A Comparison Between Dc Switching Regulators & Linear Regulators

Hello, and welcome to CUI In The Lab. I'm Ron Stull, CUI's Power Product Marketing Engineer. Today we're coming to you from my garage lab out of COVID concerns. In today's video, we'll be taking a look at the classic linear regulator and comparing it to CUI's switching regulator series. We'll go over the benefits and the differences of each and then test them out so you can see those first-hand.

So, to summarize the differences between the linear and switching regulator, we have the linear regulator to the simple device that operates by dissipating the excess power from the source. This results in low efficiency and large temperature rises inside the device. These large temperatures make it difficult to conduct large amounts of current without applying some thermal management, such as heat sinks on airflow. Switching regulators, on the other hand, are a little bit more complex internally and operate by storing the excess power from the source and releasing it in a controlled manner over time. This results in high efficiency and smaller temperature rises. The smaller temperatures make it easier to conduct higher amounts of current without the same need for thermal management that you would have in the linear regulator.

Now we're going to sit down at the bench and test these out so you can see the results first-hand. To perform these tests and make the comparisons, we've got the GL317 adjustable linear regulator set to 15 volts and CUI's 15 volt switching regulator - the V7815-1000. Both are rated for about an amp and have been soldered down to our test board here and can be selected with our toggle switch. They will be loaded with an LED array that we discussed in one of our blog posts that you can see in the link below, and we will be measuring the efficiency and the case temperature through our meters here. We have got our output voltage and output current, along with our input voltage and input current.

So now let's take a look at the differences between a linear and switching regulator first-hand. We'll begin with the linear and take a look at its efficiency and power dissipation under a few different conditions. We'll start by applying 24 volts in and about half an amp out. You can see we have our measurements up and we can calculate our efficiency by dividing our output power by our input power. Our output power is equal to our 14.99 volts times our half amp out, which gives us seven and a half watts, and we divide that by our input power of 24 volts in and half an amp in, which gives us 12 watts. This gives us an efficiency of about 62%, which is pretty bad. This low efficiency results in a lot of power being dissipated inside the regulator. In this case, we have 12 watts in minus our seven and a half watts out and this leaves four and a half watts to be dissipated inside our regulator, and as you can see this leads to a large increase in case temperature. We're already sitting about 20 degrees above ambient and the temperature is continuing to rise.

The large amount of power that's dissipated in a linear usually means that some form of thermal management is needed. In this case, we have a heat sink attached to our T02-220 case of the linear regulator, and it doesn't look like that's going to be enough, so we'll probably have to add some airflow in order to keep the regulator from going into thermal shutdown. An interesting characteristic of the linear regulator is that the input and the output current are the same. One result of this is that the efficiency calculation can be simplified to V out over V in as both of the currents will cancel out in the original P out over P in equation, and here we can calculate we have 15 volts out divided by 24 volts in and that gives us the same efficiency result of 62% that we saw earlier. Another result is that given a fixed output power, an increase or decrease to the input voltage will directly affect the amount of power being dissipated in our linear regulator. If we decrease the input voltage without changing the load, that will decrease the amount of input power and thus the amount of power being dissipated in our regulator. So here we've got 17 volts in now with 14.96 volts out, we can calculate the efficiency as 15 volts divided by 17, which gives us an efficiency of about 87, which is a big improvement over the 62% we saw earlier. We can see also this reduction in power being dissipated in the linear has resulted in the case temperature beginning to fall. Likewise, if we were to increase the input voltage, we would be directly increasing the amount of power being dissipated inside our linear, so if we take this up to 30 volts, we can estimate pretty quickly that we're at about 50 efficiency and with 15 watts coming in, that means seven and a half watts are going to be dissipated inside our linear, and we can see that temperature rise going quickly up to 62, and so we should apply some airflow to make sure that we don't go in to thermal protection. This is one of the major drawbacks of the linear - that so much power is dissipated inside the regulator that a lot of thermal management is needed if you were to get to any significant amount of current.

So now we're going to take a look at the efficiency and power dissipation of our V7800 series switching regulator. We're still currently operating off of the linear regulator and you can see that our input current and output current are the same, but as soon as i switch over to the switching regulator, you can see that the input current has dropped while the output current is still at our half amp level. This means that while our output power is unchanged, we've reduced the input power, meaning that less power is left to be dissipated inside our switching regulator and we should expect to see an increase in efficiency as a result. We can calculate the efficiency as 15 volts out times our 0.5 amps out divided by 24 volts in times 0.3 amps in, which gives us an efficiency of about 94%. This is significantly better than the 62% we were seeing with the linear regulator. The amount of power being dissipated inside the regulator is now eight watts minus seven and a half watts, which is about half a watt, and about one tenth of the power being dissipated in the linear regulator under the same conditions. What changed is our input power, by decreasing the input current from 0.5 amps to 0.3 amps, our input power dropped from 12 watts in to 8 watts in, without changing the amount of power going to the load. Because so little power is being dissipated, our case temperature is also much lower and doesn't need the same heat sinking or airflow that was needed with the linear under the same conditions. You can see here we're sitting at a very reasonable 35.8°C and not increasing very fast. Another benefit to this is that increasing the input voltage won't necessarily increase the input power. Instead, as we increase the input voltage, the input current will drop so that the output is relatively unchanged. At 30 volts in, we now have an input power of 8.4 watts, where before we were at 8 watts. This is a very small increase of 0.4 watts compared to the increase of 3 watts we saw with the linear, and means that we can operate at much higher input to output ratios without the same thermal concerns we had with the linear. In this case, we're operating at about 89 efficiency, and dissipating one watt compared to the 50 efficiency and seven and a half watts that we are dissipating in the linear, and as you can see, our case temperature is still very manageable, even without the thermal management that was applied to the linear.

So, in conclusion, the fundamental ways in which linear regulators work lead to poor efficiency and high power dissipation, which in turn leads to high temperatures and the need for thermal management, such as heat sinks and airflow. Switching regulators, on the other hand, such as CUI's V7800 series, operate in a fundamentally different way, storing energy rather than dissipating it. This leads to high efficiency and much improved thermal performance over the linear regulator.