simulate this circuit – Schematic created using CircuitLab

I have a device that enables a ground connection when I press a key on a remote control. This negative signal used to control other devices. But It has about 1k ohms resistance. When nothing is pressed, it provides infinite resistance (no ground). I cannot change this aspect of the circuit.

The problem is that the impedance of this signal is about 1k ohms, so this ground signal does not provide adequate current when used by the other device. It provides only 15ma, that is not sufficient for the second device to activate (150ma).

Both sides of this are correctly powered (12v) and working fine. So how can I provide a sufficiently good ground connection to the other device, using this restricted negative signal?

  • 2
    \$\begingroup\$ Welcome to SE EE, your question is a bit vague (at least to me), to improve it I suggest to add a schematic. What is a "negative connection", perhaps you mean "low side switching"? \$\endgroup\$ May 21, 2019 at 12:35
  • \$\begingroup\$ just added a schematic. by negative connection I mean ground connection \$\endgroup\$ May 21, 2019 at 12:55
  • \$\begingroup\$ just added a schematic Hmm, a resistor and a ground symbol, how complex. You mention 12 V, it is not in the schematic. You mention a relay which is not in the schematic. You mention a load which is not in the schematic. You will have to try a bit harder than this. \$\endgroup\$ May 21, 2019 at 13:00
  • \$\begingroup\$ sorry I`m new to eletronics, will improve the schematic \$\endgroup\$ May 21, 2019 at 13:02

2 Answers 2



simulate this circuit – Schematic created using CircuitLab

The "your controller" box on the left represents your "device that enables a ground connection when I press a key on a remote control", I drew a switch with 1k in series, but what is inside the black box doesn't matter.

When the switch closes, it will pull down the gate of the P-channel MOSFET, which will conduct. When the switch opens, R2 will pull the gate up, and the P-MOS will no longer conduct.

This switches the load in the positive supply wire, which may be more convenient than switching the ground, especially if the device that is switched must communicate with other devices, in this case it needs GND as a reference for signals.

Since the supply is 12V, it is not necessary to protect the PMOS gate (most will tolerate up to 20V).

Voltage drop on the PMOS will be RdsON * current, to minimize it try to pick a PMOS with a RdsON below 0.1R at Vgs=10V, that's easy to find.

You can also use a PNP transistor instead of the PMOS, same schematic but it will need a base resistor, which depends on the transistor's gain. For 150mA it's better not to use a 200mA transistor, so if you have some PNP transistors already which can handle 0.5A or more, post a link to datasheet if you need help picking the base resistor.


welcome to stack exchange!

First, let's think about your perspective on this problem. It seems odd or difficult if you think about it in terms of using this high impedance ground connection as a power ground. But that's not what it is for.

This is actually very typical among electronics, something called 'open-drain/open-collector' logic. This is essentially just a NPN or NMOS transistor that connects or disconnects ground through its drain or collector (which is open or unconnected if nothing is connected to the pin). Typically there is a series resistor for current limiting, though not always. Regardless, I would not continue drawing 150mA from the connection, it is probably not actually rated for that much current.

The line is a logic line, it is not meant to provide meaningful current or power, but rather simply control something else.

Fortunately, it is extremely simple to interface with open drain outputs. These kind of outputs I like to imagine like those cords that you used to be able to pull on in a bus to signal it to stop.

And indeed, we must use a pull-up resistor to interface with this. And there are few different ways to accomplish this, but the simplest is to simply use a PMOS/P-channel MOSFET. And everything is actually going to be in reverse of how you're thinking.

Your device should always be grounded and always have a solid, always connected connection to ground. This is the standard convention for automotive designs, and indeed, check any exposed metal anywhere in your car, and you'll see that it is connected to a common ground. So you should not switch the devices ground at all.

However, don't connect the +12V. It will instead connect to the drain of the P-channel mosfet. The source of the mosfet is then connected +12V.

MOSFETs are off when their gate is at the same potential as their source. So we connect the PMOS's gate to +12V but through a pull-up resistor. Ideally, it should be 10K-100K, it must be a good deal high impedance than our control line.

This will ensure the MOSFET, and your device, are normally turned off. In this state, the +12V is effectively disconnected from your device.

However, if you connect your open-drain control line to the gate as well, when it is 'on' (connecting ground though the 1K resistor), it can pull the PMOS's gate down since it can over power the higher impedance trying to pull the gate up. And to turn a PMOS transistor on, you need (usually) a potential 10V lower than its source. So like a NMOS transistor, but in reverse (which needs 10V above its source).

Like this:

enter image description here

All roads lead to Rome however, and this is but one road. There are a dozen other ways to accomplish this, this is just the first way that came to mind. Though many alternatives will involve actually disconnecting the ground, which I do not recommend in an automotive environment.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.