Source of propagation delay?

[green]

A chart and some scope shots of my bench circuit I posted on another thread awhile back. The (MUR460, no diode) is a time constant plot of two targets with the diode and with the diode shorted. The diode would allow sampling at 4usec instead of 9.5 usec. The amplitude is higher with no diode do to more voltage charging the coil causing a higher peak current. The on time could be increased to increase the peak current for the diode circuit. Don't see any advantage or disadvantage with the diode circuit if sampling with longer sampling delays for higher time constant targets. The (diode no diode) scope traces are coil volts with R damping disconnected for some diodes I had. The lower scope traces of resonance are with the scope probe disconnected from the coil laying near the drive circuit.

[VinceC]

Here is a spec for one that works... use that as your starting point.

Would the BYV27-xx (http://www.vishay.com/docs/86042/byv27.pdf) make more sense? It has an even faster recovery time.

Vince

[KingJL]

Is there any particular advantage for the 50V version over the 200V version?

In the real circuit, the reverse voltage buildup will be around 125V - 175V depending on flyback voltage. I would want the avalanche region to be higher than that... hence the 200V version.

I believe that green's MUR460 is rated as ultrafast would probably work as well. The main difference between them is what happens when reverse breakdown occurs. The avalanche diode is not damaged as long as reverse current is within specs.

[VinceC]

A chart and some scope shots of my bench circuit I posted on another thread awhile back. The (MUR460, no diode) is a time constant plot of two targets with the diode and with the diode shorted. The diode would allow sampling at 4usec instead of 9.5 usec. The amplitude is higher with no diode do to more voltage charging the coil causing a higher peak current. The on time could be increased to increase the peak current for the diode circuit. Don't see any advantage or disadvantage with the diode circuit if sampling with longer sampling delays for higher time constant targets. The (diode no diode) scope traces are coil volts with R damping disconnected for some diodes I had. The lower scope traces of resonance are with the scope probe disconnected from the coil laying near the drive circuit.

How do you collect that data in the first chart? What does the setup look like? It appears that you automated a translation table to operate with a DAQ.

Vince

[KingJL]

Would the BYV27-xx (http://www.vishay.com/docs/86042/byv27.pdf) make more sense? It has an even faster recovery time.

Vince

It has more forward voltage drop at the same current... but would probably work reasonably well. There are always trade-offs.

[VinceC]

It has more forward voltage drop at the same current... but would probably work reasonably well. There are always trade-offs.

Would this (http://diodes.com/datasheets/ds30178.pdf) be a better choice then?

[KingJL]

Would this (http://diodes.com/datasheets/ds30178.pdf) be a better choice then?

This reminds me of a time in life when I made my living designing and producing things. I then worked with one of the best project managers I have ever encountered. He had 2 large plaques on his wall... one: "Don't let the good preclude the acceptable" The other: "There comes a time in every project when it is advisable to shoot the engineers and ACTUALLY PRODUCE A PRODUCT".

[VinceC]

This reminds me of a time in life when I made my living designing and producing things. I then worked with one of the best project managers I have ever encountered. He had 2 large plaques on his wall... one: "Don't let the good preclude the acceptable" The other: "There comes a time in every project when it is advisable to shoot the engineers and ACTUALLY PRODUCE A PRODUCT".

Touche! However, when I don't have the luxury of trying many products to get a design that works well the first time I tend to try to find the best components that are readily available.

Vince

[green]

How do you collect that data in the first chart? What does the setup look like? It appears that you automated a translation table to operate with a DAQ.

Vince

I record the amplifier out with a DSO, the one in the scope pictures. One recording with no target and one recording with target. Subtract no target recording from target recording with Excel and chart data. The external trigger sets time zero when fet gate goes low.

[Carl-NC]

The reason that I asked about R1 is that when I removed it the simulation performance was actually better. I am nowhere close to being able to run a .NOISE analysis for any circuit yet, but I do know that high resistances are a significant source of Johnson Noise, and I am looking to be able to limit the amount of intrinsic noise in the circuitry ahead of the preamp.

Resistor noise depends on how the resistor is used. If you think about what you said -- "I do know that high resistances are a significant source of Johnson Noise" -- and then remove R1, you are actually taking R1 to "infinite resistance" but its noise contribution drops to zero. In general, series resistance contributes voltage noise (4kTR) and shunt resistance contributes current noise (4kT/R).

[Davor]

Is the leakage current missing from the Spice model for this component? Is that why the resistor is required? If so, should the resistor be connected to ground for this circuit?

Vince

There is a way of simulating leakage current of N-channel MOSFETS, but the models I checked against the spec sheets were lacking a lot of it. So in order to be more precise about it, you should see what specs say for Vgso and simply put a resistor in between drain and source that draws that much current at the average Vds voltage. You could do it more precisely, but a simple resistor should do just fine.
In case of a PI driving MOSFET, there is a negligible difference for a flyback with or without the resistor. A real advantage is seeing if the drain voltage ever drops to zero with a diode in series with drain, and with a pulse rate at hand. Most probably it will remain above ~50V with "normal" capacitances of a MOSFET, keeping a diode firmly shut.

[green]

Recorded scope trace of decay with and without 10 meg resistor. 10 meg to ground without the resistor and scope probe, 5 meg to ground with the resistor and scope probe. Plus the added capacitance of the scope probe. Wondering if comparing coil resonance with circuit resonance (R damping disconnected) or recording voltage drop is a good way to compare diodes.

[KingJL]

Recorded scope trace of decay with and without 10 meg resistor. 10 meg to ground without the resistor and scope probe, 5 meg to ground with the resistor and scope probe. Plus the added capacitance of the scope probe. Wondering if comparing coil resonance with circuit resonance (R damping disconnected) or recording voltage drop is a good way to compare diodes.

The only reason that I compare voltage drop is that is an indication of dynamic forward resistance, thus the effect of the diode on peak coil current. It is just a personal preference that I have... just as I prefer using insulated drain mosfets.

Maybe we are talking about different voltage drops... my reference to voltage drop us forward voltage drop at a particular current, as stated in diode specifications.

[green]

One more experiment to test diode effectiveness. Scope probe near circuit not connected to minimize circuit change. R damping disconnected. Circuit resonance: (coil connected to drain, 345khz) (coil connected to MUR460 anode, doide in series with drain, 1.23Mhz) (coil connected to 300 ohm resistor in series with 1N4148 in series with drain, 1.28Mhz). Would think the 1N4148 and 300 ohm resistor would give very good isolation. Without a frequency display I couldn't tell the difference with the MUR460. The MUR460 does a good job of isolating the drain from the coil.

[KingJL]

One more experiment to test diode effectiveness. Scope probe near circuit not connected to minimize circuit change. R damping disconnected. Circuit resonance: (coil connected to drain, 345khz) (coil connected to MUR460 anode, doide in series with drain, 1.23Mhz) (coil connected to 300 ohm resistor in series with 1N4148 in series with drain, 1.28Mhz). Would think the 1N4148 and 300 ohm resistor would give very good isolation. Without a frequency display I couldn't tell the difference with the MUR460. The MUR460 does a good job of isolating the drain from the coil.

For your coil drive circuit, with the relatively low coil charge current and relatively low flyback, the 1N4148 is a good choice because of the low reverse bias capacitance.

The reason I mentioned my preference for insulated drain mosfets in an above post is that if you use a heat sink with the non insulated drain, the heat sink makes a very effective antenna... insulated drain mosfets (like a FQPF6N60C ) reduce or eliminate this effect.