Has anyone mentioned this article yet?
BiAmp (Bi-Amplification - Not Quite Magic, But Close) - Part 1
Look carefully, with a filter before the amp, the amp only get a part of the spectrum resulting less IMD, thats right.
However this also happens with the filter after the amp. Huh, yes. Take a look. In a poweramp IMD is mainly produce by the emitter follower of the endstage because it has to deliver current making the internal re signal dependent thus IMD. However the passive filter blocks the unwanted frequencies so the amp only delivering power in the intended band, the same band as with active filter thus the IMD is the same despite the amp get the whole spectrum. Plus the IMD products that are out band is also filtered out which is not the case with active filtering
Here's a technical arguement. The dipole nature of the ESL panel combined with it's width will cause a low frequency pressure rolloff of 6dB/oct starting around 800-1000Hz. With apassive crossover after the amp, there's only one way to flatten out that natural rolloff and that is to absorb the energy above the rolloff point to flatten out the response. This means your speakers are way less efficient than they could be because a lot of your amplifier power is going to heat the inside of the marginal bass enclosure. Martin Logan also has a nasty habit of crossing over to the woofer way too high, which minimizes this rolloff problem at the expense of weird phasey midbass. With either a passive, or better yet, active crossover before dedicated amplifiers you can easily compensate for the low frequency rolloff of the panel.
My panels are crossed over at 130Hz and you can debone chickens with the dynamics from a pair of 20 watt amps driving the panels.
Also a passive crossover decouples the amplifier from the load, and for something reactive like an ESL, that's not a good thing.
Sheldon
You say it's better to use the passive filter than to use active or line level filters, THEN when the advantages of line-level is pointed out, you say you're only talking about mono vs. biamping.
Fix your attitude first, if your english is the problem, we'll be willing to listen through the broken english if you present things politely.
many of us try to be. but its tough to gauge the intent of the words when written down and some people bristle at the perceived tone. this is a pretty friendly place in general, particularly if I am friendly going in
Liching
As Andrew pointed out there will be less in band frequencies for IMD to occur within the amp. But in addition, would your dominant distortion mechanism of signal dependent re be reduced due to the decreased variation of the emitter current with the increased load impedance?
Hope this helps
-Antonio
Baloney. I've never asked for help, and I've been treated with respect.
Of course, I've never bragged about my system, nor have I made claims based solely on subjective evaluations.
Sweetie, as a snappy internet comeback, that's a complete fail.
OK folks I believe it's time to put away the swords and understand there may be a language barrier here.
Mrs. Liching, we only ask that you understand that some of what you say can be considered insulting to those who have much experience in this. It is good to bring forward ideas, it is bad to dismiss those ideas considering the experience and knowledge being offered. If you think I am wrong, please let us know where your information is coming from and why your convictions are so strong. In English, you are not communicating well with those who's knowledge is considered high.
Thank you for considering this post.
You all are invited to tell me what in my above statement, is not correct.
The question is not whether your statement is correct, but whether or not the assumption you extrapolate from that statement is correct. To convince anyone that your preferred topology is better, you need to provide some IMD measurements of each arrangement.
And, even if, with your speakers and your amps, there is an improvement with the crossover following the amps, it's a bit of a big step to generalise that to all speakers, amps and crossovers.
And, even if, with your speakers and your amps, there is an improvement with the crossover following the amps, it's a bit of a big step to generalise that to all speakers, amps and crossovers.
To be clear, I am not trying any body to convince. You are the ones that are saying that my approach is not correct so its yours to come with arguments. So is it in Scientivic world wheather its polite or not instead of calling that it is unusual in your establish environment with people who thinks to now.
I just claim that every solution is a compromise, and I'm not convinced that IMD matter at all. Of everything I have tried filter before the amp have been the best approach.
Maybe I have more IMD this way, but the other benefits I gain outperform this.
You still come up with a claim that is not backed up by measurements. I haven't saved my measurement with passive filter, so I can't show before and after. But measurements and my ears tell me what is better in my ears.
Nad room accoustics is also some sort of distortion, and measured i dB, the difference will definatly be much more than IMD. I don't consider THD or IMD to be a problem, and I have no idea how to measure it. If I can't measure it, I simply decide not to worry about it.
Maybe I have more IMD this way, but the other benefits I gain outperform this.
You still come up with a claim that is not backed up by measurements. I haven't saved my measurement with passive filter, so I can't show before and after. But measurements and my ears tell me what is better in my ears.
Nad room accoustics is also some sort of distortion, and measured i dB, the difference will definatly be much more than IMD. I don't consider THD or IMD to be a problem, and I have no idea how to measure it. If I can't measure it, I simply decide not to worry about it.
Room acoustics is distortion yes but is a linear one. That means that there are no new products in the music.
An Amp however gives non linear distortion. If there are 2 frequencies f1 and f2, the amp produces f1+f2, f1-f2, 2f1-f2, 2f1+f2 etc. This is IMD. Our ears are very sensitive for this. Noise or hiss is much more tolerable then this. Levels of minus 60db are still detected by our ears. The music become harsh and diffuse.
About the question how to measure, with a spectrum analyser you can see the unwanted products. At the output you see f1 and f2 but also the others.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.