A tube tracking console using ITB’s MkIII concepts

[rjuly]

@ruffrecords,

If I wanted to build a couple of channels to have higher gain, perhaps 70dB, say, for use with a ribbon mic...

- What would be required? Would it take another full gain block - ie three more triodes?
- Would it be possible to get lower signal to noise ratio by using two gain blocks with each adding less gain, to sum to 70dB?
- Would a different approach overall be the way to do this?

Many thanks,

- Richard

 

[ruffrecords]

If you use the twin line amp, each of the the two amplifiers can provide up to 40dB of gain. add to this the 20dB you get from the transformer and there is a total of 100dB gain available.

In a complete channel you need to subtract from this the 6dB loss in the output transformer plus any loss in the EQ which could be up to 20dB depending on the type. Taking these teo factors into account means you can still have 74dB of total gain should you require it.

I commonly set up channels for a total of 70dB gain. You get 60dB of it in the first half of the TLA (20dB transformer plus 40dB amplifier). I then allow for 10dB loss in the fader, 6dB loss in the output transformer and with the REDDEQ an insertion loss of about 16dB which means the second half of the TLA gets set for 32dB gain. This gives your channel a nominal gain of up to 60dB with the fader at zero and 10dB in hand on the fader for a total of 70dB.

Signal to noise is defined almost entirely by the very first stage. The 20dB in the transformer helps a lot because it has a very low noise figure and provides a 10 times bigger signal to the first TLA which is the dominant source of noise at high gains.

Cheers

Ian.

 

[rjuly]

If it were possible I would like 75dB of gain for this pair of channels, assuming the noise floor is low enough. I’m planning to record fairly low level acoustic room sound with a ribbon mic.
My current design has no fader to attenuate the signal in between amplification blocks, but only post direct output, to control mix levels to the 2-bus. Given that I won’t have eq in the circuit, I guess I will need to attenuate at some point in the circuit to bring the level down to 75dB. Should I use a fader or perhaps a fixed resistor?

Alternatively, might there be a lower gain combination of tubes for either or both of the gain blocks that would give me the desired gain structure with lower noise?

Many thanks for taking the time to indulge my ignorance.

- Richard

 

[rjuly]

Here is my initial attempt at a schematic for the channel (attached).
- Richard

 

[ruffrecords]

Several factors contribute to noise, some of which you can control and many of which you cannot. The actual noise is dominated by the noise of the first tube of the first preamp so it is worth buying some extras and selecting the lowest noise ones for use in this position. This is why I recommend using Sovtek 12AX7 WB types here as I have found them to be most consistent. The second most important factor is the gain setting resistor in this same stage. The smaller this is the less noise it contributes. This means the added noise is least at the highest gains (but of course it is amplified more so it sounds louder). This means there is no noise advantage in lowering the gain of the first preamp and increasing the gain of the second. The third most important source of noise is external interference and minimising it requires careful screening and minimising high impedance lead lengths.

If you want a post direct level/mix control the you could place it across the secondary of the insert return transformer (before C12 on your schematic).

You schematic looks basically OK to me. There are some minor errors. For example RV2 is wired such that it can short the output of the second stage. You can place this before the stage as mentioned above. Also there is nothing connected to the output of the first preamp. I think you meant to connect ot to T3? And I am not sure what K6 is meant to do.

I have absolutely no problem with you asking questions. There is a lot of talk in 'woke' circles about unconscious bias. It is an unavoidable fact that our thought processes are strongly affected by our prior experience and this is true for technical topics as much as any other. For this reason I enjoy hearing questions because even straightforward (to me) technical questions give clues to the person asking them.

Cheers

Ian

 

[rjuly]

Questions you shall have ;-) 'Basically OK' is far better than I expected for this first attempt.

Yes indeed lol - linking the first stage with the second is likely advisable. I think I must have removed the connection shown in the green circle at some point:



K6 is meant to be an insert on/off switch, for which I forgot also the switch...

Many thanks for looking it over. I will make some updates and come back with a new version and some questions.

- Richard

 

[C.B. - Boudio]

This is of quite some inspiration for me! Has there been some progress the last year?

 

[rjuly]

This is of quite some inspiration for me! Has there been some progress the last year?

Currently stalled I'm afraid. All my stuff is in storage - hoping to reboot this effort in the next three to four months or so.

- Richard