Yes, the experimental results show equal impedances when loaded equally. They agree exactly with the Thevenin models in my Figure 3.
If you will propose a doable experiment where ANY arbitrary equal pair of loads will not give results consistent with my Thevenin model but WILL give results consistent with yours, let's do it!
Gentlemen;
I still don't understand how the same current flowing through two linear networks with the same impedances connected in series can cause different voltage drops on them? We have: cathode load, the tube (anode to cathode), anode load, power supply, in the single closed loop, connected in series. If the current flows the same and impedances are equal that means voltage drops are equal. What else to discuss here? What and why to complicate?
It is trivial actually.
I still don't understand how the same current flowing through two linear networks with the same impedances connected in series can cause different voltage drops on them? We have: cathode load, the tube (anode to cathode), anode load, power supply, in the single closed loop, connected in series. If the current flows the same and impedances are equal that means voltage drops are equal. What else to discuss here? What and why to complicate?
It is trivial actually.
You are not addressing the flaws I pointed out in the experiments in your article. If your tests are flawed, their results are of questionable value.
Again, I ask you to address the flaws I attribute to your experiements.
I also ask you to find flaws in my analyses. You cannot find flaws in analyses by running bench tests. You can produce results that cause you to question an analysis if your test is well designed, but you can't discover what the flaws are, if any, from those results.
Again, I ask you to address the flaws I attribute to your experiements.
I also ask you to find flaws in my analyses. You cannot find flaws in analyses by running bench tests. You can produce results that cause you to question an analysis if your test is well designed, but you can't discover what the flaws are, if any, from those results.
You seemed concerned with "ground current". I don't get that. Ground is an arbitrary point from which we measure voltage. In my experiments, as with any analysis, it is the point at which the loads and input return. And that's where the measurements are done, where the load currents and input currents meet.
The way Thevenin source impedance is universally measured is to vary the load, either explicitly or implicitly, and see the change in the output voltage. That is exactly what I did. I can't find those "flaws" you want me to address.
If you will propose a doable experiment where ANY arbitrary equal pair of loads will not give results consistent with my Thevenin model but WILL give results consistent with yours, let's do it!
Let me set aside your #86 comments for the moment. Without abandoning my requests for you to address my criticisms directly, let me set them aside for the moment while we see if we can find a mutually acceptable experiment.
To you, it is important to ensure that impedance tests of a balanced Cathodyne maintain that balance, correct? Equal P & K loads and equal and oposite P & K voltages at all times, right?
To me, it is important that tests of impedance actually test the impedance that we say they do. We could reasonably ask for the impedance between the P & K, a "floating" impedance. Instead, I believe that we are interested in the impedance bewteen P and Gnd and K and Gnd. You can't just have the impedance of a single node. A Thevenin circuit has two nodes, right?
To accomplish this, I propose the following definition of impedance. If you agree with the definition, please say so. If you don't, please propose an alternative.
The impedance of a pair of nodes in a circuit is the change in voltage between the nodes due to and divided by a current flowing from a device outside the circuit, through the two nodes, and back through that device. The device may be a signal source or a load.
To you, it is important to ensure that impedance tests of a balanced Cathodyne maintain that balance, correct? Equal P & K loads and equal and oposite P & K voltages at all times, right?
To me, it is important that tests of impedance actually test the impedance that we say they do. We could reasonably ask for the impedance between the P & K, a "floating" impedance. Instead, I believe that we are interested in the impedance bewteen P and Gnd and K and Gnd. You can't just have the impedance of a single node. A Thevenin circuit has two nodes, right?
To accomplish this, I propose the following definition of impedance. If you agree with the definition, please say so. If you don't, please propose an alternative.
The impedance of a pair of nodes in a circuit is the change in voltage between the nodes due to and divided by a current flowing from a device outside the circuit, through the two nodes, and back through that device. The device may be a signal source or a load.
If loads are equal how voltages will be different? Remember, both loads are connected effectively in series with each other and with the tube. No matter how the tube distorts the signal, and how distortions depend on load, but if both loads are linear and equal voltage drops on them always will be equal, simply because they are effectively in series, i.e. the same current flows through them, and they have the same impedances.
You are knocking hard in already wide open gates.
