Forums

Question about DC- DC boost converter

Started by Daku October 1, 2011
Could some electronics guru please clarify the following ?
The following is the basic structure of a boost converter:

DC_in -> inductor --> diode --> capacitor --> output
                               |                         |
                               |                         |
                           switch                    |
                               |                         |
                            GND                  GND
The diode is forward biased in the sense that when
the switch is closed, the diode is reverse-biased and
prevents the capacitor from discharging. The question
is:
How is the maximum output voltage set, i.e., how is
maximum output voltage set ? Is it done by the reverse
breakdown voltage of the diode, since once that is
exceeded, the capacitor will try to discharge, when
the switch is on.
In article 
<bb74a697-6fcd-403d-b060-41e0e9381f7a@j20g2000vby.googlegroups.com>,
 Daku <dakupoto@gmail.com> wrote:

> Could some electronics guru please clarify the following ? > The following is the basic structure of a boost converter: > > DC_in -> inductor --> diode --> capacitor --> output > | | > | | > switch | > | | > GND GND > The diode is forward biased in the sense that when > the switch is closed, the diode is reverse-biased and > prevents the capacitor from discharging. The question > is: > How is the maximum output voltage set, i.e., how is > maximum output voltage set ? Is it done by the reverse > breakdown voltage of the diode, since once that is > exceeded, the capacitor will try to discharge, when > the switch is on.
Ain't no guru. In the given example (your basic joule thief circuit, with diode and cap in place of the LED, and not showing the trigger coil and transistor that are the switch in that case), it's not "set", though it will come to a setting depending on the input voltage and the real characteristics of the inductor. Assuming you don't exceed the breakdown of the capacitor or diode, and no current is drawn from the output, you'll land somewhere, and the easiest way to tell where is experiment. Likewise if you do draw current from the output, but the voltage will be lower - as it's not regulated. In a regulated DC/DC converter, there will be feedback from the output side which controls the switch (transistor, normally) to maintain a particular voltage (or in some cases, current) on the output side. This is fairly comprehensive, if slightly pompous, explanation of the Joule Thief that your circuit almost is: http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchCircuits-P1.html This has a better simplified explanation, and many variants if you poke around the blog: http://watsonseblog.blogspot.com/2011/08/2011-aug-01-joule-thief-simplified-1.html -- Cats, coffee, chocolate...vices to live by Please don't feed the trolls. Killfile and ignore them so they will go away.
On Sat, 1 Oct 2011 09:14:41 -0700 (PDT), Daku
<dakupoto@gmail.com> wrote:

>Could some electronics guru please clarify the following ? >The following is the basic structure of a boost converter: > >DC_in -> inductor --> diode --> capacitor --> output > | | > | | > switch | > | | > GND GND >The diode is forward biased in the sense that when >the switch is closed, the diode is reverse-biased and >prevents the capacitor from discharging. The question >is: >How is the maximum output voltage set, i.e., how is >maximum output voltage set ? Is it done by the reverse >breakdown voltage of the diode, since once that is >exceeded, the capacitor will try to discharge, when >the switch is on.
I'm not a guru and not even a designer. So the following is what has helped me, only. I've read some of your other posts and from them I am assured you have LTspice. Have you set this up in LTspice, yet? You should. It supports a voltage-controlled switch, which you can use to try out different switch periods and duty cycles. Make sure you test a variety of cases and see if you can make predictions. You will need to keep in mind some limitations on inductors that aren't air-core (Bmax at low-freq or eddy current losses at high-freq) when considering core area, volt-seconds, turns, and possible gapping or distributed gap materials (ferrite), if you actually try and build something and expect it to work according to some idealized LTspice schematic. There are some excellent PDF files available on the web, if you google for them. Look for 'boost,' and 'continuous,' and 'discontinuous,' and 'switcher,' as some example keywords. And in particular, look at the Wiki that comes up. TI writes a few good things (Unitrode series comes to mind.) http://www.onsemi.com/pub_link/Collateral/SMPSRM-D.PDF http://www.jaycar.com.au/images_uploaded/dcdcconv.pdf http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM401014 http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM401001 http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM405006 http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM406008 http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM408003 http://www.smps.us/Unitrode.html Jon
Daku wrote:

