Hey Tube Guru's

[Ed/KB1HYS]

I need someone to walk me through tube biasing. I have several tube manuals (PDFs) and I've read the front chapters,  I understand the basic concepts of biasing for class of operation, and tube operating parameters, but I get stuck trying to work the charts to determine what voltages/currents are needed to get there.  I've attached a sample chart for a 6AU6A pentode, to work as an example. Looking at this chart, I can tell that with 200 volts on the plate and no voltage on the control grid, it'd be running 10mA plate current, so that's 2 watts input.

So if I want to bias this as a class A amplifier, I need an input signal that never goes positive, and never gets below -3 volts?  like a 1 volt pk-pk input with a half volt bias on Cg1? or do I need 1.5 volts of bias regardless of the signal pk-pk voltage??

Say I want class C, do I put a 2 volt bias on the grid, and use that same 1 volt pk-pk signal?
 
The lowest grid voltage shown is -3 is that the cutoff or do you need to bias it a little higher to cutoff?

Would putting a resistor in the cathode return (like a triode would) provide the bias volts for this tube? Grid two wants 100 volts, why doesn't that change with the plate voltage values??

Ok, that's probably enough for now

 

[WA1GFZ]

3v/.01 = 300 ohms cathode resistor. Positive voltage on the cathode is the same as negative on the control grid. 

[WBear2GCR]

Two things.

The load line is usually a diagonal line going from the bottom axis up and to the LEFT.

Where you are on the loadline determines which class of operation you're in.

Class B is virtually cut off.
Class A is about half-way on.

You can't exceed the max current/wattage/voltage combination, so you can't run some tubes in Class A at "full max" plate voltage, they're not designed that way.

Then there is gain... how much you can put on the grid also depends on the gain.
If it's a follower the gain is ~1, that means in practice the input = output.

Do not forget the effect of the cathode resistor as you apply grid voltage!

Confusing, ain't it?? 

                   _-_-bear

[WA1GFZ]

The plate resistor needs to be sized so the resting plate voltage is 1/2 way between the screen voltage and full dc supply for cleanest operation. So say you want 175 volts at the resting plate. 250 -175 = 75  75 / .01 = 7500 ohms  75 X .01 = .75 watts minimum.  2 watts would provide a good safety factor. 5 watts and it would last forever.

[W3RSW]

Wait a sec...
You trying or wanting to use the 6AU6 as a class A power amplifier, voltage amplifier or very low signal voltage amplifier?  For voltage amplification using the large transconductance of this tube, you'll likely have pretty large values of plate load resistance, more like 100k to 479k not 7.5k   The 7.5k is used when running the tube near it's max. dissapation as in a power amplifier.  Big topic.  Also Grid no. 2 might have an 820k resistor from B+ in low signal application.

And then you have a load line at quiescent conditions and another load line at dynamic conditions.  hmm, back to the books?  A different book with examples might be of help.

[WA1GFZ]

you wouldn't be running 10 ma of plate current with a 470K plate resistor at 250 plate volts

[W3RSW]

xactly what i tried to say   

With 100k to 470k-  ur looking at far a smaller current, voltage amp. probably in first or second stage audio service.

[Ed/KB1HYS]

actually I just picked the example at random.  I really just wanted to get some info on tube circuit design using the charts.  Seems it is not as simple as I though.

Someone out there recommend a book that really just takes you through the process step by step??

[WA1GFZ]

Then you have to decide on the load you want to drive and design for power or voltage gain.

[KB2WIG]

  "  Someone out there recommend a book that really just takes you through the process step by step??  "

The RCA tube manuals do the step by step...  klc

[TedN]

ED,


   The RCA tube manuals do the step by step...  klc
second that

Also:

http://www.tubecad.com/articles_2003/Grounded_Cathode_Amplifier/index.html

A good starter read and NO mention of tessatura . Kinda hard to see but its more than one page. Step by step and explains how to read plate curves.

and

http://www.turneraudio.com.au/education+diy.html


 

[k4kyv]

http://www.tubecad.com/articles_2003/Grounded_Cathode_Amplifier/index.html

A good starter read and NO mention of tessatura . Kinda hard to see but its more than one page. Step by step and explains how to read plate curves.

A good web site.  I chose pdf mode and printed the article.  Looks like a real audiophile, not audiophool site, with real technical data, not snake oil.  Thanks for the link.

Now for my question.  Where can one find data on low level voltage amplifier stages, for transformer coupling and not R-C coupling?  The above articles, as well as the RCA tube manuals from the 30's to the present, give data on power amplifiers with xfmr coupling, but all the low-level stuff is for R-C.  I am mainly looking for how to calculate the optimum plate load impedance as presented to the tube plate by the transformer.  One big difference is that the transformer should have negligible voltage drop and run nearly the full DC voltage on the plate, whereas with R-C the plate resistor should drop the idling plate voltage to a halfway point between the minimum it can be pulled down and the maximum when the tube is near cut-off.  Most of the circuits I have seen show about 50k load per tube, using tubes medium or low mu tubes like the 6J5, 6C6, 6C4, 12AU7, etc.,  running at fairly high level.  The highest impedance primary, broadcast quality, output transformer I have is 20K mid-tapped to 500-ohm line.  I probably need something on the order of 100k mid-tapped.  One alternative would be to use a line to pushpull grids input transformer, such as the LS-12, wired in reverse, and use R-C coupling into the transformer to keep the DC off the primary (that was designed to be the secondary), since I suspect that winding would not tolerate much DC unbalance in the pushpull stage. But I would prefer something that would work directly into the 20K p-p load impedance, maybe a 12AU7, or pair of 6J5's not a pair of power tubes, or an old buzzard low-mu triode such as the type 37, since this would be the low-level output stage of an all-pushpull microphone pre-amp.

[W7SOE]

A good web site.  I chose pdf mode and printed the article.  Looks like a real audiophile, not audiophool site, with real technical data, not snake oil.  Thanks for the link.

Hey Don, how did you get a pdf of that article?

Thanks

rich

[k4kyv]

Go back to the home page of the site at http://www.tubecad.com/.  Scroll down the index to

Grounded-Cathode Amplifier
The grounded-cathode amplifier is approaching its centennial (2007) and it remains the building block of most tube audio equipment. It is simplicity itself, with little more than a few resistors and a triode; yet many are ignorant of its inner workings.   
    19 Jan 2003   PDF

The file's direct url is http://www.tubecad.com/articles_2003/Grounded_Cathode_Amplifier/Grounded_Cathode_Amplifier.pdf

[W2XR]

Don,

I think the bible for vacuum tube-based audio circuits is the RCA Radiotron Designers Handbook, 4th Edition (not the 3rd!)  by Langford-Smith. Every vacuum tube audio enthusiast must have this book!

Pages 517 thru 520, Section 4, "Transformer-Coupled Voltage Amplifiers" goes into this subject in considerable detail. I have designed a number of these types transformer-coupled gain stages using this as a reference, and with excellent results.

For XFMR-coupled voltage amplifiers, a good rule of thumb is that the primary impedance of the plate-to-line transformer should be at least 2x the plate resistance of the tube to be used, for lowest distortion operation, as well as reasonable voltage gain. For a 6J5, this would entail a transformer with a primary impedance of about 15K ohms; this is a commonly available XFMR, such as the UTC A-25, A-26, etc.

If you use a XFMR with a 100K ohm primary impedance, the loss of gain would be substantial, assuming the load impedance is on the order of 600 ohms or so. I have never seen an audio voltage amplifier circuit where the primary impedance of the plate-to-line XFMR was much above 20K ohm or so.

I have built a single-ended (with unbalanced DC in the primary winding) line-level preamp using 76s (very similar electrical characteristics to the 37), and the primary impedance of the plate-to-line XFMR I used was 15K ohm. This worked very well, with minimal apparent distortion.

73,

Bruce

[k4kyv]

For XFMR-coupled voltage amplifiers, a good rule of thumb is that the primary impedance of the plate-to-line transformer should be at least 2x the plate resistance of the tube to be used, for lowest distortion operation, as well as reasonable voltage gain. For a 6J5, this would entail a transformer with a primary impedance of about 15K ohms; this is a commonly available XFMR, such as the UTC A-25, A-26, etc.

That means, for a pushpull amplifier, the primary should be about 30K plate-to-plate, with midtap.  That is consistent with my observations.

Interestingly, for class A operation, a pushpull stage requires a primary impedance twice that of a single tube, since both tubes are sharing the load, 100% of the time and the full winding is used over the entire audio cycle.  OTOH, with class B, the total primary impedance is 4 times that of a single tube, since the tubes in each side take turns supplying the full output, and only half the winding is used at one time.  This makes the winding in effect a 2:1 step-up or step-down autotransformer, with the impedance from end to end 4 times that of one side.  That's why class-B driver transformers are rated as total primary to 1/2 secondary.

If you use a XFMR with a 100K ohm primary impedance, the loss of gain would be substantial, assuming the load impedance is on the order of 600 ohms or so. I have never seen an audio voltage amplifier circuit where the primary impedance of the plate-to-line XFMR was much above 20K ohm or so.

You are obviously talking about a single ended amplifier.  IIRC, the LS-51, pushpull plates to line, has a 30K primary.
 

I have built a single-ended (with unbalanced DC in the primary winding) line-level preamp using 76s (very similar electrical characteristics to the 37), and the primary impedance of the plate-to-line XFMR I used was 15K ohm. This worked very well, with minimal apparent distortion.

I'm not sure how well my 20K to line transformer would work.  That would be 10K load per tube.  Many power tubes have something on the order of 5K plate  resistance, but most smaller triodes used as voltage amplifiers have substantially higher.