Cartridge Loading.....Part II

That is not the case in this instance. 

This gives me confidence in my belief that the load on the cartridge is responsible for the sonic changes that I reported above.

I don't doubt that it could be affected. Loading the cartridge causes the cantilever to become stiffer (more work is being asked of it and that has to come from somewhere: the cantilever is thus harder to move)- and thus can introduce the possibility that even though the bandwidth of coil is unaffected, the mechanical aspect of the cartridge will be affected by that added stiffness- possibly making it less able to respond to higher frequencies.

The distinction between virtual and actual ground is a subtle one.

When an opamp inverting input is used, the virtual ground is extremely similar to a real ground as the action of negative feedback makes it so. The difference is in the error term- i.e. the output voltage divided by the open loop gain of the amplifier, together with any impairments added by the amplification system.

This difference can be extraordinarily small, so arguing that the virtual ground is not a real ground is largely facile. 

As far as the source is concerned there can be essentially negligible difference between a real ground and a virtual ground.

Claiming, arbitrarily, that it is prima facie audibly different as far as the source is concerned is not reasonable. 

To put this in context. The error term is a function of the opamp gain bandwidth product, open loop DC gain, and the internal sources of distortion,  both the intrinsic input non linearity and the output stage non linearity. With high gain bandwidth and high open loop DC gain the error voltage is extremely small in the audio band, and with low intrinsic open loop distortion at the signal levels present at the non inverting input when negative feedback is applied the resultant distortion component of the error signal is also extremely small.

These facts make the virtual ground , when opamps of sufficient quality are employed, extremely close to an ideal ground, in fact just about as close to an ideal ground as any "real" ground that you can actually construct.

A copy of a post I placed a short time ago...

 

I'm in the process of having a phono preamp made manufacturable by a well known audio company. The first thing they insisted upon was that the design would be implemented in SMD- purely for manufacturing purposes. This had it's own unique set of difficulties- for example, some of my preferred capacitors were not available on SMD, so a search for suitable items of a different technology (ceramic MLCC C0G vs Polypropylene film) was needed, together with a proof that the distortion characteristics were essentially equivalent.

The other aspect was that the preamp design investigated two possible implementations, either a voltage mode input, or a current mode input, and with the 75uS RIAA TC implemented in the first stage.

I ended up with a voltage mode input, as when designed for equivalent gain and RIAA characteristic there is, generally, no measurable or audible difference.

However, there are indeed differences in operation and implementation.

To put this somewhat in context, I designed a number of "transimpedance" amps while at ADI, most notably the AD846, together with a number of conventional opamps, so I am familiar with the concepts. The AD846 was designed to have almost perfect current conveyance properties and could be operated open loop as a transimpedance amplifier. Most opamps/amplifiers use negative feedback to achieve this goal, or have high distortion levels if operated open loop.

As I don't want to make this into a "white paper" I'll try to be brief.

1. A phono cartridge is a voltage generator (Vs) with an output impedance which is mostly a resistance in series with an inductance (R+Ls). This can be converted into the Thevenin equivalent current source- which is a scaled current (Vs/(R+Ls)) with the output impedance in parallel to ground.

If you take this current and drive it into a virtual ground, which shorts out the shunt components, and convey it to the load resistance (Rl) then the output voltage is Vs*Rl/(R+Ls). If Ls is small then the gain is completely dependent on the series resistance of the cartridge and will vary from cartridge to cartridge, and if LS is large there will be a HF roll off.

Any shunt capacitance will be essentially ignored.

If there is a resistor added in series with the virtual ground, then the current is shared between the equivalent shunt components and the series R, and the gain becomes even more variable.

Voltage mode lacks this complexity. Provided the load impedance is relatively large compared to R the gain is easy to determine.

However, the shunt capacitance is not ignored, but for MC cartridges the inductor is so small that all parasitic capacitance values are irrelevant, assuming the load R is small enough to damp out the LC resonance. 

Current mode does not experience this resonance, but correct loading Rs plus a phono stage with a suitably good ultrasonic overload margin will take care of the potential problem.

MM cartridges into a virtual ground are not rational unless there is a very large resistor in series with the virtual ground. This is because the R is large but L is even larger.

 

2. In voltage mode the input stage is non-inverting and will experience potentially substantial deviations about ground. For many opamps, particularly earlier generations than the most recent ones, this voltage will cause increases in distortion due to the operating conditions of the input transistors being changed. 

This distortion can be highly sensitive to the source impedance and the input signal level.

In the transresistance mode the virtual ground does not move, removing this source of distortion.

Modern audio IC opamps are generally designed with this problem in mind and exhibit negligible changes in distortion when operated in non-inverting mode.

These are the two main aspects of current versus voltage mode inputs.

There are other, more obscure, aspects.

I chose voltage mode, but also chose appropriate opamps to minimize the down side of the choice.

The bottom line is, when properly designed both current and voltage input designs can have equivalent performance from the perspective of distortion, frequency response etc.- but the voltage input design is more predictable and easier to specify. 

The voltage input design produced has very precise gain, extremely precise RIAA compliance and unmeasurably low distortion.

 

 Report this

Dear @mijostyn  : After reading the Wyn posts are you still thinking of the current phono stage superiority over the voltage designs PS you supported?

 

R.

Dear friends : For some of you that read for the first time the Wyn posts here next I pasted what he posted several years ago in other cartridge loading thread:

 

"" No, I did not design the AD797. That was Scott Wurcer- a colleague at ADI and, incidentally, for whatever it’s worth, also an ADI design fellow. However, I know the design quite well.
He and I were colleagues in the opamp group in the 80s. He focused on high performance relatively low frequency opamps such as the AD712 and then the AD797, amongst others.
I focused on high performance high speed amps like the AD843, 845 (at one point an audio darling), 846 (also a transimpedance design with some very interesting design aspects that I gave an ISSCC paper on) etc. etc. mostly using a complementary bipolar process that I helped develop that I believe was also used in the AD797. I also did things like designing the FET based AD736/737 RMS-DC converter and others.
I moved on to more RF, disk drive read/write, GSM, CDMA etc. transceivers, signal processing, PLL and DSP designs. ""

 

and here somerthing that he forgot to mention and that comes in that " old " thread that shows that that " myth " of tracking problems due to cartridge loading changes is a lie and nothing more:

 

""" heavy resistive loading you state could be definitively true- certainly not on tracking which is demonstrably false based on IM tests on tracking performance that I have incidentally performed as a function of load. While mechanical impact does occur as a result of electrical load- there is some back emf necessarily generated by the signal current that affects the mechanical motion, but a quick back of the envelope calculation using Lenz’s law and the 10uH cartridge suggests a 2 orders of magnitude difference between the generated signal and the back EMF for a 100 ohm load at 20kHz- certainly not enough to cause tracking issues. """

 

 

R.