Hi,
For some kind of motor and uC project, I want to modify general wallwart SMPS (to learn the SMPS too). I am planning to rewind ferrite transformer with thicker wire.
- Will use just little bit thicker wire as primary, because more thick has low resistance and might fry driver transistor.
-Maybe no need to modify feedback winding.
-Will use as thick as possible as a sencondary winding. Will wind little more turns to get around 8V. (For the motor and uC project there will be a 7805 as a supply entry.)

* will bypass or something else if there is a 5v limit feedback system with optocoupler/TL431.

Then can I double the current? Maybe TO92 driver transistor might not drive modified load, so maybe a medium power driver is needed. Please!

 

As it stands, 5 volts times 0.7 amps = 3.5 watts. Therefore, 3.5 watts divided by 8 volts = 437mA. Three times the power will get you 1.3 amps. You're looking at upgrading this thing by a factor of four. That's a long way to go with a small wall wart (WW). If I could make a suggestion; find a WW that is rated at or above your needs. Above would be better because working AT the rated output means you're going to be pushing it to its limits every time you run your project. Exceed your needs by at least 50%. Some would say double, but in general terms of engineering, 1.5 times the anticipated load is a pretty good factor.

I modified a 12 volt supply to put out 13.8 volts. Sacrificed the power output a little, but the supply I used was rated for 3 amps. I needed only 1.8 amps. Using my design rule, I'd be needing a supply capable of 2.7 amps. 36 watts divided by 13.8 volts gives me 2.6 amps output. Pretty close to what I planned on. 2.6 available amps drawing 1.8 amps, I'm almost at 150%. Actually, 144%.

Your project, as per the 1.5 amp requirement means (by my engineering factor) you should be looking for 2.25 amps capability. That's over 5 times bigger than what you have now. Remember, we're no longer dealing with 5 volts, we're up to 8 volts. And given that the original 5 volt supply is only rated for 700 mA, a factor of 5 times upgrade seems reasonable to me.

In the commercial world of electronics, standard rating is for 33% over rated daily use. That means that a power supply capable of delivering 100 watts would be classified as being rated for 77 watts. The same supply, if used in military, medical or automotive would be rated as 66 watts. Remember, this is the same supply. Just different ratings to reflect whether it's commercial grade or military grade. If the military wanted a power supply capable of delivering 100 watts they'd design one capable of delivering 150 watts.

Anyway, good night. I'm done for now. Will check back when there are other responses. Some may point out flaws in my reasoning. That's always welcome. If I don't learn from my mistakes then I don't learn.

 

That's not going to work. The transformer will be sized for the designed power output. It is not going to supply four times that power. The same will apply to the power semiconductors and their heatsinking.

 

Exactly what Albert said!

Add to that the fact that safety depends critically on the mechanical design of the windings. Even if everything could handle the power, this is not something someone who does not clearly understand the safety issues and how to manage them should be attempting.

 

What you want to do is like trying to modify a lawnmower engine to drive a family car.
Just go and find an old supply of sufficient rating, or buy a power supply, like a 19V laptop supply, and add a buck converter, like the $1 adjustable boards available on Ebay or Amazon.
https://www.ebay.com.au/itm/New-Min...C-Converter-Input-7V-28V-Output-/282516886062

This one is preset but other versions have a trim pot in place of the resistor top left in this picture.
They work very well. Have a look and see what you can find. Old 19V laptop supplies are readily available.
You could also use a 12V at over 1A plugpac to get what you want with one of these.

 

In simple terms, as has been mentioned numerous times, you cannot get more power out of a device than goes in. We sometimes refer to that concept as "overunity" and it doesn't work. The power you get out of a device will ALWAYS be less than the power in, sometimes it will be much less than the power it.

 

OK, and what happens if I used powerful driver transistor and little bigger transformer on same wallwart circuit?

 

......like the $1 adjustable boards available on Ebay or Amazon.
https://www.ebay.com.au/itm/New-Min...C-Converter-Input-7V-28V-Output-/282516886062
View attachment 152377

I cannot buy these mind blowing cheap stuff from eBay, Amazon, Aliexpress...

 

You are right, if you use higher powered parts you can make a power supply that provides more power.

 

You are right, if you use higher powered parts you can make a power supply that provides more power.

And bigger capacitors on the input and output, changes to the feedback circuit, a revised snubber network and possibly bigger output rectifiers. About the only remaining unchanged part is the input rectifier. Not so much modifying the original as building a new one from scratch.

 

Look in junk shops or for old power supplies that have been tossed out. If you can find a 9V 2A one and add a series diode or two on the output to drop the output volts, that is one way. Also, a 12V switch mode supply of a couple of amps and adjust the output volts via changing the voltage sensing resistive divider is another way. You are fighting a battle to raise the power of a supply so start with a 15 or more watt one.

 

add a series diode or two on the output to drop the output volts

I've tried that in the past using 1N4007's and they really didn't drop much voltage at all. I HAVE modified power supplies to drop the voltage output and I've gone the other way, increasing the voltage. But voltage is only half of the equation. What matters is power. Power (watts) is the product of multiplying the voltage by the current. You can modify a 5 volt 1 amp supply to give 9 volts and 0.556 amps. The power is the same. As you up the voltage you drop the amperage. However, components are rated in what they can handle. So you can't just willy-nilly change the voltage. Capacitors are a prime example. A 470µF capacitor rated for 16 volts will explode if you push 24 volts through it. So there's a whole lot more in the design and modification that needs to be considered. Albert Hall mentioned this first (I believe). You're always better off starting high and dropping voltage to achieve your goal. But yes, if you're thinking "What about Buck converters?" Yes, they do up the voltage. But they're designed to handle certain inputs and produce certain (maximum) outputs.

Another consideration is the current (amps). Since mentioning power (watts), resistors can be damaged when you exceed their rating. If you're dropping a voltage on a supply you're upping the current. In some cases it's possible to cook a resistor simply by drawing more watts than it can handle. Since 1 volt at 5 amps is the same as 5 volts at 1 amp, power wise, the same resistor should be OK in either case. 5 watts is a lot of power through a resistor. Which is why they design resistors that are rated to handle 5 watts. However, if you start modifying both volts and amps, you're changing the wattage in drastic ways. A bigger transformer capable of higher amperage and greater voltage translates to more heat. Possibly more heat than the circuit can handle. 5 volts at 700 mA is 3.5 watts. 8 volts at 1.5 amps is 12 watts. That's four times more heat than originally designed. You're talking about changing the transformer. But you're also going to have to change the resistors as well. There's no magic to it. It's math. And capacity. If you design an engine capable of pushing a car at 140 MPH and put it in a go-cart - you're looking for trouble. And you'll find it rather quickly. Not enough loading on the engine and it will over-rev. Parts will start flying off. Your power supply will do the same. Parts will be flying. And frying. That's why people here are telling you to start over with a supply capable of more than you need. It's easy to downgrade. Upgrading requires a lot more thought. And math.

 

OK, and what happens if I used powerful driver transistor and little bigger transformer on same wallwart circuit?

Printing the Darwin award certificate as we speak...………………

SMPSU design is not trivial - unless you want to make a loud bang.

Try junk shops - some appliances like camcorders use 8V power bricks.

 

Loud bang gets my vote. Be sure to film it.

 

-Will use as thick as possible as a sencondary winding. Will wind little more turns to get around 8V. (For the motor and uC project there will be a 7805 as a supply entry.)

Looking back with a more critical eye - you want to boost a 5 volt supply to 8 volts to run a motor and use a 5 volt regulator to run a microprocessor. OK, just think about that for a minute. No - really! You're going to boost the output so you can regulate it back down to where it started from just so you could run an 8 volt motor? Why not just get a 5 volt motor? But then again, a 700 mA supply isn't going to run a 1500 mW motor and a µC.

You COULD use the 5 volt wall wart for the µC and use an additional wall wart from a cordless phone. The ones I've held in my hand have been 9 volt outputs. Not sure if they'll drive 1.5 amps of current though. OR OR OR - - - if you really want - go to Xfinity (in the US) and ask for a used power supply from a returned DVR. They're 12 volts and push 5 amps. You can EASILY use a 7805 to regulate your power for the µC and have plenty of power for your motor. You could even get a 12 volt motor and REALLY spit stuff.

Well, it's your project. Do as you wish. Just let us know how it turns out. And like I said, be sure to shoot the video for us. We love a good laugh.

 

Loud bang gets my vote. Be sure to film it.

Needs a duration timer so it isn't a several Tb file because they're too dead to hit the stop button.