> Could some electronics guru please clarify the following ? > The following is the basic structure of a boost converter: > > DC_in -> inductor --> diode --> capacitor --> output > | | > | | > switch | > | | > GND GND > The diode is forward biased in the sense that when > the switch is closed, the diode is reverse-biased and > prevents the capacitor from discharging. The question > is: > How is the maximum output voltage set, i.e., how is > maximum output voltage set ? Is it done by the reverse > breakdown voltage of the diode, since once that is > exceeded, the capacitor will try to discharge, when > the switch is on.
Lets see, it's saturday, looks basic 101 electronics class work to find the error! Jamie
On Oct 1, 12:57=A0pm, Ecnerwal
<MyNameForw...@ReplaceWithMyVices.Com.invalid> wrote:
> In article > <bb74a697-6fcd-403d-b060-41e0e9381...@j20g2000vby.googlegroups.com>, > > >
> Ain't no guru. > > In the given example (your basic joule thief circuit, with diode and cap =
in
> place of the LED, and not showing the trigger coil and transistor that ar=
e
> the switch in that case), it's not "set", though it will come to a settin=
g
> depending on the input voltage and the real characteristics of the induct=
or.
> Assuming you don't exceed the breakdown of the capacitor or diode, and no > current is drawn from the output, you'll land somewhere, and the easiest > way to tell where is experiment. Likewise if you do draw current from the > output, but the voltage will be lower - as it's not regulated. > > In a regulated DC/DC converter, there will be feedback from the output si=
de
> which controls the switch (transistor, normally) to maintain a particular > voltage (or in some cases, current) on the output side. > > This is fairly comprehensive, if slightly pompous, explanation of the > Joule Thief that your circuit almost is: > > http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchC... > > This has a better simplified explanation, and many variants if you poke > around the blog: > > http://watsonseblog.blogspot.com/2011/08/2011-aug-01-joule-thief-simp... > > -- > Cats, coffee, chocolate...vices to live by > Please don't feed the trolls. Killfile and ignore them so they will go aw=
ay. Thanks for the explanation. I have created a working SPICE model of 5x DC-DC boost converter in the past which uses PWM for switching. The reason I asked is that I am thinking of some modifications.
On Sat, 1 Oct 2011 09:14:41 -0700 (PDT), Daku <dakupoto@gmail.com>
wrote:

>Could some electronics guru please clarify the following ? >The following is the basic structure of a boost converter: > >DC_in -> inductor --> diode --> capacitor --> output > | | > | | > switch | > | | > GND GND >The diode is forward biased in the sense that when >the switch is closed, the diode is reverse-biased and >prevents the capacitor from discharging. The question >is: >How is the maximum output voltage set, i.e., how is >maximum output voltage set ? Is it done by the reverse >breakdown voltage of the diode, since once that is >exceeded, the capacitor will try to discharge, when >the switch is on.
Check the position of your diode. I think you'll find that it needs to be moved. boB
On Oct 1, 5:14=A0pm, Daku <dakup...@gmail.com> wrote:
> Could some electronics guru please clarify the following ? > The following is the basic structure of a boost converter: > > DC_in -> inductor --> diode --> capacitor --> output > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0| =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0| =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0switch =A0 =A0 =A0=
=A0 =A0 =A0 =A0 =A0 =A0 =A0|
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0| =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 GND =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 =A0GND
> The diode is forward biased in the sense that when > the switch is closed, the diode is reverse-biased and > prevents the capacitor from discharging. The question > is: > How is the maximum output voltage set, i.e., how is > maximum output voltage set ? Is it done by the reverse > breakdown voltage of the diode, since once that is > exceeded, the capacitor will try to discharge, when > the switch is on.
If you dont add any further bits than you've drawn, there is no voltage control. Voltage control is achieved by not switching the switch on when sufficient V exists in the capacitor. Note to get adequate V control you need a capacitor that stores far more energy than the coil does. If, for experimental reasons, you were to use a small cap and a big coil, one single switching cycle would put a large voltage increase into the cap, making regulation unachievable. We all had to learn some time ya know. NT
On Oct 2, 2:00=A0am, NT <meow2...@care2.com> wrote:
ischarge, when
> > the switch is on. > > If you dont add any further bits than you've drawn, there is no > voltage control. Voltage control is achieved by not switching the > switch on when sufficient V exists in the capacitor. > > Note to get adequate V control you need a capacitor that stores far > more energy than the coil does. If, for experimental reasons, you were > to use a small cap and a big coil, one single switching cycle would > put a large voltage increase into the cap, making regulation > unachievable. > > We all had to learn some time ya know. > > NT
I am afraid the simple diagram I had in my initial post was lifted from the Maxim Web site. My own design is a lot complicated.
Daku wrote:

