Input decoupling capacitor - before V1 grid. Should I?

[jmaf]

Most amps connect the guitar input directly to the grid of V1. Some do it via a grid stopper, which is pretty common as well.

My question is: what are your previous experiences with decoupling the guitar from V1 input via a capacitor? Such as the Orange Tiny Terror?

Do you think it increases safety, by keeping the musician far away from a plate-grid short?

Does it negatively affect tone if for example I used a large value as to attenuate as little as possible of the audio spectrum?

When the musician touches the strings, does it still quiet the amplifier up as DC-coupled ones do?

Thanks in advance for any insights.

[defaced]

The caps I have seen in circuits haven't been physically large enough for the values they are to be high voltage caps. This indicates to me that this is purely a sonic thing. A plate-grid short has the plate and grid resistor in series with it, so you'd get a nasty bite, but it's not likely to take you out. 325v B+ on the first stage (which is high), 100k plate + grid resistance (which is low), and you get 3.25 ma of current. Usually between 10 and 50 ma and with high voltage and you start having serious safety concerns.

Ground is still DC connected, so you still get noise reduction from the guitar's string ground.

IMO, it's purely option, and in my experience, doesn't do anything to the sound (with the size caps I've used - 0.22u ). Yes, that statement does conflict with my first statement that this is purely a sonic thing. That's because I honestly don't have a good reason why they are used.

[jmaf]

The caps I have seen in circuits haven't been physically large enough for the values they are to be high voltage caps. This indicates to me that this is purely a sonic thing. A plate-grid short has the plate and grid resistor in series with it, so you'd get a nasty bite, but it's not likely to take you out. 325v B+ on the first stage (which is high), 100k plate + grid resistance (which is low), and you get 3.25 ma of current. Usually between 10 and 50 ma and with high voltage and you start having serious safety concerns.

Ground is still DC connected, so you still get noise reduction from the guitar's string ground.

IMO, it's purely option, and in my experience, doesn't do anything to the sound (with the size caps I've used - 0.22u ). Yes, that statement does conflict with my first statement that this is purely a sonic thing. That's because I honestly don't have a good reason why they are used.

Thanks for your input. Yes, you pointed something interesting out: they're normally high voltage caps...which means they're probably made to withstand some form of high voltage DC. Could it be a safety thing only since they don't change tone much?

[Tom Phillips]

A bit of additional information.
You will find series input caps in old designs that used a grid leak biased input stage. See the Fender 5C1 Champ and the 5B3 Deluxe for a couple of examples. The cap was in the circuit to prevent the pickup load from disturbing the bias of the input stage. The cap is not required when the input stage is cathode biased. If an input DC blocking cap was needed for safety, I'm sure the certification agencies would have required modern amp builders to add it.
Cheers,
Tom

[Chuck H]

He actually said that they're not high voltage caps.

I can't say as I never investigated the rating on these caps. But I would think that a cap in this position can do a couple of different things. One thing it would do is decouple the giutar output load from the amps input load. So the volume control on the guitar would no longer be in parallel with the 1M V1 grid load. Logically this also creates a phase shift too for better or worse. It does isolate the V1 grid from any other voltage that could throw off bias. This seems to keep with logic used in all other preamp circuits and makes good electrical sense. It can be used as a high pass filter. Sometimes a tube itself will leak a little and cause a scratchy guitar volume. This won't happen with the input cap, again for better or worse since this is often the best indicator of the problem. There are other consideration too but I'm not able to come up with them right now. I have considered using them at times but haven't gotten around to it. I may never since the pros and cons all seem subjective.

[Enzo]

The incredibly rare plate to grid short in a tube is the last thing on anyone's mind.

In a grid leak bias you need a cap so the outside world doesn't swamp your tender grid bias. But grid leak bias is also pretty darn rare.

The input cap is there to keep DC off the tube grid. Imagine if you have some effect pedal that has a few volts of DC offset at the output. Now how will your amp react if the input stage now has +5v there? or -5v there? A largish value cap that passes signal to the grid but blocks DC, solves that problem.

[jmaf]

Thanks Tom, Chuck and Enzo, all good points there. I will add a large cap to my next project, I liked the possibility of avoiding the scratchy pot and also effect pedals with a DC offset. I'll rearrange the 1M grid leak post capacitor, as now it is on my input jack. Appreciate your feedback, once again.

[Chuck H]

I'll rearrange the 1M grid leak post capacitor, as now it is on my input jack.

Yupper. With an input cap ahead of the grid load the tube has no bias. Tubes perform very poorly with no bias (sarcastic understatement).

[Tubis]

Well, keeping DC of the grid, keeping high voltages of your hands and keeping low freq from entering your amp.....

[jpfamps]

The input cap is there to keep DC off the tube grid. Imagine if you have some effect pedal that has a few volts of DC offset at the output. Now how will your amp react if the input stage now has +5v there? or -5v there? A largish value cap that passes signal to the grid but blocks DC, solves that problem.

Speaking to the designer of the TT, this is indeed the purpose of the DC blocking cap on the input.

[Steve Conner]

In an old thread, we theorized that an input blocking capacitor would limit the ability of drive pedals to overdrive the first stage. Any grid current would just charge up the cap and make the bias bounce around. I argued that this might make the amp seem less "pedal friendly".

One of my old homebuilt amps does have an input coupling capacitor, but it's a high-gain master volume type thing. I don't use it with drive pedals.

And, it is considered good practice in modern design to put DC block capacitors on the inputs of things, for the reason mentioned above.

[Merlinb]

In an old thread, we theorized that an input blocking capacitor would limit the ability of drive pedals to overdrive the first stage. Any grid current would just charge up the cap...

I find that hard to believe, considering all pedals have an output capacitor already! An extra cap at the input of the amp would simply reduce the effective coupling capacitance a tiny bit (assuming it has a large value).

[mooreamps]

I find that hard to believe, considering all pedals have an output capacitor already! An extra cap at the input of the amp would simply reduce the effective coupling capacitance a tiny bit (assuming it has a large value).

You've never heard of the "Jedi Cap Trick" ?? no, I suppose not.... :[

-g

[Chuck H]

The Jedi cap trick involves using the farce... Uh... Force. You have to believe in something that cannot be seen or proven.

And a little green dude messes with your mind.

[mooreamps]

Actually, it was believed to be practiced by Eric Clapton on some of his rigs. :}

-g