In the fender champ amp schematic, the 12AX7 pre amp has 2 class A stages. Both stages are biased the same.
A 1v signal comes in to the first stage and is amplified to 100v (the 12ax7's gain according to the datasheet). 100 volts hits the grid of the second stage but the second stage is biased for 1.5v just like the first stage.
So why does the second stage not clip like crazy? is there some huge grid impedance or something else im missing?
A 1v signal comes in to the first stage and is amplified to 100v (the 12ax7's gain according to the datasheet). 100 volts hits the grid of the second stage but the second stage is biased for 1.5v just like the first stage.
So why does the second stage not clip like crazy? is there some huge grid impedance or something else im missing?
Incorrect. Your misunderstanding of "amplification factor", or mu (µ) is what's throwing you off. The amplification factor is the ratio between a small change in plate voltage and a small change in grid voltage which results in the same change in plate current. It is an indication of the effectiveness of the control grid voltage relative to the plate voltage in controlling the plate current.
If a tube is said to have a µ of 100, it means that the grid voltage change required to produce a certain change in plate current is 100 times less than the plate voltage change required to bring about the same change in plate current. In other words, the grid voltage is 100 times more effective than the plate voltage in it's influence upon the plate current.
An example would be that if a plate current change of 1ma is produced by a plate voltage change of 10V, and a grid voltage change of .1V produces the same 1ma change in plate current, then the amplifiaction factor is 100. (µ - 10 ÷ .1 = 100) Emphasis should be placed on the fact that it is the change in plate and grid voltages that are important and not the individual values.
Most electric guitars pickup voltage are rated 100mV. High outputs ones are rated 1V.
On the Fender Champ Amp schematic I have, there are two inputs, High and Low. The High input goes through a voltage divider, which attenuates the voltage coming to the grid if the first tube (or the first half of the dual triode).
After the first tube there is a tone control which has some attenuation. After the tone control there is a volume control, which may attenuate on normal use.
Also, 12AX7 has an amplification factor of 100, but the actual amplification is lower than that. The actual amplification depends on the circuit topology and on the value of the anode resistor.
On the Fender Champ Amp schematic I have, there are two inputs, High and Low. The High input goes through a voltage divider, which attenuates the voltage coming to the grid if the first tube (or the first half of the dual triode).
After the first tube there is a tone control which has some attenuation. After the tone control there is a volume control, which may attenuate on normal use.
Also, 12AX7 has an amplification factor of 100, but the actual amplification is lower than that. The actual amplification depends on the circuit topology and on the value of the anode resistor.
How come its acceptable for amps to start clipping around volt? not that i am saying your wrong, because the output voltage of a guitar pickup seems to anywhere between 10mv and 5v. But just wondering.
Isn't that what the grid is supposed to do in this case? I have always viewed the grid as a current valve, hence why i thought they were called valves. Also doesn't the output voltage all depend on the plate resistor? The larger the input signal you have, the more variance you have in your output current which drops more / less voltage at the plate resistor. Unless im wrong of course.
Another question i have is if the cathode resistor adds in series to the plate resistor. Because current flows from negative to positive, wouldn't Rk, Rp and the tubes internal resistance all be in series with each other?
So lets use gain instead of amplification factor. In the champ amp, a calculated gain is about 60. So theoretically lets take 100mv X 60 = 6v, which is still higher than the grid bias voltage of the champ. As far as i know, this still causes the second stage to clip. Why does it not clip? Unless typical guitar voltage is more aroun 15 - 25 mv.
1. There is the unavoidable attenuation of the tone control and the probable attenuation of the volume control.
2. At high volumes and full volume, the amp probably clips, which some guitarists want.
I understand that some attenuation is unavoidable. But after 0v on the grid no more clipping can occur right? so lets say attenuation brought it to 3v, then there is still about 1.2v over 0v. (if were talking 100mvrms and 3vrms) if it smore like 100mvpp and 3vpp then the problem is solved.
3vpp would result in max clipping because it goes into grid current and to 0 grid volts, right?
Last edited:
Not quite true. The second triode has strong negative feedback applied to the cathode resistor.
