Phono pre amp build recommendations

[Eli Duttman]

I wanted to build a circuit with no semiconductor diodes or transistors at all (just for the hell of it). As I also didn't fancy using batteries or mercury vapour rectifiers, that meant using an AC heater supply.

Marcel,

I'm not "knocking" your fine effort. However, the Sovtek 12AX7LPS is a true 7025 equivalent constructed with a spiral wound, hum bucking, heater. While I will not make the attempt, that type combined with good grounding technique and heater bias off B+ should allow more conventional circuitry that's AC heated to produce satisfactory residual hum levels. I'm not a good enough builder to try AC heating in a phono preamp.

BTW, was the reduction in current noise obtained at the cost of adding partition noise? Was a net improvement in S/N performance realized?

 

[N101N]

kodabmx,
the Aikido stage is balanced as well. The Japanese extensively employ balanced topologies to accentuate the character of shamisen, sanshin, sasara, shakuhachi, biwa, sushi... at high frequencies.

 

[MarcelvdG]

Marcel,

I'm not "knocking" your fine effort. However, the Sovtek 12AX7LPS is a true 7025 equivalent constructed with a spiral wound, hum bucking, heater. While I will not make the attempt, that type combined with good grounding technique and heater bias off B+ should allow more conventional circuitry that's AC heated to produce satisfactory residual hum levels. I'm not a good enough builder to try AC heating in a phono preamp.

BTW, was the reduction in current noise obtained at the cost of adding partition noise? Was a net improvement in S/N performance realized?

There is no partition noise in the input stage, as the first EF86 is triode-connected. The RIAA- and A-weighted averages of the equivalent input noise voltage and current densities were 10.26 nV/√Hz and 0.2534 pA/√Hz, with a more conventional design using a 47 kohm termination and a triode-connected EF86 they would have been around 10.26 nV/√Hz and 0.5869 pA/√Hz. The lower current noise has its largest effect for high-inductance cartridges, but even then it tends to be masked by record surface noise whenever there is a record playing. So all in all, the noise in between records is a bit lower than with the usual 47 kohm termination.

I used thin shielded twisted pair cables for the heater wiring. With the power supply transformer placed at some distance from the rest, there was no audible hum at all, not even with the input open or shorted and the volume increased enough to clearly hear the noise floor. The difficult part was keeping the hum under control when the transformer was mounted on the chassis.

Thanks for the information about that Sovtek valve. It's a pity its datasheet doesn't specify the heater-to-grid capacitance. The datasheet claims a plate-to-plate capacitance of 600 pF - that must be a typo for 600 fF. The cathode-to-heater capacitance is specified to be 5 nF, which seems rather high to me; another typo?

https://drtube.com/datasheets/12ax7lps-sovtek.pdf

This 7025 datasheet has a grid-to-heater capacitance spec, but it is 400 fF, versus 2.5 fF for an EF86:

https://www.thetubestore.com/lib/thetubestore/Preferred-Series-7025-Data-Sheet.pdf

http://www.r-type.org/pdfs/ef86.pdf

 

[MarcelvdG]

An E88CC has a cathode-to-heater capacitance of 2.6 pF or 2.7 pF, see http://www.r-type.org/pdfs/e88cc.pdf , so I guess that 5 nF is supposed to be 5 pF.

 

[dreamth]

This is my variant of a moving-magnet phono amplifier with valves. Due to Miller effect in the first stage, it is not suitable for cartridges that require a very low load capacitance: you get about 90 pF with all switches off. The 47 kohm input resistance is made with a feedback technique to reduce the equivalent input noise current. I built it with point-to-point wiring, simply because I didn't want to design a PCB. The neon lights are used as overvoltage protection, they work best when there is some light shining on them. It is my first and so far only electronic circuit with no transistors or semiconductor diodes.

1 m of 50 ohm radio input cable has 75pf with the best cable available, usually 150pf/m.
Otherwise very interesting design.I'd think of d3a strapped triode to remove the cathode followers as i have a few.What's the gain of ef86 when triode strapped?D3a can provide aboiut 70x