On the use of a preamp to improve sound quality with VTA70

rjpjnk wrote:Greetings all. I see that many people are running active preamps before their power amplifiers, and I would like to understand if a preamp could be of any use to me.

For the sake of discussion, let's say I don't need phono, don't need input selection, and I don't need tone controls.

If I have only a single input source (say, a CD player), what possible advantage could a preamp (such as a PAS 3x or SP14) offer in terms of sound quality compared to just running directly into the VTA70? I already have a stepped attenuator for volume control.

I have read a lot of discussion already and posted on some less technical foums, but the only answers I got were along the lines of "Of course a preamp makes things sound better, you must be an idiot for asking". Even if this is correct, it is wholly unsatisfactory technically.  So I am hoping some of the more educated folks here could help me understand if in fact there are some sound technical reasons why an active preamp between my CD player (or iPhone) and my VTA70 could improve the sound quality.

I am an EE and have extensive electronics and signal processing background but limited audio experience, hence the uncertainty. From what I can see and measure, the first stage amplifier in the VTA70 already seems to be very high quality, and the input impedance and signal level appear to be a perfect match for many modern input sources. So no preamp gain is required, only attenuation, which a stepped attenuator seems to accomplish nicely.

Based on traditional noise figure analysis, it would seem that the addition of an active preamp in this case could only degrade the sound when compared to the direct path. But so many very smart people are using them, and sometimes paying huge amounts of money for them, so I am wondering if I am missing something.

Thank you in advance for any inputs.

corndog71 wrote:I've tried passive attenuators vs active preamps and the latter always come out on top.  With a passive pot you might gain some clarity but lose dynamics.  A good active preamp like Roy's SP13/SP14 bring the music to life in a way that no passive attenuator can.  

peterh wrote:

The main benefit of a preamp is - tonecontrols and possibly rumble and/or scratch filters.
There is also a possibility to adjust levels from various sources giving similar volumes ( this is not generally implemented but may be done by soldering fixed resistor networks )

tape output ( not that tapes are in widespread use but rather for connecting a typically
low-impedance computer ) might be of value if one wants to digitize any material.

Buffered line output will give the opportunity to locate the poweramp further away, closer
to the speakers.

To summarize , a preamp will give more functions, if they are needed is up to You!

Peter W. wrote:
A well-designed power-amp will easily deliver up to 10X its RMS rating for enough time to handle a music peak without clipping.

rjpjnk wrote:
So I have been under the impression that any input voltage greater than 1.5 peak would cause the output to clip. Is this not correct?

j beede wrote:

I believe you said that you saw 50 Vrms(p-p) at the output before continuous clipping was apparent with a 1 kHz, 1.5 Vrms(p-p) sinusoidal input.

Kentley wrote:I know you're looking for solid technical reasons for the use of a preamp, Sir. I'm not the guy for that. I believe that Peter W. may be on a rational path here.
Being a firm believer in the "some things can be measured but not heard, and some things cannot be measured, yet heard" camp, I can vouch for the across-the-board improvement of an SP-14 in the path between my DAC and my ST-120. So does our esteemed audiobill.
Our most sophisticated test instruments remain our ears - providing, of course, that they are well-connected, by patient experience and experiment, to a coherent brain.
And I sense that your ears are well-connected.

rjpjnk wrote:

Peter W. wrote:
A well-designed power-amp will easily deliver up to 10X its RMS rating for enough time to handle a music peak without clipping.

I'm not sure about this part. If this is correct, then it pretty clearly makes the case for a preamp just to get the voltage swings. I'd like to get some confirmation.

I have been under the impression that the VTA70 power amp was basically just a voltage amplifier that produced a bigger copy of whatever you put into it up to the point that it clipped. Maybe the transient response needs to be considered separately.

OK - let me give you a close analogy: Imagine the typical small town out in the flatlands. What do you very nearly always see there? What is the tallest structure around but-for the grain elevators? The water tower. Keep that in mind as we move on.

A while back when I had my VTA70 on the bench I tested the maximum power output as follows.

(I don't have my notes in front of me now, but I believe these numbers are about right from memory)

I put an 8 ohm resistive load on each channel, and connected a signal generator set to 1KHz sine wave to the input of both channels. I connected an oscilloscope across one output channel at a time and measured the output as I increased the input level. The maximum output level I could obtain before I saw clipping was about 25V peak. (Which equates to 38WRMS.) The input signal at this point was at about 1.5V peak. (i.e., 1.06 VRMS)

Sure. Feeding it a sine-wave (single-frequency signal). All good so far.

