Prev: Filter cap at dc motor input
Next: Charge Conservation - Hint of the Day
From: Daku on 21 Jul 2010 06:45
On Jul 20, 11:30 pm, Tim Wescott <t...(a)seemywebsite.com> wrote:

> Uh, understand how the stupid thing works, and proceed based on that
> understanding? That works well for me in most cases. Randomly fiddling
> with values in SPICE is usually a recipe for disaster.
I do understand the basic working of the circuit.
Also, no random fiddling with SPICE at all - a
discrete component circuit, the power MOSFETs are
driven with PWM, and the high output current driver
was designed by me as well. The input output ratio
is fixed and thus the original question.
> Are you trying to regulate the output, or just get a sorta-fixed 1:3
> output ratio? Are you using a regulator chip, discretes, or re-purposed
> jelly-bean chips (e.g. 555 timer for PWM generation)? Do you have any
> requirements for fixed-frequency, output regulation, EMI limiting, etc.,
> etc.?
>
> --
>
> Tim Wescott
> Wescott Design Serviceshttp://www.wescottdesign.com
>
> Do you need to implement control loops in software?
> "Applied Control Theory for Embedded Systems" was written for you.
> See details athttp://www.wescottdesign.com/actfes/actfes.html

From: Daku on 21 Jul 2010 11:07
On Jul 20, 9:06 pm, "George Jefferson" <phreon...(a)gmail.com> wrote:

> I assume you mean the standard SMPS boost converter? If so and there is a
> sufficient load then it should 1/(1 - D) where D = duty cycle.

One question. The duty cycle you mention here - is it of the
PWM pulses driving the power MOSFETs ?

From: George Jefferson on 21 Jul 2010 12:21


"Daku" <dakupoto(a)gmail.com> wrote in message
news:5e5a8b54-8f11-4869-9b70-0742c1365ed7(a)s17g2000prh.googlegroups.com...
> On Jul 20, 9:06 pm, "George Jefferson" <phreon...(a)gmail.com> wrote:
>
>> I assume you mean the standard SMPS boost converter? If so and there is a
>> sufficient load then it should 1/(1 - D) where D = duty cycle.
>
> One question. The duty cycle you mention here - is it of the
> PWM pulses driving the power MOSFETs ?

Of course! What else could it be?

From: kevin93 on 21 Jul 2010 13:51
On Jul 21, 9:21 am, "George Jefferson" <phreon...(a)gmail.com> wrote:
> "Daku" <dakup...(a)gmail.com> wrote in message
>
> news:5e5a8b54-8f11-4869-9b70-0742c1365ed7(a)s17g2000prh.googlegroups.com...
>
> > On Jul 20, 9:06 pm, "George Jefferson" <phreon...(a)gmail.com> wrote:
>
> >> I assume you mean the standard SMPS boost converter? If so and there is a
> >> sufficient load then it should 1/(1 - D) where D = duty cycle.
>
> > One question. The duty cycle you mention here - is it of the
> > PWM pulses driving the power MOSFETs ?
>
> Of course! What else could it be?

It is a bit more complicated with a boost supply tan a buck because at
low output currents (the meaning of low is mainly dependent on the
inductor value) it will change to discontinuous mode and the voltage
ratio is not then determined by the duty cycle. The output voltage
rises towards infinity as the output current get lower. Hence
feedback is necessary.

kevin
From: Grant on 21 Jul 2010 18:17
On Wed, 21 Jul 2010 10:51:34 -0700 (PDT), kevin93 <kevin(a)whitedigs.com> wrote:

>On Jul 21, 9:21 am, "George Jefferson" <phreon...(a)gmail.com> wrote:
>> "Daku" <dakup...(a)gmail.com> wrote in message
>>
>> news:5e5a8b54-8f11-4869-9b70-0742c1365ed7(a)s17g2000prh.googlegroups.com...
>>
>> > On Jul 20, 9:06 pm, "George Jefferson" <phreon...(a)gmail.com> wrote:
>>
>> >> I assume you mean the standard SMPS boost converter? If so and there is a
>> >> sufficient load then it should 1/(1 - D) where D = duty cycle.
>>
>> > One question. The duty cycle you mention here - is it of the
>> > PWM pulses driving the power MOSFETs ?
>>
>> Of course! What else could it be?
>
>It is a bit more complicated with a boost supply tan a buck because at
>low output currents (the meaning of low is mainly dependent on the
>inductor value) it will change to discontinuous mode and the voltage
>ratio is not then determined by the duty cycle. The output voltage
>rises towards infinity as the output current get lower. Hence
>feedback is necessary.

So what you do is work out the duty cycle needed for worst case
operation, then use feedback to reduce that duty cycle to get the
desired output over a range of conditions -- consider input voltage
range and output current range.

Sometimes a minimum load such as an LED indicator and/or resistor
can set a known low output current range if the load goes to zero
current.

The current limit around the boost switching transistor is for
startup and overload conditions, it is not for continuous output
current control[1]. Also the current sense may require a small RC
to filter the power turn-on transient.

[1] For constant current output, run current sense signal back to
the duty cycle control feedback point.

Hope these additions help OP.

Grant.
First  |  Prev  | 
Pages: 1 2
Prev: Filter cap at dc motor input
Next: Charge Conservation - Hint of the Day