Bleed Resistor to Discharge High Voltage Caps in Tube Amp

[Enzo]

Yes^^^

R128 is unambiguous, there will only be one R128. There might be several 10k resistors for example, and there can even be more than one in a power supply. We only have the one 30k in this particular page, but in general it is less confusing to use the component numbers. So R128, R130, C24, Q101, D22, etc.

[Boogie]

The A node powers the phase inverter - so I guess it needs more power than the rest of the preamp,since its like halfway between the preamp and power amp? (Technically i think it counts as part of the power amp?)

It doesn't need more power, but it gets more voltage. Power and voltage are not the same thing. They want the higher voltage on it for headroom, so the signal won;t clip. The PI is part of the power amp, yes, but the power supply doesn;t care where the tubes are, it just supplies current.

Good point, without really thinking it through, I was somehow thinking of power (= joules per second) as energy being conserved, as if it would have to be constant through a series circuit. But this is not the case at all, and the only thing that is constant at all points in a series circuit is current.

Having said that, I was also thinking how the signal level increases progressively at each amplifier gain stage, so the power amp stage would draw the most power from the PSU, and it seems maybe logical that the phase inverter would draw the next most to that (ie use more power from the PSU than the preamp stages, but less than the finals). Unless the signal level is still so low at the phase inverter that this doesn't hold true?

[Enzo]

I think you are still confusing power and voltage, and signal level with power.

A preamp stage is a class A stage, the current is always running through it. When the signal is present, that current varies with the signal, but the average remains the same. The 12AX7 needs no more power to have 80v of signal on its plate than to have 8 volts of signal.

If later stages have higher signal levels, then they will likely have higher power supply voltages too. Not because they need power, but because it give the stage more headroom. If you have a volt of signal on an input stage, 100v of power supply seems pretty roomy, but later in the amp, if you have 100v of signal, that 100v power supply would be laughable. So 300v is more like it. My rule of thumb is to expect ROUGHLY 1ma current through each triode of a 12AX7 in a guitar amp, wherever it is in the amp. The power tubes do need more current because they are doing more work.

Guitar amps don't necessarily increase the signal level at each stage. Usually there is an input stage that lifts the signal up from the noise floor. But if it then goes through a tone stack, the signal level is susbstantially reduced. And another stage brings it back up from that.

[Boogie]

I think you are still confusing power and voltage, and signal level with power.

A preamp stage is a class A stage, the current is always running through it. When the signal is present, that current varies with the signal, but the average remains the same. The 12AX7 needs no more power to have 80v of signal on its plate than to have 8 volts of signal.

Guitar amps don't necessarily increase the signal level at each stage. Usually there is an input stage that lifts the signal up from the noise floor. But if it then goes through a tone stack, the signal level is susbstantially reduced. And another stage brings it back up from that.

Yeah i think i probably am. I didnt know that about the tone controls, but it makes sense in that I used to wonder how an effects loop right before the power amp would work (like plugging into guitar pedals) when i thought it had already been "pre-amplified" several times compared to the input/pickup level signal.

So all those gain stages are really just to make nice distortion then? eg a Mesa Boogie Studio 22 with five 12ax7s, one for the reverb, one for the phase inverter, leaving three 12ax7's which is presumably six cascading stages of gain (the two "channels" on that amp are in series, not parallel) , but all that "gain" isnt really making the signal any louder? or am I still confusing power and voltage? the relationship between power and voltage would be impedance (i think? - kind of like P= V squared / R but more complex because impedance has inductive and capacitative components??) so perhaps the signal is increasing in voltage but not in power (or the other way around) as it goes through the preamp stages, because the impedance or "load" of the circuit is changing?

Something that this made me wonder - why are speaker level signals always talked about in watts (eg 100 watts into a speaker), and input/line/mic level signals in volts (eg 0.7V)?

If later stages have higher signal levels, then they will likely have higher power supply voltages too. Not because they need power, but because it give the stage more headroom. If you have a volt of signal on an input stage, 100v of power supply seems pretty roomy, but later in the amp, if you have 100v of signal, that 100v power supply would be laughable. So 300v is more like it. My rule of thumb is to expect ROUGHLY 1ma current through each triode of a 12AX7 in a guitar amp, wherever it is in the amp. The power tubes do need more current because they are doing more work.

This made me wonder about solid state amps, and even more so hybrids with valve preamps and SS power sections, I would have thought the valve preamps would run off much higher voltages than the transistor finals (though perhaps this is incorrect?), and how this applies.

[Enzo]

If an amplifier puts out some voltage into a speaker, the speaker moves. That means the power stage of the amp is doing work to move that speaker cone. Never operate a tube amp without a speaker or load. If I put an 8 ohm speaker across that voltage it will draw some amount of current. If I put a 4 ohm speaker across that same voltage, it will draw twice as much current. Current times voltage is measured in watts. SO the 4 ohm speaker will see twice as many watts. Watts are a measure of how much work the amp can do. Now maybe I should have used a solid state amp as the example there, because a tube amp doesn;t quite work that way, but it serves to illustrate. We measure amp output in watts because we want to know how much work it can do. COnsider a little Honda Civic can drive down the road a 70 miles per hour. So can a semi truck tractor. 70 mph is like the voltage. The semi truck can tow a large trailer full of stuff at 70, while the Civic probably could barely move such a trailer. The difference is the semi can apply a lot more power than the Civic.


