Can I build an AC/DC SMPS in 1 month?

[KEIN NEIN]

I need to finish to projecs in one month:

One that has a transistor in it
And one that uses arduino

I was planning to build an SMPS + DC load and osciloscope but having seen I don't have a lot of time, should I continue with the SMPS or should I just build the load (which also has a transistor)

Can you help me select the right parts?

Thanks in advance

[windsmurf]

 

[strawberry]

Simplest SMPS with built-in protection ...
To build it follow instructions in design sheet
http://www.datasheetcatalog.com/datasheets_pdf/R/D/6/RD6.shtml
Dont forget transfomer GAP size or AL(nH/turns2) found in sheet
parts available on Mouser, Digikey ...

[KEIN NEIN]

I see, thanks

But what if I wanted to make a higher power SMPS, could I just use higher rating parts?

Or would I have to opt for a difference topology

[KEIN NEIN]

Is it worth it to do something like a quasi reasonant psu just for efficiency reasons?

[schmitt trigger]

My exact feelings.

What does a SMPS+DC Load has to do with two assignments, one which uses an Arduino, and the other that uses a transistor.?

[amyk]

My exact feelings.

What does a SMPS+DC Load has to do with two assignments, one which uses an Arduino, and the other that uses a transistor.?

Indeed. I don't know what to be more disappointed with, the OP for not giving us more details, or the institution that's making these vague and not very sensible project demands.

To answer the (not that much more, but still) more specific question in the title of the thread: yes.

[KEIN NEIN]

The specific requirements of the project weren't very clear.

There are two different projects for two different subjects

One is for semiconductors and the requirement is to show a project that makes use of one or more transistors and we need to explain the role of those transistors.

And the other project is for visual programming, it consists of pretty much any project that takes advantage of arduino and we are using labview for programming

now if you want my particular preferences i'd rather the SMPS were efficient and capable of outputting more than 10A through various regulated 5V USB outputs (something like a power strip but with many USB plugs)

But I don't know how much price will increase and how viable it is

As for the arduino load if possible it should be able to stress high wattage ATX PSUS (so high current adjustable load) even if I have to tie a huge fan to it

[Brumby]

One is for semiconductors and the requirement is to show a project that makes use of one or more transistors and we need to explain the role of those transistors.

What level are we talking about here?  High school beginner or EE degree?

If it's High School level, then I'd probably suggest something like a good, old fashioned two transistor multivibrator or a Colpitts oscillator.  A SMPS seems a bit ambitious - and we have no idea of your experience with mains power, so safety is also a question in our minds.


If I've taken a "low level" view of the project requirements, then that is based on my assessment on the little information you have presented.  For your own sake, you need to be a little more informative for us to be better guided in how to assist - otherwise we are guessing.

[ejeffrey]

No, an AC (mains powered) to DC smps, especially a relatively high current one is not a suitable project for your apparent level of skill and experience.  Mains operated gear has significant safety issues and several additional issues with isolation. 

A simple non isolated DC to DC converter is.  A little transistor powered joule thief would be a simple example.  A moderate low voltage DC to DC converter of any of the standard types would also be reasonable project as would a discreet class D amplifier which is similar to a modulatable DC-DC converter.  It depends on the class level.  But my advice is to go on the simple side and make sure you understand it really well.  A joule thief presented by someone who understands the operation completely will get higher marks than someone who submits a 100 W D class audio power amplifier based on a design they found on the internet with no detailed analysis.  At least they would if it were my class.

[mariush]

I think you're pointlessly complicating your life.

If you have to make a project with ONE transistor why make a SMPS? It's like making a plane because you must do something with a wheel.

Here's a project idea: use transistor as on/off switch.

For example make a device to dry hands that's turned on without touch.
Have an arduino, an infrared led and an infrared sensor... when hand goes under fan area, it breaks the invisible infrared and your arduino then turns on a 80/92/120/140mm fan powered from 12v using a transistor.
Transistor lets you turn on a 12v device using 5v of arduino.

