One thing to remember if you want to prevent clipping, is to keep the peak signal voltage to the control grid to no more than the Cathode Self Bias Voltage if the self bias resistor has a bypass capacitor.
Example, a 12AX7 that has 1.5V self bias and a bypass cap, the grid can take no more than +1.5V peak.
In fact, the grid will start to conduct before +1.5V because of the space charge.
1.5V peak is 1.06V rms.
Now, pick how big the maximum signal at the grid can be, accordingly versus the self bias voltage you have (and remember to leave a little leeway to account for the space charge).
For example, a 12AX7 with 1.5V self bias, I recommend no more than 1.0V peak (0.707Vrms).
That really restricts the 2nd stage of 12AX7 cascade amp, to a very small input at the 1st stage grid.
Example, a 12AX7 that has 1.5V self bias and a bypass cap, the grid can take no more than +1.5V peak.
In fact, the grid will start to conduct before +1.5V because of the space charge.
1.5V peak is 1.06V rms.
Now, pick how big the maximum signal at the grid can be, accordingly versus the self bias voltage you have (and remember to leave a little leeway to account for the space charge).
For example, a 12AX7 with 1.5V self bias, I recommend no more than 1.0V peak (0.707Vrms).
That really restricts the 2nd stage of 12AX7 cascade amp, to a very small input at the 1st stage grid.
6A3sUMMER, That is a very good guide to the peak signal to be applied to the grid. No wonder I was having trouble getting a clean signal as I would of been way over the Cathode Self Bias Voltage.
I might add that this is an Audio amp and not a guitar amp as I understand there are some crucial differences.
I might add that this is an Audio amp and not a guitar amp as I understand there are some crucial differences.
Last edited:
Staying out of clipping is crucial to Hi-Fi preamp design. But the opposite is true of the majority of electric guitar preamps, which are designed to clip as and when prefered by the guitarist...
For a guitar preamp, I'm pretty convinced that exceeding this limit is actually a crucial part of getting good overdriven guitar sounds.
As the input signal amplitude approaches the self-bias voltage, grid current begins to flow, but only on positive half-cycles of input. This causes rectification effects, and both the input coupling capacitor and the cathode bypass capacitor start to charge up, shifting the bias point of the triode (colder).
The amount of bias shift varies with the strength of the incoming signal, so this produces a type of distortion that is dynamic, touch-sensitive, and musically more interesting. The output signal exhibits a time-varying duty cycle, which means the Fourier components shift and vary with time, which means the overdriven timbre varies subtly and interestingly with the players picking dynamics, and during the attack and decay of individual notes.
Along with harmonic distortion, this type of distortion also provides signal compression, flattening out the sharp initial transient of each guitar note, and increasing apparent sustain of the notes. Lack of sustain was a serious problem with acoustic guitars, limiting them to rhythm playing, and the increased sustain produced by the combination of solid-body electric guitars and overdriven valve guitar amplifiers was crucial to the evolution of blues, rock 'n roll, rock, and many other forms of popular music in the last fifty-odd years.
For decades now there have been little solid-state distortion pedals that use a pair of reverse-parallel semiconductor diodes and an op-amp to clip and distort the incoming guitar signal. This produces a precise and symmetrical clipping with no bias shifts or duty cycle modulation. And the audible effect is more monotonous, and less rich-sounding, than the untidy, non-ideal behaviour of valves driven into distortion to the point where they exhibit grid current flow, duty cycle modulation, and dynamically varying distortion! This is one of the reasons why Tube Screamers and their ilk never succeeded in displacing the tube guitar amp itself from the rigs of top-rung guitar players, only supplementing them.
This old 1995 interview with an engineer, John Murphy, who seems to have been the first person to recognize the existence of duty cycle modulation in valve guitar amps, and its audible contribution to good rock guitar timbre, is well worth a read:
https://www.trueaudio.com/downloads/EET_jlm interview-lg.pdf
Valve guitar amp design really is a different world from conventional audio engineering. Many things that are intolerable faults in normal audio, are encouraged and enjoyed in guitar amplification. Lots of THD is good. Limited bandwidth is good. Unstable bias voltages that vary with signal level are good. Flat frequency responses are bad. And so on, and so forth!
-Gnobuddy
In the 1980's I came across a "soft limiter" in Wireless World magazine.
I built one up, added a bit extra gain and some EQ and it sounded remarkably like an over driven valve amp. It was simply a couple of transistors in the feedback loop of an op amp stage. The bases were connected to a pot to alter the amount/type of distortion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.