04-03-2017, 11:34 PM
As I understand it, the self-capacitance of a ceramic cartridge in conjunction with the load resistance creates a high pass filter (differentiator). If the load resistance is high enough, say 2 M or more, then the turnover point is typically below the audio band, or at least towards its bottom end, say around 50 Hz. This is the basis for high impedance loading of such cartridges. With mechanically self-equalized (for the RIAA curve) cartridges (which would have included the majority by the late 1960s), no further equalization is then needed.
If the input resistance is much lower than 2 M, then the turnover point will lie in the audio band. This may be corrected by configuring the input amplifier as a low pass filter (integrator), with the same turnover point as the high pass filter created by the combination of the cartridge and the input resistance. This was the basis for using capacitive feedback in the input amplifier. Again with self-equalized ceramic cartridges, no additional equalization was needed. Clearly, the turnover point of the amplifier response curve must be related to both cartridge capacitance and the input resistance. As an example, the Quad 33 control unit used this approach for the C1 (ceramic) position of the disc input amplifier. The input resistance was 100 k, and the C1 input was said to be suitable for ceramic cartridges with self-capacitance in the range 450 to 900 pF, which I suspect covered the majority at the higher quality end of the range. I haven’t done the numbers, but I’d guess that the circuit may have been set for the geometric mean of the capacitance range (636 pF) with not-to-significant errors at either end of that range. H.P. Walker did something similar in his 1971 design (Wireless World (WW) 1971 May p.221ff), in this case with an input resistance of 200 k and the turnover point arranged to suit a cartridge self-capacitance of 600 pF.
An extensive commentary on ceramic pickup matching and equalization was provided by R.J.C. Burrows in a series of articles in WW, as follows:
WW 1970 February p.56ff: Ceramic pickups and Transistor Pre-amplifiers – Are they incompatible?
WW 1971 July p.321: Ceramic Pickup Equalization 1 – Myths against maths and measurement.
WW 1971 August p.379ff – Ceramic Pickup Equalization 2 – Practical low-impedance circuits.
All are nicely written articles and the second of the above cuts through the mass of bovine 3-methyl indole that had accumulated around the topic. The third finishes with the “Bailey-Burrows” control unit circuit, adapted from Bailey’s 1966 design.
These articles are readily accessible at this excellent site: http://www.americanradiohistory.com/Wire...gazine.htm.
Craig mentioned the landmark Dinsdale 1965 two-transistor series-feedback disc input and RIAA equalization circuit and silicon derivatives of the original germanium form. Bailey I think was one of the first to both use silicon and include the emitter follower. See WW 1966 December p.598ff. Bailey though did not use a bootstrapped input; evidently the natural input impedance of the silicon-based circuit was thought to be high enough. I don’t think that the bootstrapped input was used by many others, either. It was used on the Quad 33, providing a sufficiently high impedance that what was seen by the cartridge was essentially defined by the input load resistor alone, 68 k for the M1 and M2 magnetic positions and 100 k for the C1 ceramic position. Quad later claimed that its disc input was essentially resistive only, to within ±5 degrees over the audio range, thus allowing for the setting of precise cartridge loading in capacitive as well as resistance terms. In the late 1970s it offered a modified disc input board with the M1 position matched to the requirements of the Shure V15 Mk III cartridge. Input sensitivity was 3.8 mV (instead of 2 mV) and input resistance was 47 k (instead of 68 k), and there was a non-trivial parallel capacitance, whose value I don’t recall. This was featured at one of the late 1970s London hi-fi shows. In a nearby room Shure was showing its V15 III, and using a Quad 33/303 combination. I asked whether Shure was using the modified disc input board; the answer, given in something of a Gallic shrug manner, was no, although they were aware of it....Go figure!
The Quad 33 also include a passive HF rolloff after the disc amplifier in order to correct for the fact that a series feedback RIAA circuit, which cannot go below unity gain, has an inherent HF equalization error. This does not seem to have been much used by others, and a perhaps surprising omission was in the H.P. Walker 1971 design, otherwise marked by close attention to detail, and for which the presented RIAA curve clearly shows the HF error. Perhaps the error was considered to be negligible, but continuing the HF rolloff would seem to be desirable to prevent any ultrasonic noise from getting further into the amplifier.
I suppose we have digressed a bit, but the relevance is that these disc pre-amplifiers, designed essentially around the needs of magnetic cartridges, were in some cases also used for matching ceramic cartridges by the capacitive feedback technique, as we have seen in the Quad, Bailey-Burrows and H.P. Walker cases. Dinsdale though seemed to favour the use of low-impedance loading of ceramic cartridges with a de-equalizing circuit to cancel the inbuilt mechanical equalization, followed by RIAA equalization. (See WW 1969 November, p.500ff.)
In his 1976 design (WW 1976 November p.41ff), Doug Self used the three-transistor circuit, without input bootstrapping, but with the passive HF rolloff, and with the refinement of a bootstrapped load resistor for the second stage. The basic idea was hardly new, but I am not sure that it had hitherto seen much use in disc input stages. Earlier, H.P. Walker had used load bootstrapping in the tone control section of his 1971 design, as had done Quad with its 33.
An interesting case of bootstrapping at both ends, as it were, was in the signal input amplifier of the Revox A77 tape recorder. The output side load bootstrapping was done in what might be called a reverse way, AC coupling to the emitter follower base and DC coupling from the emitter follower emitter to the upper end of the second transistor collector load. The input bootstrapping allowed the auxiliary input impedance to be defined by a 1 M resistor. This high input impedance allowed the direct connection of ceramic cartridges, but I suspect its main raison d’être was to facilitate the connection of valve amplifiers, whose tape outputs were typically unbuffered and needed to look into 500 k or more to avoid undue loading of the stage whose anode from which they were drawn.
Cheers,
Steve
If the input resistance is much lower than 2 M, then the turnover point will lie in the audio band. This may be corrected by configuring the input amplifier as a low pass filter (integrator), with the same turnover point as the high pass filter created by the combination of the cartridge and the input resistance. This was the basis for using capacitive feedback in the input amplifier. Again with self-equalized ceramic cartridges, no additional equalization was needed. Clearly, the turnover point of the amplifier response curve must be related to both cartridge capacitance and the input resistance. As an example, the Quad 33 control unit used this approach for the C1 (ceramic) position of the disc input amplifier. The input resistance was 100 k, and the C1 input was said to be suitable for ceramic cartridges with self-capacitance in the range 450 to 900 pF, which I suspect covered the majority at the higher quality end of the range. I haven’t done the numbers, but I’d guess that the circuit may have been set for the geometric mean of the capacitance range (636 pF) with not-to-significant errors at either end of that range. H.P. Walker did something similar in his 1971 design (Wireless World (WW) 1971 May p.221ff), in this case with an input resistance of 200 k and the turnover point arranged to suit a cartridge self-capacitance of 600 pF.
An extensive commentary on ceramic pickup matching and equalization was provided by R.J.C. Burrows in a series of articles in WW, as follows:
WW 1970 February p.56ff: Ceramic pickups and Transistor Pre-amplifiers – Are they incompatible?
WW 1971 July p.321: Ceramic Pickup Equalization 1 – Myths against maths and measurement.
WW 1971 August p.379ff – Ceramic Pickup Equalization 2 – Practical low-impedance circuits.
All are nicely written articles and the second of the above cuts through the mass of bovine 3-methyl indole that had accumulated around the topic. The third finishes with the “Bailey-Burrows” control unit circuit, adapted from Bailey’s 1966 design.
These articles are readily accessible at this excellent site: http://www.americanradiohistory.com/Wire...gazine.htm.
Craig mentioned the landmark Dinsdale 1965 two-transistor series-feedback disc input and RIAA equalization circuit and silicon derivatives of the original germanium form. Bailey I think was one of the first to both use silicon and include the emitter follower. See WW 1966 December p.598ff. Bailey though did not use a bootstrapped input; evidently the natural input impedance of the silicon-based circuit was thought to be high enough. I don’t think that the bootstrapped input was used by many others, either. It was used on the Quad 33, providing a sufficiently high impedance that what was seen by the cartridge was essentially defined by the input load resistor alone, 68 k for the M1 and M2 magnetic positions and 100 k for the C1 ceramic position. Quad later claimed that its disc input was essentially resistive only, to within ±5 degrees over the audio range, thus allowing for the setting of precise cartridge loading in capacitive as well as resistance terms. In the late 1970s it offered a modified disc input board with the M1 position matched to the requirements of the Shure V15 Mk III cartridge. Input sensitivity was 3.8 mV (instead of 2 mV) and input resistance was 47 k (instead of 68 k), and there was a non-trivial parallel capacitance, whose value I don’t recall. This was featured at one of the late 1970s London hi-fi shows. In a nearby room Shure was showing its V15 III, and using a Quad 33/303 combination. I asked whether Shure was using the modified disc input board; the answer, given in something of a Gallic shrug manner, was no, although they were aware of it....Go figure!
The Quad 33 also include a passive HF rolloff after the disc amplifier in order to correct for the fact that a series feedback RIAA circuit, which cannot go below unity gain, has an inherent HF equalization error. This does not seem to have been much used by others, and a perhaps surprising omission was in the H.P. Walker 1971 design, otherwise marked by close attention to detail, and for which the presented RIAA curve clearly shows the HF error. Perhaps the error was considered to be negligible, but continuing the HF rolloff would seem to be desirable to prevent any ultrasonic noise from getting further into the amplifier.
I suppose we have digressed a bit, but the relevance is that these disc pre-amplifiers, designed essentially around the needs of magnetic cartridges, were in some cases also used for matching ceramic cartridges by the capacitive feedback technique, as we have seen in the Quad, Bailey-Burrows and H.P. Walker cases. Dinsdale though seemed to favour the use of low-impedance loading of ceramic cartridges with a de-equalizing circuit to cancel the inbuilt mechanical equalization, followed by RIAA equalization. (See WW 1969 November, p.500ff.)
In his 1976 design (WW 1976 November p.41ff), Doug Self used the three-transistor circuit, without input bootstrapping, but with the passive HF rolloff, and with the refinement of a bootstrapped load resistor for the second stage. The basic idea was hardly new, but I am not sure that it had hitherto seen much use in disc input stages. Earlier, H.P. Walker had used load bootstrapping in the tone control section of his 1971 design, as had done Quad with its 33.
An interesting case of bootstrapping at both ends, as it were, was in the signal input amplifier of the Revox A77 tape recorder. The output side load bootstrapping was done in what might be called a reverse way, AC coupling to the emitter follower base and DC coupling from the emitter follower emitter to the upper end of the second transistor collector load. The input bootstrapping allowed the auxiliary input impedance to be defined by a 1 M resistor. This high input impedance allowed the direct connection of ceramic cartridges, but I suspect its main raison d’être was to facilitate the connection of valve amplifiers, whose tape outputs were typically unbuffered and needed to look into 500 k or more to avoid undue loading of the stage whose anode from which they were drawn.
Cheers,
Steve
