Another Phono Amp for MM-cartridges

1. Introduction

In the early 80's the CD-player has been developed with Philips. At those times also shape had been given to a new phono MM-cartridge as the successor of the GP412MKII. It should have become the GP620, but the marketing was interrupted by the advent of the CD-player!
I was the lucky chap who could put his hand on a proto type of the 'GP620'.



When I fixed the cartridge on my old Thorens TD124 with a heavily modified tonearm, I was very disappointed. Reading the documentation much later, I saw that the inductance of this cartridge is 500 mH with a resistance of 800 Ω! My measurements: 535 mH with 816 Ω. It is not so nice to connect this to the phono input of the power amp via a 2 meter long screened cable!

2. Simulation of Cable Connection

In figure 1 is the simulation of the cartridge with a 200 pF cable connected to it. As expected the inductor in the cartridge resonates together with this capacitance. In spite of the rather large series resistance of the coil, the resonance peek is more than 35 dB at 15,4 kHz. This sounds horrible!
Even with a 1 meter shorter cable (1 pF/cm) the resonance is still in the audio range: the level is enhanced with 10 dB at 18 kHz. The resonance frequency is 21,8 kHz.

The alternative is to put the phono pre amp in the player close to the pivot of the tonearm. In case the capacitance is less than 25 pF.


3. The RIAA-correction

The RIAA-correction dictates poles at:

3180 μs (50 Hz),
318 μs (500 Hz) and
75 μs (2120 Hz).

To avoid distortion (because of auditory feed back, tonearm resonances, and rumble), it is wise to add a pole at 20 Hz. This offers a high pass roll off near 6 dB/oct so that a decent op amp (which open loop frequency characteristic falls with 6 dB/oct) could be used very well.


4. The Pre Amp

At 1 kHz, the amp should amplify about 40 dB to match the sensitivity of my power amp. For more than one reason (see other LF-designs on my web site) I choose the implementation with an OPA134.
At 1 kHz its open loop gain is 80 dB so that the feed back is nearly 40 dB. Figure 3 shows the diagram and figure 4 the simulated frequency characteristic.
The output impedance of the OPA134 (open loop 10 Ω) is much smaller than R3, and R1 = 0,01.R3, so that the curve will meet the RIAA-correction within 0,1 dB.
(Measured with record from Carl Lindsröm Gesellschaft m.b.H., Köln.: within 1 dB up till 14 kHz.)

4.1. An Update

To suppress the resonance of the cartridge induction Lc with the capacitance of the connected cable, it could be damped with a resistor Rp in parallel.
If we choose Rp = 2πf.Lc for f = 2120 Hz, and remove C3 (and R4) we get the same RIAA-correction! With Lc = 525 mH and Rp = 6,99 kΩ, f = 2120 Hz. (In fact Rp should be: 6990 Ω - Rc = 6990 - 816 = 6174 Ω, say 5k6.)
Something simular has been done already in 1961 by Tobey and Dinsdale and later by Bob Cordell, not for resonance suppression by the way!

Moreover this load of the cartridge could cause some electro-mechanical damping of the stylus as well. (The right mechanical damping of the stylus is achieved by its 'rubber' hanging.)

Figure 5 shows the effect of Rp at constant velocity of the stylus. Even the rest capacitance of the cable inside the tonearm (25 pF) does not influence the frequency characteristic.



4.2. Final Diagram

To conclude the diagram of the built-in phono pre amp is shown in
figure 7. The frequency characteristic is shown in figure 8. Compared to the original RIAA-curve in figure 4, it is only 1 dB lower over the whole frequency range without any linear distortion.
An HF roll off at about 80 kHz could be established if R4 were enlarged to 10 kΩ.

5. Listening

Compared to expensive moving coil cartridges, the GP620-with-the-new-pre-amp sounds about the same. The compliance is larger and the sound is so clear and transparent that I do not want to change with any moving coil cartridge with their typical idiosyncrasies as worse tracking. Much later (in 2018) when I finished my loudspeaker system: '2 BMRs in een Baffle', the performance of the piano concertoes of Bartók on Deutsche Grammophon 2LP Stereo 2726005 sounded comparable to a CD on my highly upgraded CD player and SSA35 amplifier.

6. Measurements

The frequency characteristic is flat from 20 Hz up to 14 kHz within 1 dB. Over 14 kHz the amplitude increases gradually to +4 dB at 18 kHz,
decreases to 0 dB again at 23 kHz and to -2 dB at 27 kHz.
The harmonic distortion at 315 Hz with a horizontal amplitude of 50 to 120 um is -40 to -35 dB second- and -46 to 40 dB third harmonic.
The harmonic distortion at 315 Hz with a vertical amplitude of 30 to 50 um is about -30 dB second- and -35 dB third harmonic.
Channel separation L ==> R: ~30 dB and R ==> L: ~40 dB up to 10 kHz.
Intermodulation 300 & 3000 Hz (4:1) with 4 - 10 cm/s (peak): -30 to -20 dB.
Intermodulation 1.8 & 2.2 kHz (1:1) with 10 - 25 cm/s (peak): -40 to -30 dB.

Used record: Abtastfähigkeits-Mess-Schallplatte DIN 45 549 from DGG ausgabe 1978 on a Thorens TD 124 record player with modified tonearm.

7. Conclusions

  •   An op amp (read: OPA134) is very well suited for implementing a MM phono pre amp.

  •   The large inductance of an MM-cartridge does not digest a capacitive load of more than some 30 pF.

  •   To avoid capacitive load of more than 30 pF, put the phono pre amp underneath the turntable close to the pivot of the tonearm.

  •   To avoid electrical resonance, the cartridge could best be loaded with a parallel resistor.

  •   No worry about noise with an OPA134 after the resistor loaded cartridge!

  •   If this resistor has the right value, it will replace the 75 µs time constant as well.

  •   This resistor could be connected directly to the cartridge if different cartridges will be used.

  •   Douglas Self underestimates this solution in his book: 'Electronics for vinyl'.

  •   An extra damping of the stylus could take place if the magnetic coupling between the stylus-magnet and the coil is strong enough.

  •   The Audio Technica AT3600L with 400 mH and 410 Ω needs an Rp of 4,9 kΩ for the 2120 kHz pole, so an extra 40,52 kΩ (18 + 22 kΩ in series)
in parallel with the cartridge, will satisfy.

  •   Compared to the 'old situation', the sound of the system has been enhanced tremendously.

8. Some Recordings

To get an idea of the quality of the cartridge and the pre amp, some recordings are presented here. Listen the 3D sound stage and the wet water in the sauna!

  1. Begin of Barefoot Boy
  2. Sauna
  3. A Dixiland band passing by

9. Last Remarks

If you would simulate the cartridge loaded with a cable of, say, 200 pF and a parallel resistor of 5,6 kΩ, you will find nearly the same curves, but it sounds totally different!
Don't ask me why.
Simulations can bring you to splendid ideas but the proof of the pudding is the eating. Always listen to what you are doing!

This pre amp could very well be followed by a BUF634 if the load is too heavy due to a long cable and/or low input impedance of the following system. For details see 'A Solid State Headphone Amplifier' on this web site.

This pre amp could be built for every MM-cartridge! The value of Rp depends on the induction Lc of the cartridge. As long as Rp = 2.π.2120.Lc and the original 75 μs time constant has been removed, it will work!

If the preamp should be used with several cartridges with different inductions (and coil-resistances), the damping resistors could be fixed directly to the cartridges.

People who think that a vinyl-record player outperforms a CD-player, should update their digital equipment. At least the clock should be replaced by a rutgerS'Clock and the DAC should be updated.

May 12 - 2013
june 27 - 2021
Herbert Rutgers.