Nice job. Thanks for the pictures showing what we hear. The pictures should help convince the "true" mono cartridge zealots. Among the "non true mono" listeners there still is the discussion of whether to put the coils in series or in parallel. The problem which you mentioned about the loading on each channel by the opposite channel being a low impedance and reactive bothers me, however I think it is what many manufacturers do in their 78 cartridges. The issue with the series connection is the low resistance and high capacitance loading both you and Joe addressed.
I wanted to avoid both issues and I wanted to be able to switch from lateral to vertical easily so I built the circuit below:
[img]
After getting started I figured I should have most of the playback EQs I might find a need for.
To address the issue you brought up about phase and amplitude skew by summing down stream I attempted to mitigate that by summing right after a ~10 dB buffer stage on each channel. I used 1% resistors all off the same tape reel to keep the gains well matched. Also all EQ is done after the summing node.
You may notice I included a 700 Hz treble cut position. I don't think that was used in any recording but I included it to give a constant amplitude responce when used with the bass turn of 700 Hz. With that one can hear what a ceramic cartridge would ideally sound like using a magnetic cartridge.
Phil
ld wrote:Prompted by thinking about the coil arrangement in 's carts on another thread, I came up with a general method for obtaining true mono from a stereo cartridge. There seems a significant (4-6dB) noise advantage versus stereo replay of a mono recording. Also described is a simple way to make it switchable between true mono and true stereo. Measurements and listening tests show a significant noise reduction, c 4-6dB, versus mono replay on a stereo cartridge.
I don't know if it's already known, but it's new to me. It has excellent results for noise improvement and sound, so I thought I'd post it anyway.
The idea is to connect L & R coils in series, and the series output is then true mono. Mono groove modulation is purely lateral movement, of course, and this method significantly attenuates noise output from vertical stylus movement. Each coil produces antiphase signals for pure vertical motion, in phase for pure lateral motion.
Although similar effects can be produced electronically by combining L+R signals later in the signal path, the method described here avoids the need for a seperate summing stage, and avoids phase and amplitude errors often arising through other methods.
This sketch shows an implementation using a DPDT switch to toggle between true stereo and true mono. I conveniently built this into the base of the arm housing under the TT I used for testing, so can switch between stereo and mono easily.
Here's a screenshot of a mono audio sample (tone), showing the effect of flipping the switch :
click to open full size
Signal amplitude increases by 6dB when the mono switch is set. Note noise peaks remain same amplitude.
Here's a screenshot of a silent groove, showing the effect on noise floor of flipping the switch:
click to open full size
Groove noise remains approx similar amplitude. But signal amplitude increased by 6dB (above), hence S/N improves by c 6dB when mono switch set.
Here's a spectrum plot of the silent groove samples, normalised for 6dB offset in signal amplitude:
click to open full size
The benefit is clear across the spectrum. 4-6dB
Here's a plot of a 150Hz mono sinusoid tone, to check harmonic distortion.
click to open full size
Harmonic distortion is more or less unaltered. There is something strange going on in the 20-50Hz range though. It remains when the stylus is lifted off, so seems hum or preamp related - this is not a very good preamp BTW.
Here's a plot that shows Vertical Modulation rejection at 1kHz is c 16dB, and across the spectrum is up to 20dB. That is how much attenuation vertical stylus noise gets.
click to open full size
I tested this on an Ortofon OM5E in a project 9C arm. The S/N ratio improvement for mono records is obvious, toggling the switch.
Sound is as good as the cartridge. Voicing changes significantly due to cartridge loading on toggling the switch. I found an interim load of 68k per channel voices well enough for both stereo and mono, but I expect much more can be done to obtain optimal loading and voicing.
In the fullness of time I expect to leave a TT set up like this, because the S/N advantage when replaying mono recordings seems significant.
Neat, eh ? Appols if this is old hat, but it's not to me !
. But that would dedicate that headshell/cartridge for mono only, and would be quite inconvenient, ESPECIALLY if one would wish for a mono connection to both stereo channels.
Italic lettering would be easily readable (Old English would be hard to read). That's for Calligraphy mavens 
See: