Lowering noise floor on fender and some Marshall style amps

[Mike Sulzer]

The second input not only pads 6 dB, it presents the guitar with a load of 136 Kohms, obviously loading down the pickups, robbing high end clarity and probably pinching off some lows too. I never much liked it until I started working for a band that specialized in Philly soul & Motown style music. That second input comes in real handy for rhythm guitar tracks, automatically giving them a period-authentic low-fi mid rangey "boxed in" tone. So... it's good for something.

Hope this helps!

Or if you have a strat wired for no tone control on the bridge pickup, that 136K load might be the easiest way to tame it down a bit when playing loud with a bit of distortion in the output stage.

[nsubulysses]

Wouldn't an easy way to lower noise floor and maintain same sound be to make all resistors in the signal path in the first gain stage (and even the 2nd gain stage if you wanna gild the lilly) metal film style which are lowest noise resistors?

[Mike Sulzer]

Wouldn't an easy way to lower noise floor and maintain same sound be to make all resistors in the signal path in the first gain stage (and even the 2nd gain stage if you wanna gild the lilly) metal film style which are lowest noise resistors?

Not if the resistor in series with the grid is the limiting factor. Other resistors:

1. cathode resistor: small value, and so less noise and possibly bypassed with a C or partially so anyway.
2. Plate resistor: In the output circuit where the signal is bigger and so noise is relatively less of a problem.

[Leo_Gnardo]

Wouldn't an easy way to lower noise floor and maintain same sound be to make all resistors in the signal path in the first gain stage (and even the 2nd gain stage if you wanna gild the lilly) metal film style which are lowest noise resistors?

That's an experiment I tried a long time ago. Plate load resistors, sure! There are some carbon film resistors that do well here too, cheaply available but now so are metal films. Cathode R's are capacitor bypassed, minimizing noise in all but the worst cases - funky old carbon comps from the age of dinosaurs.

Lastly, input grid stoppers. You would think metal film would obviously outperform carbon comps here but I've never found much improvement. Note I'm using my ears. Maybe fancy lab gear could show us a change, but it is our ears we're trying to satisfy. The rule of thumb has been to prefer carbon comps as grid stoppers, I'm guessing because they haven't much in the way of induction which could affect the way the circuit behaves in high radio frequencies. Somebody with RF experience might back me up on this - or not. In any case, the lowest value grid stop resistor that effectively cuts out RF interference while hopefully not contributing too much noise of its own is what we want to see here.

[nsubulysses]

Merlin Blencowe says input grid resistor in V1 is the biggest noise perpetrator in a lot of amps, and noise floor can be lowered by using 10K rather than 68K and also using metal film.

This could be one of those things where you measure the noise floor with test equipment and it certainly measures lower, but the audible differrence is minimal or not noticeable. a low gain amp like typical fender I would think it would have minimal improvement but if it's a 4 or 5 preamp tube high gain amp I'm sure it matters a lot more to use metal film or lower value resistors in series with signal path if possible

[Helmholtz]

Merlin Blencowe says input grid resistor in V1 is the biggest noise perpetrator in a lot of amps, and noise floor can be lowered by using 10K rather than 68K and also using metal film.

As the input grid serial resistor sees no DCV, it only produces thermal (Johnson) noise and its type (material/technology) should not matter.

[nsubulysses]

Well then he probably knows that and probably didn't say the input grid resistor for V1 should be metal film. I think he said resistors in series with the signal should be metal film for lower noise floor and I took it to mean that very first resistor as well.

[nevetslab]

Frequency Response of resistors

Lastly, input grid stoppers. You would think metal film would obviously outperform carbon comps here but I've never found much improvement. Note I'm using my ears. Maybe fancy lab gear could show us a change, but it is our ears we're trying to satisfy. The rule of thumb has been to prefer carbon comps as grid stoppers, I'm guessing because they haven't much in the way of induction which could affect the way the circuit behaves in high radio frequencies. Somebody with RF experience might back me up on this - or not. In any case, the lowest value grid stop resistor that effectively cuts out RF interference while hopefully not contributing too much noise of its own is what we want to see here.

One major limiting factor in leaded resistors is that.....the leads. Still, the construction of the resistor type does render very measurable RF bandwidth for a given value and physical size resistor. I found a couple documents that offer some good information on that here:

RF Electronics -Resistors.pdf
basics of Resistors.pdf
hf_resistor_application_guide.pdf

From the documents I've looked at, the MELF type resistor has the best RF bandwidth . MELF = Metal Electrode Leadless Face, which is the round body surface mount parts for PCB mounting. The RF limitation then becomes the PCB layout. None of this applies to our leaded resistors being deployed in our builds, but it does show how you get out into the GHZ domain with a resistor.

