Moreplay Lpad impact on circuit impedance?

[hmbscott]

The discussion on the Fixed attenuator for balance control? thread got me to thinking about the -6dB Lpad I installed on my Moreplay. I installed the Lpad using R1 = 8.66K and R2 = 75K. These are values recommended by Goldpoint for their 10K stepped attenuator, which is the vol pot unit I'm using. Using this equation for calculating the DC impedance: Z2 = R1 + 1/(1/R2 + 1/Z), where Z is the 10K of the attenuator, I calculate Z2 = 17.4K with my existing Lpad+vol pot. This was something I was unaware of until the recent thread regarding fixed attenuators.

Questions:
1) Is my calculation correct? I assume it's only valid for DC, but it's all resistors, so maybe?
2) Does the shift from 10K to 17.4K make any difference?
3) In any case, I want to increase the Lpad attenuation and will be changing the resistors, so should I pick values that so that Z2 = 10K in resulting circuit? Values I calculate would be R1 = 7.68K, R2 = 3.01K, & -13dB attenuation.
4) Or, should I use GP's recommended values of R1 = 38.3K, R2 49.9K resulting in Z2 = 46.6K, & -15dB?

I'm concerned that upping the vol pot circuit resistance from 10K to 47K might be a problem, but have no idea if it's actually important or not.

Thanks, Scott

[Paul Birkeland]

It's a good idea to keep the total impedance of what loads the circuit to 10K, mostly because letting it increase will mess with the output impedance, which means you really don't want to use Goldpoint's recommendations. 

For a -15dB attenuator, that's a 1:0.17 ratio, which you could get from an 8.2K resistor in series with a 1.8K resistor.  Putting a 2.2K resistor across a 10K pot will make the pot effectively into a 1.8K resistor, so use an R1 of 8.2K and an R2 of 2.2K.

Resistors by definition are (supposed to be at least) non-reactive components, so a 10K resistor is 10K at DC and AC.  When you get into frequencies much higher than are typically found in audio equipment, that assumption has to be treated a lot more carefully, but it's fine for padding a pot. 

[Paul Joppa]

Yes, you should be concerned ...  but not too much!    :^)  I did the numbers:

Almost all preamps have the volume control at the input, where high impedance is a virtue. Placing an attenuator at the output increases the output impedance, which is a bad thing - it limits the length of cable and the input capacitance of the power amp that be driven without loss of high frequencies. Goldpoint is implicitly assuming that the volume control is at the preamp input.

But Moreplay has the volume control at the output.  In fact, one of the reasons it uses a power tube is the 10K pot loads down the tube, requiring more drive current than otherwise necessary.  The L-pad really should be at the input of Moreplay or at the input of the power amp that follows it.

Here are a few notes:

* As PB suggests, the effect on the Moreplay output impedance is relatively small (30% or less?) as long as the L-pad input impedance preserves the design value of 10K. It's even smaller if the attenuation is less than 2dB or greater than 12dB. This will also preserve the sonic character of Moreplay, and is recommended if you want to keep the pad there.

* The input impedance of Moreplay is basically just the balance control, whose impedance varies from (roughly) 20K to 100K depending on the setting. If you leave it centered it's about 70K; and you can design an L-pad for 70Kohms - it won't track quite the same, but the balance control will still work properly over the full range. Place the L-pad before the balance control. This is my best-sound recommendation.

* (If you replace the balance control with a 100K Alps, as in the upgrade kit, its impedance will not vary with setting. You can design for 100K and it will track as intended.)

This all got very messy to calculate - you have to consider all four impedances. looking into and out of the input and output of the L-pad.

[hmbscott]

Paul & Paul, Your advice is much appreciated! I will place the new Lpad upstream of the bal pot.

FYI: My balance pot is a Goldpoint 24 step unit populated with RC55 4K3 1/4W 0.1% MF resistors, adding up to a 103.2K linear pot, which appears to me to be identical, circuit wise, to the standard pot. BTW, I replaced the kit's 33K resistor on the bal pot with a RC55 31K6Ω 1/4W 0.1% MF. So my mid point bal pot impedance is 71.2K.

An Lpad feeding the Bal Pot is supper confusing to calculate! Hoping I got it right. I'm shooting for approximately -12dB of added attenuation with the Lpad.

The R1, R2 of the Lpad I calculate to preserve the 71.2K of the pot with ~ 12dB added attenuation:  R1 = 53.6K and R2 = 23.7K. This Lpad appears to add between -11.3 and -14.8 dB of loss, -12.1dB when centered. The loss looks pretty flat, decreasing slightly for the nominally unattenuated channel when moving off center, while increasing for the attenuated channel, which seems fine.

This calculation assumes the grid impedance is infinite, while not true, it must be high otherwise the tube would be threatened by excessive grid current. The tube data sheet implies grid impedance is greater than 100K - 500K based on the maximum allowable gird drain resistor. Also assumed is that impedance of the driving device is zero. My driving devices are an Eros II (4k impedance) and an Aries II DAC (625 ohm impedance). Both are relatively low compared to the Moreplay's nominal 70K impedance. I estimate both can be neglected.

[hmbscott]

I'm still grappling with how much LPad'ing I want. And it occurred to me that one definition of "just right" would be enough to drive my Stereomour to max volume and perhaps another 6dB past that. That got me to looking at the Stereomour specs. One problem is that I can't seem to reconcile the specs with each other.

Stereomour Specs:
• Power: 3.5 Watts per channel
• Sensitivity: Full power at about 0.4v (2A3)
• Gain: About 8 at the 8 ohm tap (21.5 dBV)

Sensitivity of 0.4V with gain of 8 suggests full power is 1.3W onto 8 ohms. On the other hand, 3.5W max power at 8 ohm and 0.4V sensitivity suggests the gain is about 13. Regarding the gain spec, I calculate a gain of 8 is only 9dBV or 18dB(power), not 21.5 dbV, which would be a gain of nearly 150. Although, 21.5dB(power) would be a gain of about 12. These seeming inconsistencies have me wondering what's ground truth Stereomour specs.

The spec that I really need to know is the Stereomour sensitivity. Is it actually 0.4V? And, would an appropriate max voltage output for my LPad'd Moreplay be about 0.8V to insure I don't overdrive my Stereomour by more than ~ 6dB. By the way My Stereomour has no volume control, but rather is wired to mimic the volume at max.

While I'm asking dumb questions, I have another problem regarding my sources, which are an Eros II and a DAC. I just measured the DAC's output of using a 60Hz signal. It output 2.23 Vrms (spec is 2 Vrms). While I was at it, I measured my Moreplay output (with it's current LPad) at Volume max driven by the DAC. The Moreplay output 5.06 Vrms, presumably way past what I would want to feed the Stereomour.

My turntable is equipped with a 5mV cartridge (Ortofon 2M Bronze) and given the eros's 50dB of gain, I calculate that the eros should be outputting 1.58 Vrms. The weird thing is that if I play a track on my DAC that I also have on vinyl, the track played on vinyl always sounds about 5dB louder than what's coming from the DAC (determined by ear because I have to turn the Moreplay vol down 5dB to match levels). Doesn't that imply the signal coming from the Eros is 5dB higher than the DAC? Which if true implies the Eros's gain is also 5 dB more than 50 dB or the cart output is 5 dB greater than 5mV ... or what? The reason I care is that I want to know the input voltage from my DAC and phono preamp driving my Moreplay so I can get the LPad right. This volume difference seems to suggest the Eros drive is a lot higher than the DAC.

[Paul Joppa]

The Stereomour II spec is incorrect. The actual sensitivity is 1.33 volts rms for full power and the gain is 4 (12dB).

DAC volts are not the same as cartridge volts, unfotunately. DACs are rated for maximum output (no headroom) but cartridges are rated for average signal level. Peak signal voltage is typically around 14dB (5 times greater) for well-made recordings, though there is a lot of variation.

Incidentally, decibels are power numbers; which is why dBVolts is 20*log10(voltage ratio) while dBWatts is 10*log10(power ratio)

[hmbscott]

Very much appreciated! Oops, somehow I got volts vs power dB backwards.