Valve Itch phono

Hi,

I was just doing some math here.

If I assume the Itch is made for some 0,25 mV MC cartridge. With a 1:10 SUT (20 dB) the input grid gets 2,5 mV signal. With an amplification after SUT of 45 dB the Itch is gonna output 445 mV. Getting that up to modern line level around 2 V takes som 13 dB preamp (a 15 dB Salas 6V6 line maybe).

So the scenario with a HOMC 2,5 mV would be the same as exampled 0,25 mV+1:10 SUT.

Is that correct?

So, to my questions. Is it favourable to make all that amplification after the RIAA network, or would it be preferable to raise the input level somewhat with a SUT after HOMC cartridge? I was thinking of for example LL1550 6 or 12 dB connected floating balanced on primary.

6 dB from 2,5 mV would get around 5 mV on the grids on 6N2PEV and the Itch would output 887 mV. With a 12 dB config the grids would get some 10 mV and the Itch would output 1,78 V (around line level without amplifying pre).

What would be favourable?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Its a common reasoning trap. The difference is in the media average recorded level ways. What is 2V RMS is max on CD not where the music stored in it averages. Get your DMM out, go ACV on the dial, measure on CDP RCA out when on a favorite music CD replay. If you take an LP recording up to such a gain you plan its going to be ear splitting with an ugly hiss topping. Your average record just cruises around 0dB frequently going over and not hitting a hard ceiling. At 445mV RMS nominal you will find that your digital music will be playing along your vinyl at same knob position more or less. Having no preamp with gain, then if you got 28dB amp and 90+dB/W speakers you will be alright. Alexkosha in his FSP build for example set 56dB on 1mV cart. Then he had a 20dB Japanese tube pre and its assorted tube amp with average gain. Speakers normal towers, not horns. He asked for 6dB less gain mod because his LPs were too loud vs. his CDs.
 
I will, or rather, I dont have to cause I trust you. This had passed me in terms of gain calculations and I have seen many respectable constructors do the same. I am happy that you disturbed my circles!

From Line level - Wikipedia, the free encyclopedia
"The most common nominal level for consumer audio equipment is −10 dBV, and the most common nominal level for professional equipment is +4 dBu (by convention, decibel values are written with an explicit sign symbol). Expressed in absolute terms, a signal at −10 dBV is equivalent to a sine wave signal with a peak amplitude (VPK) of approximately 0.447 volts, or any general signal at 0.316 volts root mean square (VRMS)"

Thanks again!
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Probably a good idea. Atm I am running from Technical Specification | Focusrite and it has ASIO adjustable input and output levels so I can just tune it.

As I prefer DIY I am more after to set a standard on my equipment that follows common practices and also makes sense in my reasoning.

You can even use ECC88 on second stage for some more gain but not overly more. Personally I like the 6SN7 better there and the level was fine through my unity gain 6V6GT pre and 26dB CFA amp on 94dB/W speakers. If to opt for 6CG7 (noval 6SN7) you don't even need change base types or pinout for ECC88.

I run a very typical track, Hotel California by The Eagles on their Live CD "Hell Freezes Over", min-max mode on the Fluke 87V recorded 163mV AVG & 367.5mV MAX on a Benchmark DAC (RCA output L).
 
The average level of my flacs varies so much its impossible to name a standard. Afterwards I wonder if I should have choosen autoomatic level adjustment when ripping. Irritatingly large differences on CDs.

6CG7 is a good idea for gain adjustment. I too like 6SN7 better then ECC88 in character usually. I will look if I can match 4 halves decently.

The trafo I dug up yesterday showed to have a shorted primary, have to find another after work today.

PS. I didnt know that you were running your 6V6 pre in unity gain config.
 
Last edited:
diyAudio Chief Moderator
Joined 2002
Paid Member
-You got test gear. Give the amp 100mV pk-pk sinewave (verify on scope when the gen is connected to the amp) monitor its power output on the DSO's other channel while on a proper dummy load or some crap speaker, divide, convert to dB and you will know.
-ECC88 will be +4dB above the 6SN7 level but hiss will be probably more also since the second stage will not be that well governed by the input stage's noise spec anymore. But with 6CG7 you just swap them and you get to know all things to compare in real life. Gain, hiss, tone. They will both work properly in the same circuitry.
 
Ok, I managed to catch a stupid cold today so it was hard to look down in stuff with that nose but I opened the Hafler EL84 today. I noticed that I had made some temporary couplings that I made more solid.

All voltages seemed stable, 360 V B+ on the ECC83 floating paraphrase splitters and 325 V on the OPTs.

I let gen signal 200 mV sinus 1 kHz and on the 7,5 ohm resistorloaded OPTs I had 1,38 V. That makes 6,6 times amplification or about 16,4 dB.

The larger part of that amplification is in the splitters who outed around 950-1000 mV on the EL84 grids.

Stupidly enough I found the right channel flickering a little on the sinus so I guess I will have to track that down. It has some new 100 Hz not volume depended hum when I had it back on the music bench. Stupid me, never fix things that work:mad:

But that was OT. 16,4 dB it was which I think is quite common for most of EL84 PP ampz.

*atjoo*:dead::wchair:
 
Oki thanks! Im eager to start building a prototype. Shall I run same voltages 300 or more on both input and output pair for better transconductance on the input (meaning no resistor inbetween)? I shall see today if I can find me a suiting transformer. I have two that makes those voltages made for tube rectification and also some biggie 200 VA Anteks 2*300 and 350. They would suit a combo with 6V6 unity gain. Smaller stuff I have to measure, I have a whole bunch of salvages from old tube radios.