The preamp is a voltage amp, not a power amp. Not every stage has positive gain. You could make an amp with a row of stages, each one amplifying what came before so it grows and grows, but amps don;t usually work that way. As I mentioned, the common passive tone stack is cut only. It does not boost anything. You may want to think the center setting of each tone control is "flat", but it is not. All the way up is the least cut, turning any control down from their cuts more. SO whatever came out of the first stage will be a lot lower in level when it hits the next stage.

You could use just a couple 12AX7 stages to get a lot of gain. Gain just means the signal voltage grows. But some amps, like the peavey 5150, use maybe 5 of 6 stages instead of only 2. This is not to add gain on gain, in fact they cut the signal in half or even more between some stages. This gives them that many more opportunities to shape the tone. Imagine if we ran an overdrive stage that could clip one side of a waveform. Done all the time. SO we flip the signal and run it through anothe similar stage. Now it can clip the other half of the waveform. In between those we reduced the signal level so the secomnd stage could do something similar to the first. SO five stages don't result in insane gain, it results in a sophisticated tone. Oh OK< and no shortage of gain too. And in many amps, a gain stage before a tone stack is fed first through a cathode follower stage to reduce impedance into the stack. A cathode follower has a slightly less than unity gain, so it may be a triode, but it adds not "gain" to the signal.

You can have an amp with a signal level of say 30 volts at one point. We can drop that down to 1 volt as a line level output - like an FX send. AT the FX return, we see the same level, 1 volt, com,ing back. From there it is simple for one stage to boost the signal back up to 30v if we wish, as if it never left. Gain is easy to get, we don;t have to build up to it slowly, unles we want to.

I am fudging a bit, trying hard to be non-technical. I do that to make the concept of the amp structure maybe a bit more intuitive, rather than getting involved in a lot of math

A tube amp will have a very high voltage of signal at its power tubes, but small currents flow. 100ma is a lot for a tube, but that is only a tenth of an ampere. To drive a speaker, that signal is run through a transformer. That transforms the high voltage, low current tube signal into a high current, low voltage signal for the spaeker. A solid state amp works on low voltages, so it directly drives a speaker with its low voltage and high current capability. A tune preamp can put out a line level signal - 1 volt - and a solid state power amp can amplify that up to speaker levels. The power amp has no idea what technology the preamp used. So sure, a hybrid amp might have preamp tubes running on 300v, and a solid state output section working from +/-35v. Or do it the other way, and have a solid state preamp running on +/-15v, and a tube power amp, running on 500 volts. There are commercial examples of either.

Imagine you have a PA mixer, or hell even just a CD player, You can plug it into a Fender Twin Reverb and listen to it, or you can plug it into an 800 watt solid state power amp and speakers and listen to it. Your solid state CD player or PA mixer will play into either, it doesn't care if one uses 500v and tubes while the other uses 80vDC. Likewise you can take a tube preamp and play it through a solid state power amp and speaker, or a solid state preamp and play through the same. Again, the various things only care about signal levels, not what was inside making those signals. Putting a preamp and a power amp together in the same box doesn't really change this, though it lets us take some short cuts.

[Billy R aka DynaFreak]

To state the obvious, the main way tube amps drain off voltage is via the hot tubes when you hit the off switch.

[Boogie]

Thanks , that was really helpful , and a couple of things that should have been very obvious to me i now understand a lot better (eg what the output transformer is for).

Im ok with maths and forumlas since i have a physics degree and i tutor maths part time.

the only thing from that im still wondering now is how does a tube preamp put out 1v of signal , considering the high voltages the tubes run at, is there also an output transformer after the preamp in a tube preamp/ss power amp hybrid? or even in a tube preamp thats only a preamp with no power amp.

yeah ive read that a 12ax7 has a gain of 100 so i suppose its obvious that you dont need five of them chained just to give you more gain

[Boogie]

i always leave the standby switch alone now when turning off the amp, after i read that the standby switch needs to be on (as in "on" so the amp would make sound if the power was on, not "on" as in "on standby") for the capacitors to drain through the tubes

[Enzo]

Why don't you do this experiment: Turn the amp fully on long enough the voltages have all settled down. CLip your meter across the main filter cap to monitor its voltage. Now turn off the main power switch but leave the standby switch on. Note how long it takes the cap to discharge down to maybe 10% of the full voltage. Now turn the amp back on, let it charge up and settle. Now turn both switches off and see how long it takes to discharge the same amount.

Then you will KNOW rather than assume.

How do I get 1v output? I can reduce any signal to any lower level I want. Your volume control does exactly that. If I have 100v of signal at the plate of a tube, and I want 1v output, I make a simple voltage divider with a 100:1 ratio, et voila. Look up "voltage divider."

[Boogie]

Thanks, I will definitely try that next time i have the amp opened up