0
\$\begingroup\$

This is quite an open question, so apologies in advance if you would rather schematics and the like, if you really think this would help I can provide them

My problem: boosting from 2x AA Li batteries to 27V to supply an AC piezoelectric pump which resonates at 22kHz and draws around 1.2W.

On a stable PSU the system works great, I have a gate drive and H-bridge (which I design with help from this community so thank you to anyone who contributed to that part already!) and I get a good efficiency (as high as 92%) BUT when I swap out the PSU from batteries I see the following issue

The battery voltage drops significantly which therefore ruins my boost conversion.

I started at pumping for 1.5s then off for 40s. I shortened the pump cycle durations (this is to create pressure inside a steel canister btw) with the hope that I wouldn't be triggering the PTC fuse and increasing the battery's internal resistance - however this seems to still happen with shorter pulses

Is this inrush current? if so how do I make the current drawn more constant?

I've added 1000uF to the battery terminals so that some of the pulse comes out of their rather than the battery, which certainly helps but doesn't solve the issue

I'm using a load switch to turn on the boost converter (MT3608), should I stick to just using the enable pin? maybe the booster is still getting settled when I then use it (50ms later, maybe not enough?)

As you can tell i'm new to power supply design so maybe i'm missing something obvious

I've tried cascading two steps, battery to 5V5 then boosting to 27V but it seems like this doesn't prevent the voltage droop.

Based on consumption of the system and the expected capacity of the batteries I think I should get a lot more out of them, I know that it's not as simple as mAH as stated but i'm miles off

I would like to avoid feedback limiting the H-bridge itself since i'm using a standard MOSFET set-up and really need to keep all 4 FETs in saturation, so feeding back to reduce the gate voltage won't help me unfortunately

\$\endgroup\$
  • 1
    \$\begingroup\$ What is the voltage and current draw on the power supply? What voltage does the battery drop to? What brand/model are the batteries, and how are you connecting to them? For reference here is some test data for Energizer Ultimate Lithium AA lygte-info.dk/review/batteries2012/… \$\endgroup\$ – Bruce Abbott Oct 5 at 2:34
  • 2
    \$\begingroup\$ Simple answer: you need a bigger battery with a smaller series resistance and droop. \$\endgroup\$ – Voltage Spike Oct 5 at 5:41
  • \$\begingroup\$ The current draw can peak at 1A but on average around 500-600mA, I think there is some inductive kickback from the pump so the current jumps around - I should get a diode in there actually I need to use Energizer L91 Lithium AA, at the moment I have a reservoir cap on the battery output then it goes straight into the boost converter @BruceAbbott thanks for the test data! \$\endgroup\$ – Andrew Davis Oct 5 at 11:50
  • \$\begingroup\$ "I think there is some inductive kickback from the pump" Not the pump, the booster. Also the booster will draw a higher current when starting up as it charges its output capacitor. "maybe the booster is still getting settled when I then use it (50ms later, maybe not enough?)" Perhaps. What happens if you run the booster continuously and load switch the output? \$\endgroup\$ – Bruce Abbott Oct 5 at 19:59
  • \$\begingroup\$ The voltage droop decreases significantly, i'm going to try just using the enable lines of the boosts rather than a load switch - It seems like charging and discharging the boost converters is causing the battery to pulled down since the capacitance of involved looks like a short circuit upon turn on \$\endgroup\$ – Andrew Davis Oct 14 at 13:19

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.