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: "Phil Allison"
References: <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org>
Subject: Re: Help - Power mosfets - difficult load
X-Newsreader: Microsoft Outlook Express 5.50.4522.1200
Date: Tue, 22 Oct 2002 18:53:37 +1000
NNTP-Posting-Date: Tue, 22 Oct 2002 18:44:47 EST
Organization: Telstra BigPond Internet Services (http://www.bigpond.com)
"Bill Allison" wrote in message
Phil wrote :
> > ** Very optimistic view. One short burst to full when the motor
> > stopped could spell disaster - practically bound to happen.
> I intend to arrange the controller so that pulse width under all
> circumstances increases slowly.
** Good luck.
> > > > >
> > The output from a RC receiver is a 50 Hz train of 1 to 2 mS
> > > > > pulses and it would be easy to use an RC receiver, the pot on the
> > > > > winch and a NE544 or similar to generate 50 Hz motor drive pulses.
> > > > > I think the relatively low frequency is good for this
> > > > > the mosfets will spend lower proportion of time switching so less
> > > > > dissipated in them
> > > >
> > > >
> > > >
> > > > ** No, much more fet and motor dissipation than if a higher
> > frequency> > were used.
> > >
> > > I can't accept that.
> > ** Then read up on motor speed control.
> I've designed, manufactured and sold 20A 12v speed controllers which
> achieved a zero warranty returns rate. BTW they were switched at 50Hz.
> What I seek is advice on the additional issues associated with scaling
> things up ten to twenty fold
** What sort of response is that ??
> > All things (i.e. rise time) being equal, the less> frequent switching
> > must mean less power dissipation in the fet.
> > ** Switching losses with fets are very small at a few kHz. With
> > voltage it is even better.
> That doesn't make what I just said untrue.
** But it makes it irrelevant.
> > > Consider the limiting case of zero switching rate i.e. always on or
> > > always off - the switching losses are then zero... !?
> > ** The dissipation will be due to *conduction* losses, why are
> > hooked on switching losses ?
> Not hooked, merely know from device literature and graphs and
> practical experience at lower currents that the power dissipated in
> the fet during the switching transition will be significant enough
> that I must take steps to minimise it to avoid exceeding device
** Fets switch happily at 100 Khz and high voltages.
> > The motor heat> dissipation can only be the difference between input
> > output power> > > and will be a function of whether it is operated at an
> I should have said ...mainly a function... but anyway what you say
> below is wrong. The RMS value of a waveform at 50Hz is the same as the
> RMS value of the same waveform at any other Hz.
** But it will NOT be the same waveform - that is the whole point.
The "average" level is
> the same also. If 50 Hz works well with smaller motors (it does - very
> well) then surely considering the larger (mechanical and electrical)
> inertia of this motor it should work well here too.
** Your assertion is wrong - 50 Hz PWM is inefficient as hell with
any DC motor.
Check out ANY professional DC motor drive - they all whistle - 1
to 5 Khz is the norm for good battery efficiency. At low speeds the benefit
can be 10 times lower drain for same speed and torque.
The PWM buck regulator is the same principle, inductance in the
DC motor will keep the current almost at a steady value if the PWM frequency
is high enough and there is a freewheeling diode in place.
. . . . . . .. ................. Phil
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup