ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Cathode Bias Idling Current

12/25/1997 10:34 AM
Mike T	Cathode Bias Idling Current cathode volts/cathode resister ohms = idling current If there are multiple tubes with theier cathodes tied together using one single resister, is the value of this resister divided evenly among the power tubes before dividing by cathode current? Mike T I may have posted this note twice...if so, sorry

12/25/1997 4:59 PM
Stephen Delft	Hi Mike, If I understand your question right.... try looking at it from this direction- With no signal, a cath-biassed p-p output stage is going to settle at a certain voltage between grids and cathodes. How much voltage depends on the tubes, value of Rk, plate and screen voltages, etc.- All the usual things. In practice, if all tubes share the same cathode resistor, and there is no extra bias voltage applied to the grids , you can consider this Vg-k to be near-enough the voltage measured across Rk. Ohms law applies; this voltage will be (value of Rk:Ohms) x (total of all cathode currents:Amps) If you want to be precise, cathode current is .plate current plus screen current, so you may be out by typically 5-10mA per tube. How equally this total current divides between the two (or four) output tubes will depend on how closely the tubes' individual characteristics are matched. (I just mention this because it does sometimes happen in a C/B o/p stage that one or more tubes deteriorate and draw less current, Which causes the other tubes to pull _more_ current - sometimes enough to develop red hot spots on the plates. And although, by that point, the overheated tubes may have become permanently damaged, the original cause of the problem was the _other_ tubes which were running too cold. It can be a mistake to replace only the obviously overheated ones) Anyway, assuming all the o/p tubes are closely matched, then for any chosen cathode current PER TUBE, in otherwise similar amps, you would expect to find a shared Rk of (say) 250 Ohms for one pair of tubes, or 125 Ohms for two pairs of tubes, (83 Ohms for 3 pairs, etc). A certain plate current requires a certain cathode voltage above ground (=the voltage across Rk) If twice as many tubes, then twice the total cathode current, so _HALF_ the resistor value for the same voltage drop cross it. I don't know if this helps...or answers your question. I've tried to explain the relationships going on here, which is how I was taught electronics. If you understand how it works, you don't have to remember so many details Of course, if I have dropped a bad typo in here, then I'll have egg on my face. That's fair enough. If you disagree with any of the above, throw it back at me Cheers, Stephen

12/25/1997 4:59 PM

Stephen Delft

Hi Mike,

If I understand your question right....
try looking at it from this direction-
With no signal, a cath-biassed p-p output stage is going to settle at a certain voltage between grids and cathodes. How much voltage depends on the tubes, value of Rk, plate and screen voltages, etc.- All the usual things.
In practice, if all tubes share the same cathode resistor, and there is no extra bias voltage applied to the grids , you can consider this Vg-k to be near-enough the voltage measured across Rk. Ohms law applies; this voltage will be
(value of Rk:Ohms) x (total of all cathode currents:Amps) If you want to be precise, cathode current is .plate current plus screen current, so you may be out by typically 5-10mA per tube.

How equally this total current divides between the two (or four) output tubes will depend on how closely the tubes' individual characteristics are matched. (I just mention this because it does sometimes happen in a C/B o/p stage that one or more tubes deteriorate and draw less current, Which causes the other tubes to pull _more_ current - sometimes enough to develop red hot spots on the plates. And although, by that point, the overheated tubes may have become permanently damaged, the original cause of the problem was
the _other_ tubes which were running too cold. It can be a mistake to replace only the obviously overheated ones)
Anyway, assuming all the o/p tubes are closely matched, then for any chosen cathode current PER TUBE, in otherwise similar amps, you would expect to find a shared Rk of (say) 250 Ohms for one pair of tubes, or 125 Ohms for two pairs of tubes, (83 Ohms for 3 pairs, etc). A certain plate current requires a certain cathode voltage above ground
(=the voltage across Rk) If twice as many tubes, then twice the total cathode current, so _HALF_ the resistor value for the same voltage drop cross it.

