Close
0%
0%

Modifying a notebook power supply

Modification of a notebook power supply to output a desired voltage.

Similar projects worth following
This project was part of my Soldering Station project, but I've decided to make it a standalone project, as it's a little more detailed than I intend to detail the rest of the Soldering Station project.

IMPORTANT:

NEVER CONNECT YOUR POWER SUPPLY TO MAINS WHEN MEASURING AND MODIFYING IT.

Wait at least one minute after disconnecting it from mains before handling it again while open.

Test the power supply with a variable voltage power supply.

  • 1 × Variable output power supply Rated to at least the notebook PSU output voltage.
  • 2 × SMD resistors Values may vary with your desired voltage and the power supply reference voltage.

  • Broken image links fix

    Kuro09/13/2016 at 10:03 0 comments

    I wiped my webserver and the images got broken. Uploaded everything to hackaday.io now.

  • Modifying a Notebook PSU (practice)

    Kuro11/27/2014 at 13:28 2 comments

    It's now time to actually modify the notebook PSU. We'll need to open it, identify the feedback circuit and change the reference voltage divider.

    Let's have a look at the open PSU:

    Nice, but what is what on this power supply?

    Well, it's actually pretty easy to identify each part.

    There are two clear divisions on this circuit board, the rubber pads are even marking it, along with silkscreen and areas without components.

    The part closest to the AC power cord is the input (high voltage), the middle part is the control circuit (high voltage) and the part closest to the DC output is the feedback circuit (low voltage).

    Here, take a look at the same image with the subdivisions:

    Now that we've found our feedback circuit, it's time to search for the voltage divider we want to modify.

    Let's take a closer look:

    I've already identified the main parts on the feedback circuit. You can see the positive and negative outputs, the main controller (comparator and reference voltage) and the connection to the control circuit, through the optocoupler.

    A better view of the optocoupler:

    Nice, but how do we identify the voltage divider? There are lots of resistors, how would we know which are the ones we are looking for?

    There's an easy way to identify them. In this page, we can see how to use a variable output power supply to identify the voltage divider. That helps a lot.

    I connect the variable power supply to the power supply I want to modify and measure the voltage at the optocoupler. The voltage drop on the optocoupler is about 1V and the PSU is set for a 19V output. Just when it's about to reach 18V (19V - 1V), the voltage suddenly drops to 3V.

    This is the feedback circuit telling the control circuit to stop charging the output capacitor.

    Having the datasheet of the comparator in hands always helps. I've identified mine to be a TSM103A from ST. It's easy to see where the inputs are.

    Testing pin 3 I've read 2.4V, which is the 2.5V reference. It was not that precise, but well, there it was. Also, when I changed the voltage on the variable power supply, I could see the voltage on pin 2 changing according to it.

    The voltage divider I wanted was connected to pin 2 of the comparator IC.

    Using the continuity test, I found the resistors to be R40 and R41.

    If we compare this voltage divider to the one at the previous post, R40 is R1 and R41 is R2.

    The original values were 15kOhms for R40 and 2.25kOhms for R41. Consider the output voltage at 19V and calculate the divider, you get around 2.48V. That's pretty close to the 2.5V reference, huh.

    I didn't have any good values to change only one resistor of the divider, but I've found that R42 and R43 (unpopulated at the time) were in parallel to R41.

    I've then put two 10kOhms resistors and one 1.5kOhms resistor in parallel for the R41. That's a 1.15kOhms equivalent.

    For the R42, I've put a single 10kOhms resistor.

    Calculating for 24V, the divided voltage would be 2.475V, very close to the 2.5V reference.

    I reconnected the variable power supply and changed the voltage while measuring the output at the optocoupler. When it reached around 23V, the output changed. Feedback circuit working like a charm.

    There was one problem, though: the comparator was heating up.

    There's no problem feeding the TSM103 with 24V, as it's rated up to 36V. The problem was somewhere else.

    The only problem I could think was the current flowing to the comparator through the optocoupler, so I changed the old 330Ohms resistor to a 1kOhms resistor and tested the the circuit again.

    Bingo. No more heating.

    With care and no metal things around, I connected the PSU to the AC power cord and connected it to the mains. No blue magic smoke.

    Grabbed my multimeter and tested the output: 24.0V.

    Disconnected from the mains, put everything back into the case and tested again to make sure.

    It's done, now your power supply outputs the desired voltage.

  • Modifying a Notebook PSU (theory)

    Kuro11/27/2014 at 13:26 0 comments

    This project was part of my Soldering Station project. The idea behind it is modifying a 19V 65W notebook power supply to output 24V.

    I'll try to show in a very simplified way how a switch mode power supply works and how to change the feedback circuit so the power supply outputs the voltage you want.

    How a Switched Mode Power Supply (SMPS) works

    Let's first understand how a switching power supply works. The images below are purely educational. SMPS are much more complex, but this is enough to give a good understanding on the theory of operation.

    For a step down power supply, that is, a power supply which outputs a lower voltage than you have on the input, you have a switch and a capacitor. The name "switching" comes from this switch.

    When you turn on the power supply, the switch is open and the capacitor is discharged, at 0V.

    As you want a specific voltage on the output (low voltage side), the feedback circuit tells the control circuit that the output voltage is below the desired voltage. The control circuit then closes the switch, allowing current to flow from the high voltage side to the low voltage side.

    As the current flows into the capacitor, the output voltage starts to rise. When the feedback circuit senses the output voltage is at the desired value, it tells the control circuit to stop charging the capacitor.

    All good, but the capacitor starts discharging over time, and so the output voltage starts to drop.

