Vacuum Tube "grids"

On a sunny day (Wed, 22 Oct 2008 06:01:13 -0500) it happened "Jon Slaughter" wrote in :

No they are not. If you had read the link I provided in the discussion about tubophool amps then you would know.

No

There is a reason for the grids, and preventing secundary emission is one of them. In a zero bias drive in for example a 807, you can connect 2 grids, you have to supply power as you will drive the grids positive, they will conduct.

No

I dunno, there is this video of this guy making his own tooooooooobs, so obviously if you are tooooooobophool then you can do whatever you want, in the best case reinvent the beam tetrode and penthode, and, if you are so inclined and love grids, the hepthode etc.

Look up ECH83,

I stopped following that discussion a long time ago... I've got better things to do with my time than wade through all the posts about "I know more than you do".

All the books I have shown essentially show the grids as being the same except physically different. e.g., the screen is larger but configured exactly the same as the grid. The grid being closer to the cathode so it has a stronger effect but other than that...

Huh?

The grids have a purpose of creating an electric field to either attract or repel electrons, right? If that is there only purpose then it shouldn't matter if I mix and match(as for as for my specific application) unless some grids can only repel and others only attract?

No is not good enough.

What am I suppose to be looking for?

Everything I have seen shows the grid's being electrically identical(I do not mean they have the same electrical effects but that they behave identically. (essentially a wire mess that you apply a voltage to and it creates and electric field just as all resistors behave identically even though any two resistors might not be the same)

look, here a simple question you can answer and explain why(if you dare): If I have a pentode amplifier that doesn't use the suppression grid but I disconnect the circuit for the grid and connect it to the supression grid, will the circuit work? I'm not saying if it will work as expected... as if it was an audio amp and now it doesn't amplify properly(maybe distortion or weak or whatever).

I'm also not saying how they are generally used as I already pointed out. Just as diode is generally used for rectification doesn't mean it is its only purpose(it can be used for limiting, logarithmic conversion, etc...)

Simplying saying "No" isn't going to cut it with me. If you don't care to explain why or point to specific references that do then theres no bother to respond. I'll eventually figure it out one way or the other and if it's the hard way then so be it but at least I've learned something rather than assuming it is false and having no understanding why except some "expert" says "No". (and I've seen experts come and go and be wrong many times)

Burning another bridge?

On a sunny day (Wed, 22 Oct 2008 06:48:18 -0500) it happened "Jon Slaughter" wrote in :

Well, you are a bit over the top now aint't it? People provided references, but your above statement would put you in most people's kill file. I do not use kill files, too much work, so ... but I select from headers / subject lines / and poster names if must be.

Then why ask here? Do you want others to google for you? Wipe your rear end too?

I am by no means a tube expert, except that I have used tubes from my first days in electronics, 1958 or so, my first radio had a DL92 (battery powered tube) in it.

In my school days we learned 'tubes', to do the math, different types, reasons for all those grids, transistors were mentioned too. I designed and build tube amps, a tube TV set, high power tube linear amp, oscillators, mixers, SSB exciter, cannot remember all that stuff. But I stopped using tubes in the sixties when good transistors became available. So, anyways from a theoretical and practical POV I should be able to answer a decent question about tubes, some things I have forgotten though.

But for somebody to ask a good question he must first know some theory, maybe even history, about tubes. Your question is incomplete, vague, and you attack the teacher.

So you cried for example:

Now think for a moment, if you connect screen and control (grid 1) of a tetrode together, and put +200V on it, then the first grid becomes anode, will likely flash over to the cathode, melt and burn up.

Some info:

In fact the function of 'some grids' IS to repel, to repel electrons leaving the [over]heated anode and going for the screen, like in that EL84:

You will note the 'third grid' connected to 0V, thus presenting a barrier.

etc. use google. If you must, get some old books on tubes perhaps. Better forget the tubes, and use solid state, it will be harder and harder to find tubes anyways, even the PL519 etc high power TV tubes (never used those for audio though).

So, toooooooobophool :-)

If the OP really wants to learn some tube theory, he should find an old ARRL handbook. I still have the one that was given to me in 1975. It contains allot of good information on tube theory and it covers transistors too. He should be able to find one on ebay for real cheap:

Here's a couple:

Too bad he won't see this.....

Careful there or you'll end up in his killfile. ;-)

The positions of the grids matters a lot. The suppressor grid is very near the anode. It can't do much to control the current flow. You need to make sure that it always stays more negative than the plate. It usually is connected to the cathode to do this. If it comes out on its own pin, you can run it at other voltages and capacitively couple it to the plate.

The screen grid is sort of in the middle. Its voltage will control the current somewhat. It normally is allowed to pass about 10% of the plates current. You can run it at a lower voltage and a lower current at the cost of some gain. The screen grid is normally at the same AC voltage as the cathode so that it protects the control grid area from the AC on the plate. You can drive it with a signal to modulate the gain of the stage.

Contrary to the popular belief, it is not the grid which controls the current from the cathode, it is the electric field in volts-per-metre which the grid is partly responsible for creating.

This may seem like nit-picking, but you can deduce from the correct definition that the closer the grid is to the cathode, the steeper is the voltage gradient which it can create for a given signal voltage swing (thus the greater the amplification of the valve).

The second and third grids of a pentode are a long way away from the cathode and give very little amplification, they also have widely spaced turns which give poor control of the current. They are there for a different purpose and are custom-designed for that purpose only.

You could spend the rest of your life learning about valves and still only know a fraction of what has been discovered about them - but a good place to start for a proper understanding of the basics would be:

"Radio Engineering" F.E. Terman (McGraw-Hill 1932 or later editions)

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

Play if you like. I'm not sure what you mean by "virtually identical", but from your responses to others it seems pretty broad. I have a coat hanger that's made of wire, does that make it "virtually identical" to the screen grid of a 6L6?

Different grids within one tube are not identical, nor are different grids in different tubes. They have different effects on the cathode current, they have different effects on the plate current, and they have different responses to flowing current (i.e., you can burn them up if you're not careful).

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

Powell's books has a copy:

While you're there you may also want to see if they have older copies of the ARRL handbook -- it has a lot of good "technician level" information that's great if you're just starting out, and many many practical (if not at all audiophile) circuits to peruse.

Anything 1965 or earlier should be chock full of stuff. I prefer the

1940's and '50's books, because the equipment looks like the junk I would buy at Goodwill in the early '70's to dismember.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

What Adrian said.

One small addition: because the electrons see each grid sequentially, later grids see only what the control grid lets through. That's how pentagrid and hexagrid converter tubes work--you use the cathode, control grid, and screen grid to make the equivalent of a common-plate oscillator, so that the electron flow is modulated at the LO frequency. Subsequent grids intermodulate that with the RF input, producing a component of plate current at the IF.

For the OP: It's quite possible to make a triode amplifier by wiring all the grids together, and sometimes it's even a good idea.

Cheers,

Phil Hobbs

Ok, thats all I needed to know. That tells me that the grids are functionally identical. Doesn't matter the distance or anything because that only modifies the output which I'm not worried about(well, in fact that's why I want to know so I can actually modify the output).

If I can tie them all together that means I can play around with them and treat them the same. i.e., treat each one as a control grid using the same circuit topology and get different effects. If one grid, say, was functionally different than another I couldn't do that. Having a different geometry or physically being further away isn't a functional difference. Two random resistors are physically and geometrically different yet are functionally the same. (This is not to say that changing the resistance value won't have a signficant effect on the circuit)

What you have told me is what I assumed but needed verification. (although it's similar to Jan's "No" answer but since I have done my own reading and concluded that you have only reinforced it. Jan's "No" contradicts what I have read and he doesn't support his claim at all)

I know how the tube grids work in standard applications but I am using a pentode for audio freq amplification which normally one uses a triode. I could just disconnect the screen and suppressor but I was wondering what I could do to take advantage of them.

The first idea would seem to be to tie them all together like you mentioned as it would act like a "super grid". I do not know how it sounds but thats the whole point. There are other possibilities too such as using the screen for an attenuator and the supressor as a mirrored control. Again, I do not know how well it would work or sound but the point is that I'd like to try it but I needed to know if each grid was functionally equivalent.

On a sunny day (Wed, 22 Oct 2008 12:29:32 -0500) it happened "Jon Slaughter" wrote in :

Holly golly, whats next.

They are not.

You hears what you want to hear and disregard the rest. Unlike in that song, it does not make you a man.

Can I hear the remote sound of plonking in the distance?

Not if amplified by a triode connected penthode.

LOL wtf do I care.

or,

use

is

Suppressor grids are sometimes used for AGC or for keying/gating in pulse circuitry.

It is not too uncommon to use tetrodes or pentodes in "triode mode" by tying the grid and screen together. That doesn't mean the screen grid is the same as the control grid, in fact in beam power tubes the screen grid is sometimes better identified as a "beam forming plate". A good chunk of tube design for many years was heavily influenced on whether you wanted to pay patents to company 1 or company 2, and part of the distinction was whether the screen was a screen or a beam- forming-plate, and sometimes the words have their definitions twisted to mean something you wouldn't think they mean in an attempt to use notation to sneak around patent issues.

I have seen some applications use beam power tubes with nothing - nada

- attached to the anode plate. They use the screen or beam forming plate (which itself can have a respectable dissipation) as the anode.

There are some interesting heptode/mixer/computer tubes with multiple grids for specialized purposes. Look up for example the 6BE6 and its many variants for very common examples.

Tim N3QE

ke

ed

od

) ,

ol

or,

ate

all

hat

e

rent

e h d

=A0I

I

ed

s

creen

t y

uoted text -

If they were "functionally identical", you could swap the signals on the pins without any bad results. Trust me, that doesn't work.

If they were "functionally identical", you could swap the signals on the pins without any bad results. Trust me, that doesn't work.

---------

Depends on what you mean by bad results? Bad as in the thing smokes(which is what I consider bad) or that the output signal isn't what is desired?

I am only concerned with ruining components and not what the output signal does. I can't easily replace components but I can easily reconfigure them.

What you have said contradicts what phil said.

I probably shouldn't use "functionally identical" because you guys don't understand what it means. I've gave some examples but you don't look at it in the abstract sense. Obviously it requires context which I probably didn't make clear. For me it is functionally identical if the external circuitry that configures it can be used in each case without smoking any of the devices except for changes in "strength" of those components.

e.g., maybe I have to increase the resistor values for the suppressor grid but I still use the same topology. Then it is functionally equivalent. i.e., everything functions the same without any abstract changes. The output might be different but I don't care about that.

For example, a resistor and a capactor are functionally equivalent if they are only passing a fixed frequency. Sure they might have some different effects but if those effects are not important then either one could be used. To say they are not the same and it is impossible to switch between the two is simply wrong. To say that in some cases you cannot is a different story.

So, What you and Jan have said is at odds with what I have read and what Phil has said.

What you are telling me is that if I configure a circuit that uses the control grid and works but then later on rewire it to the supression grid but scale the values so the power dissipation is within rating then it won't work. Phil claims it will. I claim it will. It might not give an output that even resembles the original but that is not the point.

On a sunny day (Wed, 22 Oct 2008 14:35:18 -0500) it happened "Jon Slaughter" wrote in :

same function: a control grid has a different function then a screen grid, the grids in an ECH83 have all different functions (you would have found out if you followed that link). Actually there are a few basic errors in that text, oscillator pulling does _not_ help on strong signals in a super heterodyne, as the oscillator is on a different frequency. And the presence of the local osc / mixer indicates a super heterodyne. Probably written by some confused American.

So it is your English that is not functional (nor identical to main stream).

...

...

A more common way is to tie screen to anode.

The suppressor grid is often not accessible and is tied to cathode internally.

kevin

I didn't claim that. The suppressor grid is usually wound very sparsely and positioned close to the plate--its job is merely to control secondary electron emission from the plate, which leads to a nasty nonlinearity in large signal operation. Normally the plate is more positive than the screen, so the secondaries go back to the plate. However, with large signals the plate can be instantaneously below screen potential. At that point the secondaries can start to dominate conduction, leading to reverse conduction between plate and screen grid, which effectively changes the tube into a triode for that part of the cycle.

The suppressor grid is held at cathode potential, which means that an electron whipping past the screen and approaching the suppressor would slow down to zero speed if the grid were anything like as tightly spaced as the control grid--which would mean that the tube wouldn't work.

The loosely wound suppressor steers the electrons around it--just a little, in the case of the main conduction electrons, but by a lot in the case of the low energy secondaries. That prevents the formation of the 'virtual cathode' that causes the nonlinearity.

Cheers,

Phil

Pentagrid converter tubes like the 6BE6 are noisy because the electron current is split between the second grid (to run the oscillator) and the rest of the tube (to do the mixing). The current splitting is stochastic, i.e. it introduces shot noise, which is ordinarily much reduced by the (reverse biased) control grid.

Another interesting tube type is the 'remote cutoff' or 'variable-mu' pentode, e.g. the 6BA6, in which the control grid is wound with a varying pitch. The tightly wound parts cut off before the loosely wound ones, leading effectively to a modulation of the cathode area with grid bias. They make pretty good AGC amplifiers, but have poor strong-signal linearity.

Cheers,

Phil Hobbs