10
\$\begingroup\$

Is there any real difference between high- and low-side switching?

Assume:

  • Switching is for on/off control of an object (My case RPi)
  • Base/Gate can be driven to Vcc and GND
\$\endgroup\$
  • 1
    \$\begingroup\$ If you are switching a digital device with IO pins connected, these pins may act as another ground and low side switching won't work. In this case you will need high side switching or buffers. \$\endgroup\$ – geometrikal Aug 30 '15 at 21:49
10
\$\begingroup\$

The only real differences are the ground level and maximum current available:

  • Low-side switching means that the two subcircuits will have different ground levels since the switching element will have a (small) non-zero voltage drop.
  • High-side switching will have a lower maximum current limit since P-type (high-side) switching elements usually have a higher on resistance than N-type (low-side) switching elements.
\$\endgroup\$
23
\$\begingroup\$

Of course there is a difference, else there wouldn't be two different methods with different names.

If the load is floating, like a motor or solenoid for example, then high or low side switching makes no difference to the load. That is because, by definition of floating, the node only "sees" the differential voltage accross it and doesn't react to the common mode voltage.

Even with a floating load, the differences to the driving circuit for high versus low side switching can be significant. By convention, we usually consider ground the negative side of the power supply driving the control circuitry, with power then being positive. Since ground is the negative side, and other signals we may need to interact with that connect to the rest of the world will be referenced to this ground, the control circuitry is then also ground-referenced. For example, even if you're driving a 24 V solenoid, the microcontroller producing the PWM pulses will be powered by a 3.3 V rail and ground.

Since the control circuitry is sitting on the low side of the power (the ground), driving low side switches is usually easier than driving high side switches. Therefore, with a floating load that doesn't care whether we switch the low or high side, we usually switch the low side.

Another reason for using a low side switch is when one side of the load is already connected to the positive supply beyond our control. The only choice we have is to leave the low side of the load floating to turn off the load, or connect it to ground to turn it on. It can be convenient for some loads to be pre-connected to power on one side to simplify overall system wiring.

In some cases the load does care. If the load has other ground-referenced signals it has to connect to, then you usually need to keep its ground node connected to ground. In that case, you have to switch the positive power to the load whether you like it or not. Again, this is usually more complicated than driving a low side switch, but not overly so that it requires great lengths to avoid.

When switching the low side with low side control circuitry, it's pretty obvious you want to use a NPN transistor or N channel FET. However, with a high side switch you have to consider more options. N channel FETs generally have better characteristics as switches, but using one presents two problems: The gate has to slew over the switching range plus the gate on/off range, and it needs a voltage above the power rail when on. There are driver chips that can take of these things most of the time, but there are still issues.

A P channel FET is easier to switch as the gate voltage only has to range from the power voltage to around 10 V less for most FETs. PNP transistors can be even easier since you only have to draw some current out of the base to turn them on. However, turning them off quickly can be a challenge.

So, as usual, there is no universal answer, and the tradeoffs have to be considered separately for each application.

\$\endgroup\$
  • \$\begingroup\$ You can use pull-ups for high side, and switch between Z/L instead of H/L \$\endgroup\$ – Alexander M Aug 29 '15 at 22:59
  • 1
    \$\begingroup\$ @Alex: Sure, there are lots of ways of driving high side switches, each with their own set of advantages and disadvantages. One problem with pullups to turn off high side switches is that this turns them off slowly. That can be a serious issue for something like a buck converter, for example. \$\endgroup\$ – Olin Lathrop Aug 29 '15 at 23:01
  • \$\begingroup\$ @AlexanderM: Even if you ignore switching speed, often enough you'll use a MOSFET in an app where the load voltage is below the max Vds the MOSFET can take (obviously) but above max Vgs it can take (which is much lower, e.g. 20V on a 100V MOSFET), in which case you need more than just the pull-up resistor: books.google.com/books?id=FSpC6yNyNWcC&pg=PA297 \$\endgroup\$ – Fizz Oct 23 '15 at 5:23
  • \$\begingroup\$ Also, with a high-side switching PNP you often have to worry about the (high) voltage it's base is going to get raised to, especially when the driver is logic-level circuitry jeelabs.org/2012/11/12/high-side-switching \$\endgroup\$ – Fizz Oct 23 '15 at 5:48
  • \$\begingroup\$ @OlinLathrop, Thanks for the wonderful explanation. If we do use a low side switch for dc motor control, how is the Voltage Feedback circuit arranged? In an SCR drive, we take the V+ and V- bus and get the voltage feedback value by reducing appropriately (isolated, differentially amplified, precision rectified etc). Here, the V+ side is always connected to full supply, and since our control circuits are referenced to the ground, +ve end of the Voltage feedback circuit will show full voltage, while the -ve will show some -ve floating voltage when the low side switch is OFF, right? What to do? \$\endgroup\$ – Vishal Oct 24 '18 at 15:08
6
\$\begingroup\$

For an isolated circuit, no there is no great difference between high and low-side switching. For higher load currents, low side semiconductor switches (for example NPN transistors and N-channel MOSFETs) are often less lossy than their high-side equivalents, and so are preferred.

However, if the circuit is connected to external devices with their own power connections, this becomes blurred. If these external devices provide a connection to the same ground reference as the power supply to the circuit and you switch this in and out then the external devices will provide an alternate route to ground, your switching will be ineffective and you may end up damaging something not rated for the appropriate current along the way.

Similarly, if the external devices provide a V+ supply which is referenced to the same ground as the supply that you are switching, you can end up back-powering the positive voltage rail via the externally powered devices, again with undesirable results.

\$\endgroup\$
  • \$\begingroup\$ Accepted Ignacio's answer only b/c he was first \$\endgroup\$ – Alexander M Aug 29 '15 at 22:48
  • 3
    \$\begingroup\$ @Alex: If I had known you were going to accept any answer only 18 minutes after asking the question, I wouldn't have bothered writing one, let alone going into some background. I'll keep this in mind for future question you ask. \$\endgroup\$ – Olin Lathrop Aug 29 '15 at 22:59
  • \$\begingroup\$ Sorry, Olin. I should've said that your answers was very similar to Ignacio's. That was also why \$\endgroup\$ – Alexander M Aug 29 '15 at 23:01
  • 2
    \$\begingroup\$ @AlexanderM: it doesn't look that similar to me. Olin covered a lot of details that matter in practice, which were completely ignored in the answer you accepted. \$\endgroup\$ – Fizz Oct 23 '15 at 5:48
0
\$\begingroup\$

There are numerous reasons to choose one type of switching over the other.

If your circuit / load can tolerate the ground currents created when switching the load.. generally low side switching is easier and cheaper.

If your circuit can not tolerate this (too much disturbance in the ground plane of the more sensitive / lower voltage processor/logic) .. it is better to switch the load using high side methods, this allows the return current of the load to be managed separately (often higher power loads require a higher voltage power rail.. still sharing a common "ground" potential with separate return paths).

The other common reason for high side switching (mentioned by Olin) .. .the most readily available return current path for the load is the negative power rail. Example: automotive chassis used as a "ground" (DC return path) for a relay/ etc.. (this example has numerous additional advantages and risks).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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