Some versions of the Champ have a tone stack or a tone stack made up of fixed resistors. The early tweed Champs do not have the tone stack but there is a 1M pot between the two triodes and YES it does clip if you crank the pot to "12" and have a hot guitar signal going it.
But after that volume control pot the rest of the amp is inside a NFB loop. To talk in greater detail we have to agree on a scematic. I'd say the 5F1 is "the" classic Champ.
http://myfenderchamp.com/wp-content/uploads/2010/03/champ_5f1-schematic.png
Last edited:
i agree as well, i never considered the negative feedback. That changes things a lot. Honestly i feel much better understanding that guitars aren't really outputting 1v constantly. I using this knowledge to design my own preamp and i just wanted to understand why the champ doesn't clip at all volumes with 100v into stage 2. I knew it wasn't 100v but i wanted to know why.
I guess my only question is how far positive can you safely go on the grid, or should you never go past 0v?
As I said above, there is NFB. but I was not so clear abut how it works. The second triode's cathode is connected to the speaker terminal. So let's assume you are correct and the signal is at 6V but if so, then the cathode voltage is also high and what matters is the difference. The second triode is actually a difference amplifier.
Also you might wonder why your tube with mu of 100 only has a gain of 60. Again it has to do with feedback. The cathode voltage (and hence the grid bias) is not fixed but depends on the input signal because the voltage drop across the cathode resister depends on the current. So the more the tube conducts the less it is biased.
Which Champ circuit? The 5E1, 5F1 has two 12AX7 stages with 100k plate resistor and 1.5k on the cathodes with no bypass capacitors. Just a 1M pot between them. Also the second stage is within the negative feedback loop reducing gain.
The Champ before these models use a pentode rather than 12AX7 so it can sit this one out. The Champ after the above modeled have Fender's gain sucking tone stack (which give about a drop of 6dB when bass and treble are at their max and midrange dropped 20dB) in between the two triode stages.
I just bread boarded a 12AX7 with both stages at 220V, 100k plate and 1.5k cathode resistors, no capacitor bypass. I put 1.5V p-p into the input stage (close enough to 1V rms for me) and I got out a 30V p-p signal. So gain is 20 (remember no cathode bypass as the original 12AX7 Champ).
Then I attached the second stage, the signal at the 1M (I did not use a pot but a 1M resistor as I wanted the full signal to the second stage but wanted the 1M loading on the first stage) input to the second stage then appeared clipped in one half of the waveform. I then reduced the signal to the first stage till there was no visible clipping at the input of the second stage. I had to reduce the input to 100mV before the sine wave had no sign of clipping, and it just so happened the signal level at the 2nd stage input was about 2V p-p.
Both stages were biased at about 0.85V on the cathodes. So conceivably you should be able to put a 1.5V signal in before any input clipping, seems we can put a bit more than that.
Now just for kicks I checked how much signal was needed at the input of the first stage before the signal at the output of the second stage starts to distort. Well the two stages can put out 30V p-p (with a 220V supply) with a 60mV p-p sine wave at the input. Since this is more than enough to drive a 6V6 to full output any guitar should have enough signal. The extra gain, well you can use it as you will. Add bypass caps and you have more gain.
The Champ before these models use a pentode rather than 12AX7 so it can sit this one out. The Champ after the above modeled have Fender's gain sucking tone stack (which give about a drop of 6dB when bass and treble are at their max and midrange dropped 20dB) in between the two triode stages.
I just bread boarded a 12AX7 with both stages at 220V, 100k plate and 1.5k cathode resistors, no capacitor bypass. I put 1.5V p-p into the input stage (close enough to 1V rms for me) and I got out a 30V p-p signal. So gain is 20 (remember no cathode bypass as the original 12AX7 Champ).
Then I attached the second stage, the signal at the 1M (I did not use a pot but a 1M resistor as I wanted the full signal to the second stage but wanted the 1M loading on the first stage) input to the second stage then appeared clipped in one half of the waveform. I then reduced the signal to the first stage till there was no visible clipping at the input of the second stage. I had to reduce the input to 100mV before the sine wave had no sign of clipping, and it just so happened the signal level at the 2nd stage input was about 2V p-p.
