I have a HP omen laptop that operates on a 200w 19.5v power supply. I am currently lucky enough to live a nomadic life in a solar powered travel trailer and try to run all my devices on DC rather than use my inverter.

I have been unable to find a commercially made DC-DC power supply for the laptop so I have therefore tried to create my own. It was not difficult to set up a separate DC-DC power supply to output 19.5v at the required amperage, but because the laptop uses a third communication wire to identify the psu, it will not charge.

Please correct me if I am wrong, but it appears it would be difficult to trick the sensing wire into thinking my homemade power supply was a good fit for the laptop.

I am therefore thinking of wiring my diy 19.5v power supply in parallel to the output of the original power supply with the sensing wire still connected. In this configuration, I would NOT plug in the AC side off the original psu.

I am hoping that, if the DC side of the original power supply receives power injected from the output, it will power up the sensing wire tricking my laptop to using my DC-DC power supply.

Could this experiment end up ruining my original AC adapter?

Is there a chance it could work?

Anyone have any better solutions?

It is frustrating because the laptop clearly does not need 200w to charge but yet won’t charge unless it has the oversized PSU which is only necessary when the GPU works hard. I would be satisfied if I could just charge it via DC-DC then need to resort to AC and my inverter off I need full power.

Thanks in advance for any thoughts.

 

Better safe than sorry......if you are going to do that, use a pair of schottky diodes. Each anode to each supply, both cathodes connected together and to the computer's positive power input.

 

Edit: You can use the STPS30L30CT. It is a 30 volt, 15 amp, common cathode dual diode in a TO220 package.
You will have to add a heatsink to it.

 

Since he is planning on his supply back feeding the communication circuits in the original supply, the diodes would defeat the purpose.

I doubt that it would work though, the logic for the communication protocol might be run from a different secondary on the transformer.

And there is definitely a risk that the supply could be damaged. Wouldn’t do it without a schematic.

 

I am hoping that, if the DC side of the original power supply receives power injected from the output, it will power up the sensing wire tricking my laptop to using my DC-DC power supply.

It's likely the PSU sends out a signal along that line. Probably 5V or maybe 3.3V DC. To be sure - I have no idea whether yours works that way or not. But simply back feeding the PSU from your home made supply will probably do nothing.

Is there a chance it could work?

There MIGHT be a chance it would work. However, if the PSU is sending a lower Voltage on that third line and you back power it with a full 19V you could blow it out.

The approach I'd take is to strip the insulation off the PSU wires where they won't short to each other and see what voltages I find there. Then I'd have an idea where to go next. V+ & V- and Sense (at whatever voltage). Matching those voltages might work.

I said "MIGHT!" If you blow it up because of my advice - sorry - I make no claims to being an expert on this subject.

 

Since he is planning on his supply back feeding the communication circuits in the original supply, the diodes would defeat the purpose.

I doubt that it would work though, the logic for the communication protocol might be run from a different secondary on the transformer.

And there is definitely a risk that the supply could be damaged. Wouldn’t do it without a schematic.

Essentially the same thing I'm saying; you just beat me to it.

 

Thanks for all of the replies.

The third/sense wire has around 18v between it and the DC negative. Other folks on another forum suggested that by using resistors between the positive and the third wire to drop the voltage to something similar it may fool the computer to charge. I will try that and report back if it works.

 

An option that is guaranteed to work would be to feed the charger brick with 90V or more DC, assuming it is specified to work with that voltage like many laptop bricks I have seen. But I am not sure you gain much over using a AC. inverter.