    The feedback circuit notices this and, when the voltage is below the threshold, it tells the control circuit so, which closes the switch again.

    These steps keep repeating, maintaining the output voltage at the desired value. The image below is a real SMPS maintaining it's output voltage. Notice how it charges then discharges the capacitor.

    But how does this feedback circuit, which is the one we want to modify, actually works?

    There's no real secret to that. Almost every SMPS feedback circuit works by dividing the output voltage using a voltage divider, then comparing the divided voltage with a reference voltage.

    The voltage divider divides the output voltage by a known constant and feed it to a comparator. The comparator gets the value and compares it to the voltage reference. If the value is higher than the voltage reference, the comparator tells the control circuit to open the switch, as your output voltage is over the desired value. Else, if the value is lower than the voltage reference, the comparator tells the control circuit to keep charging the capacitor, because your output value is lower than the desired value.

    Let's try calculating. You want a 5V output voltage and have a 2.5V voltage reference. Your comparator will tell you if a value is higher or lower than 2.5V.

    You'll need to make 5V looks like 2.5V to your comparator. How would you do this? Using a voltage divider.

    Let's use a R2 as 1kOhm and calculate R1 to make our voltage divider. If R2 is 1kOhm, you have:

    Solve it and you get R1 = 1kOhm.

    You now have a voltage divider which halves your voltage.

    Suppose your output voltage was at 4.5V. You would be comparing the half of 4.5V, 2.25V, with 2.5V. The comparator will notice that 2.25V is lower than 2.5V and tell the control circuit to keep charging your capacitor.

    Now, suppose your output voltage keeps raising and reach 5.5V. The half of it is 2.75V, which is higher than 2.5V. The comparator will now tell the control circuit to stop charging the capacitor.

    This finishes the explanation of how a SMPS works.

    How to change the output voltage of a SMPS

    Now, with a basic understanding of how a SMPS works, there should be no problem modifying one to output our desired voltage.

    What we want to do is to change the voltage divider, changing the voltage at which the comparator tells the control circuit to charge the capacitor.

    This will be explained in the next post.

View all 3 project logs

Enjoy this project?

Share

Discussions

Yarim Berreby wrote 02/11/2019 at 13:06 point

I'm In the process of trying to do it myself, but I can't find the voltage divider, The http://polonai.se/smps/tosh/ link is down, can anyone summarize the method to find it?

  Are you sure? yes | no

o.o-colorado wrote 01/14/2019 at 17:09 point

I had the same problem when trying to get 12V out of a 19V PSU. Here is how to fix it:

My power supply is based on the LTA201P controller (it is actually a slightly modified version of the TEA1533T from Philips). When the output voltage divisor is changed to get 12V, the circuit starts to operate in burst mode at light loads. There are two solutions: 

- Connect a power resistor in parallel with the output in order to always have some load (Not efficient).

- Modify the burst mode threshold: Pin 6 of the LTA201P receives the feedback signal from the secondary side and if there is not enough load at the output, it starts to operate in burst mode to save power. According to the TEA1533T datasheet, this threshold is set by the resistor connected between pin 6 and primary ground (originally 1k in my PSU). Increasing this resistor reduces the threshold. Using a 10k resistor, this value is set to the minimum. Yet, some very small load is required at the output. So, solder a 1W 500 ohm resistor in parallel with the output and you should get a stable 12V power supply.

  Are you sure? yes | no

sale030 wrote 07/15/2018 at 12:18 point

P.S. The oscillating start only with load. Without load, voltage is stable???

  Are you sure? yes | no

sale030 wrote 07/15/2018 at 11:59 point

Hello, I made the modify my SADP-65KB with same comparator (DAS001=TSM103) to 12V for my LED strips, but problem is when I put down the voltage below 12.8V, adapter start to pump (oscillate). Some undervoltage protection, or...??? Please help.

  Are you sure? yes | no

grzegorz wrote 08/12/2017 at 12:17 point

Hello, i have the same problem as users above :-( when i change output voltage to 24V and connect small load, power supply turn's off. I think is DAP8A some protection. Has anyone solved this problem?

  Are you sure? yes | no

Adam Insanoff wrote 07/22/2017 at 13:17 point

I have the same problem like David with modifying 20V Fujitsu Siemens laptop power brick. When nothing connected it is 28V what is modified to be. But when I connect even very small load, it shuts off. Over voltage protection could be a problem with this, but I have not figured out how to bypass it yet.

  Are you sure? yes | no

grzegorz wrote 08/12/2017 at 12:18 point

Has you solved this problem?

  Are you sure? yes | no

David wrote 06/09/2017 at 11:25 point

I managed to modify a power supply which uses exactly the same comparator, the voltage divider in mine was 22k and 14k+4.3k in parallel (3.29k effective). I've replaced the 22k with 27k since I did not have 28k available and the output voltage is a bit above 23V which is enough for me. But as soon as I connect Ts100 soldering iron the voltage drops to 0 until I disconnect the PSU from the mains. Any clue what could be the cause? 

  Are you sure? yes | no

Pedro Serigatto wrote 12/20/2016 at 19:52 point

Quick question! Do you have any ideia what's that iron core transformer doing on the top side of the board? It is laying on the part of the board you described as Feedback Circuit, although it makes little sense since the Feedback uses only DC (with some possible ripple) voltage.

  Are you sure? yes | no

OiD wrote 01/17/2015 at 20:03 point

I've just stumbled on this project while I was doing the same with a Compaq 19V adapter.

I managed to get up to 21.5V before some protection cut the supply. Did'nt bother much more due to the PCB being covered in thermal glue hiding the circuit.

It's a great guide.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates