Please explain resistors in Marshall Superlite

[shocki]

If you look at this schematic
http://mhuss.com/18watt/schematics/1...perliteTMB.gif

After the first vol control you see two 470k resistors. The one that goes to ground takes away some of the signal. The result would be a weaker signal.
But what does the one that goes to the next stage do? Does it act as a grid stopper ? This would create quite a lot high end roll of I guess.

Another way is to see this is as a voltage divider just like a vol pot turned up half. The effects of changing the resistor values is explained here very nicely.
http://music-electronics-forum.com/t6412/

If you look at the 150k resistor in the Trainwreck build we don't have a voltage divider. But this still reduces the gain?
http://music-electronics-forum.com/a...ress-input.jpg

What is the sound difference between such a voltage divider arrangement and just having the resistor to ground?
Thanks for helping.

PS: I would like the change the title of the thread cause its not really acurate. But I don't know how

[Enzo]

The one to ground doesn't take away signal, not by itself. But with the other one, they form a voltage divider, a two to one voltage divider. The signal leaving the volume control is reduced by half, as you surmised. I leave it to someone else to speak of grid stopping effect.

[shocki]

But what does the 150k in the Trainwreck build do then? I thought this also reduces gain? I don't hear much of a difference when using a trim pot here though.

[Jazz P Bass]

The 150K resistor on the Trainwreck is needed for biasing the tube.
By returning the grid to the negative (ground) terminal, it establishes a 'more negative' potential on the grid as opposed to the cathode (more positive).

As to the Marshall, the two 470K resistors do indeed make a divider, but more importantly, the first resistor (attached to the Vol wiper) limits how close to ground the next grid will go, which maintains the grid bias.

[shocki]

The 150K resistor on the Trainwreck is needed for biasing the tube.
By returning the grid to the negative (ground) terminal, it establishes a 'more negative' potential on the grid as opposed to the cathode (more positive).

I just figured. It also forms a highpass filter right? Cutoff Frequency is around 50Hz.

As to the Marshall, the two 470K resistors do indeed make a divider, but more importantly, the first resistor (attached to the Vol wiper) limits how close to ground the next grid will go, which maintains the grid bias.
Mhh. I have to think about that

[pdf64]

'If you look at the 150k resistor in the Trainwreck build we don't have a voltage divider. But this still reduces the gain?
http://music-electronics-forum.com/t...ress-input.jpg'

That 150k forms a voltage divider and reduces gain.
This occurs due to the tube that feeds it not being a perfect voltage source, in that it has a source impedance of around 40k.
This is formed by the parallel combination of the tube's plate resistance (eg ~60k) and plate resistor (100k).
So, starting from the notional perfect voltage source, there's a source impedance in series, then a blocking cap of 2nF and a load of 150k (which also acts as the grid leak resistor for the following stage.
A potential divider will be formed by voltage being dropped across the source impedance as it feeds the 150k load, and at lower frequencies significant voltage will also be dropped across the 2nF cap, ie a high pass filter; the breakpoint of this is ~420Hz (add the source impedance to the load resistance for the R component).

The 150k will limit the maximum voltage swing from the preceding plate.

It will also reduce that stage's gain, which is set by the tube's gm and the parallel combination of all plate circuit impedances, ie 60k ra // 100k Ra // 150k RL

When the output voltage rises to such a point that the following stage reaches saturation, the grid to cathode impedance of the following stage becomes very low, eg 2k.
The effect becomes a bit tricky to describe, because the cathode resistor is unbypassed and much larger than usual, 10k, but it will act to further load and reduce the gain of (ie yet a further parallel component to the above) the preceding stage, for most of half the wave.

I seem to remember reading that Ken Fisher's goal was to make an amp that had a very fast response.

It may be seen that his designs appear to avoid adding series resistance to the signal path.

My take is that may be because he was trying to avoid limiting the slew rate / high frequency response, which series resistance (in conjunction with a tube's Miller capacitance) acts to restrict.
ie the only 'designed in' grid stoppers are on the output tubes (which have very low Miller capacitance), my understanding being that he only added grid stoppers to the implemented design as necessary, in order to 'tune' the amp into stability.

I think that Komet have a Trainwreck inspired amp that has a 'fast / slow' switch; on investigation this seemed just to be a potential divider in the signal path, the 'slow' setting being the lower gain, the upper resistor of the potential divider being in series with the signal path and, following the rationale, slowing the response.

Hope that helps - Pete

[shocki]

Thanks alot. I need some time to fully understand this

[pdf64]

'Thanks alot. I need some time to fully understand this'

It's taken me a few decades!

On the Marshall, the 2 x 470k resistors form a second potential divider, after that of the volume control.

They will create a minimum grid circuit resistance of 235k, even at low setting of the volume control.
So yes, they will act with the tube's Miller capacitence to form a low pass filter / reduce the high frequency response, with a -3dB breakpoint of 5k6Hz.

As the volume control is turned up, the equivalent circuit resistance will increase still further; when the control is electrically halfway (about 8 on the dial of an audio pot) it will max out at 500k, adding to the 235k, bring the filter breakpoint down to 1k8Hz.

At max volume, the tube's plate circuit resistance of about 40k will have an effect, but won't significantly alter it much compared to the low volume filter frequency; may bring it down below 5kHz but it all gets complicated.

So, as you noticed, there are 2 very different approaches exemplified by Trainwreck and the Superlite 18 watt.

The Trainwreck 'eliminate all series resistance' approach necessitates a 'more than normal' degree of high pass filtering at the main saturation point, in order to mitigate blocking distortion / bias shift.
Pete

[mhuss]

Hey - that site looks familiar.

The two 470ks are there to attenuate the signal by 1/2 (simple voltage divider). As a positive side effect, the series one also helps prevent an excess of grid current flowing in the second stage under heavy clipping, which can make for ratty sounding distortion.

--mark

[Gingertube]

That 150K Grid Leak Resistor on the 3rd stage is an unusually low value and is often referred to a a "Gain Dumping" resistor.

On the "Real Trainwreck Mojo" thread I posted this:

The 150K "Gain Dumping" resistor as the grid leak on the input to the third stage:-
The RL of the 2nd stage would be 90K (100K parallel 1M) ordinarily but with a 150K grid leak following instead of a 1M then RL becomes (100K parallel 150K) = 60K.
Gain = mu.(RL/ (RL + ra)) so the actual gain of the 2nd stage is dropped marginally from 58 to 48.
However the output signal is developed across RL so with 60K instead of 90K we are swinging 1/2 again as much signal current in the tube to get the same output voltage.
That means that the ra variation (delta ra) with signal current is larger (Tube internal plate impedance ra varies (inversely) with anode current)
AND
since RL is much lower then Delta ra vs RL is a much bigger ratio - that is important because delta ra vs RL is the distortion level.
So we have modified the 2nd stage to drop the gain by a bit less than 20% but are likely to have increased distortion by much more, at least 50% and possibly as much as 100%

Cheers,
Ian