Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
24.19.101.21
1) What are Tympanis? I did a google search, and all I found were used multipanel maggies. Were they one of the panels, made by Magnepan?2) How do studios make that same-sound-from-both-speaker effect. For instance, if I play a mono recording of a single instrument, the image will be centered between the two speakers. Often for the chorus of songs, the harmony singers will (in stereo) have the effect of coming from both left and right side of the room. If same phase and same volume = centered, do they just play the same sound out of phase?
3) Why do tweeters blow: http://sound.westhost.com/tweeters.htm I have read many of his articles. There's something here that doesn't make sense to me. Scroll down to the bottom three images with the green waveforms. The author says that overdriving an amp to where it's producing the square waveforms will destroy a tweeter due to the average power level exceeding what the tweeter is designed. That makes perfect sense - but the culprit here to be doesn't seem to be clipping. If you took a much more powerful amp and set the same gain, the average power level would be even HIGHER, since it's not chopping off the tops or bottoms. I absolutely see how having the headroom makes a big difference for quality audio reproduction, but I fail to see how that protects tweeters from excessive volume. But I doubt the pro is wrong and rookie is right, so where am I going wrong?
4) Other than convenience and being able to swap components, are there any benefits ALLXO have over PLLXO? Rod Elliot compares ALLXO to passive (amp level?) XOs.
5) Anyone have good reading material for digital crossovers? And any useful Windows-compatible software to know about?
edit - 6) If Maggies are so "power hungry" and clipping is such an issue to sound reproduction, how do so many people get great results from 40W tube amps?
I guess that's all that's been bugging me. Thanks :-)
Edits: 03/15/12Follow Ups:
Just to add a gloss on #6, you have to keep several things in mind:
- It takes a 10x increase in power to produce a 2x increase in loudness. So someone who listens at 1000 watts can only listen twice as loud as someone who listens at 100 watts. In addition, not all concerts have a super high SPL. So only the loudest sections of a handful of recordings would require the 1000 watt amp, and only if the listener cared to listen to them at a natural level.
- Most commercial recordings are compressed and/or peak limited. That also discourages people from listening at natural levels.
- Tubes overload more gracefully than transistors, so you can generally get by with half the power.
- RMS amp power ratings are a rough guide at best. SPL's are actually limited by peaks, which can be 10 or 20 dB higher than the average level. This is what determines how loud an amp can be played, not the RMS average power. Some amps have more peak output capability than others. It's isn't uncommon to hear someone say that an amp with a lower RMS power rating played louder than one with a higher power rating.
- Minor peak clipping is inaudible, so any amp can play above its nominal range. (Dangerous for tweeters, though, because of the high harmonics of the clipped waveform.)
- Subjective loudness is determined by average levels, rather than peaks. So almost any amp can play deafeningly loud. The difference is in the cleanness of the peak reproduction, whether and to what degree clipping occurs. So people will listen to say a 40 watt amp and say "Hey, this sounds just as loud as the 1000 watt amp," but that's *not* the question -- the question is whether it sounds good on the loudest peaks you play.
- Different listening distances and rooms can alter levels by several dB in either direction, so some setups require more amplifier power than others.
Bottom line, the only way to tell how big an amp you need is to try it, or to play music at the levels you like and watch on an oscilloscope so you can see if and where peak clipping occurs. Or you can measure with a Radio Shack meter and guesstimate the peaks by adding 10 dB to the readings (20 dB for percussive instruments like piano).
There are endless debates about this on forums, but ultimately, they don't mean anything, because what matters is what works for you.
I left a few for others to answer because it is 4am.
One thing, though: Don't you have one of the best ALLXOs in the market?
In any event, when one can get away with it, a PLLXO is the cleanest thing there is...and, ironically, the cheapest. The problem is that it has to be designed for the specific hardware and, at the most, with 2nd order slopes. With most Maggies the latter is not an issue. I use a PLLXO.
I have a Bryston 10B STD. I don't know where in the hierarchy of AXOs it ranks. But I like it. Of course the only think I have to compare it to is stock XO on the 1.6s.
Among the best there is in its class. If I recall well, they even avoided the use of opamps (which some people don't like). Like everything in our world, it cannot do every trick on the book, but what it can do, it does superbly.
A passive line level works for gentle slopes and predictable impedances - and is great if you like the resulting sound. Cleanest? Probably not - a 50k nominal - but varying input impedance into an undetermined circuit -is likely to offer issues ranging from high frequency oscillation (ultrasonic) to added noise levels due to better power transfer of noise in the bandstop areas of the passive XO. Active opamp or descrete circuits offer an advantage. A couple things that a store bought active analog xover like a Bryston doesn't do is allow mismatches slopes and frequency points. The passive high level XO in loudspeakers usually are mismached because they are designed to complement the electrical and driver response characteristicts. This is the part of loudspeaker design that has greatly improved their sound over the last 20 years. In addition, the internal high level xovers of a loudspeaker might also perform EQ and impedance correction (not necessarily Maggie's though). That is why I like to use DSP based XO - or custom Analog XOVER - because you can model the performance to match each driver. Not only that - from a design perspective - you don't really care (within limits) that the loudspeaker drivers themselves have an even frequency response - because you can model the correction in the signal going into the drivers.
Three most important things in Audio reproduction: Keep the noise levels low, the power high and the room diffuse.
B_ATL: "Cleanest?"
LOL! To be sure, it is tricky...but yes if the PLLXO is custom designed for a given set of hardware and with quality parts...and only for the proper application. Unfortunatly, the right combination of hardware, speakers and skills/knowledge does not happen all that often.
