## Monday, August 31, 2015

### What is meant by a “fast” power supply?

We regularly get requests for a power supply with a “fast” output. This means different things to different people, so we always have to ask clarifying questions. Not only do we need to find out what change needs to happen quickly, but we need to quantify the need and find out how quickly it needs to change. For example, recently, a customer testing power amplifiers wanted to know how quickly a particular power supply could attain its output voltage. Two ways to look at this are:

1. How long does it take for a power supply output voltage to change from one value to another value?
2. How long does it take for a power supply output voltage to recover to its original value following a load current change?

This customer wanted to know the answer to question 1. Luckily, both of these answers can be found in our specifications and supplemental characteristic tables.

Question 1 is referring to a supplemental characteristic that has a variety of similar names: programming speed, settling time, output response time, output response characteristic, and programming response time. This is typically described with rise time and fall time values, or settling time values, or occasionally with a time constant. Rise (and fall) time values are what you would expect: the time it takes for the output voltage to go from 10% of its final value to 90% of its final value. Settling time (labeled “Output response time” in the graph below) is the time from when the output voltage begins to change until it settles within a specified settling band around the final value, such as 1% or even 0.1%, or sometimes within an LSB (least significant bit) of the final value. My fellow blogger, Ed, posted about how this affects throughput (click here) back in September of 2013.

Question 2 is referring to a specification called transient response, or load transient recovery time. Whenever the load current changes from a low current to a higher current, the output voltage temporarily dips down slightly and then quickly recovers back to the original value (or close to it).
The feedback loop design inside the power supply determines how quickly the voltage recovers from this load current change. Higher bandwidth designs recover more quickly but are less stable. Likewise, lower bandwidth designs recover more slowly and are more stable. Ed posted about optimizing the output response back in April of this year (click here).

So the transient response recovery time is the time from when the load current begins to increase (coincident with the output voltage beginning to drop) to when the output voltage settles within a specified settling band around the final voltage value.

Our customer was interested in a “fast” power supply, meaning one with a settling time to meet his needs. Once we understood what he needed, we directed him to a power supply that could easily meet his requirements!

### Forums and Programming Examples

Hi everybody!

I am not sure how many of you know but we have instrument specific forums at Keysight.  You can find the power supply forums at: Keysight Power Supply Forums.  If you have questions on power supplies you can post them there and either someone here will answer them or sometimes another user has had a similar experience.

We are also in the midst of revamping our example programs to make them more useful for our customers and would like feedback.  We are interested in the following pieces of information:

1. What programming languages/IO Libraries do you use? We are thinking of concentrating on VB.NET, C#, C/C++, Labview, Matlab, Excel (VBA), and Python. Are we missing anything?

2. Any specific programming examples that would help you more effectively program your power supplies. I cannot guarantee that we will do them but anything requested will be considered.

You can either answer these questions in the comments here or you could use the forums to respond to the thread that I just created for this blog post: Power Supply Programming Example Feedback Thread.

Thanks!

## Friday, August 28, 2015

### Verify inverter MPPT algorithms with Keysight’s new PV array simulators

While I normally avoid simply promoting new products in my blog posts, when Keysight Technologies announces a new power product, I feel obligated to mention it here. After all, this is Keysight’s power blog!

So, yesterday, Keysight Technologies announced two new photovoltaic array simulators. Click here for the press release.

The two new models are the N8937APV (208 Vac 3-phase input) and N8957APV (400 Vac 3-phase input). Both are autorangers and provide up to 15 kW, 1500 V, and 30 A on their outputs. Autoranging power supplies cover more output voltage and current combinations than power supplies with rectangular output characteristics. Click here for a previous post on autorangers and here for a post on the power supplies on which these two new models are based. These models can be put in parallel to provide a single output of up to 90 kW! They complement the family of Solar Array Simulators (SAS) that have been available from Keysight for decades.

