Separate names with a comma.
Discussion in 'Amp Tech Center' started by peteb, Oct 8, 2017.
The page doesn't mention fixed bias, just cathode and grid bias and it never discusses bias current.
they kind of start with fixed bias to explain cathode bias, but they are talking about triodes, which used as preamp tubes are typically cathode biased.
heres another one I like:
why am I bringing these up?
I like these descriptions, they match my idea of what is going on in a tube.
Both sites haves the grid negative even with the positive signal voltage added to the negative bias voltage. This is what makes sense to me.
I have been absorbing amp knowledge for a long time, and I just sort of assumed that everyone who has looked closely at the bias to signal relationship also understood that the signal voltage is not to exceed the bias voltage. (but some people are saying it is not true? maybe it isn't)
I don't totally know this to be true, but the signal voltage being less than the bias voltage makes sense to me. The first link says that it is so, but doesn't really give a good reason for it. The second link gives a couple of reasons for keeping the grid negative. A positive grid is going to lead to excessive current thru the tube, and a positive grid will attract electrons and distort the signal on the grid.
To me, this is one of the fundamentals of tube operation. The negative bias keeps a lid on the current, and the positive signal opens up the tube allowing the current to flow by negating (or undoing) the negative bias.
Oh man! Angelfire! That takes me back...
BTW, this page is not "by" Angelfire. Angelfire is a web hosting service.
I like how the signal voltage is shown as a moving needle on a meter.
As the signal voltage goes up the current flow is shown to increase.
I think they should show another meter showing the bias voltage, and as the negative (instant) bias voltage approaches zero the current flow increases.
The signal to bias relationship is at the heart of every tube amplifying a signal.
Very few online amp resources, or even the amp books I have go into the detail of the actual signal bias relationship.
It seems like an important thing to understand, if you want to understand tubes.
Bias voltage is the idle, no signal voltage between cathode and grid. In cathode biased circuits there is an interaction between a signal on the grid and the voltage drop across the cathode resistor and therefore the cathode voltage. You cannot separate bias and signal voltage so you cannot measure bias and signal voltage separately. This is the concept that is keeping you from understanding how a signal voltage greater than the bias voltage can be amplified without hard clipping.
thanks for the reply Rob,
Are you sure the signal voltage can go higher than the bias voltage? I never see it documented that way, and when it is documented, as in the two links I linked, it will say the signal voltage is to be less than the bias voltage so the grid stays generally negative. The RCA tube manual says the same thing, the signal voltage is less than the bias voltage.
In what case does the signal voltage exceed the bias voltage? It is not the case in typical preamp or power tubes, in these cases the bias voltage always exceeds the signal voltage.
You say that you cannot measure bias voltage and signal voltage separately. I don't see why not. There is no problem measuring the negative idle DC bias voltage. And, there is no problem measuring the AC signal voltage on the grid, and because it is AC, and the meter is set to AC, the AC is measured and not the DC.
It is agreed that the bias voltage is the DC on the grid at idle, the signal is the AC on the plate, and I was using another term, instant bias, which is the DC and the AC added togethor.
I really don't go along with your statement that you cant measure the idle bias DC separately from the AC signal, and vice versa, and so I disagree that this is holding back my understanding.
I also don't see any evidence that the AC can exceed the DC significantly, maybe by a little momentarily, but otherwise the grid would be positive which is a no no.
Why this old topic came up again is that I found some decent sources that say that the grid is to stay negative.
Do you have any decent sources that show that it is OK for the grid to go significantly positive?
From the OP's Angelfire link, a decent description:
"As the input signal swings positive the grid becomes less negative. This allows more electrons to get through to the plate which is an increase in plate current. The voltage drop across the plate resistor becomes greater and because the B+ end of the resistor is at a fixed voltage the plate end becomes less positive. You can see right off that as the grid voltage goes up the plate voltage goes down. This is the nature of this type of amplifier circuit. When the input signal swings negative the more negative grid cuts down the number of electrons getting through and the plate current is reduced. This causes the drop across the plate resistor to decrease and the plate voltage gets closer to the B+ voltage, higher. Neither the grid voltage nor the plate voltage gets all the way to zero. If it should the amount of distortion would be very high."
^This is the key.
there should be no question here.
There may be (slight) exceptions to the rule, but the rule is the grid is negative to zero.
Check out the RCA receiving tube manual, it clearly states:
In class AB1 the peak signal voltage is equal to or less than the negative bias voltage.
(I don't see that this leaves any room for debate)
Then if you consider that the RMS voltage, the voltage that is measured or is on a schematic, is only about 71% of the peak voltage, one can see that the magnitude of the signal voltage is less than the magnitude of the negative bias voltage.
the point of this thread is not to argue about it, it is to bring up quality sourced info to make the point.
