak at crossover can be very small in the scheme of things.
Going active is attractive just because passive does degrade the sound, in comparison to the same passives at small signal level. Having less crossover, ie going biwired and biamped or from 3 way passive to 2 way passive and active sub, do all improve sound quality.
I still dont like dsp though, too much ADC, SAMPLE HOLD, DAC, business and going optical isnt worth it, more tranducers and conversion to distort the source material. I duno about VSTIs tho, maybe theyre not real time capable, not sure.
I have a Ultrafex pro multiband dsp rack unit. Its OK. havent had a desire to use the surround processor section, or the multiband. But with the bass dsp set just right, my old audax take on a very deep taught extended response without booming. Thats why i kept it!
I would desire a smaller impedance peak at Fx than Fs, and i generally avoid whacky Z at cross over. With the right tweeter and done well, the peak at crossover can be very small in the scheme of things.
Going active is attractive just because passive does degrade the sound, in comparison to the same passives at small signal level. Having less crossover, ie going biwired and biamped or from 3 way passive to 2 way passive and active sub, do all improve sound quality.
I still dont like dsp though, too much ADC, SAMPLE HOLD, DAC, business and going optical isnt worth it, more tranducers and conversion to distort the source material. I duno about VSTIs tho, maybe theyre not real time capable, not sure.
I have a Ultrafex pro multiband dsp rack unit. Its OK. havent had a desire to use the surround processor section, or the multiband. But with the bass dsp set just right, my old audax take on a very deep taught extended response without booming. Thats why i kept it!
Last edited:
No argument. But the better the speaker the less sensitive to the amp, providing it is a half decent amp. My experience is a hybrid is often the best solution. You can do things active that are difficult passive, but putting notch and impedance compensation at the driver is still necessary. Building a sub without active networks is probably only going to win in an open air venue. There are excellent results with all three approaches. The cost of a big bank of quality amps may take away from buying better drivers and crossovers. All things equal, and they never are, passive gives better value. But Siegfried used cheap amps and the very best drivers for the Orion. I for one am in no position to argue with him. Can I complain about B&W 800's? Nope.
I really could care less about any awards. It is only my wife's and my ears I have to please. All, and I do mean all speakers including the ones I make, are retched compared to the rest of the chain. You have to listen for yourself.
- Status
- Not open for further replies.