Pictured below is the front panel of the two new photovoltaic array simulator models (15,000 W in a 3 U package):
So what is a photovoltaic array simulator? It is a specialized power supply that has an output characteristic that mimics the output characteristic of a solar panel (or a collection of solar panels known as a solar array). Photovoltaic (PV) simply refers to something that generates electricity when exposed to light so solar panels are PV devices. Solar panels have an output characteristic called an I-V curve that looks something like the solid line shown below. Isc is the short circuit current, Voc is the open circuit voltage, and Imp and Vmp are the current and voltage at the maximum power point.
Solar arrays are made by taking many solar panels and connecting them in series and parallel combinations for more power. When put in series, the total voltage increases. When put in parallel, the total current increases. Solar inverters take the DC output power from a solar array and convert it from DC into AC that can be used to power AC-mains devices (like those that plug in the wall in your home). So manufacturers of solar inverters are interested in testing their inverters and using a PV array simulator helps them. Instead of connecting their inverters to a real solar panel array that operates only when there is sunlight shining on it, they “simulate” the power output of the array with a PV array simulator. This enables them to test the inverter in many different conditions that affect the I-V curve of a solar panel, such as the temperature surrounding the panel, the angle of the sun on the panel, and cloud cover. Inverters must work when solar panels are subjected to all variations of these parameters, and waiting for them to occur with an actual solar array and the sun is not practical.

The inverter manufacturers are very interested in harvesting as much power as possible from the array, so they design their inverter circuitry to include Maximum Power Point Tracking (MPPT) algorithms that ensure their inverters operate at Pmax (the maximum power point) shown on the I-V curve above. The new Keysight photovoltaic array simulators allow engineers to test their MPPT algorithms.

So the next time you see a rooftop of solar panels, or a parking lot covered with them, or a field filled with panels collecting sunlight and converting it into electrical energy, think about the inverter connected to those panels converting the DC into AC for your use….hopefully, the inverter was fully tested with a Keysight SAS or one of these new photovoltaic array simulators!

## Friday, August 14, 2015

### Not all two-quadrant power supplies are the same when operating near or at zero volts!

Occasionally when working with customers on power supply applications that require sourcing and sinking current which can be addressed with the proper choice of a two-quadrant power supply, I am told “we need a four-quadrant power supply to do this!” I ask why and it is explained to me that they want to sink current down near or at zero volts and it requires 4-quadrant operation to work. The reasoning why is the case is illustrated in Figure 1.

Figure 1: Power supply sinking current while regulating near or at zero volts at the DUT

As can be seen in the diagram, in practical applications when regulating a voltage at the DUT when sinking current, the voltage at the power supply’s output terminals will be lower than the voltage at the DUT, due to voltage drops in the wiring and connections. Often this means the power supply’s output voltage at its terminals will be negative in order to regulate the voltage at the DUT near or at zero volts.

Hence a four-quadrant power supply is required, right? Well, not necessarily. It all depends on the choice of the two-quadrant power supply as they’re not all the same! Some two-quadrant power supplies will regulate right down to zero volts even when sinking current, while others will not. This can be ascertained from reviewing their output characteristics.

Our N6781A, N6782A, N6785A and N6786A are examples of some of our two-quadrant power supplies that will regulate down to zero volts even when sinking current.  This is reflected in the graph of their output characteristics, shown in Figure 2.

Figure 2: Keysight N6781A, N6782A, N6785A and N6786A 2-quadrant output characteristics

What can be seen in Figure 2 is that these two-quadrant power supplies can source and sink their full output current rating, even along the horizontal zero volt axis of their V-I output characteristic plots. The reason why they are able to do this is because internally they do incorporate a negative voltage power rail that allows them to regulate at zero volts even when sinking current. While you cannot program a negative output voltage on them, making them two-quadrants instead of four, they are actually able to drive their output terminals negative by a small amount, if necessary. This will allow them to compensate for remote sense voltage drop in the wiring, in order to maintain zero volts at the DUT while sinking current. This also makes for a more complicated and more expensive design.

Our N6900A and N7900A series advanced power sources (APS) also have two-quadrant outputs. Their output characteristic is shown in Figure 3.

Figure 3: Keysight N6900A and N7900A series 2-quadrant output characteristics

Here, in comparison, a certain amount of minimum positive voltage is required when sinking current. It can be seen this minimum positive voltage is proportional to the amount of sink current as indicated by the sloping line that starts a small maximum voltage when at maximum sink current and tapers to zero volts at zero sink current.  Basically these series of 2-quadrant power supplies are not able to regulate down to zero volts when sinking current. The reason why is because they do not have an internal negative power voltage rail that is needed for regulating at zero volts when sinking current.

So when needing to source and sink current and power near or at zero volts do not immediately assume a 4-quadrant power supply is required. Depending on the design of a 2-quadrant power supply, it may meet the requirements, as not all 2-quadrant power supplies are the same! One way to tell is to look at its output characteristics.

## Thursday, July 30, 2015

### How do I use Python (and no installed IO Library) to talk to my instrument?

Hello everybody,

Our newest support engineer and I have decided to learn Python (https://www.python.org/) so that we can generate programming examples.  We are hearing that more and more customers are using Python so we figured that we would get on the front side and learn some Python.  This blog represents my first attempts at doing anything in Python so there are probably better methods to do this but I wanted to get this out since I am pretty excited about it.

What we are going to do is do use Python with Telnet and LAN to directly send SCPI commands to my instrument.  The major advantage of this is that with this method, you do not need to use an IO Library so you can use it in different operating systems.  Since we are going to be using Telnet with Python, the first thing that we are going to need to do is to make sure we understand how telnet works with our power supplies.  For all my work here, I will be using my N7953A Advanced Power System (APS) because it is on my desk and it is awesome.  The APS uses port 5024 for telnet.  On my PC running Windows 7, I enter the following in my command window:

 Figure 1

Once I hit enter, I get our very friendly welcome screen:

 Figure 2

Note the text “Welcome to …”, this will be important later.

To send commands or do queries, you just need to enter the SCPI command after the prompt.   The response to the query will automatically appear after you hit "Enter".  Here is how you do a *IDN? Query:

 Figure 3

Note that the query response ends with a new line.  This will be important later.  The prompt also re-appears after every interaction.  In the case of the APS, the prompt is the model number ("N7953A>").  On some other instruments , the prompt is "SCPI>".  Either way, you need to know what the prompt is so that you can account for it later.

Now that we are Telnet experts, we are going to switch to Python and send a *IDN? query to my APS.  The first thing we are going to do is to import the telnet library.  After that we are going to create our APS object by opening a telnet session to the instrument (sorry these screenshots might overflow the frame a very tiny bit):

 Figure 4

Basically, we are at the same point we are at in Figure 1.  We essentially just hit the "Enter" key.  We need to get to the equivalent to Figure 2 so that we can send a command.  That means that we need to read out all of that welcome text from the power supply.  Luckily, the Python Telnet Library has a read_until function that will read text until it encounters a predefined string.  We know that our prompt to enter text in this case is "N6753A>" so that is what we are going to use:

 Figure 5

Let’s send our *IDN? query.  You use the write command to write to the Telnet session.  Helpful tip: all commands sent through telnet need to be terminated with a newline ("\n"):

 Figure 6

So now that we sent a query, we need to read the buffer to get the response.  Remember that the query response ends with a newline so we are going to use read_until and use the newline ("\n") as our read_until text:

 Figure 7

As you can see we get the same response as before.

So now we sent our command and read the response, we need to read out the prompt again so that our power supply is ready to accept the next command sent to it.  We will just use the same read_until that we used before:

 Figure 8

Now we are set to send the next command.

When you are done with programming the instrument, you end the Telnet session with the close command:

 Figure 9

So that's an extremely basic example of how to use Python and Telnet.  I used the shell because that is what I used to figure this all out.  You can also write scripts.  As Chris and I learn more about Python, we will be releasing more examples.  Stay tuned for those.

