• EEVblog #268 – Xantrex 300V 4A Power Supply Teardown

    Teardown Tuesday time.
    What’s inside a professional 1200W Xantrex XFR300-4 300V 4A rack mount power supply?

    Be Sociable, Share!

      About EEVblog

      Check Also

      EEVblog #845 – Oscilloscope FFT Comparison

      EEVblog #845 – Oscilloscope FFT Comparison

      Dave compares the FFT modes on 7 different oscilloscopes: Rohde & Schwarz HMO1202 Series Tektronix ...

      • Wayne Osborn

        Hi Dave, nice review.

        I am guessing the 2nd relay is altering the bridge/C configuration for 115VAC operation.

        Nice to see a good deal of EMC filtering on the mains input, and the dual HRC fuses is seriously paranoid protection!

        All the best, Wayne.

      • PChi

        It’s possible that the Pactron Capacitors are part of the resonant circuit and need high ripple current capability.
        Three decades ago I repaired Power Supplies for a (poor) living. I turned on a customer returned rack mouting laboratory power supply to find that it didn’t work. Took the lid off and went to lunch. Came back and started to work on it and received an electric shock. The input filter electrolytic capacitors were still charged despite the two bleed resistors looking perfect. Both resistors had gone open circuit.
        The input inrush current consumption circuit is simple. The fun starts when the relay fails to bypass the resistor. I’ve seen a few (other models) where the resistor has fried.

      • allan

        Nice teardown, Dave. I enjoyed seeing that.

        Did you figure out why the cap. on the output side was shorted out?

        I’m not familiar with the PLCC package. Is it obsolete?

        As far as temperature sensors go…..the one on the heatsink, anyway…..what do they control and how do they work (digital or analog) ?

        I’m guessing they either turn the fan on/off, or modulate its speed.

        I noticed you have an oscilloscope in your lab. :) Could you figure out the inrush resitor value on the output side by measuring the RC time constant with an oscilloscope ? Of course, you’d have to be able to account for the individual lag effects of all the inductors between POWER IN and the capacitor bank. Easier said than done, I’d imagine.

        • Zach

          My first thought as to why there’d be a short across one cap footprint is.. they made just one board layout for all the various voltage outputs, and there’s an option for 2 caps in series for the 600V version.

      • Worf

        The fan configuration is pretty standard.

        Cool air is drawn in from the front, and hot air is exhausted to the rear. You’d have racks of equipment with the fronts facing each other forming a “cold aisle” (where A/C is pumped in), and the backs face each other as well where a “hot aisle” is formed and air for the A/C is drawn out. In this case, the air for the fans are drawn from the holes in the top cover which are closer to the front and draw cooler air over the heatsinks, which are exhausted out the vent in the back.

        (Xantrex is/was a local company here specializing in power electronics – power supplies, inverters, etc).

        For the two relays, one might be soft-start, the other might be over temperature protection – those components will get hot after all and wit hall that thermal monitoring, there needs to be a way to shut it down.

        I wonder if there’s a side circuit providing a small amount of power for the digital circuitry – keeping it alive for the thermal protection (which also powers the fans so they keep running to help cool things down) as well as the logic for the soft-start and other features.

        As for the missing capacitor footprints on the output side – I guess that output board is shared with the 600V version.

      • http://www.pulsecommercialutilities.co.uk/ Energy Broker

        Hi thanks for the blog. From a long time i have searched for it.now i got it.

      • http://www.bastli.ethz.ch Mario

        The Converter Topology is most presumably a resonant full bridge forward converter with a center-tap output rectifier (actually two rectifier circuits, one for each “half” of the transformer.) The output diodes (8 of them in total) seem to be paralleled.
        So the big thing there is the transformer and the two smaller ferrite components are two inductors.

        Nice Device, thanks for the teardown!

      • http://www.lifelongexercise.com/HDWASPFormCaptcha/css/uvnkj.asp www.lifelongexercise.com

        Hi, just wanted to say, I loved this post. It was funny. Keep on posting

      • Haus Of Richie( A+ )

        Found this very interesting. I am fixing a 1U power amp made by JB systems which has a Thermister NTC 10D-20 near the mains input which has blown. I will be putting more detail about this problem in the repair forum.


        Only recently discovered this EEV site and absolutely loving it – Cheers Dave !!!