Resistors between the preamp grid and anode?

MikeM · 2010-01-08T19:02:14+00:00

What do resistors do between the preamp grid anode do? I'm going over some Fender designs, and specifically this one http://www.schematicheaven.com/fenderamps/bassman_5f6_schem.pdf And compared to a more simple layout (A 5F1 which is my reference), the bassman has a 100 k between the anode and the grid(Or would this be seen as the anode of stage 1 in that preamp and the second anode?) Very interested in all these variables..

Gearhead

The current runs basically through Rb, your tube and Rk. Swings across Rk are kept very low by Ck. The voltage between the Rb and the anode is determined by the resistance ratio between the plate resistance and Rb. A voltage swing upward by the grid (your input side) will cause the tube to start conducting less (increase resistance) and the voltage across Rb will be more, so the voltage going into Cc is going down. The amount it goes down (assuming same tube) is determined by the voltage drop across Rb caused by the current change through the tube. Should the grid voltage drop, the opposite is true and the output voltage rises by a similar amount. In other words, a lower value of Rb will give less gain.

It is a bit surprising that your lowering Rb (called the plate resistor) gives you a nice crunch, but I am assuming you had to turn up the previous stage to do that.
The different situation you put the tube in can be assessed by drawing a so called load line in the tube characteristic. Follow some of the best classes on this topic here:
http://www.audioxpress.com/resource/audioclass/index.htm

The darkness is a bit harder to explain, are you sure that was because of the plate resistor change?

You could put a pot as Rb, but you would need to check whether it can handle the high voltage, you would have to keep it away from earth and shield the shaft you are turning. I would also put it in series with your 33k to make sure your turning the pot will not make the voltage across the tube too much. All in all not a good idea.

Gearhead

:-[ reading through the theory from my own link again, I can explain the loss of highs as well :-[

The high frequency roll-off of the stage is dominated by the lowest impedance of plate resistance of the tube, your Rb and the Rg of the following stage. You have not only lowered Rb significantly, but also put the tube in an operating point with a lower plate resistance. I would make the plate resistor at least 68K.

TomCat

I do have a lot of gain as I have two previous stages before I get to this point. P1 being the input tube and then the first cascades stage being P2. The impedance drop is at the input of P3.

nicovlogg

Gearhead wrote It is a bit surprising that your lowering Rb (called the plate resistor) gives you a nice crunch, but I am assuming you had to turn up the previous stage to do that.

On second thought that was probably thanks to the other changes in the overdrive circuit that I made. :-[

Gearhead wrote The different situation you put the tube in can be assessed by drawing a so called load line in the tube characteristic. Follow some of the best classes on this topic here:
http://www.audioxpress.com/resource/audioclass/index.htm
The darkness is a bit harder to explain, are you sure that was because of the plate resistor change?

Grand link that, thanks a lot! I reckoned that 33k was too low, but following the mod info I found it very hard to understand the instructions. Here's a sample -

"1) Change R118 (Rb) on the V101A input grid to 22-33k. This mod is more important than it seems. What you cut at the input (treble) you can never restore later in the circuit. But, it might be a god idea to cut the copper foil that goes to V101B and put another 33k from R118 and to the tube input grid at V101B. The high gain channel does like the treble roll of with the higher value 68k (33+33k)."

I cannot understand why you would bridge Rb to to the input grid with a resistor. What on earth does he mean?

http://forum.metroamp.com/viewtopic.php?f=4&t=13499&st=0&sk=t&sd=a&sid=a64a9ba4b65743b0edbbc27226ad142e&start=30

Sorry for thread derailing!

Karel-Mars

Nice thread guys 8)

The thread you quote is a discussion about grid stopper resistors. The clue is: the higher the resistor the more tops disappear.

Tubes have capacitance between all elements; but for our discussion here, consider the tube as a capacitor to ground. A small cap - so it only affects the highest frequencies. BTW It is also a resistor etc.

So if you put a signal into the tube, some high frequencies are going to disappear. This effect becomes more pronounced as you increase the grid stopper resistor. Unfortunately it is not on the diagram above. From radio manual, the grid stopper's job is to kill parasitic oscillations. Lowest value suggested is 470ohm, but most circuits use 1K or 1K5. (typically tied to the grid 1 lug on the socket for better results)

The larger this resistor is, the more pronounced the effect of the capacitance of the tube is. A lot of the time in high gain circuits, you have too much tops. So this is a handy tweak. You could also just add a 47pF/1nF cap from the grid to ground to achieve similar results (or from anode to power supply or from anode to ground), but less elegant IMO.

nicovlogg

Karel Mars wrote The larger this resistor is, the more pronounced the effect of the capacitance of the tube is. A lot of the time in high gain circuits, you have too much tops. So this is a handy tweak. You could also just add a 47pF/1nF cap from the grid to ground to achieve similar results (or from anode to power supply or from anode to ground), but less elegant IMO.

Makes sense to me. But now I'm questioning this mod - because then won't the overdrive channel be doubly dark - firstly due to the low plate voltage and secondly due to the grid stopper resistor?

Karel-Mars

Usually the plate resistor is left at 100K. The plate resistor is more associated with the ideal operating points for the triode.

See this data: http://tdsl.duncanamps.com/show.php?des=12AX7

12AX7 Application Data
Class Va Vg1 Ia Ra S
Triode 100 -1 0.5 80,000 1.25
Triode 250 -2 1.2 62,500 1.6

The second line is the one of relevance. Generally speaking, if you have a 12ax7 with 250V on the plate and a 1K5 resistor on the cathode, it is going to sound best. At 1.2mA current running, a 100K resistor is going to cause a 1.2X100K=120V drop; so you should have about 370V at the power supply where the 100K is tied to.

If you lower the plate load resistor, the anode voltage goes up (less voltage drop from power supply level). This might give the crunch, but is not recommended. Check production and vintage amps. They seldom exceed the 250V spec. You are still safe though till 300V; see max ratings below.

12AX7 Ratings
Vh Ih VaMax VhkMax PaMax IkMax
6.3 0.3 300 100 1 8
12.6 0.15 300 100 1 8

Old tube manuals used to have a table at the back called "Resistance-coupled amplifier data". Tables for each tube with listings of performance with different plate load resistors etc. See the Sylvania manual here http://www.pmillett.com/sylvania_tube_manuals_online.htm

See page 37 of the Appendix Section 1 of the 1951 manual. Under the Self Bias section Ebb 250V look at Rb 0.1 (=100 000 ohms a.k.a 100K). That is the Fender and Marshall spec. Note Rcf should be 470K (usually a volume pot - 500K in Marshall) and your gain is about 57. Note the table does not list lower values than 100K, but rather higher values (270K and 470K).

If with a 100K resistor on the anode, you still get a voltage higher than 250V, you should tweak you power supply resistor (increase value 10K/22K/33K/47K) to drop more voltage.

Another rule of thumb to: the anode load should be between 2 and 3 times the plate resistance for optimal sound/efficiency. The hifi guys carry on about this a lot. See the first table in this post: Ra is 62,500 for the 250V application. So the plate load resistor should be between 125K and 187K. 100K is close enough for rock and roll 8)

« Previous Page