In this article I discuss my ridiculous motivation for buying a new Keithley 2830-120-60 to replace my very functional Keithey 2380-500-15

2380-500-15 & 2380-120-60

While working on the IRDC3894 I spent a significant amount of time using the Keithley 2380 in Constant Current Mode.  The development kit that I was testing was configured to enable 1.2v output at 15A.  In the picture below I am pulling 1A.  You can see the Keithley set to pull 1A and it is actually drawing 0.9998A (plenty close)

While I was testing the setup I would slowly increase the current.  In the picture below you can see that I got to 5.4A with no problem.

But at 5.5A the trouble starts.  In the picture below you can see that I am asking for 5.5A but I am only getting 5.48A

And the gap gets worse as I increase the current.

So I posted on the Keithley site trying to figure out what was happening.  Was the Keithley dead?

And unfortunately there is the answer.  The load has a minimum operating voltage of 4.5v when it is in the 15A mode.

But the 2380-120-60 has a 1.8V operating voltage at 60A

And when I get it plugged in I find that it will happily deliver 16A at 1.2V

And it doesn’t start to roll over until 17A (at 1.2V)

In this article I will show you how to use a Keithley 2380 (actually two different ones) to test the output of the IRDC3894 12V->1.2V 12A buck development kit.

The Story

To this point I have written several articles about my process of designing a power supply for my new IoT device.  It needs to provide for quite a bit of power, actually 60W is what I am planning on.  I really wanted to make sure that the IR3894 chip would do what it says it would, specifically supply 12A.  The development kit is pretty simple.  There are two banana plug to  provide power to Vin and two banana plus for the load.

For this round of tests I will Keithley 2230-30-1 to provide power and I will use my Keithley 2380-120-60 to serve as the load.

The two mini grabbers are attached to to remote sensing terminals on the Keithley 2380.

After I had it all hooked up I went in 1A increments from 0 to 12A, then I went in 0.1A increments until I ran out of input power.

Here is the actual data table.  Note that I added columns to show the calculated input power.  And I calculated the efficiency of the system Wout/Win

When I plot the data there is something sticking out like a sore thumb.  WTF?  At first I assume that I typed in the wrong number when I transposed the hand written data to the spreadsheet.  So I went and looked at the data table where it appears that I typed it in correctly.  Does the efficiency really have a peak like that?

I decided to go remeasure the 5A datapoint.

Then I looked at my handwritten data sheet where I find that I transposed the last two digits of the input current. (I definitely should automate this measurement)

OK… now the plot looks way better

When I compare the plot from the data sheet versus my data on the same scale (about) they look very similar.  All seems good.

In this article I will walk you through the first steps of building a complete Type-C power supply that will look like this:

The Project

I have been working on a project that will drive several strings of WS2812 LEDs.  Specifically, a CapSense dimmable “IoT-ified” nightlight using a PSoC 6 attached to a CYW43012 attached to a string of WS2812 LEDs.   Right now, I have this thing built up with a development kit + a breadboard + 2 wall warts and it is sitting on the floor beside my bed.

When you see this picture, I’m sure that you are thinking.  “You are probably going to be sleeping on the floor beside your bed if you don’t do something better than that.”  And you would be right.  I know that I want a single PCB in a nice 3-d printed box that does all of this.  I also know that I want to use Type-C instead of a normal 12v wall wart.  When I started this I had only the vaguest ideas about how to turn Type-C into something that could drive a bunch of LEDs and a PSoC.  How much power do I need?  And at what voltages?  That seemed like the first question that needed answering.

First, I put a meter on a string of 144 WS2812 LEDs.  Wow, 5V at ~4A when the LEDs are full on.  That is 20W per string… basically 30mA per WS2812.

To make a board that can drive three strings of these LEDs I am going to require 3x20w + whatever the PSoC takes.  A bunch.  But where should I get this much power?  The answer is I am going to start with a laptop Type-C charger like this one from Apple (which I have several of)

The first/next question is, how do I tell the Apple adaptor what voltage/current I want?  It turns out that Cypress is the leader in Type-C chips and we make the perfect chip for this application.  It is called the CYPD3177-24LQXQ and is known colloquially as the EZ-PD Barrel Connector Replacement (BCR).  This is good because that is exactly what I want to do, replace a wall wart barrel with a Type-C.

Cypress CY4533 Development Kit

To get this going I start with the Cypress CY4533 development kit which you can see in the picture below.

This board has

1. A place to plug in Type-C
2. A 5 position switch to tell the EZ-PD chip to select (5, 9,12,15 or 20V)
3. Screw terminals for the output voltage
4. A header with an I2C connection to the EZ-PD chip
5. A load switch to isolate the load

Here is a block diagram

The kit quick start guide has a picture of exactly what I did next

When I turned the selector, I noticed that the output from my Apple charger was (5, 9, 9, 9, 20) and wondered why.  Yet, when I measured another Type-C power supply I got (5, 9, 12, 15, and 20).  It turns out that when you read the fine print on the side of the charger it tells you the answer.  Here is a picture of the side where unfortunately you can’t read (but I used a magnifying glass)

• 20V @ 3A = 60W
• 9V @ 3A = 27W
• 5V @ 2.4A = 12W

The kit guide gives you the answer as to why 5,9,9,9,20:

Infineon

OK.  All that is great, but how do I power my board where I need 5V@12A + 3.6V + 3.3V + 1.8V, this is where Infineon comes into the picture.  Actually, to be completely clear, Infineon came into the picture starting mid-last year when they offered to pay $10B-ish for Cypress.

Infineon makes a line of Buck regulators which are perfect for solving the first part of my problem because

1. They take high-ish voltage inputs (up to 21V)
2. They can supply high-ish currents at the right voltage (up to 35A)

These regulators are called the “SupIRBuck” and are part of the “Integrated POL Analog Voltage Regulators (Industrial)”

So, I ordered a development kit… unfortunately I  choose the wrong one, IRDC3823 which is 12V @ 3A.  However, it was close enough for me to try out.

The board came in a box with the kit and a USB stick.

The USB Stick had the Kit Guide, Datasheet, and Gerber Files. That was nice of them.

The kit it actually very simple.  It has a place to plug in your input supply (the two terminals on the left).  And it has a place to plug in the output.

The board also has a place to configure the startup time of the Buck (the little four position jumper).  When I connected the EZ-PD BCR kit to the IR3823 Eval Board, look what I got.  1.2V.  Great.

This is cool and all of that… but I have a bunch of questions that need answering

1. How do I get 5V out (instead of 1.2V)
2. Why does the the kit guide say a maximum of 13V on the input?
3. How do I configure the PGOOD signal to be compatible with the PSoC
4. How do I measure the efficiency?
5. What is all this stuff about switching frequency and what is the right number?
6. What should I choose for SS_Select and why?
7. What is an “external VCC about?
8. How do I get 5A (instead of 900mA)?
9. How do I talk to the EZ-PD chip via I2C?

All of these questions and more are deferred to future articles.