Replacing the original Sonotone 9TA in my Oz HMV Y2-D3 3 in 1, with a magnetic cart.
The Gram section of this unit (late '60s vintage) is based on the very common TA solid state chassis that HMV used in most grams at the time.
So I had a look at the service data for the TA, which is normally used with a couple of different ceramic carts, and discovered that some models had a magnetic cart option.
So how does this work? The service data contains no reference to changing the preamp. And the circuit has more in it than would be needed for just a ceramic cart input.
Here is the preamp circuit:
Thinking this might be an example of Gupta's Law, I modeled the circuit in LTSpice using the Shure Analog for the magnetic cartridge and the generally accepted model for a ceramic which is a voltage source in series with 800pF.
And it works! From 100Hz to 5kHz the two carts track each other. Outside that there are differences that nevertheless don't exceed 6dB.
Anyone ever seen this arrangement before? It's really quite elegant and it works very well with either type of cart.
@irob2345 When you say track each other is that frequency response or output voltage from the preamp, or perhaps both?
Frank
Here in the UK the Thorn '74' Chassis - used in many a unit audio and a stereogram - was configurable to work with either ceramic cartridges like the Sonotone 3559 or with magnetic cartridges such as the Goldring G850. From what I can remember, it too involved a two-transistor pre-amp and the PCBs were already capable of taking the extra components for the magnetic cartridge.
I knocked up a tiny MM amp, based around the 5532... kicad bits and the gerbers are here : https://github.com/andydoswell/Micro-phono
Here I have a Portogram portable record player with a Goldring GL72 in it, fitted with a Sanyo ceramic cartridge. I plan to bin that and fit a stereo magnetic to it, to wit a Goldring G850 as I have both microgroove and 78rpm styli for it. I modified it a few years ago when I transcribed some Berliner discs for a former colleague to have a buffered output directly from the cartridge for recording purposes, rather than the potted-down output from the loudspeaker amplifier. However, the plan now is to make use of the 'stabilised' (read: zener) 24V supply to power an old warhorse in the shape of an LM381 RIAA pre-amp, which was designed for single-rail operation: this will then feed a TL074 used as both a stereo buffer and a virtual earth mixer to mono in 'M3' mode so that will feed the internal amplifier, which is about 20W RMS.
In case you're wondering why the amplifier has such a high output, it's because Portogram record players found a niche in teaching dance classes - the variable-speed Goldring being ideal for teaching steps at a slower pace until the students got the hang of it, whereupon the speed could be increased to the correct tempo. The 'gram is designed to feed external loudspeakers, hence the high output: there is an internal speaker but it is only a 'pilot' device and so is heavily attenuated and has its own on/off switch.
@cathovisor You may find the G850 with 78rpm records may overload the input stage of the preamp.
I had this 50 odd years ago when a customer bought a music system, Hacker or Dynatron with a G850 cartridge. He had many 78 records, shelves full of them. The Deck had a plug-in head shell so I fitted another G850 into a new head shell and potted down the output within the head shell. He just swopped head shells for the different formats. The 78’s were unusable until I potted down the output.
Frank
Yes when I said they track each other, the preamp gives the same output for mag or ceramic cart, correctly compensated too.
It's a very clever circuit. I wonder did Neville Thiele have something to do with it? He worked at EMI Australia for a time.
Note the 1m resistor connected between the L and R input. You would think that it would degrade channel crosstalk performance but it actually improves perceived channel separation on a typical stereo ceramic cart. (I tried removing it, back in the day to check). Anyone know why?
When the 9TA/HC was new, Sonotone produced what they called a "velocity compensator" which allowed it to be connected into the magnetic pickup input of a pre-amp. ISTR that if you connected the 9TA into a low load (e.g. 47k) it behaved like a magnetic pickup in terms of frequency response.
@irob2345 For what little they are my engineering skills can’t help, still trying to work out how the circuit behaves for both types of cartridge.
Frank
It uses shunt current feedback. With shunt current feedback, if the source looks like a high impedance, the feedback reduces the gain to (in this case) near unity. With a low source impedance, more gain results. The two time constants in the feedback path set the compensation for each mode. As I said the circuit looks like Neville Theile's work, mystifying at first until you discover the brilliance.
In case you've never heard of him, he's the "Theile" in the Theile-Small parameters on which any worthwhile speaker design is based. He pioneered mathematical modeling for speaker enclosure design. (And the speakers in my Y2 sound very nice for what they are.)
Do you use LTSpice? I can post the schematic / model file (I think) that I used to verify that the preamp works the way it should for both carts. I drew this up because I was curious.
I'm itching to test my new mag cart on it but the Garrard 2025's headshell (such as it is) won't take it, it's designed specifically for the 9TA and is too narrow. I picked up a couple of screws at work today that will allow me to complete the modification / butchery of the headshell to take a half-inch standard mount cart.
Pity the headshell isn't unplug-able, it's truly painful to work on..
I hope to finish it tonight. Let you know how it goes.
Thanks for the explanation, at least it makes sense now. I enjoy being part of the forum, and comment if I think I appropriate but I have no involvement in electronics either job or hobby. My interests are more arts these days, something different from a lifetime of fixing things.
The 2025 fixed head shell was designed for the 9TA as you know, a really good cartridge in its day. Is there enough counter balance to compensate for the heavier mag cartridge?
Frank
I wish this TT had a counterweight! No, just a spring that you hook into different holes under the arm.
I can get it down to 6g.
I see electronics as art, so I can appreciate where you are coming from.
Re the 9TA, in the late 60s and early 70s I was a field service tech. We'd carry 9TAs in van stock for service replacement purposes. I always enjoyed it when the customer could hear the improvement over those horrid grey BSR things. The 9TA was far and away the best sounding ceramic back then.
Now of course my hearing is 12db down at 8kHz and my S/N is degraded by tinnitus that's around 14kHz and never goes away.
Unfortunately the 2025TC also got used with inappropriate cartridges.
My first stereo record player was a Bush SRP58: a portable unit with optional dust cover, it used a very simple four-transistor amplifier powered by the motor and a right nail of a cartridge in the shape of the Sonotone 2509, a high output/low compliance stereo crystal. In the mid-70s, as recording techniques changed I found my record player had a problem - it would 'jump'. I took records back to the shop I bought them from, but they always played perfectly.
The issue was the arm: it was designed for high-compliance cartridges, as I discovered when I fitted a 9TA/HC from a scrapped radiogram. Low output, but the records played fine.
In later years, I learned that Garrard and Bush had had so many complaints about this that they took the step of replacing the plastic 'headshell' with a die-cast one and fitting an uprated spring to reset the tracking weight for the cartridges affected. Because the effective mass of the arm was now higher, the jumping problem went away.
I'm sure @irob2345 is aware, but many contemporary decks made by Garrard all shared the same motor board cutout so a more upmarket changer from that era could be fitted without any modification to the cabinet.
Yes my collector friend Pete is digging out an AT6 for me.
In many ways this very expensive (in '69) high-end unit was spoiled by low-end fitments. Could have been worse I suppose, the TT could have been a BSR! And the TA series gram chassis was just HMV's standard unit that they used in low and mid-range models.
The TV chassis, although it performs well, is really a 2nd-tier design. It should have had the PU5 hybrid chassis rather than the tail-end-of-production all valve hand wired chassis it has. At least it has stabilised scan.
Yes my collector friend Pete is digging out an AT6 for me.
The AT6 is probably my favourite middling autochanger - whilst early SP25s are very obviously cut-down versions of it, the higher-quality motor and above-deck components on the AT6 conceal that the auto-changing mechanism is derived from the somewhat nasty "Autoslim" deck.
As for BSR - well, during the 60s BSR produced a heavily-modified version of the Loewy-designed UA16 for Pye that could track at 2g: this was so Pye could use the CBS-designed "Butterfly" pickup cartridge. It had a counterweight extension at the rear of the arm. Although the "Butterfly" was best known here for its use in the "Achoic" stereo record player (complete with the most bizarre stereo circuitry known to man), it was also used in a high-end Pye stereogram - my grandmother had one.
Did you get the "Butterfly" in .au?
Thank you both for your thoughtful and informative responses. Although I'm a design engineer I've had little experience with gram related issues, last time I worked with them is when I was in the service industry back before the flood!
Interested to hear about the shunt vs series feedback S/N question.
"There was an interesting follow up sequence of “Letters to the Editor”, in which J. Linsley Hood, who had use a shunt feedback circuit in his 1969 modular preamp design, attempted to challenge Walker’s findings."
Did he succeed with his challenge? I would have expected the lower impedance of a shunt topology to result in lower noise, all other things being equal. To claim otherwise is surprising and counter-intuitive.
You don't by any chance have a link to those articles?
