Re: This was really bothering me...
Good definition, but this is for max power, ie, max plate dissipation, only. The 'centre point' or operating or quiescent point, need not be at the centre for the tube to still be operating in Class A1.
The plate dissipation at quiescent is the product of the anode current and the anode to cathode voltage, with no signal applied to the grid, and it can be anywhere between a very small amount and 100% (or even more before something melts).
You still don't quite have it, but don't go overcomplicating what is really a simple thing. I don't have time to analyse all your examples, but they're no neccessary to understand what's going on in the tube/cct.
Quiescent means 'at rest', so with no signal present at the grid, the tube is (signal-wise) at rest, ie DC, so it's plate dissipation will be xx watts. It's grid will have some bias value of -yy volts, so anywhere between a grid voltage of 0 and -2yy volts it's in Class A1. If the quiescent dissipation is less than the max plate dissipation as determined by the manufacturer, then obviously the power delivered to the load will be less than the theroetical maximum possible if the tube was being run at max plate dissipation.
If you still don't get it, then hit the books again. I was correct in my earlier posts, you have the terminology wrong.
bear said:
Good definition, but this is for max power, ie, max plate dissipation, only. The 'centre point' or operating or quiescent point, need not be at the centre for the tube to still be operating in Class A1.
Plate dissipation at quiescent will be 30W if you're running the tube at maximum spec.
More or less correct within the limitation that a tube can be operating wholly within Class A1 and have the quiescent (ie no signal) operating point at just about anywhere between a very small Pd% and 100% Pd max.
The plate dissipation at quiescent is the product of the anode current and the anode to cathode voltage, with no signal applied to the grid, and it can be anywhere between a very small amount and 100% (or even more before something melts).
You still don't quite have it, but don't go overcomplicating what is really a simple thing. I don't have time to analyse all your examples, but they're no neccessary to understand what's going on in the tube/cct.
Quiescent means 'at rest', so with no signal present at the grid, the tube is (signal-wise) at rest, ie DC, so it's plate dissipation will be xx watts. It's grid will have some bias value of -yy volts, so anywhere between a grid voltage of 0 and -2yy volts it's in Class A1. If the quiescent dissipation is less than the max plate dissipation as determined by the manufacturer, then obviously the power delivered to the load will be less than the theroetical maximum possible if the tube was being run at max plate dissipation.
If you still don't get it, then hit the books again. I was correct in my earlier posts, you have the terminology wrong.
Re: Re: This was really bothering me...
As reply was posted by Brett
The quoted segment makes it clear that maximum power or not is not a consideration... so I don't know what you are speaking to.
Yes, assuming that the signal applied to the grid is *limited*.
You forgot to say that. Eh?
yes. so? My point is simply that 100% Pd is NOT the limit of the tubes ability to conduct current. The definition of this fact is what my posts have been addressing.
You seem to not "have it." The point is that my statement about class A being biased at 50% is completely correct, especially if that 50% point is taken to be the same point as a given tube's Pd. That point being != saturation or Max Signal Plate Input. Obviously one can limit the grid signal applied and get "class A" operation virtually anywhere in the possible range of quiescent currents for a given tube.
Soorrry Brett. The manufacturer does provide a "max plate dissipation" figure. That is Pd, it is NOT the same as *Max Signal Plate Input* - which *is* the figure that actually determines what the max *output* power to a load will end up being.
And your example here is only valid *if* the "-yy" point happens to be chosen at the "center point" of the loadline in your example. In a PP circuit if you chose that point too low (Pd wise)you can still run "-2yy" volts of signal, but you'll be in class AB1.
You should hit the books? Heh.
You're arguing about this for reasons that are unclear to me.
I am quite certain that the terminology that I am using is accurate and correct.
The key point I was trying to make clear is/was that Pd *does not* represent the *limit* of the tube's ability to conduct and produce power & current, which isn't always clear on an initial read of a tube manual, and might lead to a misunderstanding.
Pd is 100% of Pd. And setting the bias to 100% of Pd does yield a set up condition where you are at the maximum output point for pure Class A1 operation according to the mfr's specs. However in terms of *output signal swing* this is point also the 50% point or center point.
All this assumes that they tube's have been set with an appropriate loadline/plate resistance & output load. Because interestingly, the very same Pd point is spec's for a P-P 6L6GC for both Class A1 and AB1!! The only change comes in the B+ voltage (higher for AB1) and plate load resistance. Pd, quiescent power, is the same.