Both stages were biased at about 0.85V on the cathodes. So conceivably you should be able to put a 1.5V signal in before any input clipping, seems we can put a bit more than that.
Now just for kicks I checked how much signal was needed at the input of the first stage before the signal at the output of the second stage starts to distort. Well the two stages can put out 30V p-p (with a 220V supply) with a 60mV p-p sine wave at the input. Since this is more than enough to drive a 6V6 to full output any guitar should have enough signal. The extra gain, well you can use it as you will. Add bypass caps and you have more gain.
Last edited:
To put this in real simplistic terms, as others have pointed out, guitar pickups rarely put out 1V, the attenuation of tone and volume controls, the 12ax7 real gain (in an actual circuit) is typically much less than the theoretical gain, voltage dividers, and negative feedback, all these factors keeps things more in control.
If you've ever heard a guitar amp that doesn't clip, like with an audio amp, you'd be surprised how it sounds, very campfire like (for lack of a better term). Guitar amps should clip, while audio amps never should, the clipping (overdrive) is an essential character for guitar amps.
I have a friend that sometimes designs guitar amps, he says his sounded like crap until he made preamp stages that had at least 6 to 12dB of overdrive capability into the driver stage, and the driver stage should be able to overdrive the power stage as well.
Typically, audio amp preamp and driver stages never go into clipping, only the power stages, and then only when the driver stage is completely maxed out (swinging maximum voltage).
When an audio amps starts to clip, we turn it down, when a guitar amp starts to clip, well that's a good thing.
It's important to realize the difference in design criteria for the two different amp genres.
twystd
If you've ever heard a guitar amp that doesn't clip, like with an audio amp, you'd be surprised how it sounds, very campfire like (for lack of a better term). Guitar amps should clip, while audio amps never should, the clipping (overdrive) is an essential character for guitar amps.
I have a friend that sometimes designs guitar amps, he says his sounded like crap until he made preamp stages that had at least 6 to 12dB of overdrive capability into the driver stage, and the driver stage should be able to overdrive the power stage as well.
Typically, audio amp preamp and driver stages never go into clipping, only the power stages, and then only when the driver stage is completely maxed out (swinging maximum voltage).
When an audio amps starts to clip, we turn it down, when a guitar amp starts to clip, well that's a good thing.
It's important to realize the difference in design criteria for the two different amp genres. twystd
If you scope a guitar pickup you will see an initial attack that can be a volt maybe two with humbuckers. But the body of the signal is not all that high. So when the attack is shaved off by the clipping it is not really noticed all that much since it is full of high harmonics anyway. And what does clipping add, harmonics. It is mainly when we boost the signal when the rest of the signal is clipped that we think it as being a distorted guitar.
The results of your bread board experiment are valid but the breadboard does not represent a Fender Champ. NFB changes "everything" and you left it out of the breadboard.
I just happen to have a couple Champs. One is a 5E1 clone I built and the other a Fender Champ 600. ( I built the 5E1 because I was slightly disappointed with the 600) Just this last weekend I took the feedback off the 5E1 in order to trouble shoot a problem. While I had it disconnected I thought I'd try plugging in a guitar. It dramatically changes the amp.
Last edited:
It also makes a huge difference it you are playing chords or a single string lead. It should be clear that six strings are louder than one string.
My theory is that the champ does not distort so easily because there is less gain than you'd think because only the first triode is outside of the NFB loop.
Do you have any evidence that it doesn't clip?
IMHO, if you have a Champ that only plays clean, it needs new tubes!
If the volume pot is turned down then there isn't clipping. I was just using this as an example if the input to the tube was 60v then there would be constant clipping at almost any volume level.
Thanks for all the help guys, i have a much better understanding of this now.
My new question is if i were to make a tube distortion pedal (essentially an overdriven preamp in a pedal) what way is best to reduce the output of the second stage back to guitar level, lets say 100mv?
I was thinking of using a voltage divider but when you put the input of the next amp in parallel with the voltage divider it will change the amount of voltage dropped by the divider.
i guess It really depends on the impedance of the input section of a guitar amp, but they could all be different...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.