Happy Birthday Edmond..................
Glen what happened to your land rover restoration project!!
cheers,
Kanwar
G.Kleinschmidt said:
Yeah!
The fact that they dribble a little bit, usually smell like wee, blindly and haphazardly drive their Volvo's and Nissan Pulsars at a top speed of 6 mph and generally suck society dry via their pension benefits doesn't mean that they are silly.
Lumanauw, that's an awfully unkind thing to imply.
Glen what happened to your land rover restoration project!!
cheers,
Kanwar
Here's a thought to get us back on track. Edmond has gotten me to thinking in more detail about the Cgd of MOSFETs. Here's a question I pose to think about:
What the MOSFET datasheets do and do not tell us about the behavior of Cgd as a function of multiple parameters like Vds, Vgs and Id.
Cheers,
Bob
What the MOSFET datasheets do and do not tell us about the behavior of Cgd as a function of multiple parameters like Vds, Vgs and Id.
Cheers,
Bob
Well, that's easy. The data sheets tell you virtually nothing about Cgd, especially anything that would be helpful when designing an audio amplifier.
If you are lucky they will give you a single graph, but only under conditions that would *never* be present in an amplifier, and therefore completely misrepresentative of the actual Cgd when used in an amplifier circuit.
But we can't fault the manufacturers for that. Why should they spend thousands of dollars characterizing a device for an application it was never intended to be used for?
Edit: I am specifically referring to vertical devices in this post.
If you are lucky they will give you a single graph, but only under conditions that would *never* be present in an amplifier, and therefore completely misrepresentative of the actual Cgd when used in an amplifier circuit.
But we can't fault the manufacturers for that. Why should they spend thousands of dollars characterizing a device for an application it was never intended to be used for?
Edit: I am specifically referring to vertical devices in this post.
Hello Edmond,
Yes, you are correct. I had forgotten about these parts. They never seem to have caught on, at least in this country as their performance (in terms of transconductance and power handling) isn't much better than the lateral parts from Hitachi (Renesas). The only product I know of that uses them are some of John Curl's designs for Parasound, where they are employed as drivers for a BJT output stage. Thanks for the reminder.
Yes, you are correct. I had forgotten about these parts. They never seem to have caught on, at least in this country as their performance (in terms of transconductance and power handling) isn't much better than the lateral parts from Hitachi (Renesas). The only product I know of that uses them are some of John Curl's designs for Parasound, where they are employed as drivers for a BJT output stage. Thanks for the reminder.
By the way, the only source I am aware of that discusses the effects (and causes) of Cgd in MOSFET's (both vertical and lateral) is in a book by Ed Oxner. Ed is the "godfather" of FETs in this country and worked at Siliconix for over 30 years. He wrote 4 books on FET's, which are unfortunately now all out of print. You may be able to find used copies on the internet if you are patient.
Charles Hansen said:
Hi Charles,
I agree, the datasheets are lacking, and the single graph they give indeed does not seem to tell us much of what we would like to know. That, of course, doesn't mean the devices are no good. I also think that the datasheet for capacitances in the Toshiba verticals are lacking.
One aspect I find lacking is the fact that the Cgd is measured only as a function of Vds with Vgs = 0. As you correctly point out, that is not a very useful condition.
Maybe when I get back I'll do some lab measurements. I'm not sure I trust the way it is often modeled for SPICE, either.
For example, what is Cgd in a source follower when the drain and gate are both at the same rail potential and the device is carrying 5 amps, so Vgs might be 5 V or so for an IRFP240? The data sheet just doesn't tell us that. Better yet, what is the Cgd when the gate is a few volts ABOVE the drain potential?
You're right about Ed Oxner. A great guy. A long time ago he hepled me understand some of the intricacies of power MOSFET parasitic oscillations.
Cheers,
Bob