As reply was posted by Brett
The quoted segment makes it clear that maximum power or not is not a consideration... so I don't know what you are speaking to.
Yes, assuming that the signal applied to the grid is *limited*.
You forgot to say that. Eh?
yes. so? My point is simply that 100% Pd is NOT the limit of the tubes ability to conduct current. The definition of this fact is what my posts have been addressing.
You seem to not "have it." The point is that my statement about class A being biased at 50% is completely correct, especially if that 50% point is taken to be the same point as a given tube's Pd. That point being != saturation or Max Signal Plate Input. Obviously one can limit the grid signal applied and get "class A" operation virtually anywhere in the possible range of quiescent currents for a given tube.
Soorrry Brett. The manufacturer does provide a "max plate dissipation" figure. That is Pd, it is NOT the same as *Max Signal Plate Input* - which *is* the figure that actually determines what the max *output* power to a load will end up being.
And your example here is only valid *if* the "-yy" point happens to be chosen at the "center point" of the loadline in your example. In a PP circuit if you chose that point too low (Pd wise)you can still run "-2yy" volts of signal, but you'll be in class AB1.
You should hit the books? Heh.
You're arguing about this for reasons that are unclear to me.
I am quite certain that the terminology that I am using is accurate and correct.
The key point I was trying to make clear is/was that Pd *does not* represent the *limit* of the tube's ability to conduct and produce power & current, which isn't always clear on an initial read of a tube manual, and might lead to a misunderstanding.
Pd is 100% of Pd. And setting the bias to 100% of Pd does yield a set up condition where you are at the maximum output point for pure Class A1 operation according to the mfr's specs. However in terms of *output signal swing* this is point also the 50% point or center point.
All this assumes that they tube's have been set with an appropriate loadline/plate resistance & output load. Because interestingly, the very same Pd point is spec's for a P-P 6L6GC for both Class A1 and AB1!! The only change comes in the B+ voltage (higher for AB1) and plate load resistance. Pd, quiescent power, is the same.
bear said:
I'm not confused, you're not reading what I've written fully.
Firstly, I've already explained that the aim is to lower the plate voltage, and that the amplifier would need to be rebiased.
A cathode can only emit so many electrons per second.
With a preamp valve like the ECC83, 12AX7 you have to approach the maximum plate voltage before you can draw the maximum cathode current at 0V on the grid. That would mean toasting the plate with about twice the permitted dissipation.
With a valve like the EL34 however, it's much easier to push the cathode's emission right up to it's limit under 0V at the grid.
I believe this is why so many guitarist prefer the distortion that comes about with the amplifier running so hard. Because the power amplifiers valves begin approaching this point.
In this case, the cathode will begin to distort the signal due to it's inability to emit enough electrons before the grid ever reaches the 0V distortion point.
I think Van Halen was creating the distortion he was by rebiasing his amplifiers for much higher cathode currents. To make sure the valves didn't go too far beyond their dissipation ratings, he would have needed to use a lower plate voltage.
I can't stress this enough... I'm really not even trying to discuss how to run, bias, or configure an amplifier for Class A.
I think you'd probably have trouble selling a $50k audiophile grade Class A guitar amp to lead guitarists for more than $1000 because it's just not fitting for the sound people expect a lead guitar to make. A lead guitar is supposed to lead the rhythm, not sit with it. You'll have major trouble pulling the guitar to the front of the music if it's crystal clear.
Eeka,
I have no clue exactly what ur thinking about.
The preceeding post explained exactly how and why Pd != maximum plate signal dissipation or cathode emission.
IF the cathode emission was limited to the level where Pd was, then there could not be output for a full half cycle of signal. So clearly there is excess cathode emission available - in fact enough to run almost *any* tube's plate into hot red glowing putty. If there wasn't, then tubes could not burn holes in the plate - they'd be self limiting - they're not.
Regarding what you said about the variac voltage drop requiring a bias reset, the point was that the discussion up to this point had not included any discussion of rebiasing an amp run like this, just dropping the AC mains voltage - whereupon I suggested that the *effect* is dual; both the B+ *and* the bias are effectively being changed.
Then I said, *if* this is how you want your amp to run, and you have someone who knows enough to "rebais it" then why not change out the power tranny to set the new "proper" voltages for tone in the first place??
Price has nothing to do with tone.
People can and do make class A guitar amps some which cost more than $1,000. Since these are commercial ventures, I presume that people buy them.
