Can you hear a difference in Sound between Audio Amplifiers?

[RichB]

During a bi-amp experiment, I measured the voltage of a 100HZ sine wave to a Revel Salon with the HA-1 set at a reasonably low level -30 using a Parasound A21 amplifier. This is roughly 80 DB and a hair less than 1 watt.

The voltage single amped was 2.48.
Bi-amped it was 2.51 on the highs and 2.45 on the lows.

Since the Salons are crossed to the 3 woofers at 125Hz, both sides were producing some of the 100Hz sound.
I suppose this could be a inaccurate measurement using the RadioShack Multi-meter.

I am not sure if this is how your calculate it, but 2.45 / 2.48 = 0.9879032258064516.
Multiplied by the result * 80 DB = 79.03 this is a 0.97 DB on the low amp channel.

That may not be the right math but it might represent a measureable difference that is audible that represents a particular amplifier frequency response change when driving a reactive load at very low power.

- Rich

 

[Seth=L]

Separate amplifiers have the potential for superior power at their given specs. Most receivers have a central power supply that provides power to the amplifier, processor (dsp, video), line voltage for preamplification, display, and so on. Much like a computer processor, when the the computer is on and doing nothing the system's idle processes are consuming a fair share of power even though you're not actively doing anything. The power amplifier's job is singular, produce power for speakers.

Most receivers are lucky to pack even a 500kVA transformer and are just as likely to have a modest set of power suppy filter capacitors for those instantaneous power needs. While these receiver amps appear to perform admirably on the bench they're not typically being tested with reactive loads with real program material, which is exactly what you'll encounter when you go to use it.

So yes, amplifiers can sound significantly different depending on the use. If you listen at chamber levels you may never notice that difference. If you're doing critical listening at levels above that you probably could.

 

[Seth=L]

That may not be the right math but it might represent a measureable difference that is audible that represents a particular amplifier frequency response change when driving a reactive load at very low power.

I believe this to be the case in most scenarios. There's quite a few speakers out there that throw some curve balls at amplifiers, and if they amplifier isn't up to task it will get fussy to some degree. It may only due this once in a while and most people probably wouldn't notice it, especially as casually as many people listen to music or watch movies. Music is more likely to reveal the pitfalls of your equipment (other than subwoofers) in most scenarios. I was listening to some CDs the other day on a little Tripath amplifier and the sound was great until I got to one CD that had a lot of energy in the midrange and I started to hear a strange distortion. At first I thought maybe I had a faulty tweeter, but then I changed the amplifier out for another less powerful class A/B amplifier and turned it up louder than it was with the Tripath and the odd distortion went away. The less powerful a/b amp with a standard PSU handed the reactive load better at higher volume than the SMPS powered Tripath design that's rated for more output. Granted comparing these two amps that have different topologies and power supplies is like comparing apples and oranges, but you can still use the same concept to apply to like technologies to a lesser degree.

 

[highfigh]

It wold be so much more simple if terms like 'bi-amp' was used in the traditional way, meaning using one amplifier channel for each frequency band, with the crossover placed ahead of the amplifier in the signal path. It would also be a more efficient system because the crossover would be an active circuit with the ability to provide unity gain without the insertion losses caused by passive components. It could be tailored to the specific drivers WRT slope, phase response could be more correct, either inherently or through correction and response corrections would be able to remove peaks without the phase shift from a passive network.

I can see a benefit from using one channel on one woofer and a different channel on the other, but at a given output level, the difference comes as more headroom. In theory, the maximum output will be higher, but is that really the goal? It isn't, for me. That said, if the amplifiers in question have an input level control, or at least the one for the upper range, the sound can be balanced better and more easily. It's equalization without using a passive filter and without the effects from that.

Bi-wiring, as explained by the Monster Cable people when they started to use one pair of conductors for the highs and another for the lows, only to be combined at the amp AND speakers, makes less sense, to me because they referred to the speed of the signal, not the current. I can see using one set of smaller conductors for the highs and mids because they just don't present the difficult loads of woofers, though.

A .97dB increase doesn't seem worthwhile, to me. At some point, the speakers may exhibit thermal compression and all of these benefits will be lost, anyway. Trying to power through this makes smoke, not decibels.

 

[highfigh]

I believe this to be the case in most scenarios. There's quite a few speakers out there that throw some curve balls at amplifiers, and if they amplifier isn't up to task it will get fussy to some degree. It may only due this once in a while and most people probably wouldn't notice it, especially as casually as many people listen to music or watch movies. Music is more likely to reveal the pitfalls of your equipment (other than subwoofers) in most scenarios. I was listening to some CDs the other day on a little Tripath amplifier and the sound was great until I got to one CD that had a lot of energy in the midrange and I started to hear a strange distortion. At first I thought maybe I had a faulty tweeter, but then I changed the amplifier out for another less powerful class A/B amplifier and turned it up louder than it was with the Tripath and the odd distortion went away. The less powerful a/b amp with a standard PSU handed the reactive load better at higher volume than the SMPS powered Tripath design that's rated for more output. Granted comparing these two amps that have different topologies and power supplies is like comparing apples and oranges, but you can still use the same concept to apply to like technologies to a lesser degree.

WRT to the discussion at AVS Forum about amplifiers being held to standards that are more stringent that they are, an amp that outputs high voltage is one thing, an amplifier that has high current capablity is another. The low impedance and the reactive components make for some interesting experiments in frustration. Where one person hears angels singing with their system, another hears the voices of hell with theirs and the only difference are the room and speakers. The room, obviously can't be expected to be the same but the speaker loads, especially when one isn't designed to provide a 'safe' amplifier load, can cause some amps to bog down and distort badly if it's happy place is 8 Ohms. Receivers and integrated amps form the '70s and early-'80s are prime examples of this, but the speakers from that era tended to be very close to 8 Ohm loads. Polk was one of the first that I remember deviating from that path and now, receivers are rated for 4-6 Ohms and speakers are regularly seen with loads that are comparable.

However, the fact that the distortion may have been caused by the energy in the mid-range makes me think it could be IM distortion and this is something that's receiving inadequate attention, these days. Power is still king and I, as well as others, think that current and IM are more important than just output voltage into a "safe" load.

As we have discussed in a couple of other threads, I just bought a Parasound amp. That does things very differently, compared to the Denon AVR I was using. I thought the Denon sounded good, but I wasn't comparing it with anything. When I turned it to -20dB (I use the attenuation scale, not arbitrary numbers and ALWAYS limit receivers to -10, when they have that ability), it was loud, but my ears had kind of a "stuffed full of cotton" feeling. This was with no Audyssey or other EQ, no Dynamic EQ or anything the Denon has to make it sound better in everyone's room. Highs seemed pretty good, bass was good- extension was very good. My new amp is totally different. The highs aren't stronger, but it seems to be cleaner, if you will. Cymbals sound more like metal, not a reproduction of a cymbal, bass is more punchy, dynamics are sometimes a bit startling and when it's loud, not only is the 'stuffed eard' sensation gone, the sound is huge. I'm certain that the added height is due to the fact that my ceiling isn't treated and the additional SPL is causing the reflected sound to be more audible. One song I heard on Thursday and again today was 'Voodoo Chile', from Stevie Ray Vaughn's 'Couldn't Stand The Weather'. Huge sound- it was like being at a live show with a great mix and I was at the show he did here when he toured for this album, first row and we had seats in front of his rig. This line of amps is designed to minimize the upper harmonics and the attention to this is what I have seen when many other amp manufacturers discuss their products.

 

[RichB]

^^^
Everything you hear makes sense to me since music has peaks and if the power is not there, then you get attenuation.
Some feel that AVR's will distort badly and that may be true at some level but a peak will just get lopped of.

In the section "The Peak Power Demands of Well-recorded Music"

http://www.cordellaudio.com/he2007/show_report.shtml

This demonstration was a real "Wow" for the attendees. The Rickie Lee Jones (RLJ) cut was played at realistic, but certainly not unpleasant, levels in the relatively small hotel exhibit room on speakers with an estimated sensitivity of about 89 dB. The average power typically read 1-2 Watts, while the power on peaks often topped 250 Watts (the power display monitored only one channel, so these numbers should be interpreted as Watts per channel). On this cut, most peaks occurred with an aggressive "thwack" to a snare drum positioned dead center.

The "thwack" was likely to include a great deal of midrange energy.

- Rich

 

[Seth=L]

@highfigh

Sorry if I missed something, but what is IM distortion and WRT?

I am curious to know what you think about the Tripath amplifier. It's a Teac AG-L800 5.1 receiver that supposedly has a 50 watt per channel amp which is powered by a SMPS that has a current draw rating of 100 watts stated on the side.

Picture of the inside (might help, idk)

Compare that to the Niles SI-275, which may or may not be a rebadged and reconfigured ATI amplifier.

 

[PENG]

I am not sure if this is how your calculate it, but 2.45 / 2.48 = 0.9879032258064516.
Multiplied by the result * 80 DB = 79.03 this is a 0.97 DB on the low amp channel.
- Rich

It is logarithmic, the difference dB between 2.45V and 2.48V is about 0.1 dB.

 

[highfigh]

@highfigh

Sorry if I missed something, but what is IM distortion and WRT?

I am curious to know what you think about the Tripath amplifier. It's a Teac AG-L800 5.1 receiver that supposedly has a 50 watt per channel amp which is powered by a SMPS that has a current draw rating of 100 watts stated on the side.

Picture of the inside (might help, idk)



Compare that to the Niles SI-275, which may or may not be a rebadged and reconfigured ATI amplifier.

WRT is With Respect To- sorry.

IM is Intermodulation distortion.

I don't have any first-hand experience with Tri-Path, but my friend has and he likes them. I have been looking at them but haven't had a chance to do anything with one. I have seen that using a more stout power supply has a big effect on the output.


I think the Dayton amps from Parts Express may be from ATI, too. The 12 channel model that I have used has a standard power transformer, but the circuit board looks similar. Unfortunately for the first one I used, water got in and made some surface-mount components disappear.

 

[RichB]

It is logarithmic, the difference dB between 2.45V and 2.48V is about 0.1 dB.

Thanks.
I had a feeling that might be the case

- Rich

 

[PENG]

Thanks.
I had a feeling that might be the case

- Rich

You're welcome! If you don't want to bother with the math, you can book mark this site:

http://www.sengpielaudio.com/calculator-gainloss.htm

Try punching in the two voltages and it will spit out the dB (about 0.106), same as what I got using my own formula using Excel.

 

[Jared J. Crandall]

In the past, I had ls50s, bought a parasound a23, but decided that maybe amps made a difference, so i bought a modwright 100se...I could clearly tell that the amp changed the soundstage to be much more captivating. A couple of years later, i decided to get a nad m27 for my kef r series....once again, the amp does make a difference. Unfortunately, I am not an audiophile but still notice the difference, so I pay the difference.

 

[KenM10759]

Jared, which R series speakers do you have and why do you no longer had the LS50's?

I was telling my dealer that I'm feeling I get a little too much rolloff of the tweeters at the high end and was wondering if they had an amp (integrated or power) that could deliver more from my R500. He commented that my NAD receiver, and all NAD amps in general tend to be "a little warmer" at the top end and I should consider a Rotel. He sells, NAD, Parasound, Parasound Halo, Bryston, Rotel, Sony ES products, Classe, and just added Emotiva.

I'm still not ready to make any moves but do want to keep gathering other users impressions on the various options before I take home a couple selected option for home audition. I'd appreciate knowing what speakers are driven by and what source feeds your M27 amp.

