Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: tubenit on January 09, 2011, 09:37:56 pm
-
Does it make any difference functionally in how each of these is wired?
180k into pot into 100k VS. 100k into pot into 180k ?
and pot wired into cap to ground VS. cap into pot into ground?
I'm thinking it doesn't matter?
Thanks, Tubenit
-
Yes it matters. The resistor on top of the pot is part of a voltage divider and will affect the signal level available to the next stage. The resistor between the pot wiper and the tube grid is not part of a voltage divider and wont affect the signal level applied to the tube.
The circuit with the 180K on top of the pot will have less signal, ie, less gain, applied to the next stage than the circuit with the 100K on top of the pot.
-
How about the 1meg pot & the .001 cap (O.D. high frequency cut/bleeder) .......does positioning/order matter in that case?
-
It should not matter about the tone circuit. Putting the cap on top of the pot will prevent dc from appearing on the pot (if that bothers you) but since no dc current will flow thru the pot either way, the pot should not sound scratchy.
-
Thanks Sluckley
-
Thanks guys! I corrected both of those but was curious if it was really needed.
With respect, Tubenit
-
The resistor between the pot wiper and the tube grid is not part of a voltage divider and wont affect the signal level applied to the tube.
Before the wiper: voltage divider, gotcha but what exactally does the resistor after the wiper do? I notice on some schematics between two stages two 470K resistors are used. Sometimes as a voltage divider but sometimes "backwards" 470K to ground first, through 470K to grid and wondered why this was done. I think Doug's stereo preamp's like that.
I always thought it was a backwards volume. I can't remember but some old Fender had the two channels volumes wired backwards like that and I thought that was another way to do it(and save money on isolation resistors). I think Gibson wired the pots backwards on some LPs so when you were on both pickups turning one all the way down wouldn't kill the signal.
Ah here it is Fender's pro-amp 5C5
-
Another question comes to mind if the resistor after the voltage divider doesn't reduce the signal what does putting a 470K/500p after the volume pot do? Before, I get it, the 470K and 1M pot is a voltage divider reducing the whole signal as the 500p bypasses the 470K giving more high end. But if the 470K after the volume pot doesn't reduce the signal what is the 500p doing?
__500p___
__470K___| | _500p_
> > | |
< <---- <<----470K-----
> >
< <
| |
-
Here's Doug's stereo preamp:
http://www.el34world.com/projects/images/stereopreampschematic.gif
Between the second & third gain stages there is a 470K resistor tied to ground before a 470K resistor in series with the grid of the third gain stage.
IIUIC (if I understand it correctly :wink:) the resistor going to ground is necessary to keep current off the control grid. It is frequently called a "grid leak" or "grid return" resistor, and it is essential for the operation of a common cathode triode gain stage.
The 470K tied to the control grid of the following stage is called a "grid stopper". It forms a low pass filter with the Miller capacitance of the tube. IOW it cuts a certain amount of higher frequencies, including keeping your local radio station out of your amp signal. The math is beyond me at this moment, but I do "get" the concept. Notice that Fender used a 68K grid stopper before the first triode in virtually every amp.
BTW the effective value of the grid return resistor actually is the sum of the two 470K resistors in this example.
If you swap the order of the two 470K resistors, you get a voltage divider and a 470K grid return resistor. I think that the 470K in series with the 3rd control grid still forms a low pass filter but am not certain of that.
See page 19 of Merlin's first book re: grid return resistor and Page 32 for "Miller effect"
http://www.freewebs.com/valvewizard1/Common_Gain_Stage.pdf
These resistors also impact the impedance of the circuit, but that's a concept I still haven't worked through yet. (Page 31)
Does that help?
Jeff - I'm guessing that your 500pf cap is used to reduce the Miller effect. IOW a low pass filter is formed by a resistor on the "upper" leg and a cap on the "lower" (tied to ground) leg of a voltage divider, right? If you put a small cap in series with the resistor on the upper leg, then the AC frequencies which pass through that cap will not be affected by the low pass filter... maybe.
Hopefully sluckey will drop by and clear up whatever confusion I've created.
Chip
-
Usually a 470k/bypassed by a 500pf cap is a treble peaking circuit used to boost treble frequencies.Marshall used these in a couple of spots on a JCM800 and Plexi circuits.Some guys don't like these but they are essential to prevent mud.
-
How about the 1meg pot & the .001 cap (O.D. high frequency cut/bleeder) .......does positioning/order matter in that case?
Sluckey is right, it doesn't matter, at least to us. But I think the order of the [cap & series resistor or pot] affects the phasing of an AC signal. It might matter in test equipment or to hi-fi purists.
IIUIC (if I understand it correctly) the resistor going to ground is necessary to keep current off the control grid. It is frequently called a "grid leak" or "grid return" resistor, and it is essential for the operation of a common cathode triode gain stage.
Agfter struggling through some Richard Kuehnel writings, I relazed that it's helpful to distinguish between the AC vs. DC operation of a circuit. For DC operation, the 2X 470K resistors form a 1M grid leak resistor, as Freshtart says. AC, or signal operation of the resistors is more complex. Ohm's Law is more difficult to apply, because for AC we have impedance, nor simple resistance. I.e., the resistors impose more or less impedance depending on the frequency of the AC signal. So, for AC operation, it seems that those resistrors are acting as a filter, along with some fixed attenuation of signal generally. Otherwise the gain of the 2nd stage would probably swamp the input of the 3rd stage. Note that for AC operation, the 1st 470K resistor is in parallel with the plate resistor of the 2nd stage.
-
I.e., the resistors impose more or less impedance depending on the frequency of the AC signal.
No they don't. Pure resistance is not frequency dependant for DC or AC circuits. It's only when you introduce inductance and/or capacitance into the circuit that frequency becomes a factor.
But, for that particular circuit, you have the tube's miller capacitance as Chip stated. So the combined resistance and capacitive reactance will become freq sensitive.
-
It's only when you introduce inductance and/or capacitance into the circuit that frequency becomes a factor.
Would that also apply if it were a speaker, rather than a tube, being fed? I guess the speaker coil would have such reactance factors?
-
Yes. A speaker is inherently inductive. The voice coil is a coil of wire just like a choke.
The formula for inductive reactance XL is ... XL = 6.28*F*L, where F is freq in Hertz and L is inductance in Henries. So, as freq increases, inductive reactance increases.
But the speaker also has pure resistance. You need more math to determine the total impedance (for a particular frequency) because you're combining vectors. A speaker that is 8Ω impedance at 1KHz 'may' be 16Ω at 2KHz if you ignore the resistance of the voice coil.
-
I think the following picture helps to understand the voltage and current division.
There some input resistance (high) in the tube. Theoretically it matters
but in practice there is probably not so big meaning in which order the resistors are.
/Leevi
(http://)
-
Usually a 470k/bypassed by a 500pf cap is a treble peaking circuit used to boost treble frequencies.Marshall used these in a couple of spots on a JCM800 and Plexi circuits.Some guys don't like these but they are essential to prevent mud.
This has been a very helpful thread.
I have just been looking at a bunch of Plexi/JCM 800. Just the info I needed
Thanks