Tweakers' Asylum Tweaks for systems, rooms and Do It Yourself (DIY) help. FAQ.

---

[ Follow Ups ] Thread: [ Display  All  Email ] [ Tweakers' Asylum ]

---

[ Alert Moderator ]

---

My experience with current sensing MC phono pre's

Posted by Jon Risch (B) on December 4, 2002 at 19:46:39

In Reply to: A current input MC phono preamplifier project posted by RJM on December 4, 2002 at 06:37:01:

I came up with a version of this back in 1979, tried to get it patented, but apparently, due to a 'similar' Shell oil patent on a sesmic sensing circuit for finding petroleum, it was deemed to be too risky to try and patent.

However, I did tweak it out for personal use, and worked with the circuit for years, finding out what did and didn't work so well.
I will now go through as many of the details as I can think of right now, please realize that I have not verbalized about this for over 15 years.

First, several conventional paradigms are turned on their heads with this kind of approach. You need to turn your thinking around, to focus on the CURRENT related aspects of op-amp performance.

For instance, what is important with normal voltage sensing circuits is low voltage noise, and low input offset voltage. With this circuit, low current noise, esp. at LF's, and low absolute input bias current and low input offset current are paramount.

Op-amps such as the 5532/34 are not optimal for this type of application, because they are not low in bias current, or as low in current noise as many other op-amps. In point of fact, FET input based op-amps are a much better choice, as most of them have very low current noise and fairly low bias current and current offset.

Another aspect is that the op-amp sees the feedback resistor as a portion of the total load on the op-amp, therefore, you do not want to exceed it's capabilities in this regard. Op-amps such as the OPA-2604/604, that have excellent load drive capability, and have low current noise and low current offsets are a good starting point.

For the OPA-2604, I would not go below 300 ohms for the feedback resistor, and one saving grace is that the signal level, and consequnet current output demanded from a typical low Z MC phono cartridge, are not that great. In other words, the application at MC signal levels is such that you can effectively load the op-amp as if it were only going to see a very low level signal and consequent current output level. You still want to go easy on the op-amp, they still sound their best when 'lightly' loaded.

The coil resistance of the MC phono cartridge is going to create an input resistance, and so, you are not really operating in true current sensing mode, but rather, the op-amp is providing a virtual ground, and with the addition of the MC cart coil resistance, it is now operating partially in voltage mode, and partially in current sensing mode. The ideal situation in terms of operating primarily in current sensing mode would be a very low resistance MC cart coil, say, 3-5 ohms.

However, there is a fly in this ointment. As you go lower and lower in MC cart coil resistance, the input bias current and the current offset of the circuit become more and more critical. Op-amps with a method for balancing the input currents using an external trimpot can balance this out, or one can use an equivalent resistance in the path between the postive input terminal and ground. A trimpot here will also work, as long as it is bypassed with a resistor of higher balue, so that if the pot wiper loses contact, even for a split second, that the op-amp does not lose it's ground connection completely.

As an example, if you have a 10 ohm MC cart coil resistance, then a 50 ohm resistor to ground, with a 50 ohm pot connected in fail-safe rheostat mode can be used to trim out any offset fairly low. These are then bypassed with a film cap to lower the AC impedances.

Another point is that as you go above about 100 ohms of MC cart coil resistance, this cirucit is starting to devolve into voltage mode more and more. The higher impedance of the MC cart coil creates a much larger input resistor for the total circuit.
The other down side of a higher MC cart coil resistance, is that they do not tend to respond to current sensing loading very well. That is, they tend to sound as if they are rolling off in the HF's, due to a lot more electrical damping than they really want to see. This sounds like the MC cart is "underwater", as it is highly overdamped.

I have found that given the real world limitations of op-amps, and what most MC phono carts like to see as a load, that the optimum range of MC cart coil resitance is from about 5 ohms to about 35-40 ohms.

Now for some real neat and tweaky stuff!

