Reducing voltage of laptop power supply

Well not really unbelievable. Is there reference against which the output voltage is compared to in a switching power supply?

Mikek

There might be. Cheapskates might rely on the mains voltage being stable and predictable enough and effectively divide it down from that.

You can't tell until you've got inside the power supply and looked, but a cheap circuit is more likely to use a bandgap reference at about 1.2V and compare it with a divided down version of the output. If you could find the divider you might be able to bodge that. It's not likely to be a cost effective exercise.

Sticking a 6.8V zener in series with the 19V output might work, but it would need a 25W zener, on a heat sink, so that wouldn't be cost effective either.

Just buy a 12V switching power supply.

The Australian version of Newark lists this for about $US 20. It can deliver up to 5A

I power all sorts of stuff off random laptop bricks, mostly the old-fashioned ones with the 5/2.5 mm coaxial connections.

One good approach for light-duty use is to buy an eBay buck module rated at about three times your expected maximum current. (Chinese amps are on the small size.)

Cheers

Phil Hobbs

Unlikely. There are too many possibilities, and most are not going to be hackable.

An external regulator would work. Details depend.

Or just buy a 12 volt power supply.

The whole of the power supply is designed for providing 19V. If you try to change that to 12V, you would need to go through the design process again, to assure it will work correctly.

To answer your question, there is typically a resistor divider that feeds a portion of the output voltage to the controller chip. This is compared to an internal reference (or it could be a separate reference, but that's not important). To change the output voltage, you change one or both of those resistors. If the circuit has capacitors in parallel with the resistors, make sure you get the data sheet for the controller part and adjust the capacitors if required.

Why not just buy a 12V supply (there's plenty of them around)? If you buy off eBay or Alibaba, be sure to derate the power output by 2 or 3 or 4x.

That is what I do. There are buch and a lot of boost modules, some are buck/boost. You can even get them with a volt meter and amp meter for very little.

I highly doubt that a 19V laptop is going to run at 12V.

You MIGHT get lucky ?

I would bet it won't work.

boB

That's because you are an ignorant idiot.

Not a good bet. The off-the-shelf 12V supply I posted a link to runs from 90V AC to 264V AC, and Toshiba lap-top supply is probably going to be equally flexible.

Pushing out 12V rather than 19V just means running at a different mark-to-space ratio. It shouldn't be a problem.

Some of us do engineering by using design principles and reading data sheets. A lot of people here just like to hack. I think I know which camp you are in.

The simplest solution might be to find someone else with a scrap 12v laptop PSU that wants to swap it for a 19v one! A bit like the lateral thinking solution to height of a tall building using a barometer.

Offer to give the barometer to the janitor if he can tell you how tall the building is!

How near to 12v does it need to be and at what current? You could daisy chain an external regulator onto it.

It's almost 100% certain to be controlled with a shunt feedback regulator like the TL431.

So you can change a single resistor and it will provide 12V instead. Provided that the power supply designer did his job and made sure there is enough phase margin at lower output voltage.

Change RFB1 in linked schematics:

Do not change RFB2 since that is directly used for the feedback loop stability.

Cheers

Klaus

Given that 19V is a *very* common laptop supply voltage, I wouldn't be surprised if they used a fixed 19V supply chip.

no.

If you're lucky the aux winding on the main flyback transformer Produces enough output to keep the controller running when the output is reduced by 35%

Then all you need to do is find the feedback path (usually through an opto-isolator) and modify whatever circuit is driving that to trigger at 12V instead of 19V - it might be a zener diode. or it could be using a TL431 etc...

If you're unlucky the auxillary winding isn't going to provide enough power and then it's not going to work unless you can modify the transformer.

I would. On-chip voltage references aren't all that tightly specified, so you have to have a twiddle mechanism to get exactly 19V out. Once you have that twiddle mechanism you can use the same chip for a range of supply voltages.

I do exclude the National Bureau of Standard voltage reference chip - it does have a very tightly specified output, but it is a stack of Josephson junctions running in liquid helium.

with a built in opto-isolator?

I'm just going to use the 19V to run the heaters, and add a 12V regulator to power the Arduino and the 12V relays I have. I have tested with all 5 heaters in series on 19V (6 Watts) it is not enough heat, so I'll probably go with the 30 Watt configuration using four heat strips. Although if I parallel two and parallel 3, then put those in series, it is 31 watts and I can have a little better heat distribution. ( I have 5 heat strips) Mikek

You will do well to provide better insulation around the bits you are heating. I believe you are heating the space between the buckets, which would be your insulation, but if you are heating it, it ain't insulating. Maybe you need a third bucket, or can wrap the entire thing with corrugated cardboard. That includes a bit of air gap that would be better than nothing. In fact, if the buckets will nest enough, the cardboard could be the insulation between them, with the heating strips between the cardboard and the inner bucket.

You don't need to worry so much about distributing the heat. The heat will flow, from the heaters, to the outside. The inner part will not have so much temperature variation, unless that heat flow is through the part you are trying to heat. If the top and/or bottom are not insulated, then the heat will flow through the core to the top and/or bottom, creating a temperature gradient. So insulate all sides.

Not trying to criticize, but I think I would have started with a box. The heaters could be near the bottom, and all sides insulated, with a door on a hinge. Line with even a thin layer of styrofoam and you can keep it warm with a 7W night light. Do they still sell 7W, incandescent night lights? lol

I'm trying to keep everything food grade. So, the buckets that were food buckets. The heaters are, not food grade are kept from the water, so should not contaminate. I laid some thin copper tape, put the heater on and covered it with another piece of copper tape. I offset the twp pieces of tape to get more area, but I'm not happy with the heater reaching 147*F. I think I may need more heaters (or a different type) to get the required Wattage but a lower temperature. My original plan was to use a 5 gallon bucket and foam in a 3.5 gallon bucket for insulation, but I decided against it as, I didn't see a way, to keep it food safe. Actually, now that I have the two buckets that seem to be a good solution, I could put this all in a foam insulated 5 gallon bucket. Maybe something for the future development. Mikek