Grid Current

If they glow, this is the place to be
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#1 Grid Current

Post by Paul Barker »

Inspired by Nick's ideas on driving A2 output valves via the cathode (Cathodyne or Grounded Grid) I have penned a 211 design which would make use of Danbury Interstage transformers.

As Nick showed if you choose a driver with the same current as the output stage the transformer doesn't see the stress of the current imbalance. for linearity I am not suggesting to regap the Danbury transformer, rather to give it an easy life. I suppose it would possibly be beneficial however to regap it a little, and perhaps use a pinstripe core with part Nickel, though impractical. But that is drifting from the point. In the circuit below don't copy the transformer connection. I haven't paid attention to making the connection so that the two currents cancel. Simply connect both ways and the way without obvious distortion due to 200mA of dc current is the one you want. I deliberately made it a design with is low (for the 211) voltage in order to explore the method within grid current regions.
Image

Added 4 watt version for the 801A

Image

It is a shame I can't find 211 grid current curves. One day someone will measure it.

I have an issue with this form of connection, but in theory there is always an issue with A2 operation because of grid current.

Nick's idea does not imunise us from the distortion of grid current.

As the cathode sinks negative relative to the grid, the grid is positive and so attracts current just as it does with grid drive. at the time this happens the anode voltage approaches the grid voltage which makes the grid exponentially more attractive to current.

I can't model things with Spice it is a step I haven't climbed.

One of these days I will write out long hand what actually happens in the grid anode interaction as we push these A2 valves into the steep climbing exponential area.

but my hunch is the effect is exactly the same whether we drive the grid or the cathode.

however the real benefits of Nick's connection is the lower input capacitance and therefore better HF performance. This is the main benefit of all GG operation wherever in the amplifier you put it.

Just a quick recap of current flow for those who are lost.

Cathodes are heated directly in this case, which cretaes a situation where when there is a proximal positive pole electrons leave the cathode and aim for said positive pole. In grid current situations a proportion of the electrons heading for the anode are sidetracked to the grid. In Nick's scenario because in AC terms the cathode potential bobs below ground (due to the AC effect the transformer has) the grounded grid is positive and so electrons divert to it.

I propose that we are simply swaping problems to other places.

The now significant factor is the dc resistance of the transformer. Lower equals less distortion during grid current.

Am I right or wrong or misguided.

Anyone fancy building it?
Last edited by Paul Barker on Fri Jan 25, 2013 6:36 pm, edited 1 time in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#2

Post by Paul Barker »

I know I keep throwing circuits up but not building anything.

I only throw up stuff I would build when other areas of life allow. right now they don't.

But I can still enjoy this hobby vicariously.

in the circuit above I forgot to mention the transformer is connected 3.75k : 3.75k so DC resistance is 200 ohms. I had to minimise that as explained in the text.

For 25 watts power output we need 150v peak to peak into the 211, we can get that from 1v RMS with headroom.

I chose the parallel 4P1L because of it's mu and I tried the 300b curves etc and found that to achieve my objectives without a high mu input valve (which I detest) this circuit was optimal. The parallel 4P1L at this exact operating point comes highly recommended and less distorted than for example the single 2a3. But with my desired Mu.

I tried to make the 801A fit the need, yes you could use it at 70mA and tolerate some DC in the interstage transformer. It would otherwise fit the criteria and quite likely sound great.

the 30 is just acceptable, the higher mu 76 or 56 could be used and I believe the sonics of that as an input valve would be extremely rewarding. It would give you more sensitivity.

the 27 would also be a good substitute.

I would change a few things for you if you wanted to use 27 or 76.

The 6J5 or 6C5 or 1/2 7n7 6sn7 would bring sensetivity right up, but with modern sources not required. Reasonably good sounding valves but very plain after hearing the 76 for example. Use 9 pin equivalents and you are alone! 8)
Last edited by Paul Barker on Fri Jan 25, 2013 9:27 am, edited 1 time in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#3

Post by pre65 »

Some scenarios (on the interweb) with 811a using cathode drive still show the grid having a bias, or is the 811a a special case ?

Mind you, those scenarios may be for ham radio use.
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#4

Post by Paul Barker »

Yes you still have to bias the valve.

I selected a operating point of 700v -20v 100mA. the bias is achieved by the 200ohms dc resistance of the interstage transformer.

While you were typing Phil I was adding detail to my second post you may need to go back a read that for completeness.
Last edited by Paul Barker on Fri Jan 25, 2013 2:18 pm, edited 2 times in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#5

Post by Paul Barker »

pre65 wrote:
Mind you, those scenarios may be for ham radio use.
that is what cathodyne operation is all about. HF! In their case it is a necessity.

At the moment I can only see two benefits to us. HF performance, and this clever use of the interstage.

but as you are learning the operating conditions have to be carefully matched. the dc conditions of the interstage determine which valves and at which operating parameters you have to work with. I have done that work for you.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#6

Post by pre65 »

Ah yes, I can see that now.

I had not realised that the Danbury interstage could withstand 80ma if used with an 811a, but now I am even more tempted to do something along those lines aka DTB's latest project.

Is that 200R of the interstage a value you have measured and is the transformer wired 1:1. Are Mikes interstages not interleaved ?

