[SF]

Hi everyone;

I was making some measurements last night at the speaker output of my 14BST last night at different volumes, and I had some surprising results.
To summarize, at about 11:00 clock on volume, my voltmeter was showing around 8-9 volts in the AC mode (without eliminating DC by the cap circuit). This was the max I went because I live in a condo, and this was very loud. Let's suppose at 12:00 clock the Vrms goes to 12 volts for ease of calculation. On transients the indicator would jump, but my volt meter doesnt have the resolution to say anything about Vpeak. So I assume I am measuring Vmean or Vrms. My speakers are nominal 8ohms, but they dip to 4-5 ohms for much of the frequency range (B&W804S). This calculates to 1.5 Amps at 8 ohms, or 3 Amps at 4 ohms. The calculated power is only 18 Watts at 8 ohms or 36 Watts at 4 ohms! I do not have an SPL meter, so I can't tell you what the sound level pressure was, but the music was LOUD (wife and kids ran into the bedroom and closed the door). Now, if the volume gain is linear, at full volume, I would be using 36 Watts. My Amp is rated at 600 wpc, so this doesn't make sense. The volume gain is likely to be logarithmic, but I wonder if anyone knows at what rate the power output changes either at the pre-amp level or at the amp level. According to the schematics volume is controlled by:"10K (log)" variable resistor, but that still leaves the question: if I go from 9 O'clock to 12 O'clock, how many fold have I increased the volume? How about from 12 O'clock to 6 O'clock?

Also, on the DAC stage of the BCD-1, code 253 from the remote sets the gain to -3dB. Does anyone know how this is achieved? Will it be sonically superior to have 0dB or the -3dB? The latter gives you more play on the volume control, which is nice when you are adjusting volume with the remote.

If anyone has insight into these questions, I would very much appreciate your input.

Thank you for reading.

Shafie

[James Tanner]

Hi everyone;

I was making some measurements last night at the speaker output of my 14BST last night at different volumes, and I had some surprising results.
To summarize, at about 11:00 clock on volume, my voltmeter was showing around 8-9 volts in the AC mode (without eliminating DC by the cap circuit). This was the max I went because I live in a condo, and this was very loud. Let's suppose at 12:00 clock the Vrms goes to 12 volts for ease of calculation. On transients the indicator would jump, but my volt meter doesnt have the resolution to say anything about Vpeak. So I assume I am measuring Vmean or Vrms. My speakers are nominal 8ohms, but they dip to 4-5 ohms for much of the frequency range (B&W804S). This calculates to 1.5 Amps at 8 ohms, or 3 Amps at 4 ohms. The calculated power is only 18 Watts at 8 ohms or 36 Watts at 4 ohms! I do not have an SPL meter, so I can't tell you what the sound level pressure was, but the music was LOUD (wife and kids ran into the bedroom and closed the door). Now, if the volume gain is linear, at full volume, I would be using 36 Watts. My Amp is rated at 600 wpc, so this doesn't make sense. The volume gain is likely to be logarithmic, but I wonder if anyone knows at what rate the power output changes either at the pre-amp level or at the amp level. According to the schematics volume is controlled by:"10K (log)" variable resistor, but that still leaves the question: if I go from 9 O'clock to 12 O'clock, how many fold have I increased the volume? How about from 12 O'clock to 6 O'clock?

Also, on the DAC stage of the BCD-1, code 253 from the remote sets the gain to -3dB. Does anyone know how this is achieved? Will it be sonically superior to have 0dB or the -3dB? The latter gives you more play on the volume control, which is nice when you are adjusting volume with the remote.

If anyone has insight into these questions, I would very much appreciate your input.

Thank you for reading.

Shafie

Hi Shafie,

I will get a explaination for you from engineering on your volume control question - that ones over my head!

On the CD Player gain question.. it is done in the digital section of the CD Player.

james

[SF]

Thank you James.

Re BCD-1 DAC -3dB gain, is there an anticipated loss in sonic performance with this setting?

Shafie

[James Tanner]

Thank you James.

Re BCD-1 DAC -3dB gain, is there an anticipated loss in sonic performance with this setting?

Shafie

Hi Shafie,

Well now you've really opened a can of worms  depending on who you talk to.

The problem with digital volume control is that you lose resolution as you reduce volume. The better chips today claim you can reduce volume substantially before there is any affect on performance so the 3dB reduction we allow is erring on the side of caution.

james

[SF]

Great. That's good to know.

Thank you very much James.

Shafie

[bummrush]

so how do the burr brown ones fit into all this,or is a case of implemention,

[James Tanner]

so how do the burr brown ones fit into all this,or is a case of implemention,

Hi,

Sorry not sure what you mean by 'burr-brown ones?'


james

[mcgsxr]

I believe he is referring to Burr Brown DAC chipset, not sure if that is what is used in your products or not.

[James Tanner]

I believe he is referring to Burr Brown DAC chipset, not sure if that is what is used in your products or not.

HI,

We use Crystal DAC's and the BDA-1 External DAC uses a Burr Brown SRC (sample rate converter).

james

[James Tanner]

Hi everyone;

I was making some measurements last night at the speaker output of my 14BST last night at different volumes, and I had some surprising results.
To summarize, at about 11:00 clock on volume, my voltmeter was showing around 8-9 volts in the AC mode (without eliminating DC by the cap circuit). This was the max I went because I live in a condo, and this was very loud. Let's suppose at 12:00 clock the Vrms goes to 12 volts for ease of calculation. On transients the indicator would jump, but my volt meter doesnt have the resolution to say anything about Vpeak. So I assume I am measuring Vmean or Vrms. My speakers are nominal 8ohms, but they dip to 4-5 ohms for much of the frequency range (B&W804S). This calculates to 1.5 Amps at 8 ohms, or 3 Amps at 4 ohms. The calculated power is only 18 Watts at 8 ohms or 36 Watts at 4 ohms! I do not have an SPL meter, so I can't tell you what the sound level pressure was, but the music was LOUD (wife and kids ran into the bedroom and closed the door). Now, if the volume gain is linear, at full volume, I would be using 36 Watts. My Amp is rated at 600 wpc, so this doesn't make sense. The volume gain is likely to be logarithmic, but I wonder if anyone knows at what rate the power output changes either at the pre-amp level or at the amp level. According to the schematics volume is controlled by:"10K (log)" variable resistor, but that still leaves the question: if I go from 9 O'clock to 12 O'clock, how many fold have I increased the volume? How about from 12 O'clock to 6 O'clock?

Also, on the DAC stage of the BCD-1, code 253 from the remote sets the gain to -3dB. Does anyone know how this is achieved? Will it be sonically superior to have 0dB or the -3dB? The latter gives you more play on the volume control, which is nice when you are adjusting volume with the remote.

If anyone has insight into these questions, I would very much appreciate your input.

Thank you for reading.

Shafie

Hi Shafie;
 
A volume control generally works a logarithmic taper with 10% of the final Voltage value reached at 50% rotation, so a small adjustment at low volume may only make a difference of a few tens of miiliVolts.  At volumes over 12:00 or 1:00 on the control, however, it can make a difference of several Volts. 

Another thing to keep in mind is that the Voltmeter is broadly measuring the average Voltage at the output terminals of the amplifier.  Generally, there is a substantial difference, 10-20dB, between average and peak levels in recorded music.  Even a 10dB difference is over a factor of 3 in Voltage, and 20dB is a factor of 10.  Thus, if you are measuring roughly 10-12 Volts average, the peaks could be between 35-120Volts! 

Keep in mind also that power is calculated as the square of the output Voltage, divided by the speaker impedance.  Thus, a doubling of Voltage means a quadrupling of power, and a multiplication of 10 in Voltage means a factor of 100 in power!  If you see the red clipping LEDs flashing that means you are reaching over 600 Watts on peaks, (in fact, on musical peaks, it would be well over 750 watts). 

In general, a rotation of one 'clock position' on the control, (from 11:00 to 12:00, say), may give you a 6dB increase in volume.  That means a quadrupling of power, so multiplying 4X4X4X4X4X4X4 eventually becomes a LOT of power, even if you start at a few milliWatts at lower levels.
 
I hope the above is helpful, but please let me know if you have any other questions.  Thanks for choosing Bryston!
 
Sincerely,
Chris Russell
Bryston Ltd.

[SF]

Dear Chris and James;

Thank you for your responses. Now my measurements make sense. The numbers have important implications. The 804S is rated at 200W unclipped. If transients are 10-20 dB higher than average, that would suggest, taking the more conservative 10dB rating, that unclipped reproduction requires 9x200 Watts = 1800 Watts. That implies that my 14B ST is likely to clip before I even come close to the 200W unclipped program of the 804s. They say you can never have too much power and this certainly proves it. To do it the other way, for unclipped signal, the average power usage, with 10dB head-room for dynamics, would be 600Watts / 9= 66 Watts! If at 12 O'clock I'm at 10% voltage, that would 10% x (SQRT(600Watts x 8 ohms)) = 7 Vmean or 6 Watts. At 1:00, double voltage and quadruple power, it would be 24 Watts, at 2:00 it would be 100 Watts. This would be past the 66Watts + appropriate head room for transients (taking the 10 dB conservative increase for transients). According to these calculations, anything past 2:00 clock would put the Amp into clipping range -- way before my 200 Watt speakers give up. Now, if you want full head room for proper representation of transients using 20dB rating, that translates to (20dB = 100-fold power increase) 600 Watts / 100 = 6W! That is the power at 10%Vmax, which is accomplished at 12:00 clock. So for my amp/speaker combination, going past the 12:00 clock my lead to compression of transients and unfaithful reproduction of dynamic music. This is very important information, and it shows that to fully utilize my current speakers, I have a long way to go in terms of power (somewhat surprising that the 600 wpc amp would clip before the speakers would). That means 28BSST(2) before I even think of the 500W B&W802Ds, which theoretically could never achieve their full potential with any available Amp design!

Alright, this was a very valuable exercise.

Again, let me thank you both for taking the time to respond with hard numbers to crunch. I appreciate it vey much -- another reason to love Bryston.

Best regards,

Shafie