Questions about channel interference when sharing 1st stage

[psychepool]

I have an amp I want to build, and I want to add channels while minimizing the number of tubes.
I've seen a lot of cases where the 1st stage is shared in multi-channel amps, and the amps I've made previously also have that configuration, so I'm going to refer to it and apply it.

However, in this case, I was curious about whether there was interference between channels, so I asked this question.

The amp I'm planning to build is an Orange Rockerverb.
Like Dual Dark, I plan to put two gain channels and if possible, add a clean channel as well.

I couldn't find an official circuit diagram for Dual Dark, but looking at the tube layout, it appears to be comprised of completely independent channels.
It is presumed to be composed of a Rockerverb preamp and a TH30 (Dark Terror) preamp.
I'm thinking that the circuit might look something like 'image_01'.


But as you can see, these two preamps have about 98% identical circuitry except for the tone stack.
I don't think there's a need to duplicate made of common parts, so I'm thinking about channel switching like 'image_02'.


I don't think there will be any problems with the operation itself, but what I'm curious about is channel interference.
If one of the two channels is selected, the signal appears to be partially connected to the ground through the other channel's coupling cap and resistance (1n->220K->220K/470pF chain), which affects the tone of the channel in use. This raises the question of whether this will actually have a valid impact.

The reason I'm asking about this is because there are cases where problems have occurred with the amplifiers I've previously made, and there are cases where there are no problems at all.

'Image_03' is SloRectoTwin and VH4, respectively.
Since it is a circuit that shares the 1st stage, it is a configuration that could cause the problems I mentioned, but I did not experience any special problems while using it.
Of course, it is possible that a certain amount of tonal interference was taken into consideration at the time of design and the remaining parts of the channel were tuned.

However, in the following cases(image_04), there was serious tonal interference.
This is a Bogner Ecstasy-based amplifier that I made. I take steps to mute unused channels by connecting them to ground through a relay.
However, when using the clean channel, the 4.7n coupling cap of the dirty channel was connected to ground, causing a serious muddy tone problem in the clean channel.
By removing this mute switch, I was able to get a sparkling clean tone again.


As such, channel interference when sharing the 1st stage seems to be a case-by-case issue, so I am curious to know what the results will be in the Orange circuit I am considering.
Since the fairly high resistance of hundreds of K is connected to the ground, it is thought that there will not be enough tonal interference to have a special effect. However, on the other hand, since something is connected anyway, I feel like there will be some degree of tonal interference.

What do you guys think? Please give me your opinion.

[acheld]

When a triode stage is "shared" with two (or more) additional triode stages, the AC load line will be much different than if the stage is not shared.  In effect, you've introduced a voltage divider in the output which your circuit needs to handle.   

The technical term is impedance matching, a subject I failed in school.   Rich Kuehnel addresses this topic in some of his books, but it is a slog, I'm not gonna lie.

My own approach is to route the signal to the chosen channel via switches (or relays).  I don't want to amplify a signal I'm not going to use and this avoids having to re-calculate load lines . . .

Using shared stages IS possible, but you should really calculate AC and DC load lines for the stage in order to know what is going on in your circuit.

[pdf64]

We don't use impedance matching in our application, rather we work with voltage signals, so impedance bridging provides the best transfer.
A typical common cathode 12AX7 stage has an output impedance of about 40k.
Impedance match that and the output voltage halves. Oh wait, that pretty much describes the stage driving into a typical Fender opto trem, so it does get used, but it's probably just coincidence, the 50k linear trem intensity pot happens to provide a good user interface.


Whatever, a 40k source impedance loaded by a 4n7F cap creates a low pass filter with a corner frequency of about 850 Hz. So yes, that'll result in a muffled tone.
Hence that perhaps wasn't a good method of muting the unused channel.

Provided the constraint of the 40k output impedance of the shared input stage gets properly considered, then there's usually a way to accommodate for it and work around it.

[psychepool]

I thought it was a simple band filter problem, but it turned out to be a bit more complicated.

All of these problems seem to be solved by disconnecting the 1n coupling cap of the unused channel with a switch, but simply disconnecting it with a switch is expected to cause popping noise. If it doesn't cause a big pop, I'm willing to accept it and apply it to some extent, but I think it'll probably cause a pretty big pop.

So I thought of two ideas for switching this part.

The first is to use an optocoupler.
I have never actually applied switching using an optocoupler, but I have heard that it significantly alleviates popping by preventing sudden short circuits.  However, I am concerned about tone loss, and I am a little doubtful whether popping will also be significantly improved in this case.

Second, connect a resistor to ground behind 1n-220K chain.  Originally, a 220K resistor was connected to the ground behind the 1n-220K chain, so change it to 470K and attach a separate 470K to the rear of the 1n-220K to prevent DC disconnection when the switch is off, and when the switch is on, 470K is in parallel, similar to the existing value. It's about adjusting it accordingly.
I think it's an effective method, but I still think that the 1n->220K->470K chain will act as a band filter for other channels in use.  However, if we think about the case of guitar's tone knob at full, even with a value of 22n->500K, there is not much difference compared to the case without a tone knob, so I think it might be worth a try.

What do you think about the above methods?

[AlNewman]

I can't see anything labelled mute switch in figure 4, but I can see you are dumping a lot of signal to ground either way.

[acheld]

Non-definitve comments:

Opto-isolators (many names for these) -- most commonly the X-Vive VTL5C1 (or 3 or 4) is what is used by folks on this forum.   Look at Sluckey's  Tremonator circuit (here:  https://sluckeyamps.com/trem/trem.htm ).   These are not really switches per se -- they are variable resistors which can have a very high (10M - 400Mohm) resistance when "off."  They do not pop in my experience.   It is worthwhile paying a bit more for the X-Vive branded ones, I've had good experience with them, and "variable" experience with non-branded ones.

The Tremonator circuit is a perfect use case for these.  (And it sounds fantastic!)

Your second drawing shows the use of switches.  This would be your better choice IMO.    On or off, no in-between.

A simple DPDT switch would work fine.  Popping is generally suppressed with ceramic caps (only) IF you need them; I normally don't need pop suppression.

A little more complicated choice would be Omron low signal relays -- these are really useful DPDT switches that use low voltage DC to operate the relay.  I use these frequently.  Relays allow switching from a stomp box, very easy to implement and does not introduce long cable runs where signal is exposed.    I have not had significant popping with these when used in Dumble style amps.   Hoffman (this forum's sponsor) sells these with associated boards and power supply very inexpensively.  They work great.  Highly recommended.

[psychepool]

Will either of the above methods cause 'perceptible' tonal interference on the channel I'm using? Even if there is tonal interference, I think it can be considered as long as it is not severe.

I think that using an optocoupler will have less of an impact, but even if I use a simple contact switch without using an optocoupler, won't a big pop occur at that location?
Since I'm not expecting completely quiet anyway, I'd like to choose one that doesn't interfere with the sound unless there's a very loud pop noise.
This is because pops are likely to occur to some extent at other switching points, even if they are not in this location anyway.

[acheld]

Optocoupler, LDR, relays, switches -- dealer's choice.  These won't affect tone. 

IIRC, there are versions of the SLO-100 using relays and the same circuit using LDRs over on TAG.

IF you will want to flip a switch while playing -- for example, engage an OD channel or pre-amp boost -- that would argue against using simple switches, and for using VTL5Cx or relays (allowing you to make a change without having a long signal path, Fender Rev/Trem circuits notwithstanding).

[psychepool]

Since English is not my native language, I used some incorrect expressions.
Basically, it is a footswitchable channel, so I plan to use a switch that can be remotely controlled, whether it is an optocoupler or a relay.
What I mentioned 'Simple contact switch' is referring to a relay. I think I used that expression because the control itself is performed remotely, but the switching itself is done through a physical contact point.
Sorry for the confusion.

I've heard a lot about optocouplers having tone loss, but from what you said, it seems like it's some sort of superstition.
If I use an optocoupler with appropriate specifications, including the product you mentioned, there may be no problem.

I'm trying to decide which of several amps to make.
If I make this circuit, I think I would test that switching points in the following order.

1. Test by applying a toggle switch to each point of each channel
2. If there is no significant tone or functional problem even if the location is connected compared to the unconnected state, leave both channels connected without applying any switch and finish.
3. If there is a problem with tone or function, check whether popping noise occurs when switching to the toggle at the corresponding position.
4. If there is no popping noise, finish by applying a relay.
5. If popping noise occurs, apply an optocoupler. If the popping problem is resolved with the optocoupler, finish as is.
6. If the popping noise is not resolved with the optocoupler, apply the 2nd method uploaded above: resistor to ground as a test.
6. If it is effective, finish it as is. If it is not effective or there is a problem with tone or function, apply a relay or optocoupler again at the risk of some popping noise (or just make it with 2 channels of clean/dirty like the original Rockerverb).

In fact, it is a concern that can be avoided by using one more tube, but it is difficult to increase the number of tubes according to the build plan. A small difference complicates the problem.
In such a limited situation, it is a good choice to give up my greed, but I have never succeeded in giving up my greed while making an amplifier. :)

Thank you for everyone's response. It helped a lot!

[acheld]

I have never succeeded in giving up my greed while making an amplifier. :)

Oh yes, we know about this!   

Honestly, you won't have problems with popping with either the relays or the VTL5C1 ("optocoupler").   All the Dumble style amps use the relays with impunity.   

Do check out Hoffman's little 5VDC power board and his relay boards.  They are small, rugged and have stood the test of time.