I don't know if this helps...or answers your question. I've tried to explain the relationships
going on here, which is how I was taught electronics. If you understand how it works, you don't have to remember so many details :-)

Of course, if I have dropped a bad typo in here,
then I'll have egg on my face. That's fair enough. If you disagree with any of the above, throw it back at me :-)

Cheers, Stephen

12/26/1997 9:35 AM
Mike T	Hi Stephen, Good to hear from you. How are things in NZ? Things are OK here, they've stabalized a bit, although my New Years Eve gig just got cancelled for lack of ticket sales. Bummer. Anyway, you did understand my question, but let me try and give you specifics as to what I need to know, and see if I understand you correctly. I'm building an amp that I want to bias into Class A. I have obtained a 300-0-300 250 mA PT and a 4300 ohm primary 60watt OT, and I want to use it with either (2) EL34s or (2) 5881s. I burned out the part of my DMM that reads current, so i have to find another means short of buying another DMM. EL34s have a max current of about 150mA, so I would need to bias them at about 75mA per tube. If I were to use a 250ohm rK, and read the cathode voltage at 35 volts, that would be interpreted to be (25/250)/2 = 70mA per tube, right? Thanks. Mike Tremante

12/26/1997 9:35 AM

Mike T

Hi Stephen,

Good to hear from you. How are things in NZ? Things are OK here, they've stabalized a bit, although my New Years Eve gig just got cancelled for lack of ticket sales. Bummer.

Anyway, you did understand my question, but let me try and give you specifics as to what I need to know, and see if I understand you correctly. I'm building an amp that I want to bias into Class A. I have obtained a 300-0-300 250 mA PT and a 4300 ohm primary 60watt OT, and I want to use it with either (2) EL34s or (2) 5881s. I burned out the part of my DMM that reads current, so i have to find another means short of buying another DMM. EL34s have a max current of about 150mA, so I would need to bias them at about 75mA per tube. If I were to use a 250ohm rK, and read the cathode voltage at 35 volts, that would be interpreted to be (25/250)/2 = 70mA per tube, right? Thanks.

Mike Tremante

12/26/1997 7:59 PM
Stephen	Yes...you have a small typo there, but I assume you mean (35/250)/2 = 70 mA I usually do the mental arithmetic as (35 x 1000) ----------- =140mA TOTAL 250 but it comes out the same either way More resistance = less current Less resistance = more current . But it's not quite a linear relationship, so if you tweak the resistor value, measure volts again and recalculate total current for the new resistance value to check it is ok. BTW, I hardly ever use the current range on my DVM. I put 1 ohm resistors in all output tube cathode leads, Then 1mV across 1 Ohm = 1mA through it...35mV indicates 35mA. Doesn't upset bias or tube curves significantly, and you can at any time check the individual tube currents without disturbing the circuit conditions. Things in NZ are ..."middling-fair" I'm doing more housekeeping and cooking...but hey, it's mid-summer and there are fresh raspberries in the garden for Christmas dinner. Could be worse Cheers, Stephen

12/26/1997 7:59 PM

Stephen

Yes...you have a small typo there, but I assume you mean (35/250)/2 = 70 mA
I usually do the mental arithmetic as
(35 x 1000)
----------- =140mA TOTAL
250
but it comes out the same either way :-)

More resistance = less current
Less resistance = more current
.
But it's not quite a linear relationship, so
if you tweak the resistor value, measure volts again and recalculate total current for the new resistance value to check it is ok.

BTW, I hardly ever use the current range on my DVM. I put 1 ohm resistors in all output tube cathode leads, Then 1mV across 1 Ohm = 1mA through it...35mV indicates 35mA. Doesn't upset bias or tube curves significantly, and you can at any time check the individual tube currents without disturbing the circuit conditions.

Things in NZ are ..."middling-fair" I'm doing more housekeeping and cooking...but hey, it's mid-summer and there are fresh raspberries in the garden for Christmas dinner. Could be worse :-)

Cheers, Stephen

12/27/1997 9:21 AM
Mike T	I just fired it up and something is wrong. I get a fully distorted sound at all settings with barely audible volume. I'm getting about 350 V at the plates which is OK, and 25V at the cathodes with a 250rK, so that comes out to 50mA per tube, too low. I'll try a lower value rK as soon as I find the other problem. Is there a way to tweak the cathode voltage? In other words, will the voltage at the cathode remain the same if I change the resister, and if so, is there a component that regulates the voltage at the cathode? I'm using a Hammond OT with taps for 4, 8 and 16 ohms, and have hard wired it according to their spec sheet for 8 ohms. Could be a problem there. But I'm going to have to set this project down and come back to it, because my old shop is not easily accessable to me at this time, and I need the space, as this disassembled amp has been occupying my primary workspace for a few days now. After I think about it for awhile though, I'll probably come up with the solution for this distortion problem, as it is probably real dumb, as they usually are (for me, anyway). I have a couple other amps I can gig with (if I can get some gigs). I've read about that method for reading mA as voltage with a 1 ohm resister at the cathode, but never used it bacause all my amps, except for the Pro Jr are cathode bias. Wouldn't installing a 1 ohm resister from cathode to ground upset the bias if wired in parallel to the bias rK (path of least resistance)? Or are they wired is series with the 1 ohm first from the cathode and then temporarily jumped to ground from where they meet and then read accross the 1 ohm resister? Summertime. Man. It's sloppy, wet, and cold up here. A few months of winter to go, too, and January and February are usually the worst. But I'm used to it. Mike

12/27/1997 9:21 AM

Mike T

I just fired it up and something is wrong. I get
a fully distorted sound at all settings with barely audible volume. I'm getting about 350 V at the plates which is OK, and 25V at the cathodes with a 250rK, so that comes out to 50mA per tube, too low.
I'll try a lower value rK as soon as I find the other problem. Is there a way to tweak the cathode voltage? In other words, will the voltage at the cathode remain the same if I change the resister, and if so, is there a component that regulates the voltage at the cathode? I'm using a Hammond OT with taps for 4, 8 and 16 ohms, and have hard wired it according to their spec sheet for 8 ohms. Could be a problem there. But I'm going to have to set this project down and come back to it, because my old shop is not easily accessable to me at this time, and I need the space, as this disassembled amp has been occupying my primary workspace for a few days now. After I think about it for awhile though, I'll probably come up with the solution for this distortion problem, as it is probably real dumb, as they usually are (for me, anyway). I have a couple other amps I can gig with (if I can get some gigs).

I've read about that method for reading mA as voltage with a 1 ohm resister at the cathode, but never used it bacause all my amps, except for the Pro Jr are cathode bias. Wouldn't installing a 1 ohm resister from cathode to ground upset the bias if wired in parallel to the bias rK (path of least resistance)? Or are they wired is series with the 1 ohm first from the cathode and then temporarily jumped to ground from where they meet and then read accross the 1 ohm resister?

Summertime. Man. It's sloppy, wet, and cold up here. A few months of winter to go, too, and January and February are usually the worst. But I'm used to it.