Strangely enough I won an auction for a couple of 76 the other day, and I have four 4P1L to hand so things are looking good. :wink:

EDIT I see Paul has added to his previous posts which answers my questions. :)
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#7

Post by IslandPink »

Paul Barker wrote: I chose the parallel 4P1L because of it's mu ...
Bearing in mind the 4P1L is a single pentode, do you suggest the heaters are run separately on seperate windings or do you suggest joining them in parallel ? To what extent will matching of the valves be important ?
"Once you find out ... the Circumstances ; then you can go out"
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#8

Post by pre65 »

Paul, on your drawing, I cant make the figures for the 30 add up.

If you are dropping 80v (170-90) through a 32K resistor that should be 2.5ma but you have got (at the cathode) 2v 0.06A (60ma).

According to the data sheet 2.5ma is about right.

Not trying to pick holes, just wanting to understand the basics of the circuit.

And what about the choke between the 30 and 4P1L, is it a proper grid choke, or what ?
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#9

Post by Paul Barker »

pre65 wrote: I had not realised that the Danbury interstage could withstand 80ma
EDIT I see Paul has added to his previous posts which answers my questions. :)
Just in case this one not answered.

The whole point of Nick's thought is near zero DC flux.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#10

Post by Paul Barker »

pre65 wrote:Paul, on your drawing, I cant make the figures for the 30 add up.

If you are dropping 80v (170-90) through a 32K resistor that should be 2.5ma but you have got (at the cathode) 2v 0.06A (60ma).

And what about the choke between the 30 and 4P1L, is it a proper grid choke, or what ?
Just my doodling , reminding myself of the filament requirements (2v 0.06A). I realised it was ambiguous to the reader, but the picture is of my doodles. I put these issues on only where I don't know them so that I don't have to look for the valve data when building it.

Yes I just drew a grid choke because I have some. You could substitute a resistor, I haven't looked into the max for the 4p1l maybe as low as 100k so you would have to see into that remember there are two in parallel so half the published figure (if it is translated to English). Here the grid choke looks advantageous. But 100k should be tolerated. then your coupling cap decision is made by you, this is the one -3db high pass filter you can control, your interstage and your OPT are the two elements you don't have control over unless you are wealthy enough to dictate terms!

so you need to push it down to 2 hz I suggest.

I also just drew a resistor load for simplicity. It could be replaced by a solid state ccs or Nick's pentode load. Or an SRPP or Kimmel Mu stage or anything you like.

start simple and dress it up if you wish to.
Last edited by Paul Barker on Fri Jan 25, 2013 12:27 pm, edited 1 time in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#11

Post by pre65 »

I got confused about the interstage, Nick was warning about bifilar windings.

Quote.
"But remember as long as its not a bifilar interstage, you can offset the secondary DC current by the primary current."
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#12

Post by Paul Barker »

Yes there are reasons why not to transpose the windings of the bifilar. Some other IT's also perform better one way round. I am not sure if this is with that behaviour or against it. but certainly bifilar is a one trick pony. It has one massive advantage and all the rest is disadvantage. It is not a cheap fix all. It is a cheap solution to one problem alone. If it fixed everything every 1:1 interstage transformer would be bifilar.

The 100k load gives problems the coupling cap of 1uF would be -3db at 3.5hz. So I suggest 1uF at the very least.

The grid choke looks more favourable against that fact.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#13

Post by Paul Barker »

IslandPink wrote:
Bearing in mind the 4P1L is a single pentode, do you suggest the heaters are run separately on seperate windings or do you suggest joining them in parallel ? To what extent will matching of the valves be important ?
Just from what others have found it seems they do have to be selected but they work very well indeed in pairs with very good distorion results.

I haven't spent any time thinking about the filament, it would surely be better as two individual supplies but I expect parallel would be fine. Series is naturally out of the question.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#14

Post by pre65 »

So, this is an interesting project Paul.

I have a PSU and heaters for the 811a, then I've got a couple of 76 on the way, four 4P1L in the box and a Danbury interstage on another project which I can rob.

The only thing I have not got (of any value) is grid chokes.

I must look at my wood pile to see what is left for breadboarding. :D
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#15

Post by Paul Barker »

By the way Phil on the interstage you have 4 X 400ohm dc resistance windings.

All four inseries is a 1600 ohm dc resistance choke. I no longer recall the inductance. Plenty for an anode choke not nearly enough for a grid choke.

Two in series on each side is 800ohm and 1/4 of the inductance.

Two in parallel (our forumla) on each side is 200 ohm and 1/4 of the former quater of the inductance.

Because when we are in grid current we want as low a dc resistance as possible we are using 200ohm.

We therefore have to find a driver to suit and within all the other constraints and issues involved the arrangement chosen is optimal having considered 300b and 801a I landed in the lap of what is published.

the 300b was pretty favourable except it's low mu. You would have to basically put one of those horrible high transconductance first valves together and I won't put my name to it.

But I am forgeting the fashion here is to have a 2a3 or 6v6 gain stage active preamp. In which case you don't need the 1v RMS sensitivity I have designed it for.
Last edited by Paul Barker on Fri Jan 25, 2013 12:47 pm, edited 1 time in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
Post Reply