It is trivial as well: if loads are equal output impedances are equal. If loads are not equal, impedances are not. Because in case of concertina you can't split output and load impedances, because they are interconnected.
And indeed I measured between plate and ground and cathode and ground. Simultaneously.
A Thevenin circuit has two nodes. Two Thevenin circuits have either three or four nodes, in this case, the former since the return leads are referenced to the same point (ground).
Please proceed step by step. You invited me to propose the test, right?
To you, it is important to ensure that impedance tests of a balanced Cathodyne maintain that balance, correct? Equal P & K loads and equal and oposite P & K voltages at all times, right?
To me, it is important that tests of impedance actually test the impedance that we say they do. We could reasonably ask for the impedance between the P & K, a "floating" impedance. Instead, I believe that we are interested in the impedance bewteen P and Gnd and K and Gnd. You can't just have the impedance of a single node. A Thevenin circuit has two nodes, right?
To accomplish this, I propose the following definition of impedance. If you agree with the definition, please say so. If you don't, please propose an alternative.
The impedance of a pair of nodes in a circuit is the change in voltage between the nodes due to and divided by a current flowing from a device outside the circuit, through the two nodes, and back through that device. The device may be a signal source or a load.
To you, it is important to ensure that impedance tests of a balanced Cathodyne maintain that balance, correct? Equal P & K loads and equal and oposite P & K voltages at all times, right?
To me, it is important that tests of impedance actually test the impedance that we say they do. We could reasonably ask for the impedance between the P & K, a "floating" impedance. Instead, I believe that we are interested in the impedance bewteen P and Gnd and K and Gnd. You can't just have the impedance of a single node. A Thevenin circuit has two nodes, right?
To accomplish this, I propose the following definition of impedance. If you agree with the definition, please say so. If you don't, please propose an alternative.
The impedance of a pair of nodes in a circuit is the change in voltage between the nodes due to and divided by a current flowing from a device outside the circuit, through the two nodes, and back through that device. The device may be a signal source or a load.
I think your approach of driving one node is flawed because if you drive the plate node with a current, you do not get an equal but opposite voltage appearing at the cathode.
see the attached sims which suggest that the outpuut Z of a 6dj8 concertina is around 82 ohms in both the plate and the cathode. I used your current injection method (using 1A for ease of reading). I then looked at the voltage waveform at the plate and cathode to confirm proper behavior and phasing. Next I confirmed my results with the concept that Zout=Vopen circuit/Ishort circuit.
dave
Please prove before how dV caused by dI on equal impedances connected in series can be unequal. I.e. that Ohm's law is wrong, or definitions of current, voltages, and impedance are wrong.
Please see Letters to the Editor under Online Services on the Linear Audio website. You'll see a letter that responds to Stuart Yaniger's article and describes a way to determine Cathodyne plate and cathode impedances.
Because the P and K currents of a triode with no grid current are always equal and opposite, these currents through identical plate and cathode loads will yield equal and opposite voltages at the P and K. It does not follow that the impedances of the plate and cathode of the triode driving them are equal. In fact, you will find all over the place expressions for the impedance Zp looking into the plate = rp + (1 + u)*Rk, and the impedance Zk looking into the cathode = (rp + Rp)/(1 + u). You can see that these are not equal when Rp = Rk. I will leave it to you to reconcile this fact with the assertion that Zk = Zp in a Cathodyne where Rp = Rk.
Because the P and K currents of a triode with no grid current are always equal and opposite, these currents through identical plate and cathode loads will yield equal and opposite voltages at the P and K. It does not follow that the impedances of the plate and cathode of the triode driving them are equal. In fact, you will find all over the place expressions for the impedance Zp looking into the plate = rp + (1 + u)*Rk, and the impedance Zk looking into the cathode = (rp + Rp)/(1 + u). You can see that these are not equal when Rp = Rk. I will leave it to you to reconcile this fact with the assertion that Zk = Zp in a Cathodyne where Rp = Rk.
Dave did define it, no need to ask him to repeat it, and he defined it the standard way: open circuit voltage divided by short circuit current.
Well, yes, that is the definition of Thevenin source impedance. You generally don't short the output of a power amp to test the source impedance because of current limits, you vary the load a small amount and see how the output changes.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.