Isolator and PWM Control in SMPS (and maybe more assorted questions)

[Rulycat]

There's a lot I don't understand about power supply design.

I was looking on Hardware Secrets at this page, to do with SMPS:
http://www.hardwaresecrets.com/artic...Supplies/327/7

Am I correct in the knowledge that the PWM provides pulses of voltage to the MOSFET, switching it on SOME of the time to reduce the average voltage of, say, 12V, to something lower like 5V (after the caps smooth the pulses of 12V)?
And the voltage output is compared against the reference voltage to change how often the PWM chip switches the MOSFET?

What's the isolator for? Is it to prevent the output voltage directly reaching the PWM chip and frying it?

After next week going to be purchasing this book:
http://www.amazon.co.uk/Switching-Po...dp_ob_title_bk

I've not read either edition, so should be interesting.

Thanks for any help, as always.

[ben7]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

There's a lot I don't understand about power supply design.

I was looking on Hardware Secrets at this page, to do with SMPS:
http://www.hardwaresecrets.com/artic...Supplies/327/7

Am I correct in the knowledge that the PWM provides pulses of voltage to the MOSFET, switching it on SOME of the time to reduce the average voltage of, say, 12V, to something lower like 5V (after the caps smooth the pulses of 12V)?
And the voltage output is compared against the reference voltage to change how often the PWM chip switches the MOSFET?

What's the isolator for? Is it to prevent the output voltage directly reaching the PWM chip and frying it?

After next week going to be purchasing this book:
http://www.amazon.co.uk/Switching-Po...dp_ob_title_bk

I've not read either edition, so should be interesting.

Thanks for any help, as always.

An SMPS uses a high frequency transformer which is smaller and much more efficient than an old iron-core transformer. Also, as you mentioned, it uses PWM which decreases low-load losses even more. (whereas an iron core transformer will be hot and constantly drawing power with no load)The isolater is to help prevent any high voltage from shocking the user if there is a power surge, or something goes wrong in the circuits.

Generally iron core transformers are about 50-60% efficient, SMPS transformers can be 80-90% efficient if designed correctly.

[budm]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

The opto is used for isolation between the Primary AC side and the Isolated DC side of the circuits to prevent electrical shock when the user make contact with the isolated DC side. The SMPS in the primary side looks at the output voltage of the secondary side and adjust its pulse width accordingly, the feedback for this action is coupled through the opto, the output voltage is compared with the ref voltage (TL431 is the most commonly used for the ref) and it will determine how hard to turn on the LED inside the opto which is in the Isolated DC side, the photo-transistor on the receiving side of the opto will conduct more or less based on how much the LED is turn on (it is called current transfer ratio, if you look up the spec of the opto). The opto can withstand up to 2500v before insulation breakdown that will cause the conduction of the current between the AC primary and Isolation secondary side. The secondary side circuit ground is also tied to the chassis, the AC safety ground is also tied to chassis ground to divert any fault current to safety ground.

[Th3_uN1Qu3]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

Generally iron core transformers are about 50-60% efficient, SMPS transformers can be 80-90% efficient if designed correctly.

Linear regulators are typically 30 to 60% efficient. Transformers, be they iron core, or high frequency ferrite (as used in SMPS) are typically 90 to 99% efficient. Regular mains transformers have to be big because of the low frequency needing a large number of turns, and since there are a lot of turns hence a very long wire, then the wire also has to be heavy gauge to achieve low resistance that can carry the current without burning up (remember Ohm's law: P=I^2*R). This alone does NOT forbid the efficiency of a mains transformer to be below 90%, because if you use enough copper the efficiency will be high - the way in which it is manufactured dictates it instead.

Because iron core transformers are so big, and thus have surface area to dissipate quite a bit of heat, some manufacturers run cores into saturation, or use wire that is too thin, and the losses increase. Microwave oven transformers are the worst offenders. But the transformers found in PA amplifiers for example, are wound to be quite efficient.

With the use of high frequency transformers like in SMPS, two problems get solved simultaneously. The numbers of turns is drastically reduced, so that a shorter wire is required to do the same voltage transformation. Hence the wire resistance also reduces (since the length is reduced), thus the wire can also be made thinner and still be able to carry the same current it would in a regular mains transformer.

[Rulycat]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

So is it the switching of the MOSFETs which gives the high Hz?

Linear power supplies are easy for me to understand - there's just something about switchers which confuses me.

[ben7]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

So is it the switching of the MOSFETs which gives the high Hz?

Linear power supplies are easy for me to understand - there's just something about switchers which confuses me.

Yes.

Its feeding the transformer with a high frequency signal. It regulates the output voltage by changing the width of the pulses, which lets more or less energy go into the transformer.

[budm]

Re: Isolator and PWM Control in SMPS (and maybe more assorted questions)

Another benefit for running at high frequency is smaller capacitance for filter caps.