0
\$\begingroup\$

I'm running a small 12V-500V DC-DC converter (shown) on a purely resistive 50mA load. I was getting weird behavior off my bench 12V DC supply so I plugged in the SLA battery shown. I connected a multimeter directly to the battery's terminals and read 13V. It holds that voltage for single-digit mA loads on the high side of the converter, but when I step up to the higher load my high-side ammeter reads 50mA but the battery's voltage collapses. The multimeter (digital, so only updates a few times per second) shows it mostly hovering about 3.6V, but occasionally cycling to other values from 0 to almost 12V for one update.

SLA and DC-DC converter

I actually tested two different batteries and got the same behavior. Then, to make sure it wasn't a surge condition, I put a 3A fuse in-line. Then, to make sure the batteries weren't just drained, I connected a 4Ω power resistor directly to each one and confirmed no voltage drop and a sustained 3A through the resistor.

These are SLA batteries designed for 120V AC UPS (which can draw 250W for 5-10 minutes), and my high-voltage load that's killing the battery and converter should not be more than 25W. How is that terminal voltage drop even possible given that feeding the converter isn't blowing a 3A fuse and can sustain a direct 3A test? Evidently there's some dynamic here of which I am ignorant, so what could be going on?

Amendment: As another test, to determine if this DC converter has "spikey" demand, I added a 5µF polypropylene capacitor in parallel with the 12V power supply, and put a multimeter on the capacitor's leads. This eliminated all voltage cycling on the supply side, but it did drop the 12A voltage instantly to just under 6V when load was connected. Then I put the volt meter on the high side of the DC converter and saw the voltage cycling from 0 to 500V while the high-side ammeter showed it barely holding 50mA. I could understand high-side problems if the converter were just underrated for the load, but I am still so confused as to how it can affect low-side supply side voltage like that. Especially with nothing heating up or blowing.

\$\endgroup\$
  • \$\begingroup\$ Further musing: Could this be something like a power-factor problem? \$\endgroup\$ – feetwet May 25 '16 at 22:52
  • \$\begingroup\$ No. Power factor only applies to AC loads. \$\endgroup\$ – duskwuff May 25 '16 at 23:08
  • \$\begingroup\$ You said you were getting weird behaviour off the bench supply ... what current was it supplying, and what was weird about it? \$\endgroup\$ – Brian Drummond May 25 '16 at 23:34
  • \$\begingroup\$ SMPS have been known to draw high current as 'pulses' under overload or restart conditions , but that usually blows a fuse, or the SEPIC IC is destroyed. \$\endgroup\$ – Sparky256 May 26 '16 at 0:12
  • \$\begingroup\$ @BrianDrummond - I thought the bench supply was rated for 10 amps, but even though it never blew a 3A fuse it was also dropping and cycling voltage on non-trivial high-side loads (though not as much as the battery). It's like something in that power converter is able to drain voltage without pushing current limits. If I could get any fuses to blow or any component to fail I would understand. But nothing heats up, nothing bad happens, and even though I've stressed it a bunch of times it continues to function as expected at very low high-side power. \$\endgroup\$ – feetwet May 26 '16 at 1:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.