SE draws max current at idle???

[jeff]

 I hear this all the time. Is this fact or fiction? I cannot understand this, could someone please explain.
If your SE amp is drawing max current at idle and your input swings from 0V to 1V to 0V(basically same as idle) to -1V how can the tube be drawing max current at idle?
When it swings from 0V to -1V doesn't that rdeuce the current flowing through the tube, so when it swings from 0V to 1V what happens? If it's at max at idle how can it reproduce both the pos and neg???
I can understand if the tube is biased to 100% of the max dissipation rating at idle but I can't understand how a SE amp can draw max current at idle. That means it can only reduce current to get a full wave???????
 Someone please explain is this just a myth or am I missing something.
Thank you
Jeff

[jeff]

I usually hear this in referecne to: SE amps don't sag beacuse they draw max current at idle. But is that really what's going on? Or is it that the amp is at idle, and when there's an input signal from pos to neg to pos to neg..etc... the power tube draws more than idle and then less than idle and then more then less...etc... but SE amps don't sag because the power supply's filter caps can keep up with this.
I just don't get it.

[jazbo8]

For SE operation, the maximum current is twice the idle current, the minimum current is zero, so the average current over one cycle is the same as the idle current, the power supply basically sees a constant load which results in no sag - all of the above assumes ideal operating conditions and using perfect tubes, but it's pretty close in real amps.

[stratele52]

You do not talk about how SE amp work but Class A amp . This is same for class A  push pull amp

[jazbo8]

You do not talk about how SE amp work but Class A amp . This is same for class A  push pull amp

Don't make it harder than it is... SE amp operates in Class A, no? How often do you see Class A push-pull guitar amps?

[tubeswell]

For SE operation, the maximum current is twice the idle current, the minimum current is zero, ...

the hypothetical maximum current is 2x idle current, but in reality the plate voltage never gets down to zero and the plate current never gets to 2 x idle current. In fact  because of inefficiencies in the circuit, peak plate current usually doesn't even get to 1.5x idle current. But for a max DC current rating, 1.5x idle current would be a safe bet.

[jjasilli]

Per the tube charts, say for a 6L6 in Pentode mode: http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6L6GC.pdf Page 4.  Average Plate Characteristics.  Pick any one of the curves,  for an example.  For any given plate voltage in the +/- 300VDC area, the tube can be biased to operate at idle at a certain point in the middle of the curve.  The slope of the middle of the curve is nearly horizontal.  Hence, as signal swings the voltage a bit around the operating point, the amount of current draw (vertical axis) remains nearly unchanged under idle vs. signal conditions.  Also, see:  http://diyaudioprojects.com/mirror/members.aol.com/sbench102/pent.html


But, at the far left of the tube curves is a knee.  The 6L6 curve is near vertical between 0 - 25 plate volts, and not yet very horizontal between 0 - 200 plate volts.  If the tube were biased to idle at a point within that low voltage range, then the swing of signal voltage would more substantially affect current draw, because of the steeper slope of the curve.


Also, screen current varies with voltage.  So, a large screen resistor will cause current swings and affect the operation of the tube under signal conditions.

[HotBluePlates]

Per the tube charts, say for a 6L6 in Pentode mode: http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6L6GC.pdf ...  Pick any one of the curves ...

Also, screen current varies with voltage.  So, a large screen resistor will cause current swings and affect the operation of the tube under signal conditions.

Make Life Simple.

Go to Page 2, compare "Push-Pull Class A₁ Amplifier, values for two tubes" (take the 270v plate condition) to "Push-Pull Class AB₁ Amplifier, values for two tubes" (take the 470v plate condition).

Class A:
Zero-Signal Plate Current + Zero-Signal Screen Current = 134mA + 11mA = 145mA
Maximum-Signal Plate Current + Maximum-Signal Screen Current = 155mA + 17mA = 172mA
Average-Current Change from Zero-signal to Maximum-signal = 172mA/145mA - 1 = 11.7% increase

Class AB:
Zero-Signal Plate Current + Zero-Signal Screen Current = 116mA + 5.6mA = 121.6mA
Maximum-Signal Plate Current + Maximum-Signal Screen Current = 210mA + 22mA = 232mA
Average-Current Change from Zero-signal to Maximum-signal = 232mA/121.6mA - 1 = 90.8% increase

G.E. already converted the currents into correct average current for you, allowing quick comparison. Current nearly doubles in the Class AB case, and that increased current pulled through supply impedance equals sag.