Many guitarists *do* play through "crystal clear" amps and rely upon other means of creating tone. (And, btw, there is nothing that says that an overdriven class A output stage remains "crystal clear.") Some do rely upon the distortion of the output stage and/or speakers run just below melt down.
_-_-bear
I have no clue exactly what ur thinking about.
The preceeding post explained exactly how and why Pd != maximum plate signal dissipation or cathode emission.
IF the cathode emission was limited to the level where Pd was, then there could not be output for a full half cycle of signal. So clearly there is excess cathode emission available - in fact enough to run almost *any* tube's plate into hot red glowing putty. If there wasn't, then tubes could not burn holes in the plate - they'd be self limiting - they're not.
Regarding what you said about the variac voltage drop requiring a bias reset, the point was that the discussion up to this point had not included any discussion of rebiasing an amp run like this, just dropping the AC mains voltage - whereupon I suggested that the *effect* is dual; both the B+ *and* the bias are effectively being changed.
Then I said, *if* this is how you want your amp to run, and you have someone who knows enough to "rebais it" then why not change out the power tranny to set the new "proper" voltages for tone in the first place??
Price has nothing to do with tone.
People can and do make class A guitar amps some which cost more than $1,000. Since these are commercial ventures, I presume that people buy them.
Many guitarists *do* play through "crystal clear" amps and rely upon other means of creating tone. (And, btw, there is nothing that says that an overdriven class A output stage remains "crystal clear.") Some do rely upon the distortion of the output stage and/or speakers run just below melt down.
_-_-bear
People, EVH lies: he had his plexis MODDED!! and he used boosters before the amp. With preamp mods (think plate resistors... 100k is usual, but some prefers 220k, much more gain), different speakers and of course, high output pickups and mainly STOMPBOXES you can obtain that kind of sounds from a relatively soft plexi...
And for that class A debate, go read the Aiken pages on "class A".
http://www.aikenamps.com/
Biasing a tube at 100% dissipation is because in the center line of its operation there isn't the half dissipation, but the whole one! At an end of the line there is max V and zero I, on the other there is max I and minimum V. So the max dissipation power is in the middle! on the borders you have zero dissipation! That's because switching transistors are so efficient: they operate between points of zero dissipation, and the only power dissipated is when it changes its state, a short time.
In a tube amp, 70% max dissipation is typical in class AB amps, 'cause it had to increase when the maximum output is given. In a class A amp, the common bias is for about 90%, because you should bias for the full 100%, but I fear that if you left your amp unplugged for a long time, with some not-so-good tubes (think Chinese), the tubes will cook.
And here in Italy musicians are very poor, so they tend to ask to the technicians more lifespan from the tubes than asking for that "funny 100% thing I saw on that site on the Net"...
And for that class A debate, go read the Aiken pages on "class A".
http://www.aikenamps.com/
Biasing a tube at 100% dissipation is because in the center line of its operation there isn't the half dissipation, but the whole one! At an end of the line there is max V and zero I, on the other there is max I and minimum V. So the max dissipation power is in the middle! on the borders you have zero dissipation! That's because switching transistors are so efficient: they operate between points of zero dissipation, and the only power dissipated is when it changes its state, a short time.
In a tube amp, 70% max dissipation is typical in class AB amps, 'cause it had to increase when the maximum output is given. In a class A amp, the common bias is for about 90%, because you should bias for the full 100%, but I fear that if you left your amp unplugged for a long time, with some not-so-good tubes (think Chinese), the tubes will cook.
And here in Italy musicians are very poor, so they tend to ask to the technicians more lifespan from the tubes than asking for that "funny 100% thing I saw on that site on the Net"...
Giaime said:
What?
Zero dissipation? Transistors are so efficient?? Eh?
Lost me here.
In a tube amp, 70% max dissipation is typical in class AB amps, 'cause it had to increase when the maximum output is given. In a class A amp, the common bias is for about 90%, because you should bias for the full 100%, but I fear that if you left your amp unplugged for a long time, with some not-so-good tubes (think Chinese), the tubes will cook.
In class A push-pull the dissipation for the two tubes sums as a nominal "constant", each tube is not dissipating "constant" power as the signal swings. It's a "see-saw". One goes up, the other goes down.
With any class of linear operation the signal swings can exceed the rated nominal Pd "center point" value for the tube. In every case the tube is passing *higher* power and current at signal max, and *less* power and current at signal min - it's not "zero" on both ends. The tube is a valve that merely controls the passing of the electricity - like a water valve in that respect. Less is less, more is more.
_-_-bear
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.