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: firstname.lastname@example.org (Bill Allison)
Subject: Re: Help - Power mosfets - difficult load
Date: 31 Oct 2002 15:08:24 -0800
References: <email@example.com> <firstname.lastname@example.org> <3DB81206.E28CA99F@rica.net> <email@example.com> <3DB963BE.939694C2@rica.net> <firstname.lastname@example.org> <3DBC2B63.146AA440@rica.net> <email@example.com> <3DBD434D.3BA1BB54@rica.net> <firstname.lastname@example.org> <email@example.com>
NNTP-Posting-Date: 31 Oct 2002 23:08:24 GMT
Winfield Hill wrote in message news:...
> Bill Allison wrote
> > Tony, John
> > I know that discontinuous low frequency operation will be inefficient,
> > but...
> > 1) Nobody has disagreed with my contention that those large low
> > frequency impulses will be *very* useful in overcoming stiction and
> > allowing the motor to be controlled smoothly at start-up and at very
> > low speed. ...
> You could force a 50% 50Hz duty cycle on top of the high-frequency
> PWM, but only for the lowest PWM ratios, to create vibration.
> > 2) *Provided* I can keep within fet ratings, I don't care too much
> > about efficiency - the unit is going to be powered up for only 10
> > seconds at most every 5 minutes at least so heating isn't a huge
> > problem and irrespective of frequency, at full throttle (i.e 100%
> > duty cycle) losses will be due only to RDSon.
> I showed in detail how you'll be heating the FETs with 4x higher
> dissipation by using a 50Hz PWM rate, and running them near or past
> their specs in the 25% to 100% PWM region. A FET junction's thermal
> time constant is well under a second, more like 0.1 second, so your
> 10-second operating time may not be of much help, if you intend to
> keep the FET junction below the recommended temperature limits.
Indeed you did, for which I'm grateful, however those calculations
were based on stall current, which won't pass through the fet if I use
shunt resistance as Steve S suggested or some other current limiting
I just downloaded an-1033.pdf which I somehow missed until now. A
software program called HEXRISE referred to therein is said to be able
to accurately calculate and graph junction temperature for any
waveform. I've downloaded it to see how it and my project get on.
(It's an Access 2000 file, and i've got Access 97 - I've stepped off
Bill Gates horrible upgrade bandwagon - but I can get around that).
> > If tests show this isn't do-able (fets-wise) at 50 Hz then circuitry
> > will need to be devised to produce a 500Hz PWM drive from the 50Hz
> > control signal but I want to avoid that complication if possible.
> A easy way to do this would be to use a pot-operated PWM-generator
> circuit, and run the pot from your RC servo output shaft. :-)
Feels too much like cheating, and from that wry smile I guess seems
the same to you!
> - Win
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup