Source of propagation delay?

[KingJL]

I had seen a lower resistance for R1 in another thread. It worked better with the 10meg, so I left it there. Probably don't need it.

Don't need it... that is why it worked better with 10meg. Lower resistance bleeds off the reverse bias charge buildup at the cathode of the diode. If the proper ultrafast diode is used ( I use BYV28-200 in the actual circuit... can be simulated with a model for a BYG20J ) this reverse bias charge buildup reduces the capacitance of the diode from the time that the coil discharge pulse returns to 150V-100V until the diode id forward biased by the TX cghrge current from the mosfet. The result is that you can reduce the TX curcuit capacutance to that of the diode (or thereabout). This behavior is disputed by some... but it works in the real world for me... I think the results depend on the characteristics of the diode.

[Davor]

Actually 6Meg simulates MOSFET leakage better. In case you wish to see a drain voltage simulated properly with a MUR460, you should keep the resistor. You should be able to observe it as SPA11N60C3 is modelled with all the necessary capacitances.

[green]

I tried 100k, 10meg, and 100meg in the spice model. Couldn't see a difference with 10 or 100meg. With 100k the coil decay didn't go to zero and stay there. If it's constant probably wouldn't matter in the real circuit. It looks better with high resistance in spice.

[VinceC]

I don't think R1 would contribute to noise. The voltage on R1 is over 200 volts when the coil volts is at zero Isolated from the amplifier. I'm not a EE so could be wrong. The cap could be either place for the model. Putting it external allows someone to see the value looking at the schematic.

This is obviously where my ignorance will shine! I have no idea if the leakage current of the diode allows "reverse" noise propagation through the diode. Perhaps one of our more knowledgeable members can dispel my crazy concept.

Vince

[VinceC]

Don't need it... that is why it worked better with 10meg. Lower resistance bleeds off the reverse bias charge buildup at the cathode of the diode. If the proper ultrafast diode is used ( I use BYV28-200 in the actual circuit... can be simulated with a model for a BYG20J ) this reverse bias charge buildup reduces the capacitance of the diode from the time that the coil discharge pulse returns to 150V-100V until the diode id forward biased by the TX cghrge current from the mosfet. The result is that you can reduce the TX curcuit capacutance to that of the diode (or thereabout). This behavior is disputed by some... but it works in the real world for me... I think the results depend on the characteristics of the diode.

If your concept is correct, what diode characteristics are really required to maximize this effect?

Vince

[VinceC]

Actually 6Meg simulates MOSFET leakage better. In case you wish to see a drain voltage simulated properly with a MUR460, you should keep the resistor. You should be able to observe it as SPA11N60C3 is modelled with all the necessary capacitances.

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

[VinceC]

I tried 100k, 10meg, and 100meg in the spice model. Couldn't see a difference with 10 or 100meg. With 100k the coil decay didn't go to zero and stay there. If it's constant probably wouldn't matter in the real circuit. It looks better with high resistance in spice.

I actually found that infinite resistance gave the best results! But if what Davor is saying is correct, the Spice Model may be a bit too ideal!

Vince

[KingJL]

If your concept is correct, what diode characteristics are really required to maximize this effect?

Vince

I am not sure of all the needed characteristics, but speed is extremely important... a fast diode is not fast enough. I haven't tried many diodes in an actual circuit. I tried a 1N1004 and a BYV28-200 in my HAMMERHEAD 2. The 1N1004 did not produce the desired results... the BYV28-200 did... my damper resistance went from around 470 ohm to 820 ohm and I can sample down to 6.5 usec. I plan on using the BYV28-200 (I have 9 of them) in my current Minipulse Plus build.

[KingJL]

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

It depends on the model used... I tend not to use the LT Spice built-in mosfet models. I usually get the Mosfet spice models and the Mosfets from Fairchild.

[green]

I am not sure of all the needed characteristics, but speed is extremely important... a fast diode is not fast enough. I haven't tried many diodes in an actual circuit. I tried a 1N1004 and a BYV28-200 in my HAMMERHEAD 2. The 1N1004 did not produce the desired results... the BYV28-200 did... my damper resistance went from around 470 ohm to 820 ohm and I can sample down to 6.5 usec. I plan on using the BYV28-200 (I have 9 of them) in my current Minipulse Plus build.

Maybe I'm wrong, but it sounds your referencing the diode. We were referencing the 10meg resistor from the diode to ground. A fast diode is needed to speed things up. I have tried the 1N4004 with the same result as you got(didn't work).

[VinceC]

I am not sure of all the needed characteristics, but speed is extremely important... a fast diode is not fast enough. I haven't tried many diodes in an actual circuit. I tried a 1N1004 and a BYV28-200 in my HAMMERHEAD 2. The 1N1004 did not produce the desired results... the BYV28-200 did... my damper resistance went from around 470 ohm to 820 ohm and I can sample down to 6.5 usec. I plan on using the BYV28-200 (I have 9 of them) in my current Minipulse Plus build.

For the non-EE, what is "fast" in your mind?

[KingJL]

For the non-EE, what is "fast" in your mind?

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

[golfnut]

But wouldn't a preamp design that reduces the time to settle be preferable?


Its tough in a single stage. The easy way is two amps in cascade, each amp saturates less and so recovers from large inputs faster.

[VinceC]

But wouldn't a preamp design that reduces the time to settle be preferable?


Its tough in a single stage. The easy way is two amps in cascade, each amp saturates less and so recovers from large inputs faster.

From what I have been reading, this is not and "either or" but rather you MUST have both to achieve the full advantages of either portion.

Vince

[VinceC]

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

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