So I have been under the impression that any input voltage greater than 1.5 peak would cause the output to clip. Is this not correct?

No, and yes. Remember the water tower.  The pumps that fill the tower = the RMS power of your amp. Half-time of the Superbowl, those pumps could not begin to meet the demand when very nearly every toilet in town flushes, and every sink is turned on. BUT - the reserve water in the tower maintains the peak flow and nobody runs dry. THAT is the PEAK demand against the average demand - and that is what the power-supply of a well-designed amp will handle for such peaks. But, yes, should the water in the tower become exhausted, and the demand continue, there will be problems - in an amp, that is clipping. So, as you ramp up power on your bench test into a dummy load using a single-frequency sine-wave signal, you are not allowing the tower to fill. And the amp will clip.

rjpjnk wrote:Ok, so it sounds like you are saying that the transient response may be significantly higher than the max level obtainable is a steady state response. JBeede mentioned this as well. I will attempt to measure the transient response to see "how tall that water tower is". Thanks,

But in addition to this, we need the input signal to also contain these transients. You mention that having a preamp that can swing 10v allows this. But the preamp cannot produce these transients unless they were also present in the original source signal. The fact that the preamp is taking a signal in the 0-2 volt range and amplifying it to the 0-10 volt range does not create anything new.

Do you understand the energy involved in sound?

a) To double the volume, you need 10 times the energy.
b) Every 10 dB = double the volume.
c) If your P/A = 30 Db, and you are operating at 1-watt on average, you will need (theoretically) 1,000 watts for a peak.
d) Your lowest volume sound may be only 1/10 watt (76 dB)

So I don't see how having this additional stage of amplification changes anything.

If your passive attenuator  can give only 2 V into the amp, the amp will never clip, and no reserves will ever be needed. So, if your  speakers deliver 86 dB at 1 watt, and you have a 38 watt amp, the loudest peak you can achieve will be 102 dB. And the softest passages will be close to unlistenable. Remember, you are covering a dynamic range of 40 dB, spanning the loudest passage requiring 10,000 times the energy of the softest.

If you want to be sitting on the 10th row and 30 feet from the Bombard Pipes 2 V is not going to cut it. Your peaks will want to be 116 dB - and, no, that will not damage your hearing.  

solderblob wrote:So far this discussion has been about a CD player or dac directly driving an amp vs inserting a line preamp before the amp.  How about when a phono preamp is used as a single input to the amp using a stepped attenuator?  

Is the phono preamp essentially equivalent to a dac in the sense of this discussion?

dave

Bob Latino wrote:Peter,

Re: > "If your passive attenuator can give only 2 V into the amp, the amp will never clip"

The VTA amps all need about 1.3 volts in for full output. At 2.0 volts IN all the VTA amps will be overdriven and be clipping at least to some extent ... See the photo below of the sinusoidal output on an oscilloscope showing how the output on the top and bottom of the signal is "clipped" off.



Bob

Peter W. wrote:Once again, please note the interpolations.

rjpjnk wrote:Ok, so it sounds like you are saying that the transient response may be significantly higher than the max level obtainable is a steady state response. JBeede mentioned this as well. I will attempt to measure the transient response to see "how tall that water tower is". Thanks,

But in addition to this, we need the input signal to also contain these transients. You mention that having a preamp that can swing 10v allows this. But the preamp cannot produce these transients unless they were also present in the original source signal. The fact that the preamp is taking a signal in the 0-2 volt range and amplifying it to the 0-10 volt range does not create anything new.

Do you understand the energy involved in sound?

a) To double the volume, you need 10 times the energy.
b) Every 10 dB = double the volume.
c) If your P/A = 30 Db, and you are operating at 1-watt on average, you will need (theoretically) 1,000 watts for a peak.
d) Your lowest volume sound may be only 1/10 watt (76 dB)

So I don't see how having this additional stage of amplification changes anything.

If your passive attenuator  can give only 2 V into the amp, the amp will never clip, and no reserves will ever be needed. So, if your  speakers deliver 86 dB at 1 watt, and you have a 38 watt amp, the loudest peak you can achieve will be 102 dB. And the softest passages will be close to unlistenable. Remember, you are covering a dynamic range of 40 dB, spanning the loudest passage requiring 10,000 times the energy of the softest.

If you want to be sitting on the 10th row and 30 feet from the Bombard Pipes 2 V is not going to cut it. Your peaks will want to be 116 dB - and, no, that will not damage your hearing.