SMPS circuit

[StraightView]

Hello everyone, I made three smps circuits copied from diagrams found in the internet like in picture below, but none of them worked, the problems I faced is either no current generated at transformer's output or as soon I connect the main power (220v), a short circuit happen and damaged the transistor or mosfet, I'm putting a single thin wire as a fuse, it gets burn out,
also according to my poor electronic knowledge, when I look to those smps diagrams I see that there's a short circuit of high voltage because the power goes directly to transformer's premary which is connected to transistor's collector and goes through the emitter which is connected directly to the ground! so is this not a short circuit?!!.
I appreciate any help solving this problem, thank you.

 

[betwixt]

As I scrolled down my first thought was this is a far better design although still far from optimal. Then I reached the photo of layout and my heart sank!

The layout is fairly critical, it doesn't have to be perfect but you really should keep the wiring lengths as short as possible. 1cm is a good maximum space between components.

Yes, the heat sinking is required, the ratings of the transistor are at a particular temperature and as it gets warmer, the ratings drop. The idea behind heat sinking is the temperature rise is minimized. Conducting heat into the heat sink doesn't make it vanish, it spreads it out to allow air currents to carry it away more easily.

It will work without the input filter coils but it is advisable to fit them. They serve two purposes:
1. they help to stop spikes on the AC power lines reaching the PSU and
2. they help stop the spikes generated inside the PSU getting back into the AC lines where they would cause radio interference.

Note the winding polarities on the coils, they should share a common core. Because the AC current flows through one on the way to the rectifiers and the other on the return path back to the AC supply, the currents are effectively generating cancelling magnetic fields. The core should be ferrite, pretty much useless at 50 or 60Hz but very effective at higher frequencies so they let the low frequency AC line voltage pass through but cancel out the higher frequencies by producing opposing magnetic fields.

 

[StraightView]

As I scrolled down my first thought was this is a far better design although still far from optimal. Then I reached the photo of layout and my heart sank!

The layout is fairly critical, it doesn't have to be perfect but you really should keep the wiring lengths as short as possible. 1cm is a good maximum space between components.

Yes, the heat sinking is required, the ratings of the transistor are at a particular temperature and as it gets warmer, the ratings drop. The idea behind heat sinking is the temperature rise is minimized.

I didn't know the distance between components mattered, but I'm kind of forced to make the distance between components longer depending on how I'm used to building the circuits which I'm not sure if it will work, because that way I can make changes to the circuits easier, but having said that to me you and the other brother here in this post regarding the necessity of making the distance less than 1cm, I will take that into account from now on.
By the way, should the maximum distance between the components be 1cm only in this type of circuits, or is this related to all electronics circuits?

 

[KlausST]

I didn't know the distance between components mattered

Short distance is a general requirement for (nowadays) electronics design. How can you not know?
And especially SMPS design requires this. What SMPS design tutorials did you read?

I'm kind of forced to make the distance between components longer

What's the idea behind "longer distance"? In decades of electronics design I never heard about such a requirement.
Makes no sense to me.

*****
Another important requirement for SMPS design is a really solid GND plane.

Klaus

 

[StraightView]

Short distance is a general requirement for (nowadays) electronics design. How can you not know?
And especially SMPS design requires this. What SMPS design tutorials did you read?


What's the idea behind "longer distance"? In decades of electronics design I never heard about such a requirement.
Makes no sense to me.

*****
Another important requirement for SMPS design is a really solid GND plane.

Klaus

I didn't know that the distance between the components matters because I never studied electricity/electronics in my life.

Also you don't understand what I said, I make long distance between components when I build the circuits that I have never built before, the circuits that I try to make it work, because with this way I can easily make modifications, change components......, but when the circuit work successfully then I rebuild again but properly.

 

[betwixt]

The reason wiring length is important is because no connection is perfect. You may think the wire linking the components together has the same voltage at both ends but in practice that is far from reality. Wires, and it applies to PCB tracks as well, have two undesirable properties, resistance and inductance.

In a SMPS circuit is it normal to have pulsed high currents flowing between nodes. Imagine say 1 Amp passes down a wire and it has a resistance of 1 Ohm, there will be a voltage drop between one end and the other of (V=I*R) = 1 Volt. The example exaggerates real resistance which will in most cases will be smaller than 1 Ohm but you can see how unanticipated voltage can appear and upset circuit operation. Even tiny voltages where they shouldn't be can cause serious problems.

The inductance of the wire/track is also a problem. Inductance exhibits itself as a barrier to rapidly rising and falling signals reaching their final voltage quickly. SMPS use fast switching on and off of a transistor to 'chop' the current passing through a transformer. They are efficient because the transistor works like a switch, either not passing current or not having any voltage across it, both (W = V*I) meaning power loss is very small. If anything slows down the change from on to off or from off to on, the transistor passes through a linear state where power dissipation can be extremely high.

If you ever move on the high frequency circuits you will quickly learn that even 1cm wires can be wwwaaayyyy too long! Some of the circuits I work on have tuning 'coils' that are just 3mm long straight PCB tracks.

Brian.

 

[danadakk]

Prototyping methods :

Manhattan - https://www.qrpme.com/docs/K7QO Manhattan.pdf This is RF work, a lot of SMPS applicable

manhatten prototyping - Google Search

manhatten prototyping - Google Search

https://www.ebay.com/itm/143826734716 https://www.kanga-products.co.uk/ou...lder-Pads-sheet-of-150-Double-Sided-Pads.html https://www.qsl.net/n5ib/IC Pads - Manhattan Style.pdf

Google "manhattan prototyping" and in browser select images to look at.

Veroboard :

The Lost Art of Strip Board Prototyping

A timeless technique to wire up circuits for prototyping or functional applications.

www.nutsvolts.com

Google "veroboard layout utility", software available to help


Regards, Dana

 

[KlausST]

I didn't know that the distance between the components matters because I never studied electricity/electronics in my life.

Also you don't understand what I said, I make long distance between components when I build the circuits that I have never built before, the circuits that I try to make it work, because with this way I can easily make modifications, change components......, but when the circuit work successfully then I rebuild again but properly.

I did understand what you said.

The thing is: you can´t build an SMPS circuit working reliably with long distance connections.
It may work. Work by accident. This is not what I call "reliably".
It may work now, but may refuse to work the next day. (killing itself by stray inductance high voltage spikes)
It may work now, but not according EMI/EMC regulations. (So it´s not allowed to be operated in public. It may harm other devices - maybe your pacemaker. Or it may refuse to operate correctly when a WiFi is nearby)
It may work, but not with the expected performance.

It´s like saying: I fist build a house of cards, and when it´s rigid enough I build it with concrete.

Klaus