[SoulFetish]

I thought it was intended to have the Hi or Lo function when used with one instrument, or add the two inputs through 68K resistors when two instruments were used.

right.
You get the option of attenuation if using a single instrument. Or, if you happen to be the two people in history who like putting two instruments in the same channel, you get a miserable carbon comp resistive mixer. Unless, I'm missing something else?

[Chuck H]

The hi/lo inputs, as well as normal/bright, normal/trem, omnivore/vegan, etc. are what they are. They are designed in situ for usability with one or two instruments if you choose or to be used as options for a single instrument. They are not intended to be ideal, only options. Back when multiple inputs were common it was also common to have limited amplification on stage and it was not unusual to have a need for an extra player in a single amp. It was never meant to be ideal for anything. Only a potentially better option for whatever came up for the player/s. In other words, those inputs simply are what they are and there is no ideal or "correct" way to use or even consider them.

JM2C

[Justin Thomas]

My experience with using both inputs on the same channel, Hi & Lo, for two guitars playing at the same time:

Just don't. It's terrible. One guitar is louder than the other immediately, even if you put the LP in input 2. When you start twiddling Volume & Tone controls it seriously becomes a huge overly-interactive fluster-cuck. It was terrible.

Having an amp with two separate channels (a la VINTAGE Fender, not that "channel switchy" stuff) is SO much better! In this case, it genuinely is very much like having a crude PA system. Granted, it does prefer some rather sensitive & efficient speakers to keep up with the more complex signal, but that's no problem for me.

Justin

[Chuck H]

I just think it was originally a cheap added feature. That's why it's not a great feature for much. More of a selling point at the time these amps were made. And always less than a good option. Just something to put in the print.

[Leo_Gnardo]

My experience with using both inputs on the same channel, Hi & Lo, for two guitars playing at the same time:

Just don't. It's terrible. One guitar is louder than the other immediately, even if you put the LP in input 2. When you start twiddling Volume & Tone controls it seriously becomes a huge overly-interactive fluster-cuck. It was terrible.

"Just don't. It's terrible." How right you are! At one memorable rehearsal with my college band, the singer plugged her mic into the unused channel on my bass amp. Holey Hannah, she sounded like she was gargling! Ohhh it was way awful. And funny! "We gotta get you your own system." So she bought a Vox Super Beatle amp - to sing through. Still laughable, but at least we didn't have the gargle effect. Oh the good ol' days. Shelley, if you're out there, I luv ya! Gawd she could belt it out, yes indeedy. Gal had a lot of spirit. I hope she still does.

[Justin Thomas]

Hey Leo,

I meant don't use both inputs on the same channel. No issues with using one input from each. And the BEST PA sound I ever heard was my friend running a 58 into my Spitfire clone feeding a CL Special homemade 4x10 column. The acoustic player never realized he was using a mic until he asked me to give him a bit more high end. When nobody even realizes you're milked, THAT is a good PA sound.

And every keyboard player I've run through one of my vintage Fenders immediately notelices the MUCH better sound than plugging into a PA - warm, without the clicky attack and abrasive harsh sound. Sure, maybe you could spend a few hours eqing it out, or you can just play evertmtthing through a vintage Fender, twiddle a couple knobs, and GO!

Justin

[SoulFetish]

The hi/lo inputs, as well as normal/bright, normal/trem, omnivore/vegan, etc. are what they are. They are designed in situ for usability with one or two instruments if you choose or to be used as options for a single instrument. They are not intended to be ideal, only options. Back when multiple inputs were common it was also common to have limited amplification on stage and it was not unusual to have a need for an extra player in a single amp. It was never meant to be ideal for anything. Only a potentially better option for whatever came up for the player/s. In other words, those inputs simply are what they are and there is no ideal or "correct" way to use or even consider them.

JM2C

I get why vintage amplifiers where designed with the interfaces they had. But it was for the benefit of the market, not the end user. Manufactures were trying to fill the needs of a growing music industry and wanted to get as many people (playing a variety of instruments) into a single model as they could get away with. It makes sense for them. But for the end user, it's about as inefficient an offering of options as you could get. Fender maybe being the worst offender.
The players and techs have had to carry around the dead weight of Fenders insistence on high/low inputs and "Normal" channels for last 60 years. I know, there's always going to whispers and rumors of the one guy who actually used the low input on a blackface normal channel. I actually met him, he came in the shop a few weeks ago, and that poor bastard has been hauling our vibrato channel around inside his amp his whole life while it makes heat and takes up too much space on the front of his amp. Hey, and good for him though; takes all kinds.
I can concede though, that it isn't the feature that I think is (basically) useless. It's the function and application. I got the chance to play an early MKII Marshall today after a good size repair job (shorted PT). I never played or heard one before and I really dug the overdriven sound. But there is so much gain in the front end of the amp that I thought that was a great example of where a low input is probably needed and definitely useful if you want any kind of clean headroom, and both of them sounded good... IMO