Load Slammer

Sep 07, 2019 7 Replies

formatting link



Looks silly to me. It's apparently not polarity protected. They recommend soldering it to the board under test. Looks like it's designed to break at $1800 per pass.



I'm about to test some switchers on a new board, for noise and transient dynamics, but I don't need this.



Would be a little useful with a lowish jitter trigger and adjustable pulse width or pulse pattern. Do some ChipWhisperer stuff -- supply glitching.

But considering the ChipWhisperer is open, and a lot cheaper, yeah...

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website:

formatting link

Reminding you about our programmable current-source design and performance in AoE x-Chapters, section 4x.26, updated:

formatting link

We show a 2.5A max version, with 25ns response time, but with bigger MOSFETs you can easily make them for 20A, 50A or whatever. The MOSFETs in higher-current versions will have higher C1 (Ciss) and hence need lower R3 values, or C2 R2 setup for slower (50ns, 100ns) response, or maintain 25ns speed with a BUF634 gate-drive buffer and R3=6.8, etc.

Low cost. Use with a programmable waveform generator, to fully characterize your power-supply's transient response.

There's a PCB to help build and test prototype designs.

formatting link

Thanks, - Win

For modest-current switchers, I can just connect a 50 ohm square wave generator to the switcher output and see the basic transient response. I also inherited a couple of Kikusui power supply load boxes, which can dissipate hundreds of watts and can do square-wave load modulation.

But just a power mosfet and a few load resistors would work.

The new board that I'm getting on Monday uses a few TPS54302 switchers. I've never got WebBench to run, and don't care to learn Tina, so I guessed the compensation and will verify it experimentally.

You can also just inject a tiny square wave into a switcher feedback pin, when it's driving its normal load, and see what the loop does.

I designed the TPS54302 into my RIS-788 bee-hive monitor board, but was shocked not to be able to purchase enough for the first production run. So I had to change to TPS54202H. The H was a special '54202 version that hadn't been wiped out of stock.

Thanks, - Win

I have a B&K 8601 and an 8610.

Or use a VNA, or something designed for the purpose. I add a 50ohm resistor to the upper feedback resistor, with test points, to make connection painless.

There seems to be lots of 54302's around now. Great little part.

Step response tells me all I need to know, and just needs a function or pulse generator, and a scope. The scope checks for noise on the power rail, and startup behavior, too.

I only have one knob to turn on the TPS54302, the value of one capacitor.

formatting link

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required