The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Winfield Hill
Subject: Re: Help - Power mosfets - difficult load
Date: 28 Oct 2002 10:06:41 -0800
Organization: Rowland Institute
References: <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com> <3DB81206.E28CA99F@rica.net> <firstname.lastname@example.org> <3DB963BE.939694C2@rica.net>
X-Newsreader: Direct Read News 2.98
>> At these low impedances (low voltage and high current) I think
>> it is impractical to build effective snubbers. The best you
>> can do is drive the gate hard and fast, and use some hefty
>> bypass capacitors between the non switched end of the schottky
>> and the source of the fet. And tie the motor lead that is
>> unswitched to this cap also on its way back to the battery.
>> Your circuit is going to produce some whopping big transients
>> in the battery leads, and you need to shunt those with a very
>> local capacitor. If you do that, I think the drain voltage will
>> be well behaved, and the turn on and turn off transients so fast
>> that the total heat generated will be low compared to the high
>> current conduction losses. The big risk is that the wiring
>> inductance back to the battery will produce unintended multiple
>> edges when you try to turn the fet off while something like 100
>> amps is flowing in those leads.
> The transient problems you fear are easily avoided with proper
> gate-drive wiring: the gate-driver IC's ground and its bypass
> capacitor must be connected directly with a very short leads to
> the FET's source using a Kelvin connection. Note the FB180SA10
> provides two source connections, one for the high-current path,
> the other for the twisted-pair gate drive. In this fashion, as
> the high-current source path experiences huge di/dt inductance
> voltage spikes the gate driver follows, so the gate drive isn't
> compromised and the gate isn't damaged. In this situation, gate
> capacitance is your friend. Note, in cases where extreme source
> spikes exist, such as likely will happen here, the driver IC can
> be isolated from the rest of the system with RC connections,
> --/\/----, 0.1
> _+-||-, _______ FB180SA10
> 1.0k | \ | 3.3 | |
> --/\/--+---| >-- | --/\/\-------G D===========
> | |_/ | twist | | braid
> '-||--+----+--------------S S===x===+==== BATT -
> logic 0.001 |_______| |
> gnd ----------------------------------------'
> These insure that the driver IC is powered and gets its input
> logic command, and that both remain undisturbed (as seen by the
> driver) when the high-current source connection (point x) goes
> through violent switching gyrations. In this way the FET gate
> drive (whether just turned on or off) is also undisturbed, as
> seen internally by the FET.
> I have used this scheme to quickly switch 900A.
>> Inductance in the motor leads is not only harmless, (if the
>> diode is right up against the fet) but actually beneficial, and
>> you may want to increase it by wrapping the motor leads ...
> He'll have negligible load capacitance, and up to 200uH of
> starter-motor inductance, plus wiring, so I don't really see
> much of an advantage to adding more. :-)
John, I assume you saw this posting? I added a ground ref wire.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup