Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Linear Vs Switching Regulation.

Status
Not open for further replies.
Hero999 said:
The controller is normally powered from a simple zenner and resistor regulator but if you need more current then build a very small power supply using a small 1.2VA mains transformer, rectifier and smoother capacitor.

Yea, I was going to use a LDO and try to Limit it to 30V, I'll just hook up the unregulated VCC to the mosfet.

According to the data sheet, it requires the use of a PMOS type switch, Im assuming they mean P type MOSFETs right? Would it matter which mode the mosfet is in (depeltion or enhancment mode)

https://www.electro-tech-online.com/custompdfs/2006/12/uc3573.pdf

The book uses this IC, and uses formulas to calculate for certain things, However I dont see any formulas on the data sheet, for example, Whats Cbulk?
 
I know it's a bit late in the game for you, but you might check out "The ARRL handbook for radio communications 2005". It has complete project details to construct a 13.8V 40A switching power supply. It gives circuit description, design details, schematics, parts list with part numbers, pcb pattern (I think you can even order one pre etched)....You get the idea. They even tell you how to construct all the magnetic components and how to test and adjust it. Sorry to ramble on, I just thought you might find it usefull.
 
Overclocked said:
Yea, I was going to use a LDO and try to Limit it to 30V, I'll just hook up the unregulated VCC to the mosfet.
You don't need a LDO if you're reducing the voltage to your IC from 340V to 30V.

Overclocked said:
Im assuming they mean P type MOSFETs right? Would it matter which mode the mosfet is in (depeltion or enhancment mode)
Enhancment mode is normally required and is also the only option for power MOSFETs.
 
Hero999 said:
You don't need a LDO if you're reducing the voltage to your IC from 340V to 30V.

Well I Kinda wanted to Keep efficiency high, if there is a large Vdrop across the Regulator for the IC, then that equals wasted Power. It seems that with a HV regulator (for example, LM317..) can only have a 40V difference between the inputs and outputs. The greatest voltage that might be seen is 130VAC (183V Peak) while the Lowest voltage is 90Vac (127V peak). A 40V Difference would be practically useless for these voltages, So what if I used a Zener to drop the voltage first, Then use a regulator?

The greatest current draw the chip consumes is 12mA, So At most I would need a 20mA Supply for the IC. In the book they show a transistor with a zener and a resistor as a regulator, I might use that (as someone suggested) instead of a prebuilt.

Also, The IC uses a Current Sensing Resistor in series with the MOSFET, Does this resistor need to be a power Resistor? The book does not say which type to use. In the book it gives a part number, so I looked it up

https://www.newark.com/jsp/Passives...2010+.020<++1+/displayProduct.jsp?sku=20C1711

I guess that doesnt Count as a Power Resistor? Then again, the author is using 3Amps for this calcs, but Im using 10Amps (err 14 Amps Peak or so). I guess it wouldnt hurt to use one.
 
It doesn't make any differenct if it's a LDO regulator or not. Low drop out just means it will regulate properly when the input voltage is very close to the output voltage dropping 130V to 40V still wastes 90V and will result in the same power dissipation at a given current.

The IC you've selected isn't suitable for this application which is a bad idea anyway because the output won't be isolated from the mains input.

Mains powerd SMPs use a ferrite isolation transformer and an optoisolator or pulse transformer for the feedback loop. A special controller is used which maybe powered from 5V zenner when it's started then powered from the big ferrite transformer onece it's running.
 
Hero999 said:
The IC you've selected isn't suitable for this application which is a bad idea anyway because the output won't be isolated from the mains input.

Mains powerd SMPs use a ferrite isolation transformer and an optoisolator or pulse transformer for the feedback loop. A special controller is used which maybe powered from 5V zenner when it's started then powered from the big ferrite transformer onece it's running.

Whats the other reason this IC is bad for this application?
 
It's designed to be used as a buck step-down DC-DC converter using a single inductor. An isolated mains SMPs needs a controller that will work with a ferrite transfromer in either push-pull or flyback configureation.
 
Hero999 said:
It's designed to be used as a buck step-down DC-DC converter using a single inductor. An isolated mains SMPs needs a controller that will work with a ferrite transfromer in either push-pull or flyback configureation.

So then What if I used a Transformer (ie, a 12.9V 10Amp) and then used the switcher to regulate it? (just an Idea)

Or I could go with this IC as you said (however this one doesnt have MOSFET drivers..)
https://www.electro-tech-online.com/custompdfs/2006/12/ucc25705.pdf

The only problem is the transformer. I could get it custom made
 
Overclocked said:
So then What if I used a Transformer (ie, a 12.9V 10Amp) and then used the switcher to regulate it? (just an Idea)
That would work but you've still got the big and bulky low frequency transformer and huge filter capacitors.

Or I could go with this IC as you said (however this one doesnt have MOSFET drivers..)
https://www.electro-tech-online.com/custompdfs/2006/12/ucc25705-1.pdf
That will do the trick.

The only problem is the transformer. I could get it custom made
That's true but it shouldn't be too hard to make, just use plenty of insulation between the primary and secondary windings.

Have you considered modifying a n old PC PSU or even rebuilding it to your specification?

If you can find the datasheet for the controller then you'll have nearly all the parts you'll need at a lower price than you can buy them separately.
 
Hero999 said:
That would work but you've still got the big and bulky low frequency transformer and huge filter capacitors.


That will do the trick.


That's true but it shouldn't be too hard to make, just use plenty of insulation between the primary and secondary windings.

Have you considered modifying a n old PC PSU or even rebuilding it to your specification?

If you can find the datasheet for the controller then you'll have nearly all the parts you'll need at a lower price than you can buy them separately.

You know, I do have an Old PSU I was using as a reference to part Placement. The only problem is there are MOSFETs (or some other device) That has a part number but I cant find a data sheet. I did however find the Data sheet for the controller IC.

I looked in the book and they have a design using a transformer, It doesnt seem to hard to design. Which ever design I choose (Ie, Isolated) I will still need the EMI Filter along with the bridge rectifier. I see this circuit layout in ALL SMPS's Ive taken apart.

Thanks.
 
I wouldn't worry about finding a datasheet for the MOSFETs, they'll have a high enough power and voltage rating and it's easy to find out the pin out. The source-drain connection should read about 0.6V to 0.7V using the diode test function on a multimeter and the gate should read open circuit from all other conections in both directions.

Just out of interest what controller did for find in your SMPS?
 
Hero999 said:
I wouldn't worry about finding a datasheet for the MOSFETs, they'll have a high enough power and voltage rating and it's easy to find out the pin out. The source-drain connection should read about 0.6V to 0.7V using the diode test function on a multimeter and the gate should read open circuit from all other conections in both directions.

Just out of interest what controller did for find in your SMPS?

Sorry I didnt get back to you about this right away. The PSU I have uses TL494CN 16pin DIP as the controller. If I could figure out the Freq. at which it ran then I could probably use the transformer (there are 3, one big one and 2 Smaller Ones.

The PSU was made by Raidmax (as marked on the Transformers-Probably indicating they are custom made..). I could go take apart a working one and measure the freq. across the Transformer or the Transistor (or mosfet). It seems the Mosfets are also Isolated from the actual switching signal, theres a optoisolater between the board (Intrestingly, the board is Seperated. Theres the AC side and then the actual switching side. Inbetween there is Just Empty space). Youd figure they would make use of that space, although it may have been done to keep EMI down.
 
Where are the smaller transformers?

They might be mains filters but it's more likely they're pulse transformers used to driver either the MOSFETs or as a feeback for the low voltage side but an opto isolator is probably more likely for the latter.

It's probably a good idea to find out the frequency of the transformer but you still might want to rewind it because you might want a different output voltage and you might not require all the other outputs. Even if you end up rewinding it, I strongly advise keeping a record of the number of turns, wire gauge and phasing.
 
Hero999 said:
Where are the smaller transformers?

They might be mains filters but it's more likely they're pulse transformers used to driver either the MOSFETs or as a feeback for the low voltage side but an opto isolator is probably more likely for the latter.

It's probably a good idea to find out the frequency of the transformer but you still might want to rewind it because you might want a different output voltage and you might not require all the other outputs. Even if you end up rewinding it, I strongly advise keeping a record of the number of turns, wire gauge and phasing.

The big one and smaller ones are in between the 2 heatsinks where the fan blows in.

The main filter...is well near the main.. (EMI filter type Toroid). Im thinking if I could find the indutance of the EMI filter, I could also find the Freq of the Osc (ie, In my design of 100Khz, I need a Inductance of 900uH in the EMI filter) So then it maybe the drive for something else...

Now that Ive read how many turns are actually Needed, Winding a transformer doesnt seem to hard. Its just Id have to keep the windings Tight.
 
You can't work out the switching frequency from the mains filter, you need a frequency counter or oscilloscope to monitor the oscillator.

Also I forgot to mention that the primary is likely to be a bi-filar winding; watch out for that.
 
Hero999 said:
You can't work out the switching frequency from the mains filter, you need a frequency counter or oscilloscope to monitor the oscillator.

Also I forgot to mention that the primary is likely to be a bi-filar winding; watch out for that.

Yea, I noticed theres this big thick wire coming from the transformer. Its a bunch of wires soldered together.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top