Are audio companies all involved in a huge conspiracy? - Page 45

It would not be 200 wpc at any given time anyway but if the amp is 200wpc and he has 2 channels going to each speaker (1 to the woofer and 1 to the tweeter) then it may be true that 400 watts are available to each speaker at any given time. Does that mean that its going to be twice as loud? No. From what I understand bi amping is pretty much a waste of time when a crossover is involved. You may get a little bit cleaner sound if it is a high quality speaker.
Edited by Bond 007 - 2/7/13 at 2:43am

So 200W won't fry his tweeters, because the crossover is "burning off" the low frequency energy that would otherwise destroy the tweeters. Of course, if you send an amount of energy the tweeter can't handle, it will go up in smoke anyway.

Certainly, that is true.

The limit to the power delivered to most speakers is the voltage applied to them. Passive bi-amping does zero towards increasing the voltage available to the speakers.

IMO, passive biamping is an audio dealer scheme to sell more power amplifiers to people who don't need them and can't benefit from them once they buy them. It is a loss for the consumer. The only excuse is that most audio dealers are too technically unsophisticated to know what they are actually doing. You just don't find a lot of really good EEs in retailing or repping and most of the few of them that exist are too busy or scared to do right.

IMO, that otherwise sensible AVR manufacturers are indulging in this weirdness does not put them in a good light if technical integrity in the marketplace means anything.

If they wanted to extend the utility of their AVRs they would add amp bridging, which actually makes an audible difference and does not cost much more to implement.
Edited by arnyk - 2/7/13 at 3:36am

Precisely. And it is that crossover that would need to be bypassed if biamping were to actually make a difference.

Somehow I knew you would set the record straight arnyk. I wasn't real sure myself. Good info.

So the speaker does not "see" 200 watts for the lows and 200 watts to the highs because of the passive crossovers?

Heinrich S, I don't want to sound rude but you have to read up on the topic. There is only so much that can be explained in layman's terms. Maybe this helps a bit: http://science.howstuffworks.com/environmental/energy/question501.htm

With one power amp the speaker "sees" 200 watts for the lows and 200 watts to the highs because of the voltage limit of the power amp it sees.

With two power amps the speaker "sees" 200 watts for the lows and 200 watts to the highs because of the voltage limit of the power amps it sees.

There is no significant difference with passive biamping because the power amps amplify the same audio signal whether there is one power amp or two.

With active biamping, the power amps receive and amplify two vastly different signals. There is no crossover in series with the speaker drivers, which can waste power, big time.

Win/Win!

Available doesn't mean used. In normal content, the power required for the frequencies the tweeter reproduces is about 25 percent of the power required for the rest of the audio band, so unless he plays white noise at very high levels, the tweeter will not receive high power. Ever. Because what you play doesn't have it.

I also suspect some misunderstanding that if I have a 200 watt power amp, I am feeding 200 watts to the speaker all the time. Not true. If we can agree that, all things being equal, a more powerful amp can make a speaker louder, we're halfway home. The speaker gets louder with higher voltage/more power input. That means it cannot always be receiving full power from the amp, because sometimes you turn down, and the speaker is quieter, and sometimes the content is not at full tilt, and the speaker is quieter. Actually most of the time. So when you''re listening mighty loud and the sound is 95 or 100 dB in room, you're likely using all of 5 watts or less, regardless of whether the attached amp has 10 total watts available or 1000.

And btw, as I understand it the high pass element connnected to the tweeter as part of the crossover does not burn off energy - - it'd likely need cooling fins if when biamping the tweeter element was getting 4 times the power it was actually using and converting the rest to heat. Instead the high pass presents such a high impedance to the amp below the crossover point that the amp does not make significant current (see Ohm's law) and the power delivered to the tweeter in the lower frequencies is inadequate to harm it.

Finally doubling power NEVER EVER doubles the perceived sound level. As long as the speaker stays linear (doesn't compress) doubling power adds 3 dB. That's going to sound "about one notch" louder, to most folks. Twice as loud is usually defined as 10 dB, and requires 10 times the power. Thus, for example if you're watching a movie with relatively loud dialog at 85 dB, that has explosions that use the max available headroom at 105 dB, the dialog is using 1/100 (that's right one percent) of the power that will be used for the explosions.
Edited by JHAz - 2/7/13 at 8:29am

Here is the schematic of a typical crossover:



Here is how the tweeter side of the crossover breaks down:



And here is what it looks like made out of real parts:



The two resistors that form the attenuator connected together in the upper left hand corner are indeed grey ceramic high-powered resistors that are designed to handle real amounts of power. However, as you say the high pass filter blocks most of the power so that the power resistors don't have to be the finned kind.

Not a reply to Arny, he know this very well, but for those of you with close to no understanding of filters:

the capacitor does not like low frequencies as it will be charged up, but it has no problems changing/discharging fast.... so it will be working for letting high frequencies in towards the tweeter.

the coil is chaging it's magnetic field when a current is run though, so until done that will be blocking... low frequencies will keep going in the same direction and then it will be charged and work as a short circuit and not let low frequencies go through the tweeter....

So it's two components with opposite function, but they're also put in opposite ways.... capacitor in series with the tweeter, coil in parallel, so both with do the same.... getting rid of low frequencies.

