HELLO TUBES WORL,
Cathode Bias with a Counstant Current Source
so, i been doing some reading on cathode bias with CCS and found the artical on tubecad web site.
As i understand from reading the artical. Why is this not more common in tube amp designs? as it seem to be a win win if you want class A with more wattage compare to SET type tube amp?
Cathode Bias with a Counstant Current Source
so, i been doing some reading on cathode bias with CCS and found the artical on tubecad web site.
As i understand from reading the artical. Why is this not more common in tube amp designs? as it seem to be a win win if you want class A with more wattage compare to SET type tube amp?
Many things that look ok on paper and even on the bench then really don't sound good enough. I really don't like CCS and similar "restrictions" at the power stage. A good power supply with fixed bias (DC coupled cathode follower or source follower or fixed bias with interstage transformer) is the best solution for me. This is especially true for low power (10W or less) amplifiers where I tend to avoid feedback as well.
The main reason is that I want to use all their power and I don't want hard clipping.
It works remarkably well, though requires a big cap for bypassing. I have done it many times and generally bypass the cathodes to each other using back to back capacitors to create a differential output stage.
My experience is that in this differential configuration going the extra mile to fixed bias only offers marginal improvement.
A theoretical issue which has been raised is that this only offers DC balance in the output stage and that AC balance maybe worse than with resistive self bias - but I have not found this to be an issue personally.
Shoog
My experience is that in this differential configuration going the extra mile to fixed bias only offers marginal improvement.
A theoretical issue which has been raised is that this only offers DC balance in the output stage and that AC balance maybe worse than with resistive self bias - but I have not found this to be an issue personally.
Shoog
CCS installed.
The amps are PP kt120 with IT phase splitter.
CCS are set at 86ma per tube, didn't have time today to give it a good listen.
Will report back as soon as I get more playing time.
The amps are PP kt120 with IT phase splitter.
CCS are set at 86ma per tube, didn't have time today to give it a good listen.
Will report back as soon as I get more playing time.
Had a check at cathode voltage when running at continuous hi power, you may be surprised !
The worst situation is while you measure max power 😱
I usually add a Zener of appropriate power in parallel with each CCS at about 1.5 times the quiescent cathode voltage.
Yves.
The worst situation is while you measure max power 😱
I usually add a Zener of appropriate power in parallel with each CCS at about 1.5 times the quiescent cathode voltage.
Yves.
Your CCS need to be well heatsinked. In many cases this can be just bolted to an aluminum case.
I generally use DN2540 or the IXY 600V or 900V chips for robustness.
I have also had success with TL783 which is a 120V version of the LM317.
I mess about with direct coupled amps and in this configuration the initial voltage spikes at switch on can go to the +B which will take out any lesser CCS. It can be very frustrating when one blows.
Its generally better not to have your CCS the weakest link in the overall chain.
Shoog
I generally use DN2540 or the IXY 600V or 900V chips for robustness.
I have also had success with TL783 which is a 120V version of the LM317.
I mess about with direct coupled amps and in this configuration the initial voltage spikes at switch on can go to the +B which will take out any lesser CCS. It can be very frustrating when one blows.
Its generally better not to have your CCS the weakest link in the overall chain.
Shoog
thanks for all the advice. as i do need it.
this is what i use from k&k
http://www.kandkaudio.com/accessories.html
Cascode Current Source Kit attachable to an external heat sink (up to 20 watts with adequate heat sink)
this is what i use from k&k
http://www.kandkaudio.com/accessories.html
Cascode Current Source Kit attachable to an external heat sink (up to 20 watts with adequate heat sink)
I've been running differential power amps with CCSs on the output stage for almost 10 years now. If all the bases are covered the work and sound great.
Check out the 47 P-P and Tabor schematics (pentode) and the SS Tabor (MOSFET) on my web page for ideas.
Check out the 47 P-P and Tabor schematics (pentode) and the SS Tabor (MOSFET) on my web page for ideas.
Thanks Gary!
I try to process this as much as I can so I don't screw things up.
Kevin at k&k have been very helpful, but my lack of understanding this CCS stuff vs fix bias and the traditional resister and bypass cap. So I figure I post something on here to get a better understanding.
I try to process this as much as I can so I don't screw things up.
Kevin at k&k have been very helpful, but my lack of understanding this CCS stuff vs fix bias and the traditional resister and bypass cap. So I figure I post something on here to get a better understanding.
Well, the CCS, as its name implies, tries to keep constant the current wich flows thru it what could be the voltage applied accross it.
The Ohm's law (I = U / R) states that the current (I) must be equal to the voltage (U) divided by the resistance (R) of a circuit.
So, a CCS may be seen as a resistor wich the value adapts itself "magically" to keep the current constant still satifiying Ohm's law.
For exemple, a 1 amp CCS to wich a 1 volt is applied looks like a 1 ohm resistor, but if 2 volts are applied accross it, its resistance "magically" changes to 2 ohms in order to maintain a 1 amp current thru it.
Similary, if the current thru the CCS tries to rise (or drop), its resistance increases (or decreases) until the current is restored at its original value and consequently the voltage accross the CCS also changes.
Does this help ?
Yves.
Last edited:
The bypass capacitor is the essential "spring" element into which the tubes cathode works and allows a signal voltage to pass. Without it the CCS blocks all AC signal from been generated by the output tube by imposing a fixed DC bias voltage and no AC signal to pass, the DC voltage appears amplified at the plate.
The bias on the tube must be enough to accommodate the full signal swing, plus the dropout voltage of the CCS. in the case of the LM317 this is about 3V. Its not usually a problem to generate a few extra volts of bias in an output stage, but in a preamp stage where bias might only be 1V - it will often need a negative rail. At this point fixed bias might be just as easy to implement.
Shoog
The bias on the tube must be enough to accommodate the full signal swing, plus the dropout voltage of the CCS. in the case of the LM317 this is about 3V. Its not usually a problem to generate a few extra volts of bias in an output stage, but in a preamp stage where bias might only be 1V - it will often need a negative rail. At this point fixed bias might be just as easy to implement.
Shoog
I fully agree . . .
Mmmh . . . Assuming that the anode current is equal to the cathode one (wich is very near of the truth for a triode, and "almost" true for a penthode) the CCS in the cathode voids any current change in the anode and thus in the load so that no power appears.
But, the voltage on cathode is NOT fixed at all, it follows the grid changes (the signal indeed) so that the tube no longer see it.
With a cap all goes right as long as the mean current stays constant that is -roughly- as long as we stay in class A.
When this is no longer true (i.e. at hi power level), the "spring" capacitor is more expensed than compressed and the mean voltage at each cathode rises.
"In doubt, measure !"
Yves.
Yves,
That's a great explanation on CCS device, exactly what I need to know and can understand.
Shoog,
Thanks for explaining how this works.
But here's a simple question for me if youguys don't mind.
So, to make my PP running I'n class A, I can just ignore the bypass cap? As long as I don't clip the amp right?
Rightnow the amp is bias with a pair of CCS in parallel at around 80ma per tube.
All sounded pretty good at the moment, but I want to make sure I'm not going to BMW anything up.
That's a great explanation on CCS device, exactly what I need to know and can understand.
Shoog,
Thanks for explaining how this works.
But here's a simple question for me if youguys don't mind.
So, to make my PP running I'n class A, I can just ignore the bypass cap? As long as I don't clip the amp right?
Rightnow the amp is bias with a pair of CCS in parallel at around 80ma per tube.
All sounded pretty good at the moment, but I want to make sure I'm not going to BMW anything up.
Is there a seperate CCS for each cathode - if so then they each need a bypass cap.
If there is a common CCS for the two tube cathodes (differential pair) then you don't need a bypass cap since the opposite cathode acts as the "spring".
Shoog
If there is a common CCS for the two tube cathodes (differential pair) then you don't need a bypass cap since the opposite cathode acts as the "spring".
Shoog
- Status
- Not open for further replies.