Mike

12/27/1997 4:28 PM
Stephen	Mike, 50mA per tube should be working _close_ to optimum - so that's unlikely to be the direct cause of your problem. Check screen volts. Check that pins 1 and 8 are linked on EL34 sockets. 1 Ohm current monitoring resistors are NOT fitted or jumped from cathode(s) to ground, and NOT in parallel with 250 Ohm Rk. Either of these is a recipe for large tube currents and rapidly approaching meltdown For "permanent" installation, diss each o/p tube cathode lead and insert a 1 Ohm resistor in each lead. So one end of each 1 Ohm goes to each pin8 (6L6) or pins 1+8 (EL34) Other ends of 1 Ohm R's join together and go to the top of 250 ohm shared Rk. Bottom of 250 Ohms goes to ground as usual. Imagine a Y shape....Any bypass cap remains across the 250 ohm. Additional biassing effect of the 1 Ohm R's is not usually significant - it's less than the normal tolerance of your 250 Ohm Rk. Measure separate tube currents _across_ 1 Ohm R's, NOT to ground. You can make temporary current sniffers (bias probes?) from octal plugs and sockets glued together, with all pins and sockets wired like to like, except for pin 8 and socket 8 which are linked by a 1 Ohm resistor. Extend thick wire stubs or tags out from pin 8 and socket 8 so you can clip on your DVM leads. Make preferably two of these adaptors. You can plug them between the o/p tubes and the sockets, and monitor individual currents without doing any mods to the amp. You can buy these things, but they are easy enough to make. The 1 Ohm R can go inside, but if it fails, replacement will be a pain. I use tiny 1/4 watt metal film 1 Ohm R's, and mounted them on the OUTSIDE of the adaptors, directly across the meter clip-on points. Makes a useful emergency fuse, and easily replaced if it blows or drifts off-spec. I am sure this must be in Keen's tube FAQ somewhere. If this is not included, I could do a drawing and a tidier explanation, if people think it would be a useful addition. For same tubes, in this kind of p-p setup, the no-signal volts across the shared Rk is _roughly_ constant. It's a balancing act - a kind of negative fedback which stabilises the total cathode current (as long as your tubes are similar and in good condition) As V g-k is approx constant, the value of Rk controls total cathode current. . Fender did apply a separate grid bias _and_ cathode resistor(s) on some of their later amps, but these seem to be generally disliked by amp fanciers - whether because of the mixed bias, or for other reasons, I don't know . Cheers, Stephen. .

12/27/1997 4:28 PM

Stephen

Mike,
50mA per tube should be working _close_ to optimum - so that's unlikely to be the direct cause of your problem. Check screen volts. Check that pins 1 and 8 are linked on EL34 sockets.

1 Ohm current monitoring resistors are NOT fitted or jumped from cathode(s) to ground, and NOT in parallel with 250 Ohm Rk. Either of these is a recipe for large tube currents and rapidly approaching meltdown :-)

For "permanent" installation, diss each o/p tube cathode lead and insert a 1 Ohm resistor in each lead. So one end of each 1 Ohm goes to each pin8 (6L6) or pins 1+8 (EL34) Other ends of 1 Ohm R's join together and go to the top of 250 ohm shared Rk. Bottom of 250 Ohms goes to ground as usual. *Imagine a Y shape*....Any bypass cap remains across the 250 ohm.

Additional biassing effect of the 1 Ohm R's is not usually significant - it's less than the normal tolerance of your 250 Ohm Rk. Measure separate tube currents _across_ 1 Ohm R's, NOT to ground.

You can make temporary current sniffers (bias probes?) from octal plugs and sockets glued together, with all pins and sockets wired like to like, except for pin 8 and socket 8 which are linked by a 1 Ohm resistor.
Extend thick wire stubs or tags out from pin 8 and socket 8 so you can clip on your DVM leads. Make preferably two of these adaptors. You can plug them between the o/p tubes and the sockets, and monitor individual currents without doing any mods to the amp. You can buy these things, but they are easy enough to make.
The 1 Ohm R can go inside, but if it fails, replacement will be a pain. I use tiny 1/4 watt metal film 1 Ohm R's, and mounted them on the OUTSIDE of the adaptors, directly across the meter clip-on points. Makes a useful emergency fuse, and easily replaced if it blows or drifts off-spec.
I am sure this must be in Keen's tube FAQ somewhere. If this is not included, I could do a drawing and a tidier explanation, if people think it would be a useful addition.

For same tubes, in this kind of p-p setup, the no-signal volts across the shared Rk is _roughly_ constant. It's a balancing act - a kind of negative fedback which stabilises the total cathode current (as long as your tubes are similar and in good condition)
As V g-k is approx constant, the value of Rk controls total cathode current. . Fender did apply a separate grid bias _and_ cathode resistor(s) on some of their later amps, but these seem to be generally disliked by amp fanciers - whether because of the mixed bias, or for other reasons, I don't know .
Cheers, Stephen.
.

12/28/1997 12:30 AM
Mike T	Thanks, Stephen. Cheers to you too, and happy new year. Mike

Page 1 of 3

Next>

Last Page>>