Question about Marshall schematic

[Nickos]

Hi!
Can someone help me with this schematic?
http://www.drtube.com/schematics/marshall/cd0109-iss9.pdf

There are 2 zener in the first part of the circuit... what are these for? is a sort of clipping or limiting?  ???

thank you,
Nickos

 

[recnsci]

is a sort of clipping or limiting?  ???

Yes, sort of. It protects input from large tranisents or severe overvoltge by limiting (clipping) input voltage to aprox +/-10V range

edit: btw, it's not "effect" in any sort of way, for normal levels in guitar world it's pretty much transparent.

 

[Nickos]

It's strange because few marshall amps have this protection...

Thank you! 

 

[johnR]

It's common to use input protection llike this on solid state amps, but it seems like overkill on a valve amp. Maybe it's to protect against high voltages appearing at the input in the case of a faulty valve and/or leaky blocking cap, or perhaps they thought someone was likely to connect a speaker output to the input.  Whatever, that 10k series resistor will generate a significant amount of thermal noise.

 

[recnsci]

... or perhaps they thought someone was likely to connect a speaker output to the input... 

You should hear this story: this guitarist has heard "from reliable source" that "pros" get their "pro tone" by running output from one Marshall head into input of another. And yeah, he would probably use speaker out if he had two JMPs by any chance (I don't know how much would unloaded JMP head put at output, or would it survive unloaded condition). So, you can never be too cautious around people with magnetic pickups on their instruments (unless it's bassplayer; we 4-stringers are smart, educated and well behaved people ... not).

BTW, since I'm less than familiar (understatement) with thermionics, how sensitive are tube amps to abuse? I allways had this vague feel of fragility about them.

Whatever, that 10k series resistor will generate a significant amount of thermal noise.

On paper, yes. In practice, I dont think that would be really noticable. BTW, can anyone confirm: when I analyse inherent noise (that is, stuff not picked by induction from environment) of pickup, I use LCR aproximation of pickup and add noise sources same as would be if DCR was plain ol' resistor, and calculate noise spectrum from there? Is this right aproach?

 

[Nickos]

So...
do you think that they operate in the guitar range? are they cutting transients?
I find hard to believe that a guitar pickup would generate 10V but... you never know  ;D

 

[johnR]

this guitarist has heard "from reliable source"

Those "reliable sources" are often good for a laugh.

On the subject of thermal noise, the 10k resistor will make noticeable noise on high gain settings. A typical guitar pickup will have a dc resistance of 5k to 10k (some high output humbuckers are higher), and some series inductance plus a small parallel capacitance. That is in parallel with the cable capacitance, which for a moderately long cable is typically in the order of 1nF. The total source impedance is the combination of all those. At high frequencies it's usually some way below 10k ohms with the guitar volume on full, so adding 10k in series with that will make a noticeable difference to the source impedance seen by the grid of the first valve. That source impedance is the main determining factor in amplifier noise level at high gain settings (speaking from practical experience as well as theory).

To put this in perspective, the old Super Lead and Master Volume Marshalls had a 68k resistor in series with the input, but despite the lower gain of those amplifiers it was still an annoying source of thermal noise IMO (annoying enough for me to remove the resistors in my old 2204s). Nowadays I make part of my living designing and building custom guitar amps, and thermal noise on input stages is something I pay a lot of attention to.

 

[johnR]

So...
do you think that they operate in the guitar range? are they cutting transients?
I find hard to believe that a guitar pickup would generate 10V but... you never know  ;D

No. The input valve however will clip (at least on the negative half cycles) at one or two volts peak, which a high output humbucker on a guitar with heavy strings, played hard, can exceed.

 

[Nickos]

Ok thank you very much.

 

[recnsci]

So...
do you think that they operate in the guitar range? are they cutting transients?
I find hard to believe that a guitar pickup would generate 10V but... you never know  ;D

Oh no, certanly not. As I said, those zeners pretty much don't exist for guitar level stuff. Guitar pickup (ol' school pasive stuff) will pump out 10-100 mV, bass can stretch to 1 V or so I've read. Onboard preamps and guitar pedals can reach 9 V max in theory, in reality it's probably 4-5 V at best. There are devices with 18 V supply, but even those can reach 18 V peak2peak (and that's only in theory), and those zeners will conduct with something like 19-20 V pp.

 

[johnR]

Guitar pickup (ol' school pasive stuff) will pump out 10-100 mV, bass can stretch to 1 V or so I've read.

The high output Kent Armstrong humbucker on my test guitar was giving slightly over 2V peak on a low E chord just now on my scope. The other pickup, an Armstrong PAF+ type with parallel coils (hence around 6dB lower output than normal) , gave 500mV peak. Single coil pickups with light strings will give a lot less. There's quite a wide range of levels from different guitars. It makes it tricky to design a preamp that can produce a clean signal from high output guitars when required, but still have enough gain available (along with low enough noise) for plenty of overdrive with low output guitars.

 

[recnsci]

The high output Kent Armstrong humbucker on my test guitar was giving slightly over 2V peak on a low E chord just now on my scope.

W O W

4 V p-p ??? jezzzus, I had no idea. Yes, dynamic range of guitars certanly presents muchm more of a chalenge than I thought so.


About that 10K being an issue or not, my logic was this: I have passive V/V/T Jazz bass, with 250K volume pots, and I regulary use it with volume 6dB down (essentialy to reduce pickup interaction) which means about 100K in series with output, and I have no noise issues. So I figured 10K would not be much of a problem. Don't know, maybe I'm not annoyed by noise that much or maybe those thick strings induce so much output that noise is masked.

 

[johnR]

About that 10K being an issue or not, my logic was this: I have passive V/V/T Jazz bass, with 250K volume pots, and I regulary use it with volume 6dB down (essentialy to reduce pickup interaction) which means about 100K in series with output, and I have no noise issues. So I figured 10K would not be much of a problem. Don't know, maybe I'm not annoyed by noise that much or maybe those thick strings induce so much output that noise is masked.

With the volume pot down 6dB the output impedance of the guitar will be at its highest, but don't forget that's shunted by the capacitance of the cable (assuming there isn't an active preamp in the guitar). A 1nF cable capacitance will have a reactance of about 16k ohms at 10kHz, so the source impedance seen by the amp is quite a bit lower at high frequencies. You're right that 10k ohm won't make a huge difference on that volume setting, but at full volume it could mean 3 or 4dB of additional noise (and near zero volume the difference would be much bigger). I play heavy stuff with lots of gain, so the odd few dB of extra noise can be a dealbreaker for me.

 

[recnsci]

I play heavy stuff with lots of gain

<slap on a forhead>
Yes, being a bass player I conveniently ignored some situations.

 

[johnR]

I play heavy stuff with lots of gain

<slap on a forhead>
Yes, being a bass player I conveniently ignored some situations.

With the gain set low enough for a clean sound, thermal noise at the input won't be so noticeable, and it's often swamped by noise from later amplifier stages. For the typical Marshall user, that isn't always the case.

 

[okgb]

is there a defined purpose for the 68k ?
other than knocking down the level ?
Some amps use small caps on the input as well ,
although this doesn't seem common but serves
a noise / impedance function ?

 

[rodabod]

is there a defined purpose for the 68k ?
other than knocking down the level ?

That's a grid-stopper. Forms a low pass filter in combination with the Miller capcitance seen at the grid to reduce HF / RF.