Or instead of fan , think of it as alarm... have hall effect sensor on door edge, magnet on door... if door opens magnet goes away so sensor reports that and arduino reads that and plays alarm on piezo speaker using transistor.

could use a 555 to generate sound but arduino and pwm would be easier

see

https://www.murata.com/en-sg/support/faqs/products/sound/sounder/char/sch0001
https://electronics.stackexchange.com/questions/15629/driving-piezo-buzzer-from-mcu-pin
http://hades.mech.northwestern.edu/index.php/Driving_a_piezo_speaker_with_a_PIC
https://www.eevblog.com/forum/projects/how-to-connect-a-piezo-speaker-to-a-microcontroller/

Kill two rabbits with one project.

[KEIN NEIN]

Electric engineering degree, and yes I'm aware of the risks involved when working with mains voltage

[KEIN NEIN]

I guess I'll come back whenever I've made up my mind, I have to read the PSU cookbook but as of right now I believe I'll opt for a flyback PSU bc I believe it can power a few PSU devices and it's not so complex in design

Correct me if I'm wrong

Thanks anyway

[KEIN NEIN]

I am aware I am overcomplicating otherwise I wouldn't plan on building an SMPS don't you think so?

[TimNJ]

I would not rush working on a mains switch-mode power supply. Seriously, even though you may have some ideas about the dangers of working with mains, if you have never really worked on something like this before, now is not the time to rush it. There are lots of things to be cognizant of, from both design and test standpoints. Doesn't mean you shouldn't do it in the future, of course.

[Brumby]

Just a word about "being aware" of the risks of working with mains voltages.....

I've been "working with" mains voltages for more years than I care to admit, but this has been with fairly simple situations - mains wiring, switches, fuses, transformers and such.  When there is a problem, these things can often be successfully diagnosed without power applied - and when power is applied, the measurements required are often simple and few.

Diagnosing an SMPS is more likely to be more involved.  You will be making more measurements in a more confined space with more places for probes to slip.

I approach SMPS diagnosis with an order of magnitude greater care than "regular" mains circuitry.  Not doing much SMPS troubleshooting will influence my caution, but it's still worth being more attentive.

[KEIN NEIN]

understood, in that case could you tell me how to build an arduino DC load or osciloscope capable of drawing <10A (<50A if possible)

I have watched two videos already

[KEIN NEIN]

or a buck converter maybe?

[pigrew]

or a buck converter maybe?

My suggestion would be to build a 5V to 3.3V buck converter, using Arduino to implement the PWM & feedback loop. You would need a few capacitors, resistors, an inductor, and diode (maybe a second MOSFET, too). Perhaps use a pMOSFET as the pass transistor (but a BJT could work, too).

-Nathan

[KEIN NEIN]

Aside from mouser electronics and digikey which other stores do you recommend? (from mexico ofc)

[KEIN NEIN]

or a buck converter maybe?

My suggestion would be to build a 5V to 3.3V buck converter, using Arduino to implement the PWM & feedback loop. You would need a few capacitors, resistors, an inductor, and diode (maybe a second MOSFET, too). Perhaps use a pMOSFET as the pass transistor (but a BJT could work, too).

-Nathan

thank you

Modern buck converters use two MOSFETs for syncronous rectification right?
I didn't know I could do that with arduino, where can I find more info on that?

[pigrew]

Modern buck converters use two MOSFETs for syncronous rectification right?
I didn't know I could do that with arduino, where can I find more info on that?

Yes, synchronous rectifiers use a second MOSFET to improve efficiency (to avoid diode losses). You can just look at the wiki page for buck converters for help. Building a synchronous rectifier is difficult, and I would suggest making one using a Schottky diode first.

I've never used Arduino (except for my 3D printer), so I don't have any good references for you to look at. However, you'll be needing to make 50~200 kHz PWM, which is certainly within the capabilities of most microcontrollers.  Getting the feedback loop right may be tricky. I'd suggest starting with doing analog simulations of the inductor+transistor to see how they respond.

[KEIN NEIN]

Understood, let me see