Since the current through the feedback resistor is exactly the same as the current being sensed at the virtual ground point, and therefore, is identical to the current coming from AND FLOWING THROUGH the MC cart coils, you can place a "frozen" MC cart assembly/coil in series with the feedback resistor, and via the feedback network, cancel out much of the non-lineraties of the MC cart coil core. Most MC carts have a ferrous core that the coil is wrapped on, and by placing another cartridge in the negative feedback loop, you can cancel out most of the magnetic distortions that tend to occur within the cartridge, mostly odd order harmonic distortion, which tends to compress and harden the sound a bit.

Placing this frozen (that is, the moving assembly is held still in place by flowing two-part epoxy all around the coil assembly portion) cartridge in the feedback loop also tends to cancel out (to a 1st approximation) any other non-linearities within the cartridge, even if it does not have a magnetic core for the coils.

You need to keep the channels straight, that is, make sure that the proper channel is used to cancel out the proper channel of the cartridge.

This techinqque can absolutely transform an MC based playback system into a world class device. Details emerge from the music, ambience galore is recovered, subtle musical lines and details, that singer in the background you never even knew was there, the proverbial lifting of many veils of obscuration occurs.

However, the downside is cost. You literally have to 'burn' an entire MC cartirdge assembly to do this. IF you happen to have a used MC cart the same as the present one, or one with a broken cantilever or stylus, then implementing this just as easy as making a nice tight signal path for the connection of this MC cart assembly into the feedback loop with the overall signal path a short as possible, so that you do not introduce any instabilities.

Due to the minute amount of inductance present in even a very low impedance MC, you will need to parallel a small (10-47 pf) cap in parallel with the MC cart coil, so that the extreme HF's do not get unstable.

Of on another tangent:
Another tweaky way to implent the current sensing circuit, is to use a balanced version of the virtual ground front end. You would provide a virtual ground for each side of the MC phono cart outputs, the so-called "hot" and the so-called "ground". In reality, most all phono carts are balanced devices, they have an output that does not REQUIRE a ground reference in order to pass along the signal, and so, you can run EACH LEG into a virtual ground current sensing op-amp.

In theory, you could take a 2nd frozen MC cart assembly, and place it in the feedback loop of the second virtual ground input, and gain that advantage as well, but I think that this probably exceeds the bounds of most folks resources. When I was experimenting with this circuit, I had access to all the used MC carts I needed to try out even this over the top implememntation, and while it did gild the lily, the single frozen cart provided 90-95% of the improvement.

If you do NOT use a frozen MC cart assembly in the feedback loop, then use the balanced double virtual ground version. It has very high hum immunity, and is slightly cleaner than the single-ended version, and is worth doing as the extra op-amp is cheap. You run the two outputs of these into a differential input, which ideally would consist of a full instrumentation style (3 op-amp confiuguration) differential amp, although, since the source impedacnes are defined and under user control coming out oif the virtual ground input op-amps, you can "get-by" with a single good op-amp operated in differential mode use.

Some other general comments re this kind of MC phono circuit.
Think of the virtual ground portion of the phon stage as a seperate circuit, and than add-on/create a normal RIAA EQ stage, which can be one more op-amp, or more ideally, two op-amps to provide the full amount of gain neede (even after the current-to-voltage conversion, the signal is still rather small, about as strong as a typical hot MM cart).

One other point. At these low signal levels, the power supply noise should be as low as possible, and most IC regulators are too noisy to provide a power supply that is as clean as is needed.

You can get around this by placing an RC filter after some conventional regulators, and using a regulator of a slightly higher voltage to account for the losses through the RC network.
A smallish resistor, and a huge amount of capacitance, along with plenty of film bypasses will do the job.

I hope that this helps shed some light on the more esoteric details of implementing a current sensing virtual ground MC phon preamp circuit.
If there are any Q's re the above, or further tweaks, I will answer the best I can.

Jon Risch

This post is made possible by the generous support of people like you and our sponsors:

Atma-Sphere Music Systems, Inc.

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---