 

[Jared J. Crandall]

I haven't spent the time and am not good enough to distinguish differences in music between two amps, but what I do notice are the immediate differences like soundstage, clarity, and affects to the audio spectrum (lows, mids, and highs). I did have the ls50s, even used them for home theater, but was curious about "upgraded" to a larger speaker, so the kef R series seemed the closest to the ls50s, so I bought R500, R200c, etc. I have used a parasound a52, outlaw 7125, and nad m27. I didn't notice a difference between the a52 and outlaw, but I did notice a change going to the m27. The difference may have been going to a Class A/B amp to a Class D, but a review from hometheaterreview claimed that the a51 sounds very close to the m27 but a little weightier. The outlaw and a52 for sure seemed weightier, but the NAD m27 distinguished itself because it seemed to have a more pure sound and soundstage, but for me, the vocals just popped with the nad. I do for sure think the NAD made the difference with the mid-high frequencies, although I am using movies as my source, I haven't compared the two amps using music. As I mentioned, I was hoping that I could keep the outlaw ($1000) and save a great deal of money, but the NAD just made the small difference that was big to me.

One reserve that I had is that I use a pre/pro with room correction, so I thought that a change with the calibration would obviate any amp change I could make i.e. the room correction would alter or correct the sound so much that an amp change wouldn't make a difference since room correction can be so volatile, however, I did room correction a few times with each amp and the differences remained each time. I do think the the ls50s and R500s have such low distortion that an amp makes a difference. My rule-of-thumb has been to only upgrade when a change is readily apparent, as opposed to searching for those differences, and the amp change met that criterion.

Let me know if you need more descriptions and info.

*fyi, I've also had the emotiva xmc-1, and I think it had a similar effect as the NAD m27--clarity, soundstage, etc

*marantz av7702mkii, Nad m27, R500, R200c, 800ds, e301 (height).

 

[KenM10759]

Thanks! I agree, these speakers are very low distortion even when played LOUD. They do seem to punch above their weight and I love that they sound the same at "easy conversation" levels as they do at "wife beating me about the head" levels.

This helps.

 

[Jared J. Crandall]

Thanks! I agree, these speakers are very low distortion even when played LOUD. They do seem to punch above their weight and I love that they sound the same at "easy conversation" levels as they do at "wife beating me about the head" levels.

Thanks! I agree, these speakers are very low distortion even when played LOUD. They do seem to punch above their weight and I love that they sound the same at "easy conversation" levels as they do at "wife beating me about the head" levels.

This helps.

I just figured out a good description: the nad increased the black around each sound. Maybe that is the noise floor, but that is the best way to describe the difference I found.
his helps.

Thanks! I agree, these speakers are very low distortion even when played LOUD. They do seem to punch above their weight and I love that they sound the same at "easy conversation" levels as they do at "wife beating me about the head" levels.

This helps.

 

[AcuDefTechGuy]

In the past, I had ls50s, bought a parasound a23, but decided that maybe amps made a difference, so i bought a modwright 100se...I could clearly tell that the amp changed the soundstage to be much more captivating. A couple of years later, i decided to get a nad m27 for my kef r series....once again, the amp does make a difference. Unfortunately, I am not an audiophile but still notice the difference, so I pay the difference.

Did you do a volume level match comparison within seconds between one amp vs the other?

Or just base this on your memory?

 

[Goliath]

I certainly heard a difference in my Parasound HCA 1000A vs my closeout Crown XLS 402D. The Crown was easily a lower floor noise amp.

I could pick it out blind 100% of the time.

Can you give us more details regarding the listening test? Single blind, double blind, Mushra/Triangle?

Level matching method? Quick-switched, or slow-switched? Music selection, number of trials? Etc? Was this a proctored blind test, or performed by yourself?

 

[Jared J. Crandall]

I don't think that was relevant because regardless of how low or high the volume was, the benefit was apparent. I am not an audiophile but I attempt to disprove that the two amps sounded different but the difference was undeniable. Just unfortunate that it cost me $2450 more than the outlaw amp..like i said though, I went from a budget a/b outlaw, and the parasound a52, to a class d, but some reviews say that the nad m27 is similar to a51 but the matter has a little more weight

 

[Jared J. Crandall]

Can you give us more details regarding the listening test? Single blind, double blind, Mushra/Triangle?

Level matching method? Quick-switched, or slow-switched? Music selection, number of trials? Etc? Was this a proctored blind test, or performed by yourself?

I second this notion. I went from parasound to nad m27 and I didn't need to be scientific because the sound difference was very apparent. I think it was the noise floor difference, because the background has been very black.