Luckily, many Maggies prefer the shallow slopes that a PLLXO can do well...but that's just a part of it.
In addition, I have to say that some kind of measuring capability is needed to fully calibrate. Anyone who tries a PLLXO simply on the basis of the theoretical component behavior will miss some of its great potential. As an example, careful selection and matching of the caps and resistors will make a huge difference compared to other sets of similarly rated quality parts.
I can relate to your reservations about the impedance. I had them at first. A bit of luck helped me in that the equipment I have already allows for lower impedance matches. I can run the mid/bass panel in 1st or 2nd order. The higher impedance imposed by 2nd order in my case ( <5k ohms for 1st vs <53k ohms for 2nd) causes a 6db signal loss...but no loss of clarity that I can perceive.
In general, one can go and throw some numbers relative to the interface specs, and typically wind up with a PLLXO bi-amp that can be clearly superior to a speaker-level xover but miss the mark against a good ALLXO.
Or, one can go further by choosing, matching and measuring behaviors, and further refining accordingly. Then one winds up with something that can't be beat...but it is a lot of work and it is not for everyone.
Still, thank heavens for the good ALLXOs. They maintain signal strength. They can do sharp slopes when needed and allow for multi-drivers much more easily. Some can even do quite nifty digital tricks for those who don't mind extra digital steps. Furthermore, ALLXOs can do most of this without a care for what they are hooked into (my PLLXO will easily need a redesign if I change some components).
ALLXO can do all this in manners that are more practical for most everyone...and do it superbly.
However, the parts count...count. If one can get away with a proper PLLXO implementation, this shows.
Well let me be arrogant enough to try to answer your tough questions. :)
1. Tympanies were a multipanel system that magnepan sold when Andyr was in his 80s and could still hear.
Here is a link to the specs on them and the different models. For some reason this is under the "articles" section which IMHO makes no sense.
http://www.integracoustics.com/MUG/MUG/articles/speakers.html
2. Different vocals in left and in right speakers.
3. I read that to mean something like "just because an amp is powerful doesnt mean it wont clip". It is just less likely than a lower powered amp, but even big amps can clip, so just because you have a huge amp doesnt mean you can throw caution to the wind. FWIW there have been quite a few inmates over the years that post about their big huge amps clipping and blowing tweeters when they dont think that should happen. Usually the poor fuse gets blamed....
4. A few that active have are being able to buy them store bought, having the ability to run tubes, and being able to drive longer cables or tough loads better than most sources can drive amps directly.
5. That link I gave you over on the pc audio forum has a download for cplay. It is a great sounding player and I managed to get the author to add vst plugin capability. Now you can use a pc crossover program that works as a vst plugin to get crossovers. Allocator, Acourate, and Audio Lense come to mind on the pc side. Look for a post by Playmate entitled something like "Poor man's DEQX".
6. I am not sure they DO get great results. Well my 40 watt tube amp was nothing to write home about on the mags. Anyhow it is system/ room dependent. Tube amps clip more gently so they will do less damage when clipping IMHO. But clip they will. I am in the minority in that I dont think mags NEED current. I think they work better with voltage and IIRC tube amps are better at this.
Afterwards we discovered faith; it's all you need
Praising Andy will get you nowhere.
He will still attempt to steal your A21s because he heard them driving a pair of 3.6s once and liked what he heard. He thinks the Frankie's tweeters can use a dedicated A21 each. You have the pair, so watch out! : - ))
re: #3
No, I meant for the same gain, a higher powered amp (not clipping) will have a higher RMS power than a lower powered amp that clips, and I supposed that is more likely to cause damage (i.e. the damage is simply a function of power being supplied, and an amp clipping off the top of the waveform actually lowers the average power - so buying a bigger amp with more headroom does not save the tweeters if you like earsplitting SPLs)
Too many variables to answer quickly. Here's the gist of it.
Two amps with the same sensitivity at the same gain setting will produce the same power. If you ask the amps to put out more power by turning up the gain, the high power amp will do so. The low power amp will clip. A clipped signal sends DC to the tweeter. This will heat up the voice coil very fast and eventually burn it. Because a tweeter has such a small voice coil (compared to a woofer for instance) it doesn't have much capacity to dissipate that heat and burns quickly. Generally speakers will handle more than their rated power (at least briefly) so long as it is undistorted. But feed a speaker a distorted (clipped) signal and it won't last long.
Of course if you surpass the speakers rating, even with a clean signal long enough, it will burn as well.
If you look at a driver as a circuit, then there is a limit to the current it can take. So, that is a limitation, power rating of the cable or ribbon,
If you look at the motion limitations of a driver, you will find yet another set of ratings. They may not be stated but they exist and in the case of woofers one of these, cone excursion, has to be stated.
So, yes, too much clean power can still kill a driver, based on power rating of the circuit alone. You may also find that a driver pushed beyond its mechanical limits can be ruined at well under its total power rating.
Now bring a clipping amplifier to the party. Say its max RMS is 100 watts and, because your driver can take 300 watts, you crank that amp up to heaven. Guess what... it will exceed its maximum rating very happily and merrily distort for long enough to push the driver to madness. In doing so, it may start issuing such a horrific signal shape & impulse as to exceed the driver's mechanical limits. As the power is also getting closer to the electrical rating of the driver, things get even more stressful. If the amp does not give up, the driver will.
Tweeters just happen to be more limited but all driver have their limits.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: