Hi, my name is Anston Lobo. And I'm the systems and applications engineer for the Simple Switcher group at Texas Instruments. Today we will discuss simple selection for your Wide V-in DC-DC power module and Industry 4.0 Smart Systems.

Today we will cover the system challenges of an Industry 4.0 system, the technical details that go into selecting the right passives for your power module, and understanding the performance and solution size comparisons for a power module.

Many Industry 4.0 applications that we see today use dedicated power rails, Robotics, RCNC machines, and pick-and-place machines, Machine Vision in their NDT and fall detection, ethernet for communication, PLC motor controls, and other machinery.

The power tree inside these systems typically constitute a main high voltage DC-DC - power source, backed up by an auxiliary power supply in green. Anything on this chart in green represents an opportunity that Simple Switcher can help you with. The common rails that we typically see are 5 volts, 3.3, 24, and 1.8.

A typical system challenge is stepping down a 24-volt input to a 5-volt or a 3.3-volt output rail. Further, point of load power may be provided by stepping it down to 3.3, 1.8, or 1.2 volts. These rails typically don't exceed 5 to 6 amps. And the design challenges are keeping the solution size small, having highest efficiency possible, a low BOM cost, and a fast solution.

Here are a few reference design numbers to aid you in your design. Today we will discuss the considerations from module designers' perspective-- when a module is constructed-- determining the external component selection, a layout example, and the hidden advantages of using a power module. Some of the considerations that DI takes when they make their modules are looking at the design inputs, the silicon selection, the right package selection, internal component selection-- such as inductor, stability components, and programming components-- and setting the safe operating limits-- both electrically and thermally-- and finally, the external component considerations.

Let's take an example where we have to supply a 24-volt rail and step that down to 5 volts. Since this 24-volt rail will be prone to transients, it is wise to use a wide V-in power module. Typically, the input voltage range with respect from 18 volts to 42 volts. And a wide V-in power module would go from 4.5 volts to about 60 volts.

The commonly supported output voltages are 2.5 volts, 3.3 volts, and 5 volts, and up to an output current of up to 2 amps. The goal is to achieve a small solution size, low EMI, and have a pin for synchronizing to an external frequency clock. Additionally, we are looking for low BOM count and the need for the solution to operate over an industrial temperature range of minus 40 to 105 degrees centigrade.

The silicon selection is the first step. We select the LM46002 because it meets our design inputs. The V-in range is from 3.5 to 60 volts on this regulator. And the regulator supports an output of 1 to 28 volts.

It is a synchronous peak current mode regulator, with a peak output current of 2 amps-- or a max output current of 2 amps. The IC features precision enable, a default frequency of 500 kilohertz-- which can be adjustable to synchronize from 200 kilohertz to 2.2 megahertz-- a power good output, internal compensation, and an operating junction temperature of minus 40 to 125 C.

The packet selection is just as important. To achieve the highest possible thermal efficiency, we've chosen a BQFN package, which yields a 12 to 13 degrees C or watt, as opposed to an LGE package, which has a much higher thermal resistance. This can yield up to a 50% increase in life of the part.

The axis to the signal pins is kept simple and spaced appropriately to make routing easy. It was also designed to make solder inspection of joints easy. The power module package facilitates a high temperature reflow of up to 260 degrees C for the BQFN package.

The internal component selection for a module surrounding the regulator consists of selecting the right inductor, selecting the right compensation components for stability and transient response, and selecting the right programming components, such as bypass caps, soft start and power good, and compensation. To select the right inductor, we evaluate various corner V-ins, such as worse case V-in of 60 volts and a worse case low V-in of 12 volts.

We also evaluate this over different switching frequencies, such as 500 kilohertz and 1 megahertz. By using the equations you see on the right, we have estimated that the best inductance, that is the average of these values, is 10 Microhenry. Now that we have selected out inductor, let's evaluate the appropriate frequency range for this module.

The two red lines on the graph denote the current limits of the regulator we've used. The high side current limit is the top red line, and the low side current limit is the bottom red line. The goal is to keep the valley current-- which is defined as the trough of the inductor ripple-- below the low side limit and keep the crest of the inductor ripple-- denoted as peak current-- below the high side limit. As you can see, at 2 megahertz, there is plenty of room on the peak current limit side but hardly any room on the valley current.

If we change the switching frequency to 500 kilohertz, we increase the room on the valley current limit, at the expense of the peak current. However, there is still plenty of room here and room here. Hence, this is considered safe, this is considered safe.

If you drop the frequency lower, to 200 kilohertz, we have now reached a very minimal gap between the peak current and the limit line. This is too close for comfort, whereas the valley current has plenty of room. So the ideal frequency appears to be 500 kilohertz.

If we take this one step further and draw a graph showcasing the different frequencies, for the peak inductor current at 500 kilohertz and at 200 kilohertz, for the same 10 Microhenry inductor, we can see that at 500 kilohertz, a greater number of V-outs is made possible, because the lines of the peak current lie well below the regulator's peak limit line. Whereas if you drop the frequency down to 200 kilohertz, we can no longer do the 7.5 volt V-out option. And the 5 volt option scrimmage is too close to the limit line, making this an unsafe operational bet. Hence, 500 kilohertz is the best frequency from a peak inductor current standpoint.

If we look at the valley inductor current, again, the goal is to stay below the red limit line. At 500 kilohertz, all of these V-outs are possible. But at 1 megahertz or a higher frequency, two of the V-outs are no longer possible, because they breach the red line. And the third is on the limit line.

Any time these V-out ripple lines intersect with the red limit line, they cause the regulator to go into current limit. And output voltage regulation is affected as a result.

The 10 Microhenry shielded inductor that is used inside the module has the best in class performance and strikes a good balance between V-in, V-out, and the frequency range. It meets our CISPR22 class B radiated EMI spec and is a powered iron core for soft saturation. The powdered iron core distributes the air gap uniformly within the core, so that there is no rapid drop-off inductance once the saturation current of the inductor is reached.

As you can see over here, at 48 V-in and 24 V-in, we are well below the CISPR 22 class B limit line. These results were performed in a 10-meter chamber.

This slide illustrates what setting the safe design limits would look like. As you can see, the switching frequency range varies depending on V-out and V-in. You can use this table to determine what is the safe outfitting frequency for your application. Since the V-in, V-out, and switching frequency is constrained by high side current limit, low side current limit, slope compensation, and minimum T-on and T-off, it is prudent that this table is used in selecting your switching frequency.

Operating within the design limits will yield a switch node waveform that looks like this. Notice how the switch pulses are regular, and the V-out is a straight line. This indicates normal operation.

This is when the module is operated outside the design limits. You can see a droop and an overshoot on the V-out. And the switch node starts to skip pulses. This is good, this is bad. If you see this on your oscilloscope, you need to adjust either your switching frequency or your compensation values to push the module back into the stable operating region.

The thermally safe operating area of the module is determined across three airflow parameters-- 200, 100, and natural convection. As you can see, without any airflow, at an ambient temperature of 105 degrees, as we increase the output current, the module begins to derate. That is, it can no longer supply the full rate of 2 amps at 105 degrees C. Hence, it is important to understand the derating of the module over temperature and with varying V-in, V-out, and airflow.

The internal block diagram of the module looks like this. The area in blue is the regulator. The inductor is placed within the module. And anything within this black box is the power module, denoted by LMZ36002.

There is a built-in 100 kiloohm power good pull up resistor, so that an external resistor is no longer necessary. IMP compensation is placed internal to the regulator. The default switching frequency of 500 kilohertz is set by the resistor. And by placing all of these passive components inside the module, it makes placement of your module easy with the minimal required BOM.

Let's dive into the power module to see the construction. We use a thermally enhanced copper leadframe, which will result in a low temperature and a long life power solution. The leadframe is shown here in orange. And the regulator is the black IC in the center, the blackish gray IC in the center. We have integrated the bypass capacitors on the top here and the compensation frequency and soft style capacitors as well.

Next, we place the inductor on top of the IC, which is placed on top of the leadframe. The inductor is a high performance inductor and shielded. The solid copper leadframe is plated with a nickel, platinum, and gold coating.

Finally, an over-molded plastic process is used to encapsulate the power module. And the power pads are created so that they are thermally superior and allow easy heat dissipation. You can see that the pin edge is made of copper and is divotted toward the side, so that a solder filet is formed. All signals are accessible from the parameter.

When you design your power module, you can expect a solution size no greater than 185 square millimeters. Look at how small the solution size is. It is the world's smallest 60-volt power module.

When we look at the data sheet, here are some of the important considerations that we have to look at. Is it wide V-in, 4.5 volts to 60 volts input? Does it meet your output current requirements? This is a 2 amp module.

Does it meet your output voltage requirements? Does it synchronize to an external clock? And this is a complete integrated power solution, which means very few external components are required. The external component considerations no longer exist as in a regulator, because we have taken it upon ourselves to integrate the inductor within the module, providing you with the easiest possible power solution.

Finally, the only components that you have to place are a voltage selection resistor-- based on these values-- the C-in and the C-out capacitors-- based on these values. These tables are all stated in the data sheet.

Other optional components that you can leave open or you can populate them based on the customer requirement is an input under voltage lockout, or UVLO setting-- this is to prevent the module from powering up, unless a minimum V-in has been reached-- a soft start setting, to limit inrush currents through the regulator; a frequency setting, to change the default frequency of the module. 78.7 kiloohm equates to 500 kilohertz. The module has a built-in default of 1 megahertz. So if the RT pin is left open, the module will switch at 1 megahertz.

To recap, the hidden advantages of a module are optimized design time, low circuit debug needs, quick time to market, low part counts, greater reliability, and a small size. This will lead to less engineering design time, less debug time, less power supply design time, a small footprint, fewer components, and a higher life of the part.

For more information, please visit ti.com/powermodule. Thank you for watching.