• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

basic question on plate curves

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I thought i knew how to do this but i need help biasing my preamp triodes :rolleyes: I have searched the forum and the internet but so far i have found nothing that helps me on using plate curves.

I selected a poor operating point for my 2 parallel 12ax7's with an HT of 150V..
I chose a bias of -1.6V corresponding to 0.4/0.5mA plate current per tube.

This gave me a cathode resistor value of 4K per tube, and an anode resistor value of about 75-95K for both the tubes.
I have bypassed each tube cathode with 150uf.
However when i built it and tested it i got about 0.5V accross the cathode resistors, i dont know why?

(I used to have a good web page on this but i cant find it anywhere, if there is a blatent thread i missed on this subject please link me to it :) .)

Regards
Craig
 
Go to Steve Bench's Site. Scroll down to the first table, and look for Loadlines Pts. 1 - 5 in the extreme right hand column. That'll explain how to set Q-Points and predict gain, output voltage swing, and estimate THD.

This gave me a cathode resistor value of 4K per tube, and an anode resistor value of about 75-95K for both the tubes.
I have bypassed each tube cathode with 150uf.
However when i built it and tested it i got about 0.5V accross the cathode resistors, i dont know why?

Who knows? Might possibly be funkey tubes and/or a section of the 12AX7A is dying on you. Still, you will often see considerable variation from nominal design values when working with VTs, more so than with solid state. Then, again, VTs -- both as active devices, and in circuits -- are also considerably more "forgiving" than solid state. Back in the "good ol' days", remember, there were no such things as laser trimmed, ultraprecision metal film resistors with tolerances guaranteed to be plus/minus 0.1%. +/- 20% was quite common, +/- 10% premium, and +/- 5% considered to be "ultraprecision".
 
Craig405 said:
I thought i knew how to do this but i need help biasing my preamp triodes :rolleyes: I have searched the forum and the internet but so far i have found nothing that helps me on using plate curves.

I selected a poor operating point for my 2 parallel 12ax7's with an HT of 150V..
I chose a bias of -1.6V corresponding to 0.4/0.5mA plate current per tube.

This gave me a cathode resistor value of 4K per tube, and an anode resistor value of about 75-95K for both the tubes.
I have bypassed each tube cathode with 150uf.
However when i built it and tested it i got about 0.5V accross the cathode resistors, i dont know why?

(I used to have a good web page on this but i cant find it anywhere, if there is a blatent thread i missed on this subject please link me to it :) .)

Regards
Craig

Craig405,

I own the book "Designer's Tube Register" volume 1 common low-power triodes compiled by Tom Mitchell. This is an indespensible referance book for me I can look up almost any operating condition for many of the small nine pin family of triodes I have learned alot from it highly recomended!

Now a days I use LED's to bias as I find them far superior dynamicly to resistors there is a thread on the forum you could read it your interested.

SET12
 
You're calculating something incorrectly. Are you confusing B+ voltage (HT?) with plate voltage?

The B+ voltage needs to be split between the plate voltage of the tube and the drop across the plate load resistor. Typically around 50/50%, but usually not less that around 30% of B+ for either the tube or load resistor.

It looks to me like for two parallel 12AX7 sections and a total current of 0.4 - 0.5mA with a grid voltage of -1.5, you need a plate voltage of about 125V. (Plate voltage is measured plate-to-cathode...)

If HT means B+, you would only have 25V across the plate resistor, so it would have to be quite small (25/4.5ma) = 5.5 kohms.

As a sanity check you can look at the resistance-coupled amplifier charts: look at chart #9 in this document:

http://www.pmillett.com/tubedata/HB-3/Receiving_Tubes_Part_1/Resistance-Coupled_Amplifiers.PDF

For two parallel sections, cut both the plate and cathode R's on the table in half. Based on the chart (first row of 180V, 0.22M [220k] plate load [per section]) it looks like your plate load is OK but you need a smaller cathode resistor... which makes sense: The Rp of a 12AX7 section is around 100k, so two parallel sections you'd want at least around 100k plate load resistor.

Hopefully that makes some sense, sorry it sounds confusing...

Pete
 
Thanks for the info Set12, i might consider an LED once i get the bias point working correctly. Are they noisier than resistors though?

Pmillet, I am now using 2K cathode resistors on each half of the 12AX7 as opposed to the 4k i had before. This has given me a configuration very similar to one in your table for 12ax7 at 180v, rp of 220k driving 1 meg (values halved in my circuit as you mentioned as i am //'ing tubes so rp is still 95k ).

I only get 0.45V accross each cathode resistor (0.22mA ish) and 100V on the plate.
This puts the tube in a pretty poor place on the plate curves, i think the term is, 'cut off'.
Whats going on, im baffled :confused:

Regards
Craig

PS i have the ability to draw basic load lines now :), cheers for the link miles.
 
I just haven't really found them to be noisy as I use them for my tube power amp and especially for my phono stage which was the first place I ever saw a diode being used for bias; a very small Schottky barrier diode that had a forward voltage drop of .410 volts with 1 ma. .

I took them out and substituted a resistor and bypass cap. And gave that a listen, Well! As fast as I took it out as fast as I put it back! It wasn't even a contest! The dynamics were tremendously better! And if I hear some noise there is no way I'm going to give the dynamics up!:whazzat:

The trick to using LED's is using the right one! one which is in its linear current range which would be a small one check out the LED biasing thread on this forum and BTW LED's are superior to the Schottky.

I built a 50lb phono stage with an ultra fast 900 joule power supply 66db of gain using what I refer to as a super SRPP gain stages split RIAA and my noise floor is extreamly low for this gain which I was told would be impossible to achieve.

Although the 12AX7 is one of the most linear tubes around I like current through it if you want a truly low noise floor try a current source like the CS4 from Bottle Head this is a kit that is a very proven design talk about dynamics!

I am into a thing called Dynamic Conductance! this is when you take the (transconductance X the load) take the 12ax its trans is 2000 or .002 times a 100K load this then has a conductance of 200 take this times the CS4 load of up to 5 million ohms and you now have a figure of 10,000 of course their is other loading the other advantage is the noise floor which is way lower because its over the current source :D think of this as a Supercharger :nod:

Now think of how little current is needed across the CS to produce the voltage you need :D

Now if a fly lands on the grid the Earth is going to move!:bigeyes:

SET12
 
Set12 i have been convinced :). Your talk of improved dynamics is particularly attractive, thanks for the info.
I have biased my gain stages with LED's and all is well now....
Apart from the motorboating problem that has cropped up and the fact that i cant hear my guitar lol, at least the bias is sorted.

Brett those links are exactly what i have been trying to search the internet for, thankyou.

Thanks
Craig
 
compact7 said:
Hi all,
Newbie here.
Can I use LEDs to replace biasing resistor in SRPP circiut?
My Dac got 2 biasing resistors for each L & R ch. Can I replace the L & R cathode resistors with 2 LEDs or I must replace all the 4 resistors?

Thank you.

You can bias the upper and lower portions of an SRPP but the upper would require additional parts since the parts are probably on a PC board it would be easist to replace the Lower just make sure the current is in the LED of choice is in the linear range for best results meaning if your looking at currents of a milliamp it should be pretty small to maintain its forward voltage drop.

And like I've said I'll never go back to resistors:smash: unless it needed in a feedback loop.

Good luck!

SET12
 
Hi everyone,
I resurrecting this thread rather than starting a new one, I'm having the same initial probelm as the OP in that the measured bias is way off what is calculated for a 12AX7.
The OP cured his problem using LED's it says.... But I'd really like to know what can cause this issue. We are talking single digit mA's here so surely the resistor type isn't an issue?

Any comments appreciated!
 
Hi all,

I'm having the same issue the OP originally had, my bias is way off what is calculated even in a standalone test circuit.
THe OP solved with LED's but surely this shouldnt be necessary? With a 12AX7 current is single digit mA's so surely the resistor type wouldn't be an issue?
Any other things it could be??
I'm B+ 240V, Ra 100K, Rk 2.2K, Rg 1M, and a 68k grid stopper. I'm getting 0.8V bias but should be getting -1.9!
 
68k grid stopper?
That's a really huge value.
Why?

What is the DC voltages measured on pin 1 and pin 6 of the 12AX7?

Have you tried other 12AX7s in the circuit? Do you get the same or similar voltages from all of them?

0.8V across a 2.2k resistor would mean Ia of 0.36mA (360uA). That would mean only 36.4V dropped across the 100k resistor, which means the plate voltage (pin 1 or pin 6) would be about 204V. Is that what you're seeing?

Loadline for 240V B+ and 100k Rp attached.
 

Attachments

  • 12AX7A_240V_100k.jpg
    12AX7A_240V_100k.jpg
    249.9 KB · Views: 95
@jerryamp

There are a number of good load line plotters available on the Internet. Here is a spreadsheet from Merlin Blencowe's excellent Valve Wizard website that I use often:

http://valvewizard.co.uk/LoadLinePlotter.xls

Using that spreadsheet, attached is the load line plot using the component values you indicated. The operating point is at the intersection of the blue plate line and green cathode line.

This is a usable operating point but whether it is what you intended is another question. I ran a simulation in LTspice and it gives the same results (within reasonable tolerances). Plate current is about 0.7 mA, and 0.7 mA flowing through a 2.2k cathode resistor is going to give a bias voltage of about 1.5 volts.

It's not clear why you are getting a bias (cathode) voltage in the range of 0.8 volts; that figure doesn't make any sense given the component values you specified. Posting your schematic might help.
 

Attachments

  • 12AX7 Load Line.png
    12AX7 Load Line.png
    68.3 KB · Views: 107
Thanks for the reply!

Plate (anode) voltage is circa 150V. The calculator I used for an expectation of bias at -1.9 showed 148V I think.

Is 68k high for a grid stopper??! I thought it was pretty standard. That of course, could be an issue? it's sitting in the loop between the grid leak and the valve grid...

I'll go down to 33k grid stopper and see if that helps. Calculators don't ever let you put that value in... which I get but it must surely have some effect?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.