Cathode-bias is just that, a resistor-generated positive voltage present at the tubes cathode.
Fixed-bias is different, it's an externally supplied negative voltage applied to the control-grid (*)
With cathode-bias, the control-grid is typically at ground-level voltage (0Vdc), so raising the cathode ABOVE ground with a positive voltage, makes the control-grid "negative" by inverse relationship, creating the negative bias upon the control-grid.
(*) but an externally supplied negative voltage can actually be applied to the cathode, ala' MusicMan™ amps.
Link and page # please?
That's referring to a fixed bias power tube. I'm talking about cathode biased preamp and power tubes.
In an ideal situation where the cathode resistor is fully bypassed and the signal on the grid is symmetrical, there will be no interaction between the signal on the grid and the voltage drop across the cathode resistor. This is the case for the example that Fun with Tubes used to explain cathode bias. They only briefly hit upon non-bypassed cathode biasing at the very end. The other article took such a beating in translation that I didn't read it.
The main thrust of this thread seems to be the grid voltage going positive with respect to the cathode voltage and whether that is permissible. Given that virtually every guitar amp allows the grid to run positive of the cathode on the triodes, it is not only permissible, but it is encouraged. That is where a lot of the character of an amp comes from. For clean operation, the increase in grid current with positive signals becomes a problem and this is probably why they suggest keeping the grid voltage below the cathode voltage.
Triode grid current doesn't suddenly start when the grid-to-cathode voltage hits zero. It is there even when the tube has no signal on the grid and that is the basis for grid leak biasing. The grid current increases exponentially with grid-to-cathode voltage with the knee kind of around zero volts. The grid current bucks the positive grid signal causing non-linear operation since the grid current changes in a non-linear fashion with grid voltage. The only thing that happens with increasing grid signals beyond the bias is that the tube operation gets more non-linear (that is until the extreme when blocking distortion occurs and the tube runs out of gas).
Yep, the basis for WHY / HOW grid-leak biasing works is the control-grid's slight capture of electrons. However, unless grid-leak biasing is actually being employed, the OTHER bias voltages, cathode or fixed, tend to swamp that slight grid-current under normal operating conditions.
good discussion all, thanks
I have only briefly read thru it, and wont be able to read thru it in detail until later.
just quick comments:
Link and page number please:
the RCA manual is online, but everyone should get one, they are only a couple of bucks. To put the quote in perspective, it is the universally accepted definition of class AB1 versus class AB2. class AB1 the peak signal does not exceed bias voltage and class AB2 does. but Aiken amps says that typical amps are AB1 and it would take significant changes to the amp design to make an AB2 amp.
R.E. cathode biased versus fixed bias and preamp versus power amp. I agree that most documentation focuses on power tube bias and fixed bias. I was reading recently in Aiken that what goes on in the power tube is basically what goes on in the preamp tube.
Tenover, you are right on on the thrust of the thread. you are saying that the preamp signal voltages regularly exceed the bias on the preamp tubes. Could you please provide some examples with some numbers. What I have found from the newer schematics that show the signal voltages at the grids is that the signal voltage throughout a hot rod deluxe preamp is less than one volt and the bias on all preamp tubes is around 1.5-2.0 volts, the relationship holds true, the signal is less than the bias.
there is a reason for the negative grid, if you idle biased a power tube with a positive grid, the tube will conduct lots of current and red plate. Would, do preamp tubes redplate as well when the grid is positive?
For any of you who may not have completely finished memorizing all 762 pages of the RCA .v30, here's a copy:
(if you quote from it, please kindly note the page thx)
Such a basic question, showing complete lack of any fundemental understanding of triode set-up; loading and biasing.........yet you write so much about the subject, most of it also showing a complete lack of any fundemental understanding of triode set-up; loading and biasing. To copy snippets from text-books and post them here is easy, but your own added hypothesizing
thanks Ludwig, I will find the page for you.
"complete lack of understanding....blah blah blah blah blah"
A big lesson I've learned in life is to meet someone halfway, or better yet, go beyond half way.
When someone begins a comment with "your complete lack of understanding.." it is a complete non-starter for me.
for instance, if I am debating with someone I always strive to find the common ground that we can both agree on and then explore the differences beyond that. There is a term for it. First, you say I agree with you on this, then you say, but then I disagree with you on this. Its much more effective than starting off by saying "you are completely wrong". Relationships with people are very complex and dynamic, and when one person only tries to look at something from their own perspective without trying to see the other persons perspective at all, it shows a lack of communication skills and a lack of effort in communicating.