We are currently caught up in a technical renaissance, where new solutions based on new technologies are creating new functionalities to address new application spaces (as well as empower old ones). However, there is a lot of power infrastructure required to support all of the electronic systems involved, and it must all be evaluated, verified, and managed to ensure proper, safe, and optimum performance.
Test and Evaluation is a “Red Queen's Race” where you have to run as fast as you can just to stay in the same place. In the area of power, this race has been complicated by challenges presented by new materials like wide-bandgap semiconductors, and the advanced power topologies they both enable themselves and inspire in silicon to keep up. Being able to test and evaluate the latest power conversion and battery-charging solutions is vital to staying competitive.
We reached out to Jeremy Lo, Senior Engineering Manager, and David Holt, Senior Director of Product Management & Sales, at BK Precision, to talk about some of the challenges they are seeing in the power test space, and some of their latest solutions to address them. The company’s latest release, the 9140 family of triple-output multi range DC supplies, provides three isolated and individually controllable outputs.
The 2U half-rack device delivers up to 300 W of power with low ripple and noise, or 100 W per channel. A 4.3-inch LCD, output terminals that support sheathed banana plugs or spade-lug connectors, and advanced list mode programming, data logging, and extensive protection, address a wide range of test applications. Enhanced features include list sequencing and step triggering for synchronizing events, and list mode programs can be assigned to run on one or multiple channels simultaneously. Output sequencing allows users to link the output states between multiple channels with configurable on/off delays (Figure 1).
EE: In a recent discussion we had the point came up that there's no electronics without electricity, and power is a fundamental analog process. What are your thoughts on that?
David Holt: I think you summarized it pretty good. We need power in order to run the electronics, and of course, there's different ways to measure that. You can use power meters. You can use shunts. There's a lot of different approaches, and we support those different approaches. And batteries, of course, are a way to store the energy, and using your analogy, it's not enough to store it. You have to be able to take it back out. You have to consume it. In that regard most engineers want to know, "How much power have I put in? How much power have I taken out?"
They also want to understand things like power density. How much can I store for a given space used? And they want to know things like, how does this battery storage system work in an environment like extreme heat, extreme cold, and so on? How much energy can I take out at one time without damaging it? So there are lots of factors to deal with.
EE: here are common threads. You know, the battery storage or uninterruptible power system at a server farm is different in physical size and scope than an electric vehicle battery, but as you're saying, when it comes to getting the power in and out, aren't they pretty much the priorities in both application spaces? Or is it really depending on the application space, how much each aspect is important?
Holt: Well, in an automotive it's very difficult because of the weight and the size. The chassis has to be able to support that, and that affects the overall performance of the vehicle. Having the availability of more power is what gives you the distance and range on the vehicle, but right now in current physics, everybody that's making electric cars, there's no major breakthrough yet in battery technology. It's right around the corner, it seems, but they're still using lithium-ion batteries and you get a certain power density there, and they're heavy and they take up quite a bit of space in the vehicle. Right now the technology is getting into the 300, 400 mile range, but they take quite a bit of time to charge. So, it is the classic problem, getting the energy in and back out.
The newer vehicles are running 800 volt systems now, because that lets you get the energy in and out, and these are production vehicles I'm talking that will be online in 2021, like the Porsche and some of the new Korean cars. So it's still the same battle, and from the battery manufacturer's point of view is they've got challenges. The users of these batteries are the people that integrate them. EV powertrains are very demanding systems (Figure 2).
They have different test challenges. If you're looking at each individual cell and the battery stack, does the vendor you picked supply battery cells before you assemble the package, and assemble them into a stack? Do these things meet my incoming inspection needs? That's a challenge that a lot of people are taking on, more and more people are taking on because it's not just cars. There are now electric aircraft as well that are coming online, of all different types.
EE: How much of that proofing is worries about counterfeit cells and how much of that is to just ensure production quality of what you're receiving?
Holt: It's for both. I mean, you have to. Being an electrochemical process, there are a lot more variables than if you're just dealing with chips. There are things that happen, even during manufacturing, where the manufacturers could be very well-intended and doing everything right, and a slight change in the recipe can screw things up. I think a lot of people on incoming inspection, they even have their own internal rules. They might say, "No. We have to fully cycle that battery. We got to charge it up and completely discharge it several times on a given lot of batteries before we're happy," and that's real-time stuff. It's very time consuming. The other thing they to do is monitor the temperature. So they'll peek under the tent of something we're looking at right now, where we're able to monitor the temperature while you're doing that type of cycling.
EE: Well, power management is thermal management, right? I mean, they're hand-in-hand.
Holt: Yes, and interesting enough, there's a lot of hobbyists and makers, entrepreneurs. We get a lot of phone calls. They're experimenting in their basements, and sometimes they're working with lead-acid batteries, which is okay, but it's also dangerous. If they are charging a stack of lead-acid batteries and they forget about it in the basement and they go to dinner, and they come home, and their house has burned down.
EE: And the neighborhood with it.
Holt: This happened to one of our customers, and we were pretty shocked by that, and I'm like "Wow." And they're like, "Totally my fault. I really made a mistake," and they buy the latest power supplies. "I want to do it again." And I'm like, "I hope you take more safety precautions." And they said, "Absolutely."
EE: Oh, my goodness.
Holt: Well, as you probably know, when you're charging lead-acid batteries, the gas that comes out of there is hydrogen.
EE: Hydrogen, kaboom. My father was an auto mechanic, and I'm a jack-leg wrench-turner, but I remember when he was charging batteries, it was dangerous, and those were just 12-volt car-start batteries.
Holt: Yeah. There's a lot of people that are experimenting with that, and they don't like to talk about exactly what they're doing because they think they've found the Holy Grail. It's interesting.
EE: Well, put it this way, if you have storage, you have the Holy Grail, right?
Holt: Yeah. It's a storage element, and there are people working with every level of it. They're different chemistries and different types, and even flywheel energy. It's just all over the map, capacitors and then on the other side, the load; you want to discharge your storage system, but you want to discharge it in a scientific process so you can evaluate what's going on.
EE: How much of that is one-offs and how much of that is automated test?
Holt: It depends on the customer, but we provide a free software that's written by Jeremy. It's a really nice piece of software that lets you connect up one of our power supplies, one of our loads, and automate that process, and gather all that data.
EE: The battery is among the last link in the chain to be digitized. Now that power has been digitized, IoT is possible, but once we can digitize the battery, putting monitoring chips in each cell is one step, but, if we ever get to "a solid state battery," we could probably read the charge state directly out of the material of the battery itself in some way.
Holt: Battery management units are in the marketplace, and they come in different layers and flavors. There are also companies that sell the complete service. They'll come in and put the battery monitors on each one of your batteries, monitor the batteries, and replace defective batteries. They'll do all that for you. Then there are companies that only make a compact battery monitoring module, and they'll communicate with Bluetooth or WiFi. They gather all that data, and analyze the battery packs in real time.
We were at a trade show, and one of these companies right across from our booth was using BK equipment to evaluate their battery management system, and they're using our software tools, and they're using our power supplies and loads to measure and monitor all that stuff. I'm like, "Well, that's really cool." And they had very kind words to say about our products.
EE: It's a gestalt now, isn't it? There's no such thing anymore as a solid vertical, as any given solution, it's all part and parcel. Everybody leverages everybody else. Right?
Holt: It seems. Yeah. That's an interesting thought. It's crossing all of these silos into all kinds of markets. It's like the traditional vertical market. I'm in the metal industry or I'm in the battery, and it's much broader. It's interesting to watch and see. It can become a challenge to make electronic test and measurement equipment that suits the need because the needs are... I'm surprised all the time, who's using our power supplies, and for what. Another example, we have a lot of people that do some type of metal plating, and they use our supplies, and when you see the power supplies after they've been in that environment, it's like, "What did you do to this power supply?" It's a very toxic environment.
Now our customers are looking for a power supply that's more lab grade, that has more features, that's a little more scientific, not just"... You know. They want something that gives me more data, more control, and so we just announced that in our 9140 series, which has true three channels. There are three-channel power supplies, and it's two 30-volt channels and a 5-volt channel, so it's not really a true three channel. So the 9140 series is a true three channel, but in the same size box that our two 30 and five volts was in, and loaded with features, and for not that much more money, and also LXI. It's a stellar product. Jeremy could tell you more about it.
EE: Go ahead, Jeremy, please.
Jeremy Lo: So tagging on with what Dave said, it's one of our first power supplies in this form factor where it's a natural true three-channel. Dave already said that most power supplies in the market are not truly three channels because usually the third channel is a fixed output. Five volts, 1 amp, more like an auxiliary output, but ours is truly a multi-range, with independent channels on each, one, two, and three, and now we have the capability to do series or parallel, or tracking for two channels or three channels, which we couldn't do before. So you couldn't really combine that together internally or put it together in series without doing some kind of manual setup.
EE: What would you say your favorite feature, this power supply that no BK Precision power supplies ever had before?
Lo: We've put in a lot of work to put functions in that we've never had before. One of it is the list mode. You set your current, your voltage, and then your time, and then you have a list that kind of runs internally in the loop, so that you could kind of just let it program and output based on whatever you configure, but now we enhanced it a little bit. We've added triggering features into the list so that you could essentially create output triggers for each step of the list. That's something that we haven't done before.
Additionally, you have a block of memory, in our case, the 9140 has 10 blocks of memory to store 10 different programs, but any of the channels can utilize each of those list memories to run the same program, or run a different program. You can daisy chain the programs into different sequences. We have a parameter where you could link different list programs together, and essentially you could have multiple sets of sequences that you pre-store or pre-write into the memory and monitor each channel separately while you're doing it.
We added a monitoring functionality, so that if any user were to be running all three channels, and at any given point, they could change the display of the power supply, and by the way, the 9140 is our first color-display power supply. We really wanted to take advantage of the color and the space that we could work with, and so we added a monitoring display so you could monitor the voltage and current output in real time.
EE: How much hand-holding are you finding needs to be done with your customers who are used to working with these analog systems? How many requests for technical support are you getting from your customers with these systems, or are you finding them adopting them very smoothly?
Holt: That depends on the customer. I'd say the majority of customers rely on our user manuals, and we're working harder to make sure that's more accessible, and has what they need in there, along with our application videos, and I think they'd pick it up pretty quick. But then you have customers that are new to electronics. They're physics majors, or they're mechanical engineers, and they're doing battery charging. They don't understand the concept of constant voltage, constant current, and they don't understand that it follows Ohms Law. They're trying to force current into the battery, into the source.
We're sometimes surprised at what they don't know, and what they don't understand, but at the same time, that's our market. Those are customers that are reaching out to us to say, "I bought it. Help me," and we help them, and we even get occasionally university students that are working on their lab experiments. We recognize we're doing their lab for them and say, "Well, we've got to politely encourage them to... We give them a few tips, but we're not going to give them the whole...
Lo: I was going to say, we create application notes or FAQs, or information not too heavy to read, but good enough for someone who's new into electronics or power supplies or DC loads, or most of our general test equipment, to kind of get a feel of what is it, how to use it, how to get started, and we try to put some focus on that so that the customers like students, or those who are just trying to set up something in their lab, or hobbyists, it's really for them to get started. But we also have customers who really know what they're doing, and they just want all the bells and whistles, and they do fine with our products. We don't really get complaints or any issues that are major with them.