## Wednesday, July 29, 2015

### Battery drain test on anniversary gift clock

Last month, on June 2, 2015, I celebrated working for Hewlett-Packard/Agilent Technologies/Keysight Technologies for 35 years. During the earlier times of my career, on significant anniversaries such as 10 years or 20 years, employees could choose from a catalog of gifts to have their contributions to the company recognized. That tradition has been discontinued, but I did select a couple of nice gifts over the years. During my HP days, one gift I selected was a clock with a stand shown here:
I have had that clock for decades and it uses a silver oxide button cell battery (number 371). I have to replace the battery about once per year and wondered if that made sense based on the battery capacity and the current drain the clock presents to the battery. I expected the battery to last longer so I wanted to know if I was purchasing inferior batteries. These 1.5 V batteries are rated for about 34 mA-hours. So I set out to measure the current drain using our N6705B DC Power Analyzer with an N6781A 2-Quadrant Source/Measure Unit for Battery Drain Analysis power module installed. Making the measurement was simple…..making the connections to the tiny, delicate battery connection points was the challenging part. After one or two failed attempts (I was being very careful because I did not want to damage the connections), I solicited the help of my colleague, Paul, who handily came up with a solution (thanks, Paul!). Here is the final setup and a close-up of the connections:

I set the N6781A voltage to 1.5 V and used the N6705B built-in data logger to capture current drawn by the clock for 5 minutes, sampling voltage and current about every 40 us. The clock has a second hand and as expected, the current showed pulses once per second when the second hand moved (see Figure 1). Each current pulse looks like the one shown in Figure 2. There was an underlying 200 nA being drawn in between second-hand movements. All of this data is captured and shown below in Figure 3 showing the full 5 minute datalog along with the amp-hour measurement (0.28 uA-hours) and average current measurement (3.430 uA) between the markers.

Given the average current draw, I can calculate how long I would expect a 34 mA-hour battery to last:

34 mAh / 3.430 uA average current = 9912.54 hours = about 1.13 years

This is consistent with me changing the battery about every year, so once again, all makes sense in the world of energy and electronics (whew)! Thanks to the capabilities of the N6705B DC Power Analyzer, I now know the batteries I’m purchasing are lasting the expected time given the current drawn by the clock. How much current is your product drawing from its battery?

## Friday, July 24, 2015

### “Adaptive Fast Charging” for faster charging of mobile devices

In some of my previous posts I have talked about USB power delivery 2.0 providing greater power so that mobile devices can be charged up more quickly with their USB adapters.  A key part of this is these devices are incorporating adaptive fast charging systems to accomplish faster charging. So how does this all work anyway?

Let’s first look at the way existing USB charging work, depicted in Figure 1.

Figure 1: Legacy standard USB charging system

When the mobile device is connected to the USB adapter, the mobile device first determines what kind of USB port it is connected to and how much charging current is available that it will be able to draw in order to recharge its battery. The mobile device then proceeds to internally connect its battery up to the USB power through an internal solid state switch that regulates the charging via the device’s internal battery management. However, a major limitation here is the amount of available current and power. Today’s mobile devices are using larger batteries. Up to 4 Ah batteries are commonly used in smart phones and over 9 Ah capacity batteries are being used in tablets. Even with later updates that increased the charging current to 1.5 amps for a dedicated charging port, this is a small fraction of the charging current and power these larger batteries can handle. As one example, a 9 Ah battery having a 1C recommended maximum charging rate equates to a 9 amp charging current. This requires overnight in order to significantly recharge the battery using standard USB charging.

The shortcomings of legacy USB for battery charging purposes has been well recognized and the USB Power Delivery 2.0 specification has been released to increase the amount of power available to as much as 100 watts. This is accomplished by greater voltage, up to 20 volts, and greater current, up to 5 amps. For a mobile device incorporating this, together with an adaptive fast charging system, is able to charge its battery in much less time. This set up is depicted in Figure 2.

Figure 2: USB adaptive fast charging system

With adaptive fast charging, when the mobile device is connected to the USB adapter, after determining that it has compatible fast charging capabilities, it then negotiates for higher voltage and power. After the negotiation the adapter then increases its output accordingly. A key thing here is the mobile device will typically incorporate DC/DC power conversion in its battery management system. Here it will efficiently convert the adapter’s higher voltage charging power into greater charging current at a voltage level comparable to the mobile device’s battery voltage, to achieve much faster charging. Now you will be able to recharge your device over lunch instead of overnight!