> On Oct 1, 12:57 pm, Ecnerwal > <MyNameForw...@ReplaceWithMyVices.Com.invalid> wrote: > >>In article >><bb74a697-6fcd-403d-b060-41e0e9381...@j20g2000vby.googlegroups.com>, >> >> >> > > >>Ain't no guru. >> >>In the given example (your basic joule thief circuit, with diode and cap in >>place of the LED, and not showing the trigger coil and transistor that are >>the switch in that case), it's not "set", though it will come to a setting >>depending on the input voltage and the real characteristics of the inductor. >>Assuming you don't exceed the breakdown of the capacitor or diode, and no >>current is drawn from the output, you'll land somewhere, and the easiest >>way to tell where is experiment. Likewise if you do draw current from the >>output, but the voltage will be lower - as it's not regulated. >> >>In a regulated DC/DC converter, there will be feedback from the output side >>which controls the switch (transistor, normally) to maintain a particular >>voltage (or in some cases, current) on the output side. >> >>This is fairly comprehensive, if slightly pompous, explanation of the >>Joule Thief that your circuit almost is: >> >>http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchC... >> >>This has a better simplified explanation, and many variants if you poke >>around the blog: >> >>http://watsonseblog.blogspot.com/2011/08/2011-aug-01-joule-thief-simp... >> >>-- >>Cats, coffee, chocolate...vices to live by >>Please don't feed the trolls. Killfile and ignore them so they will go away. > > > Thanks for the explanation. I have created a working SPICE model of 5x > DC-DC boost converter in the past which uses PWM for switching. The > reason I asked is that I am thinking of some modifications.
http://www.youtube.com/watch?v=5JSr19x7kco something to look at.. and a few others there. Jamie
Daku wrote:

> On Oct 1, 12:57 pm, Ecnerwal > <MyNameForw...@ReplaceWithMyVices.Com.invalid> wrote: > >>In article >><bb74a697-6fcd-403d-b060-41e0e9381...@j20g2000vby.googlegroups.com>, >> >> >> > > >>Ain't no guru. >> >>In the given example (your basic joule thief circuit, with diode and cap in >>place of the LED, and not showing the trigger coil and transistor that are >>the switch in that case), it's not "set", though it will come to a setting >>depending on the input voltage and the real characteristics of the inductor. >>Assuming you don't exceed the breakdown of the capacitor or diode, and no >>current is drawn from the output, you'll land somewhere, and the easiest >>way to tell where is experiment. Likewise if you do draw current from the >>output, but the voltage will be lower - as it's not regulated. >> >>In a regulated DC/DC converter, there will be feedback from the output side >>which controls the switch (transistor, normally) to maintain a particular >>voltage (or in some cases, current) on the output side. >> >>This is fairly comprehensive, if slightly pompous, explanation of the >>Joule Thief that your circuit almost is: >> >>http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchC... >> >>This has a better simplified explanation, and many variants if you poke >>around the blog: >> >>http://watsonseblog.blogspot.com/2011/08/2011-aug-01-joule-thief-simp... >> >>-- >>Cats, coffee, chocolate...vices to live by >>Please don't feed the trolls. Killfile and ignore them so they will go away. > > > Thanks for the explanation. I have created a working SPICE model of 5x > DC-DC boost converter in the past which uses PWM for switching. The > reason I asked is that I am thinking of some modifications.
Start with this video, this guy seems to be doing a good job explaining things. http://www.youtube.com/watch?v=Gc2eMBbfk7o&feature=related Jamie