The title of this thread (which you started) is "biamping with X-over after the amp is better." But "you're not trying to convince anybody?"
You're off to a terrific start at DIYaudio.com. You join four days ago and already you've made 78 posts in just a couple of threads and irritated more than a few folks.
I think it might be a good idea if you just perused this excellent forum for a few months (without posting) to get a feel for the knowledge level of the folks here. Just a suggestion.
Dave.
Using an in line crossover before the amps passive or active would slpit the sound spectrum going to each amp resulting in an allocaation to each amp to reproduce part of he sound specrum. Keeping the passive xovers after the amps is ok to protect the drivers as well as take advantage of any compensation the manufacture put in if these are high end xovers. not using a crossover before the amp is ok if you have the power to spare since not using an xover before the amp results in all amps getting full sound spectrum signals to amplify and of that reproduction out of band material based on the design of the passive xovers will be shunted to ground if 12db or greater xovers are used. In the case of 6db xovers the shunting to ground will occur in the output stage of the power amp as reproduced power takes the path of least resistance which would send music that is unpassable thru the cfrossovers to be sent thru power resisters that are shunted to ground within the power output stage of the amp....assuming we are talking SS amps. Liitle more complicated for Tube gear.
For me, the main question is, if passive crossovers had not been invented, would anyone seriously propose them, now?
Apart from the obvious disadvantages, their 'design' seems to attempt to juggle too many interacting (conflicting?) variables, and it's hard to believe that, no matter how High End you label them, or how expensive their components, or how much time goes into their refinement, that they can ever be optimal. It's like comparing the Baird television system to RCA's electronic version. In one stroke the RCA system rendered Baird's most 'High End' efforts obsolete.
As some wise person said:
Apart from the obvious disadvantages, their 'design' seems to attempt to juggle too many interacting (conflicting?) variables, and it's hard to believe that, no matter how High End you label them, or how expensive their components, or how much time goes into their refinement, that they can ever be optimal. It's like comparing the Baird television system to RCA's electronic version. In one stroke the RCA system rendered Baird's most 'High End' efforts obsolete.
As some wise person said:
How will you sell a speaker box without passive filter? You specified the filter and let the customer arrange that? A mass as result is guaranteed.
The one thing that puzzles me: is it in fact correct to put the same loudspeaker filter before the amps, I mean does it display the same characteristics?
I ask this because a loudspeaker has an impedance of say 8 ohms while the entry of an amp is easily 20k ohms.
I'd expect that this will change the effect of the filter. Or am I wrong there (my electronics mathematics are 25 years out of date, sorry not to calculate it myself!) ?
I ask this because a loudspeaker has an impedance of say 8 ohms while the entry of an amp is easily 20k ohms.
I'd expect that this will change the effect of the filter. Or am I wrong there (my electronics mathematics are 25 years out of date, sorry not to calculate it myself!) ?
Passive Crossover Network Design
This chap Rod Elliott suggests that because of impedance interactions, passive crossovers suffer far more from voice coil heating effects than active systems, and that there is nothing that can be done about it, 'High End' or not. So if your speakers sound great when cold and at low levels, they simply can't when warmed up, and vice versa. The effect is less extreme for active systems, because the filter frequencies and slopes aren't being affected, only the relative levels between woofer and tweeter. If this is the case (does anyone say it is not?), is this not a great disadvantage to the passive-after-the-amp configuration?
This chap Rod Elliott suggests that because of impedance interactions, passive crossovers suffer far more from voice coil heating effects than active systems, and that there is nothing that can be done about it, 'High End' or not. So if your speakers sound great when cold and at low levels, they simply can't when warmed up, and vice versa. The effect is less extreme for active systems, because the filter frequencies and slopes aren't being affected, only the relative levels between woofer and tweeter. If this is the case (does anyone say it is not?), is this not a great disadvantage to the passive-after-the-amp configuration?
Mrs Liching,
So, if you know how to measure, why not put your idea to the test and inform us about your findings? If there is an improvement in IMD, it should come out.
As long as you haven't done so, I remain unconvinced that what you are doing is a sound and/or sane idea, for all the reasons set out above.
vac
This is a DIY forum, so commercial considerations are seldom important.
However, if you think about it for a moment, it's not really hard to see some engineering solutions that work.
- Go into collaboration with an electronics manufacturer to produce the electronics you require. Linn and Naim did this over a quarter of a century ago. You could have it done as a a complete package, xover and amps, or simply xover and let the customer supply their own amps. This need also not be a 'name' brand, as there are plenty of design and manufacturing houses in the world that could supply a complete product to specification, complete with the speaker brand's name on it, whether as a module to be placed inside the speaker cabinet or as a separate component.
- Build it as a complete system as Meridian have been doing for nearly as long. As have ATC, whom have long offered active and passive versions of the same speaker. Legend in Tasmania also offer active an active version of speakers, using a DEQX as the active xover, or passive if you prefer that option.
I can't recall seeing a passive xover that couldn't be duplicated in terms of transfer function at line level, but plenty of active xovers that could not be duplicated passively. Think about what you can accomplish with a DSP based unit very cheaply and easily. Show me a passive xover that could begin to do what a DEQX could.
Now as for you claims that the passive biamping scheme reduces IMD, SHOW US THE MEASUREMENTS. For someone who is supposed to be a professor, it should not be difficult to back it up with real data, rather than handwaving assertions.
Last edited:
OK. This article suggests that domestic listening levels are not likely to induce much of an effect from voice coil heating.
Hot Stuff: Loudspeaker Voice-Coil Temperatures | Stereophile.com
Yet B&O include compensation for it in the Beolab 5 active speaker.
www.abt.com/documents/6162/1680009_specs.pdf
and Harman have a patent for it, using a predictive model of temperature, rather than a direct measurement:
Patent US20020118841 - System for using digital signal processing to compensate for power ... - Google Patents
So is it something we DIY-ers should be building into our systems..? (the active xover people, at any rate)
If you accept 110dB pk at the listening seat and further accept that average listening levels are ~20dB below that peak level for loud/realistic reproduction, then that equates to the speaker producing an average level of ~87dB @ 2.5m
A speaker of 90dB/W @ 1m will need an average power input of ~3W to give those average listening levels.
Heating of the voice coils in domestic speakers, used sensibly, is not an issue.
It's "party time" that causes problems for domestic type speakers.
If one wants "party time" on alternate week ends, then buy suitable speakers for that duty.
If one wants good domestic reproduction of music and sound tracks accompanying film, then buy those speakers suited to that duty.
A speaker of 90dB/W @ 1m will need an average power input of ~3W to give those average listening levels.
Heating of the voice coils in domestic speakers, used sensibly, is not an issue.
It's "party time" that causes problems for domestic type speakers.
If one wants "party time" on alternate week ends, then buy suitable speakers for that duty.
If one wants good domestic reproduction of music and sound tracks accompanying film, then buy those speakers suited to that duty.
Last edited:
Thats a very good idea to put the original passive filter BEFORE THE AMP. The characteristics as designed by the manufacturer is still there. The only think is to shut off the passive filter with the same impedance as the speaker, lets say 4 ohms before connecting to the input of the Amp. Its even better then the use of an active or digtal filters as suggested in some postings. Active filters ALWAYs add distortion and noise. Digital filters needs A/D and D/A conversion. These gives a lot of problems. It seems that these disadvantages are overseen in the most postings.
However a filter, active or passive, before the amp, does not cancelled out the IMD signals produced by the amp that are located out band. The passive filter after the amp does.
Last edited:
I presume your passive crossover also includes impedance matching components that null out resonances in the drivers (or so I have read!). You would have to remove these if using a resistive dummy load I would think. Your output impedance from the amp would be lower than that of the crossover, so the damping would change..? It all sounds quite complicated!
Thats right thats why I prefer to keep the whole passive filter original in the box. However the idea to use the original passive filter before the amp is much better then an active or digital filter.
Last edited:
A preamp gives 0,7 volt at full volume. That is about 100mW at 4 ohms. Most preamps can handle this and one doesnt play full volume. If it gives problems use the heaphone output since most heaphones have a low impedance.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.