(In the bass filter you change the order for the opposite effect. )

I thought the attenuator was to get the tweeter level correct when it is significantly more sensitive than the woofers (which must almost always be the case . . .) FWIW, for a while in my car the mid/highs were bypassed with just a cap to filter out lows . . .

Agreed. I hope I conveyed no other purpose for it.

That is pretty common in simpler and cheaper speakers. There are speakers with no visible crossovers. The inductance of the woofer voice coil cuts off the highs, and the tweeter is made in such a way that its natural frequency response is right for a tweeter with crossover.

Maybe y'all could answer a couple of basic questions while we're on this topic:

1. What is the shunt RC (22 ohms, 6.8 uF) doing across the woofer? Peaking the response a bit in the lows, natch, but why?

2. When simulating (SPICE), do you use a simple resistor to model the drivers, or are better (RLC) models available from the manufacturers? In the past I've just used a resistor, but now and then made more complex models to try to match measurements, but the latter can become a lot of work.

Curious, thanks - Don

That is a zobel which compensates for the impedance of the woofer rising with frequency, usually due to voice coil inductance.

http://en.wikipedia.org/wiki/Zobel_network

http://www.silcom.com/~aludwig/Sysdes/Thiel_small_analysis.htm

http://www.silcom.com/~aludwig/Sysdes/Thiele_equiv_circuit.gif

Thanks!

Drat, I know what a Zobel is, and have used similar circuits in the RF/mW/mmW world, should have caught that.

The driver circuit derivations and equivalents look the same as the ones in my grad acoustics text book. The one I never hoped to have to use again. However, it seems like a lot of drivers provide T-S parameters these days so there is hope...

It's definitely marketing only, but it gives them an opportunity to suggest other uses for 11amps internally when a customer might only need 5. I've actually seen a shopper ask a sellrat just that question.

My (Onkyo) AVR digitises all signals after the input selector and does all processing in DSP, so it costs them a couple of lines of code to implement, and bridging would take no more effort and cost the same trivial amount (amortised across a squillion units). I'm sure many other units from many manufacturers are similar.

Exactly. The hardware is there, so why not use it, well that's the logic. It is arguable whether passive biamping actually constitutes a worthwhile use. Technically, it is not.

All AVRs that I've seen the service manuals for have a similar architecture. FWIW the DSP is preceeded by a massive source selector that includes an ADC for analog inputs. The DSP is followed by an analog volume control chip which is what drives the power amplifiers. The analog volume control chip seems superfluious, but it is there in every case I've investigated so far. It doesn't hurt the SQ because its performance is very good.

You mentioned two dates, 1975 and 1985, and I suspect your $30 number is referenced to the dollars of that day. If so, FWIW, indexed to today's $, that would be in the range of $70.

Mine is done in the digital domain with the only "processing" post DAC a 20dB gain opamp in the LFE.

Don't remind me! At one time I was quite the experimenter with fairly large SS power amps. One little mistake and 8 output transistors are toast!

Just for reference I was looking at the specs for some modern output transistors. They were about $5 each in small quantities. 50 MHz instead of the 3 MHz parts I used back in the day. They had 4-8 times the reactive load handling ability. Where I used to use 4-8 devices costing $50-70 each in today's dollars, today I would only need to use 1 each $5 part that mainstream manufacturers are probably paying $1 or $2 each for in production quantities. And people wonder why AVRs line up 5-11 power amp boards like cord wood and sell them for a few $100.

Where are all the DBT's of DAC comparisons? I see there are a few CDP comparisons, but nothing much on external DACs. Anyone have a link to a DBT or a few credible sources on DACs that are independently verified?

Did you try Google?
http://www.matrixhifi.com/molingordo5_pc_dac1_beh.htm
http://www.matrixhifi.com/contenedor_conver1.htm
http://www.matrixhifi.com/contenedor_dac1.htm

IMO Matrixhifi is suspect and I wouldn't rely upon their testing. http://www.avsforum.com/t/1336902/how-do-i-verify-or-debunk-the-claims-of-the-upgrade-company/1560#post_21959400

YMMV

You mean this one?: http://www.avsforum.com/t/1290427/berkeley-audio-alpha-dac/120#post_19727968 Looks like he fizzled out when questions got more specific.

This is the problem. It's often hard to decipher what is relevant and reliable from the irrelevant - what DBT/ABX tests are reliable accounts and can be trusted? Because it really gives the anti-science camp even more leverage than they probably deserve.

He's a professional NVH engineer whose baseline test conditions report, requested by his (former) Matrix fellows, makes for interesting reading. He also answered questions via PM and on another forum.

In other words, if there are no published reliable accounts of DAC DBT testing then one cannot dismiss their claims out of hand. They could just turn the tables on you and you would be defenseless.

Not all DAC DBTs are published online. If you are that skeptical of the results published by others, there's only one option left for you, it's time to DIY.

But shouldn't we all be skeptical unless the results were independently verified? Or should we accept the results by default? If the DBT's aren't published online then that's great. So no one will know ... except for the relevant parties involved in the testing. It doesn't sound like a particularly credible stance to hold, although I do believe that DACs largely sound the same under controlled testing. It's just difficult to prove or at least show evidence of such a thing.

The anti-science brigade will demand evidence of DBT testing results of which none can be found and round and round we go .. So although their position may be weak the alternative doesn't strike a whole lot of confidence either.