Phono 3 input capacitance

[A.N. Beal]

Dear John,

I've just checked mine. With the test tone 1kHz 5cm/s on the HiFi Sound test record I measured 900mV out, which equates to 180x overall gain, the same as the HFW 2007 test, so (if I have understood the measurements correctly) your Phono 3 is achieving more than 50% extra gain overall than mine.

Where is this difference in gain most likely to be happening? I am particularly keen to find out if any of the difference is be happening on the input valves, as their gain has a major effect on input capacitance.

Regards,

Alasdair

[John Caswell]

The only way to measure gain is to inject a known level signal into the unit and measure the output.
In my case I inject a 1kHz 5mV signal from both an AWA G232 and G233 hi - quality audio oscillators, measured on a Marconi 2600B video voltmeter plus 3 separate DMM, and simultaneously measured the output on two AWA F242A audio analysers plus another DMM. I repeated this several times for repeatability.
I suggest that you are assuming that the test record plus pickup cartridge gives you 5mV - too many variables!
You need to measure the input voltage accurately.

John

[A.N. Beal]

Dear John,

After sending my last post the penny dropped that I had rushed my testing and hadn't checked the cartridge output. It is a Goldring G900 with a Paratrace-retipped after-market stylus, so the output is probably different from the original.

I only have a multimeter, not all your proper test gear but this morning I took a bit more time over the exercise. I first disconnected the tonearm leads and measured the cartridge output on these with a 5cm/s 1kHz signal from the test disc. I then reconnected them and measured the Phono 3 output from each channel. The measurements were: 6.7mV/950mV (L), 7.9mV/990mV (R). (For interest, the 'other channel' outputs measured 30mV(L), 32mV(R), confirming excellent stereo separation of -34dB.)

The G900 has published internal resistance and inductance of 700 Ohms and 560mH, giving an approximate total internal resistance at 1kHz of 4.2kOhms. If the preamp input is 47kOhm resistance in parallel with total (say) 250pF capacitance, at 1kHz this is 43.7kOhm, so connecting the cartridge to the Phono 3 would be expected to reduce the measured output by 43.7/47.9, giving net 6.1mV(L), 7.2mV(R). Based on these figures I calculate the overall gain of my Phono 3 as 155(L), 138(R).

If these measurements and calculations are in the right ballpark, they suggest that my Phono 3 has about half the overall gain of yours. (Mine has JJ ECC83s valves.) If I have made a mistake somewhere, I apologise. However if my figures are right, it would be useful to have an idea where the gain difference most likely to be happening in the circuit. If it is in the valves is it likely to be mostly in one stage or is it likely to be roughly equal in each of the three stages?

Regards,

Alasdair

[John Caswell]

I have looked against the disparity in gain and of course the answer is fairly simple.
Alistair has a standard Phono 2/3 whereas I have a Phono 3 with battery bias.
Replacing the first stage cathode resistor with a battery is effectively the same as bypassing the cathode bias resistor with a big fat capacitor, it gets rid of the degeneration (feedback) around this stage and as a consequence the gain is pretty near doubled.
Have a Google on it there are pages and pages about it.

John

[Greg.]

What type of battery are you using, John? In the past I found sound changed according to the type of battery used.

[A.N. Beal]

Dear John,

Thanks for the solution to 'the Great Gain Mystery'. Now to try to tie this back to the initial question. The measured gain on your battery bias stage 1 was 50 and your overall gain was 280x; the measured overall gain on my standard Phono 3 was 146x (average of channels). If this difference is basically occurring on stage 1 of the amplifier, then I assume that the gain on my stage 1 must be approximately 50x146/280 = 36.1.

Plugging that back into your input capacitance calculation would give Miller effect capacitance = 3x37.1 = 81pF. Adding your measured capacitance of 19pF for the board, the cold ECC83 and phono socket = 100pF. I measured the capacitance of the signal cable currently being supplied as about 20pF, so this would give a total input capacitance of 120pF. Phew!

Does that calculation look correct to you? If so, it's a useful result for moving magnet users.

Best wishes,

Alasdair

[A.N. Beal]

Dear All,

I have done further measurements and tests, using different valves (courtesy of Stewart Gebbie at www.ValveTubeGuitarAmps.com). With JJ ECC83s valves I measured overall gain (average of channels) as x159; with JJ 5751 it was x143; with Electro Harmonix 7025 valves it was x133. With board etc. capacitance pf 19pF and internal signal lead capacitance of 21pF combined with Miller Effect capacitance these give total input capacitances of 128pF, 120pF and 114pF respectively.

My tonearm leads have capacitance of about 75pF and the recommended input capacitance for my Goldring G900IGC cartridge is 150-200pF. I have the Phono 3 set to standard RIAA equalisation and capacitor C9/10 fixed as 220pF. With the JJ ECC83S the sound is excellent apart from rolled off extreme treble. The JJ 5751 valves gave more extended extreme treble and much brighter sound but it was a bit too much, perhaps because (according to Stewart) the 5751 requires a different bias setting from an ECC83. The EH 7025 valves gave a clean, clear sound with more extended treble than the ECC83s but without the exaggeration of the JJ 5751.

The EH 7025 seems to be a very good valve and it is in current production at a reasonable price. Its lower gain usefully reduces the input capacitance of the Phono 3, straightening out the top end response of the G900IGC. If the signal wiring capacitance could be reduced to 10pF (by using twisted wires instead of screened cable?) this would reduce total input capacitance to about 100pF, which is a good target for modern high quality magnetic cartridges, as it makes it possible to achieve total capacitance including leads of less than 200pF. (It is always easy to add more capacitance.)

Regards,

Alasdair

[A.N. Beal]

Dear All,
I have now put in over 100 hours with the Electro Harmonix 7025 valves in the first two stages of my Phono 3 and can confirm that they are excellent. In addition to reducing Miller Effect capacitance and evening out the top end response of my Goldring G900IGC cartridge, they also give a very clean, clear, smooth sound. The EH7025 valves are new current production, so they provide a good alternative to chasing increasingly rare and expensive NOS ECC83 valves.
Regards,
Alasdair