[jjasilli]

Yes, the p 2 Charts show that (but not why) this is true, which at first blush might seem hard to believe.  After all, when the amp is putting-out more volume, common sense seems to require that it must be doing more work and drawing more current.  Why is this not so?  Jeff asked for an explanation.  That resides in the curve plots.


Furthermore, the statement SE amps don't sag because they draw max current at idle is not strictly true under all conditions.  To reiterate, it is true only under "usual conditions" in which the tube is biased for clean operation in the middle of the curve.  It is not true for an operating point in the low voltage range.  Again, the reason why is shown by the slope of the curve at the different operating points of the tube.

[HotBluePlates]

Yes, the p 2 Charts show that (but not why) this is true, which at first blush might seem hard to believe.  After all, when the amp is putting-out more volume, common sense seems to require that it must be doing more work and drawing more current.  Why is this not so? 

Without a guided tour, plots aren't much help; and you need to draw a load line at a selected supply voltage just to get what the values of current are at idle and peak conditions. And then you have to go through the math to interpret what's happening. All that is already done by the listed conditions, so...

Look again at the numbers I posted (but now condensed for quick reading): Class A has a maximum output power average current which is 111.7% of idle current (or an 11.7% increase), while Class AB has a maximum output power average current of 190.8% of idle (a 90.8% increase). This captures the large change of average current from zero output to maximum output.

Then a Big Current Variation * Same Resistance = Big Voltage Drop, compared to small current variation * Same Resistance = small voltage drop.

This stuff is best taught in person, especially if the audience doesn't have a strong grasp of fundamental electronic principles.

[PRR]

> SE draws max current at idle??

Whoever wrote that wrote it wrong.

The ideal class A (mostly SE) amp draws the same current at any power output.

The *dissipation* is max at idle, yes. That is not the same thing.

And be careful when you think of "max current". We pick a tube which can do 100mA, then idle it at 50mA. So that, with signal, it can go (ideally) up to 100mA and down to zero mA. While the current changes at audio rate, the am-meter can't follow such fast swings, so it reads the average, 50mA. (In real life, zero current is hard and distorty, so we may swing 90mA to 10mA, or even less swing. Still ideally 50mA average.) (Tubes are not ideal. In general, average current wants to rise 10% with big output. However at BIG output the current is determined by the load, not the bias and drive. Use a too-high load, the average current will decrease at large-signal, and may rise 20% with a too-small load.)

The dissipation is max at idle. The tube runs cooler when putting big power to a load. I have run a creaky 6550 on a dummy load and heard this happen. It warmed up, or cooled down, it creaked. Creak at turn-on, again at turn-off, like an old VW. When idling zero signal it was hot and quiet. When I beat big power to the dummy load, it creaked for a few seconds; again when I quit beating it. Thermometer confirmed cooler when dumping power to load.

Think of a Class A amp this way. You want to wash cars. You have a fire-hose, but no valve. You run the fire-hose FULL blast while waiting for a customer. Run it to a gutter to drain. When you get a dirty car, you use a shovel to deflect a little water, or a lot of water, as needed, onto the car. You can divert 0% to 100% of the water to the car. When not washing, or doing a light spray, there is a maximum amount of waste water to the gutter drain. When you have my big filthy tractor and divert 100% of water to the tractor, 0% of water goes to waste-drain.

That's over-simplified non-audio. For some reason, audio is usually measured as Sine Wave. Imagine your wash-customers demanded a tremolo-wave of more/less/more/less water. You waggle the shovel. When you do this, you get half the fire-hose water on the car. So the maximum car-water is 50%, the other 50% wasted.

A Class A non-Sine amplifier will do better than 50% efficiency. I got over 70% putting square-waves through that 6550. However the reference in Audio is usually "clean". If an amp can pass a Sine wave cleanly, it will pass any normal Audio signal cleanly. Guitar amps are different; nevertheless, we normally measure them as Sine Power even if we then turn to 11 and beat the strings to get timber-saw-tooth waves.