Granted, but oscillation is all about circuit Q and this is affected by particular Quality factors of the component reactances.
Ignoring resonant frequency and circuit change, if the lead inductance is preceded or rather interspersed with a "lossy" resistor,
Q can fall and hence oscillation be damped or even stop. I have seen this so a couple of times with L-Mosfets when looking
at the output scope trace before and after shifting the resistor closer.
You could argue that resonance was just being shifted up well beyond Ft but with the values in play, I don't assume that is the case.
Oscillation frequency was initially ~ 40 MHz and only 4V IIRC. I should add that this arose from PCB design issues.
Wider evidence is that the stopper is still not ideal and Cordell now recommends just 47 ohm stoppers with added Zobels of about
47 ohm/100 pF to the Drain on VMosfet gates.
I am not arguing against anything except a small misconception, and explaining why it is better to put the resistor closer to the device, because Andrew asked for it. I understand that getting the resistor inside the parasitic network rather than outside will decrease Q.
I feel like I'm being misunderstood, or not understood at all. Fortunately, feedback (har har) remedies all problems. Am I speaking clearly or do my posts read like nonsense?
PS. I don't like to use smileys but if I did my post would be smiling humorously.
- keantoken
I feel like I'm being misunderstood, or not understood at all. Fortunately, feedback (har har) remedies all problems. Am I speaking clearly or do my posts read like nonsense?
PS. I don't like to use smileys but if I did my post would be smiling humorously.
- keantoken
Kean,
you are losing me.
You and the others all jumped in and said that the stopper adds real resistance to the load seen by the driver.
This view took no account of oft repeated advice to locate the stopper at the base/gate of the output device.
You, particularly, state that reducing the length of lead between the stopper and the gate does not reduce the inductance between the gate capacitances and the stopper. The others are coming round to using just that analogy to support the view that stopper location is critical to it's effectiveness.
I cannot in your explanations see you specifying why the stopper location is important. Parasitic capacitance does not help me understand any better.
you are losing me.
You and the others all jumped in and said that the stopper adds real resistance to the load seen by the driver.
This view took no account of oft repeated advice to locate the stopper at the base/gate of the output device.
You, particularly, state that reducing the length of lead between the stopper and the gate does not reduce the inductance between the gate capacitances and the stopper. The others are coming round to using just that analogy to support the view that stopper location is critical to it's effectiveness.
I cannot in your explanations see you specifying why the stopper location is important. Parasitic capacitance does not help me understand any better.
I never said that reducing the lead length to the gate does not reduce the inductance between the gate capacitances and stopper. If I did, could you point it out? Picture this:
A wire goes from driver to gate/base. This wire is the same length whether there is a resistor in the middle of it or not. The resistor can either be placed next to the driver, or next to the gate/base. In the first place, the stray lead capacitance is mostly on the side of the resistor that connects to the gate/base. The currents flowing through this parasitic capacitance don't encounter the resistor in their path to the gate/base. The inductance of the gate can freely interact with this capacitance and resonate. In the second place, the resistor is closest to the gate, minimizing the exposed lead area between the resistor and gate, which minimizes the stray capacitances directly to the gate. Instead, this surface area is relocated to the other side of the resistor, but the current through this capacitance must flow through the resistor before it encounters the gate.
- keantoken
A wire goes from driver to gate/base. This wire is the same length whether there is a resistor in the middle of it or not. The resistor can either be placed next to the driver, or next to the gate/base. In the first place, the stray lead capacitance is mostly on the side of the resistor that connects to the gate/base. The currents flowing through this parasitic capacitance don't encounter the resistor in their path to the gate/base. The inductance of the gate can freely interact with this capacitance and resonate. In the second place, the resistor is closest to the gate, minimizing the exposed lead area between the resistor and gate, which minimizes the stray capacitances directly to the gate. Instead, this surface area is relocated to the other side of the resistor, but the current through this capacitance must flow through the resistor before it encounters the gate.
- keantoken
I still cannot make your parasitic capacitances fit what I think is happening.
I know my thinking is wrong and needs at modification for the very least. I can understand the reduced inductance and the real capacitances. Only now have you finally confirmed my original contention that locating the resistor direct on the shortest gate lead leg length does it reduce inductance.
I know my thinking is wrong and needs at modification for the very least. I can understand the reduced inductance and the real capacitances. Only now have you finally confirmed my original contention that locating the resistor direct on the shortest gate lead leg length does it reduce inductance.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.