Searching \ for ' [EE] PWM frequency' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/pwm/index.htm?key=pwm
Search entire site for: 'PWM frequency'.

No exact or substring matches. trying for part
PICList Thread
'[PICLIST] [EE] PWM frequency'
2002\03\19@090737 by Augusto de Conto

flavicon
face
I'd like to know what is the
best PWM frequency? Why?

A good article that explains it
can also help.

Thank

Augusto de Conto
spam_OUTaugustoTakeThisOuTspamautomacao.eng.br

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\03\19@093525 by Madhu Annapragada

picon face
Depends on the context of the application. In motion control, for example,
you might
want the switching frequency to be above the audible range so 20KHz will do
fine. If
you are working with power conversion then there are other factors to
consider.
As regards articles, you might want to consider the application first and
then do a Google
search on +PWM +Application.

Madhu

>I'd like to know what is the
?best PWM frequency? Why?

>A good article that explains it
>can also help.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\03\20@145332 by Peter L. Peres

picon face
>I'd like to know what is the
>best PWM frequency? Why?

The best PWM frequency is usually the highest frequency you can afford
when you consider the efficiency of any chokes or coils you use the PWM
voltage on (they have a maximum efficiency rated frequency). Also because
high frequency requires lower value filter capacitors and inductors for
the same smoothing, and the design should be cheaper like this.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\03\20@153013 by michael brown

flavicon
face
> The best PWM frequency is usually the highest frequency you can afford
> when you consider the efficiency of any chokes or coils you use the PWM
> voltage on (they have a maximum efficiency rated frequency). Also because
> high frequency requires lower value filter capacitors and inductors for
> the same smoothing, and the design should be cheaper like this.

Peter,

I don't disagree with you, but I have a question.  Don't FET's generate most
of their internal heating during the transition phase (turning off or
turning on, just not quite all the way on or off yet).  If so, wouldn't this
make a case for a lower PWM frequency?

michael brown

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\03\20@171402 by Madhu Annapragada

picon face
To clarify that a little bit, FETs generate switching losses during the rise
and fall times of the input PWM waveforms. So faster your rise and fall
times, the smaller the switching loss (although too fast and you have some
nasty side effects like dv/dt induced turn on of gates, activation of the
parasitic BJT in the FET etc).
Now if your PWM frequency is very high and the rise and fall times to the
gates are not that high, then you spend a lot of time in the linear region
and you have a lot of power loss. You can alleviate this by making your gate
voltage rise times faster (keeping other considerations like DV/Dt and DI/DT
effects in mind) and still keep a higher PWM input frequency.
Madhu

{Original Message removed}

2002\03\20@172624 by John Dammeyer

flavicon
face
I'll second that.  The easiest way to check it out if you have a scope
(or you can do the math too),  is to apply a 1% duty cycle at a given
frequency.  Note that the width of that pulse changes with the applied
frequency.  I use 1% but if you want better PWM resolution then use the
smallest increment you wish to have.  During the ON time of that 1% the
current through your inductor should reach some desired value.  If it's
a motor the ideal is the maximum current the motor winding can handle.
IF that takes 1ms then your PWM frequency could be set at 10Hz.  Now
with a 1% PWM value you get full motor torque for 1ms every 100ms.

At 50% you get full motor torque for have the time that power is applied
to the motor.    Using a higher frequency just means you won't develop
the rated power through a motor or solenoid until the PWM percent is at
a point where full current can develop.

John


> {Original Message removed}

2002\03\20@174451 by John Dammeyer

flavicon
face
For motor control,  the ideal is up above 15KHz simple because the noise
(whine is more like it) can drive people to do unatural acts.)  But I've
just done some PWM on a 10A 48V solenoid where I drive it at 100% until
the current approaches my cuttoff point and then I bring the current
down to about 5%.  That pulls the solenoid in really quickly with
maximum torque but holds it at a much lower current using far less
power.

John


> {Original Message removed}

2002\03\21@145426 by Peter L. Peres

picon face
>Peter,
> I don't disagree with you, but I have a question.  Don't FET's generate
>most of their internal heating during the transition phase (turning off
>or turning on, just not quite all the way on or off yet).  If so,
>wouldn't this make a case for a lower PWM frequency?

Yes, but usually there are coils and/or capacitors downstream and they are
more important to limit the switching frequency than the transistors.
Coils and transformers cost much more than the active elements in a
commercial design (because they are custom made) and they are more
important for size and efficiency gains. So you push the frequency up as
far as you can while checking for coil core material efficiency curves and
capacitor ESR and self resonant specs all the time. Going to another core
and capacitors may allow you to double the switching frequency and reduce
physical volume 2 times. You will also need better switches for this, and
better RFI suppression, but that is just by the way.

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservKILLspamspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...