Cathode biasing question

[macula56]

I know very little about cathode biasing but I want to build a Tweed Super. The schematic shows a single 250 ohm cathode resistor for the power tubes and I want to know what values to use for separate cathode resistors. Would I double it or cut it in half? Thanks as always, JMac

[terminalgs]

I know very little about cathode biasing but I want to build a Tweed Super. The schematic shows a single 250 ohm cathode resistor for the power tubes and I want to know what values to use for separate cathode resistors. Would I double it or cut it in half? Thanks as always, JMac

you'd double the resistance, and you could 1/2 the power rating.


There isn't much reason to do this, and there are several advantages to a single shared cathode circuit (cheaper, cleaner and easier wiring,  only one bypass cap, convenient place to elevate center-tap of heaters (thought by some to lower noise),  and thought increases possibility of even-order harmonics).   

[sluckey]

Why do you want separate cathode resistors? Doing so will also require two bypass caps. The only reason I can think of to use separate cathode resistors would be for bias checking or tube matching purposes. If this is your reason, why not use one cathode resistor/cap but connect a 1Ω resistor from each cathode to the ungrounded side of the cathode resistor. Then just measure the voltage across each 1Ω resistor to determine current flow thru each tube for bias calculations and tube matching. Here's a pic to show you what I mean...

[llama]

convenient place to elevate center-tap of heaters (thought by some to lower noise)...

Sorry to highjack, one question about the heater centertap..

terminalgs,  Could one connect the 100R artificial center tap of heaters to the single cathode?

I suppose its connected between the cathodes and the resistor for a 14 to 25vDC jump? (depending on tube and circuit)
-Thanks, Scot

As you were.

[macula56]

thanks for the quick replies Gentlemen.

[sluckey]

Could one connect the 100R artificial center tap of heaters to the single cathode?

Yes.

[dude]

Why do you want separate cathode resistors? Doing so will also require two bypass caps. The only reason I can think of to use separate cathode resistors would be for bias checking or tube matching purposes. If this is your reason, why not use one cathode resistor/cap but connect a 1Ω resistor from each cathode to the ungrounded side of the cathode resistor. Then just measure the voltage across each 1Ω resistor to determine current flow thru each tube for bias calculations and tube matching. Here's a pic to show you what I mean...

Can you tell me, how to calculate the cathode bias using a 1 ohm R as in your drawing.  Wouldn't I want to know the mA's, like maybe a 6V6 would like to see 25mA's or less in a single-end operation? I assume if the mA's are too high, one would increase the bias resistor. But how are the mA calculated? Also that 1 ohm R should probably be at least a watt (the main bias R would be higher, like 3 t o5 watts)?

Thanks,
al 

[sluckey]

Wouldn't I want to know the mA's, like maybe a 6V6 would like to see 25mA's or less in a single-end operation?

yes

I assume if the mA's are too high, one would increase the bias resistor.

Yes, or you could lower the B+ to the plate or screen.

But how are the mA calculated?

Measure the millivolts across the 1Ω resistor. Multiply by 1. That's the mAs flowing thru the resistor and that's the same current flowing into the cathode of the tube.

Also that 1 ohm R should probably be at least a watt (the main bias R would be higher, like 3 t o5 watts)?

I use one watt because it's easy to find. But you don't need anything nearly that powerful. Let's take a bad ass example. Say you measure 100mV across that 1Ω resistor. That equates to 100mA flowing thru the resistor and into the tube. 100mA is 0.1A. Power = Current squared times resistance, so, .1A x .1A x 1Ω = .01W. Double that .01w for a safety factor and use a .02w 1Ω resistor. Good luck finding that resistor. Bet you lose it before you can solder it into your amp.   

Here's how I would calculate the idle power dissipation for the schematic shown above. I'll just do one tube but the procedure is the same for the other tube...

1. Measure plate voltage to ground.
2. Measure cathode voltage to ground.
3. Measure millivolts across the 1Ω resistor. Mentally convert to mA. This is cathode current.
4. Subtract cathode voltage from plate voltage. This is the real plate voltage.
5. Multiply real plate voltage times cathode current. This is the tubes idle plate dissipation in watts.

The above procedure assumes that all the cathode current also flows in the plate. This ignores the small screen current, but I'm OK with that. Your actual calculated plate dissipation will always be slightly less than the real plate dissipation. If you need gnat's ass accuracy you'll need to measure screen current and subtract that from cathode current to get the gnat's ass plate current.

[lego4040]

I like Steve's method and I also like this method which Uncle Doug explains very well in his video. It helps me personally cause its like having someone next to me showing what to do. Watch part one and two https://youtu.be/w9B0Rhr_Y-E

[dude]

Wouldn't I want to know the mA's, like maybe a 6V6 would like to see 25mA's or less in a single-end operation?

yes

I assume if the mA's are too high, one would increase the bias resistor.

Yes, or you could lower the B+ to the plate or screen.

But how are the mA calculated?

Measure the millivolts across the 1Ω resistor. Multiply by 1. That's the mAs flowing thru the resistor and that's the same current flowing into the cathode of the tube.

Also that 1 ohm R should probably be at least a watt (the main bias R would be higher, like 3 t o5 watts)?

I use one watt because it's easy to find. But you don't need anything nearly that powerful. Let's take a bad ass example. Say you measure 100mV across that 1Ω resistor. That equates to 100mA flowing thru the resistor and into the tube. 100mA is 0.1A. Power = Current squared times resistance, so, .1A x .1A x 1Ω = .01W. Double that .01w for a safety factor and use a .02w 1Ω resistor. Good luck finding that resistor. Bet you lose it before you can solder it into your amp.   

Here's how I would calculate the idle power dissipation for the schematic shown above. I'll just do one tube but the procedure is the same for the other tube...

1. Measure plate voltage to ground.
2. Measure cathode voltage to ground.
3. Measure millivolts across the 1Ω resistor. Mentally convert to mA. This is cathode current.
4. Subtract cathode voltage from plate voltage. This is the real plate voltage.
5. Multiply real plate voltage times cathode current. This is the tubes idle plate dissipation in watts.

The above procedure assumes that all the cathode current also flows in the plate. This ignores the small screen current, but I'm OK with that. Your actual calculated plate dissipation will always be slightly less than the real plate dissipation. If you need gnat's ass accuracy you'll need to measure screen current and subtract that from cathode current to get the gnat's ass plate current.

Thanks ,seems I've done this before and forgot.

Anyway, I got 14.2 watts dissipation for one 6V6. BTW, I did subtract the screens.

Isn't this a tad high for one 6V6 at idle?

I guess I could lower the plate voltage or up the cathode R as you say. Yeah, I still trying to get this Vibro Champ to sing, I'm close. I've made changes that "should have got me there" so I'm thinking the dissipation is too high.  The V Champ sings with the bass low at 3 and not pass 7 on the volume.  Bass up to 5 or more (or volume up to 9 or10) then I get what I think is blocking distortion. Pretty bad distorted bass riding all over the place.

This Vibro Champ was made in 1967 (110v wall current) with the stock value bias R to run at B+ of 365vdc, I got 412vdc....? So that stock 470 ohm bias R is way too low with my voltage, I'm driving the balls off that RCA 6V6.  I don't want to use Zeners to lower B+ so I need to get the dissipation right at 412 volts with the bias R.

I got 14.2 watts diss. at idle with a 750 ohm R, so I'll try a 1K. That ought to do it.

I just want to make sure I have that 6V6 biased right before I waste time changing the gain. I was going to raise V1a Cathode R to 2K2 from 1K5, but before that I'd like to get the dissipation to around 12 watts then hopefully I'll lose that ratty distortion where I mentioned.

I know, just play your amp but this is a good way to learn, then I won't have to ask a ton of question here, ha. 

Thank Sir Sluckey,
al
 

[sluckey]

Anyway, I got 14.2 watts dissipation for one 6V6. BTW, I did subtract the screens.

Isn't this a tad high for one 6V6 at idle?

It's common for a single ended amp to idle at 100%. How did you determine screen current?

[dude]

I have a 750 ohm 3 W screen grid resistor installed (not in schematic but thought it good to add).

Measure the voltage on one side of the screen R (pin 4) and then the other, subtract and I got like 2v, so 2v/750= .00266
Then substrate the screen current from the plate current. Sound right?

OK, I'm at 14.2watts diss. so if I get it to 12 watts or a hair under, what are my chances that the 2 watt dissipation is causing my (distortion) I'm talking about?

The amp plays fantastic at bass 3, vol.  6 or 7 any higher on either and ratty bass distortion starts. I did both gain stages at 2.2uf/1K5, changed the tone values to .047uf and .022, from .1 and .047. Changed the 250pp to 220pp, tone stack (amp a little too trebley)). Took the 330pp off grid to cathode of 6V6.

And took that small 5 watt OT out and put in a larger vintage paper 10k (was sounding awesome in a tweed Princeton even dimed, basically the same as this Vibro champ). Just forget I changed the OT.

al    Edit: just measure the dissipation with a 1K, 11.2 watts, on the money but still that ratty distortion, as above. Also the Amp is not as
               loud as with the 750 ohm .
               Also why do the screens have more voltage 401v than the plate 394v both measured at the tube...? I thought the plate was always
               then the screen...?
 

[dude]

Somethings strange here with voltages. Might have something to do with the bass distortion...?

I lowered the B+ with two 5W 15volt Zeners in series on PT CT.

It lowered the B+ to 364v pretty much what the Vibro Champ schematic says.

With that B+ here are the voltages at the 6V6:

Screen      354v
Plate         335v
mA             44mA
Cathode     21v
 
Why is the plate voltage so low....?  Should be a few volts less than the B+, it's 20 volts...?
The resistance between the OT's  plate and CT (amp off of course) is 640 ohms (single ended OT)

With such small resistance why the big voltage drop on the tubes plate...?

Something wrong with this (Vintage OT I swapped in)....? Slucky I know you said take it out, could it be the problem all along causing the bass distortion when vol or bass is up high...?

I just can't understand why the 20 volt drop from tube plate and screen with only 640ohms resistance....?

al

[dude]

I guess the load of the 6V6 is causing the voltage drop just on the tubes plate, pull the 6V6 and all the voltage are inline but high (as no load). No tube and the screen is a 3 or 4 volts less than the B+.

I was checking the plate voltage off the 6V6 to bias, I would think because of the load on the tube's plate that's not correct. I should be using the B+ as my plate voltage and not the 6V6 plate to calculate the bias correctly.

OK, thanks for all your help.

The bass distortion might be the replacement OT, I'll lower the gain on the first stage with little a higher cathode resistor, if that doesn't work I can always try a lower gain preamp tube like a 5751.

Thanks Sluckey for the bias calculation,

al

[sluckey]

I should be using the B+ as my plate voltage and not the 6V6 plate to calculate the bias correctly.

No. You had it right before.

[HotBluePlates]

OK, I'm at 14.2watts diss.

Congrats, you're at 100% plate dissipation for a 6V6.

Really for a 6V6GTA, but unless you have a tube from the 50's or earlier, it probably carries a 14w plate dissipation rating. And even if you use a 6V6 old enough to really have had a 12w plate dissipation rating, it will stand this fine for many-years. Unless the normally-gray plate surface is glowing cherry red, you're fine. The tube will actually dissipate less heat in this circuit when you drive it as hard as possible.

... what are my chances that the 2 watt dissipation is causing my (distortion) I'm talking about?

The amp plays fantastic at bass 3, vol.  6 or 7 any higher on either and ratty bass distortion starts.

Probably zero chance.

You've already said the amp sounds fine with bass and/or volume down. That pretty much rules out a bias issue (usually, very-wrong bias would cause distortion at all signal levels).

You've identified blocking distortion as a possible culprit. Blocking is 100% about applying a drive signal whose peak voltage exceeds the absolute value of the bias voltage, causing grid current in the output tube, which charges up a too-big coupling cap, shifting bias (but only while the too-big drive signal exists) and eventually recovering when the drive signal is removed. You can keep trimming coupling caps downward, unless they make the amp too bright and/or keep the bass control low and/or don't turn the amp up so high.

Reducing coupling cap value is all about reduce the RC time constant of the cap and resistance to ground, shortening the recovery period after grid-shift due to too big drive signal. You could also reduce the value of the resistance to ground, but that will also reduce the effective voltage gain of the prior stage (somewhat attacking the problem 2 ways).

f course, maybe you don't like the amp's distorted sound as much as you'd expected (that's happened to me a bunch). Or maybe you don't like the sound of that speaker (happened to me, too). You could try plugging a different speaker into the amp and seeing what it sounds like (my stock Epi Valve Jr sounds amazingly better just by unhooking the internal speaker and plugging the amp into a 2x12 with a pair of Scumback speakers).

Screen      354v
Plate         335v
mA             44mA
...
Why is the plate voltage so low....?  Should be a few volts less than the B+, it's 20 volts...?
The resistance between the OT's  plate and CT (amp off of course) is 640 ohms (single ended OT)

With such small resistance why the big voltage drop on the tubes plate...?

Voltage (Drop) = Resistance x Current. You have 750Ω resistance to the screen, and only 640Ω through the OT to the plate. But your calculated screen current earlier was only ~2mA while your plate current is 44mA. 750Ω * 2mA = 1.5v, while 640Ω * 44mA = 28v.

And this is typical of most guitar amps because the idle screen current is often very much lower than idle plate current.

[shooter]

fwiw,
I'd copy paste this thread for future cath/bias builds

I just finished 2 amps, both cathode biased, 1 se, 1 pp, both have great sound, to a point, then they get ugly distorted for me, but I believe there are players out there that would say, "dude that screams just like I've been looking for."  So, keep some pretty good notes that will get you back to  that "rev 3 sound" when you find yourself not liking rev 11

[PRR]

> Really for a 6V6GTA.... a 6V6 old enough to really have had a 12w

They are the same rating. Same tube. (The "A" means heater warm-up is coordinated for series-string use, but I think that's just an agreed-on spec on what was the commonest design.)

The basic assumption changed when TV came in. (Because TV sets can go-wrong more violently than record-players, and more careful design is needed.)

12 Watts if you are just going to throw parts together and ship without testing.

14 Watts if you carefully worst-case check your design (or test every unit produced, as we do here).

A safety/slop-factor was eliminated, for the benefit of designers with extra-sharp pencils.

FWIW, it has been reported that Fender runs a modern 6V6 at 17 Watts in their new "600".

[dude]

Thanks for the info, the amp sound pretty good to me.  Actually better with the Weber sig8 then a re-coned 10" from a Silver face Super Reverb.

I tried several speakers GH1230, Red Fang, Weber blue there all great each having there own character. Tone is in the eyes of the holder. The amp really sounds incredible, like you don't turn the bass all the way up same with the treble. With the lower cathode caps and one drop in values the tone circuit, the amp is there.

Lowering the first stage cathode resistor to 2K2 from 1K5 really took that over-driven bass gain out, also a 5751 works well, I can play P90s, strats singles, even humbuckers and dial in a great tone.

So if I never turn the bass pass 5 or 6, what's the big deal if the amp sounds good. 

I also put in a 5V4 and that tight'n things up nice, the Zeners keep the high voltage down to around 380 (5V4) from 412, with a 5Y3 368. So, I've extremely happy but most of all I learned a ton working a simple circuit so when I get into more complicated circuits I'll have a good reference point to fall back on.

This amp is so light and a great big sound, all old stock tubes. I can carry it to a gig so easy for an old guy, ha, just mike it up. I can even get away without an overdrive pedal by working the guitar volume and p/u selections. So many sounds, just turning the treble a number or two, bass knob too I can get totally different tones.

I'm stoked and I only paid $250 for this 67 vibro Champ.

Thanks again all,
al

[Willabe]

So if I never turn the bass pass 5 or 6, what's the big deal if the amp sounds good.

Through the years I owned and played through a BF Dual Showman, BF Super Reverb and (still own) a BF Princeton Reverb.

I've never played any of the 3 with the bass past 3 to 3 1/2. Even with a strat. (Now I never play the lead PUP alone either.) And I always had a nice full, round sound, not thin.


All the guys I played with in blues clubs that played Fenders never set the bass past 4 either, that I can remember.


                 
                      Brad        

[dude]

So if I never turn the bass pass 5 or 6, what's the big deal if the amp sounds good.

Through the years I owned and played through a BF Dual Showman, BF Super Reverb and (still own) a BF Princeton Reverb.

I've never played any of the 3 with the bass past 3 to 3 1/2. Even with a strat. (Now I never play the lead PUP alone either.) And I always had a nice full, round sound, not thin.


All the guys I played with in blues clubs that played Fenders never set the bass past 4 either, that I can remember.


                 
                      Brad     

+1

[lego4040]

Like shooter said, I'm gonna put this link in my favs for reading. Glad you got amp tweaked to your taste.

[HotBluePlates]

... So if I never turn the bass pass 5 or 6, what's the big deal if the amp sounds good.  ...

 

I'm excited by your saying this. I had visions of myself when I started this hobby, tinkering needlessly with a perfectly-good '67 Princeton Reverb.

Best bet is fiddle the knob while not looking at the numbers, until what comes out of the speaker sounds best to you. Besides, guys (including me) tend to turn the bass up too much when playing solo probably to make up for the missing bottom end from a bass player. The same settings sound unusably muddy in a band setting; and you might play brighter in a band/recording setting than you'd think would be good hearing the guitar/amp alone.

[Fresh_Start]

HBP - thanks for great post on blocking distortion and coupling caps.  Hadn't made that connection.

Also that The V Champ sings with the bass low at 3 and not pass 7 on the volume.  Bass up to 5 or more (or volume up to 9 or10) then I get what I think is blocking distortion. Pretty bad distorted bass riding all over the place.

I just want to make sure I have that 6V6 biased right before I waste time changing the gain. I was going to raise V1a Cathode R to 2K2 from 1K5, but before that I'd like to get the dissipation to around 12 watts then hopefully I'll lose that ratty distortion where I mentioned.

I know, just play your amp but this is a good way to learn, then I won't have to ask a ton of question here, ha. 

Thank Sir Sluckey,
al

 Lowering the first stage cathode resistor to 2K2 from 1K5 really took that over-driven bass gain out, also a 5751 works well, I can play P90s, strats singles, even humbuckers and dial in a great tone.

While this thread focused on biasing the power tube, there's something else worth mentioning. When you raised the cathode resistor of the first gain stage to 2K2 without making a proportional decrease in the cathode bypass cap, you lowered the frequency cutoff point of that stage and more subsonic frequencies were being amplified throughout the amp.


IOW 1K5 in parallel with 25uf is the same as 2K2 in parallel with 17uf (not a common cap value, I know ;).

I was surprised to learn that the "standard" 1K5 || 25uf amplifies way below the frequency spectrum of our guitars.  Low E is 82 Hz, but 25uf provides full gain way below that level. If you look at the chart on page 26 of the following article by Merlin Bledsoe, you'll see that 10uf ought to do just fine. IME 6.8uf comes darn close and if you need a bit more bass, turn it up! 


http://www.valvewizard.co.uk/Common_Gain_Stage.pdf



My point is that lowering that preamp cathode resistor back to 1K5 increased your gain for guitar frequencies but simultaneously lowered the gain for inaudible low frequencies that can cause blocking distortion,nasty rumble, etc. 


Also, I've read that it's a good idea to cut unwanted bass frequencies right at the start of our preamps.


Hope this epilogue is helpful. I was puzzled do a minute about the 1k5 vs 2k2 comments and had to figure it out.


Chip

[HotBluePlates]

> Really for a 6V6GTA.... a 6V6 old enough to really have had a 12w

They are the same rating. Same tube. ...

Yes, of course. I wanted to say he was okay and not really get into the "Design Center" system vs. "Design Maximum" system of ratings. 

HBP - thanks for great post on blocking distortion and coupling caps.  Hadn't made that connection.

You're welcome, but the info is incomplete; I recommend finding a good reference on blocking distortion and reading it.

A bottom-line takeaway is that in the overwhelming majority of guitar (and audio) amps, when the output tube momentarily has 0v grid-to-cathode (the result of '- bias voltage' plus '+ signal peak'), the tube will almost certainly be distorting. If the tube hasn't run into some other limit or curve-crowding first, the distortion will be because the grid then tries to draw current when it becomes positive of the cathode. Either that current will cause the coupling cap to store a charge which momentarily shifts the tube bias, or it will overwhelm the ability of the previous stage to deliver power (signal voltage times grid current draw) and cause that prior stage's output to collapse and therefore distort.

If blocking distortion sounds bad, but other distortion mechanisms in the output tube sound good, then you have a limited set of tools to reduce the blocking effect. The main one is to do the opposite of what you want with a filter cap in the power supply: get the cap to store a charge for less time. So you make C smaller, or R smaller or both. However, the resistor from output tube grid to ground (or bias supply) is part of the a.c. load of the previous stage, and is in parallel with that stage's plate load resistor. So making R smaller also reduces the a.c. load of the prior stage, and will reduce its gain some amount. This is a central tradeoff in all voltage amplifier stages.

[dude]

Fresh_Start,

I did lower the 1st stage cathode cap, to 2.2uf, from 25uf, then to help with the Neil Young blocking bass distortion I when to 2K2, from 1K5. It helped but probably a better way to get rid of that ratty bass distortion is to put a grid stopper R to the grid of the 6V6, there's a 220K grid leak there but no grid stopper. Not sure of the value but maybe 1k5 to 4K2.

This grid stopper might be a better way (get rid of the ratty bass distortion on high vol & bass settings, after you worked the tone stack) then to lower the gain at the first stage by raising the R.

At least that's what I get from Aiken.

Several ways to skin a cat, don't fix it if it ain't broke and don't get crazy trying to get blood for a turnip....

al