DAC/CD Player output levels

hi, and yes this is a familiar CD issue which many suffer from but many I suspect are blissfully unaware of and blame the format more than perhaps they should.

I successfully used on-line attenuators on my Vintage Luxman Amp a few years ago, quite possibly from the same supplier (Lancashire based from memory), I had them custom spec'd but cannot recall the value.

They made the sound more relaxed, more open and even and less closed in. The overall dynamics improved albeit initially the snap and bass punch appeared to suffer. I say initially as once used a little more I stopped noticing this. Why? In my opinion I was accustomed to a certain sound but once tuned in I forgot about it.

My current CD player has an output voltage quoted as 2.3VRMS. My current amp has CD sensitivity rated at 565mv with other inputs at 311mv. in broad terms not a million miles off yours although I can attenuate my inputs by 10db which I do, in the region of about 5db I suspect.

I would recommend you persist in your listening I suspect they will grow on you and probably run in too.

good luck

JamesD wrote: The technical bit....  To the best of my knowledge the attenuator is simply a potential divider network. This means extra series resistance is being put into place between the DAC output circuit and the input circuit of the amp. The amp's input circuit comprises of a 22k-ohm resistor to ground, followed by a dc-blocking circuit comprising of a 10uF capacitor and a 82k-ohm resistor to ground. The 'centre tap' of this R/C network is fed into a simple op-amp buffer stage, which then drives further circuits in the amp. I wonder if the extra resistance that the attenuator is putting into the circuit is disturbing the operation of this dc-blocking circuit, i.e. it is filtering out more of the lower frequencies. Any help/advice is welcome...

A series resistor into an input which will have some (unknown) input capacitance will form a low pass filter - the opposite of what you're thinking. How much input capacitance? We don't know, and if there is no physical capacitor, then it is the input capacitance of the op-amp, which won't be much. If there were a physical capacitor there would be some high frequency roll-off, and in my experience can impact the bass (but I'm not going to write an essay on that right now).

As for the brightness without the attenuator: there comes a signal level with all op-amp inputs which saturates the input long tailed pair (LTP) - daft as it sounds but it is true. The input signal then has to drive the voltage amplifier stage (VAS) and its associated miller capacitor, which isn't a very nice load, and can result in the brightness described.  So 2.5 volts is too "hot" for it, and 600mV is OK, nothing new there.

An attenuator is simply an L-pad comprising a series resistor followed by a resistor across the signal.

The resistance (and hence impedance) should be kept as low as possible and if we look at the standards for equipment matching, 22k is the recommended minimum load resistance so we can make the series resistor 22k.

The value of the resistor across the signal has to be calculated to give the required attenuation taking into consideration the load - but the load can vary between different amps. If we took the 22k standard, then to get -10dB (divide by 3.16) the total value of the resistance across the signal would be (Rin/Rp) + 1 or 10k which is near enough.

So we should then take away the amp input impedance which is 22k||82k or 17.3k, which means the resistor across the signal would need to be 23.6k, or 22k as the nearest obtainable value.

The trouble with attenuators is if the load isn't stated, you have no idea what the attenuation will actually be in practice.

I will assume the attenuation of this 10dB attenuator is into "free air" and if so the amplifier input impedance appears in parallel with the resistor across the signal. And therefore it isn't 10dB anymore but a higher figure.

Complicated isn't it?

Thanks Drewan. The amp does actually have a balanced input which could work well with the Majestic.

Thanks for the explanation Graham. It is complicated but I do understand the maths. And yes, having another "unknown" RC filter working the "opposite way around" on the input could harm the higher frequencies. Hence I'm wondering whether to do some work inside the amp in the form of a dedicated circuit for the DAC input in the form of a potential divider directly connected to the non inverting pin of the buffer circuit (in place of the 82k resistor to ground). All assuming that the attenuators don't suit me. Or I make a unbalanced to balanced adapter and use the balanced input on the amp...