1. Home
  2. Electronics Design
  3. 12V DC-DC converter square wave problems (and solution)

12V DC-DC converter square wave problems (and solution)

In a previous post I mentioned a small DC-DC converter using 12 VDC and an IRS24531 driving two dual MOSFETs FDS6930B and two custom built ferrite transformers with an E187 core. I tested one set of transformers that I built, which have about 348 uH inductance and 25 H leakage inductance. The ones in this board have about 500 uH and only 2 H leakage.

I wanted to see the current draw of the circuit, so I applied 12 VDC, and the current shot up to 200-400 mA with no load. I was able to get it to run at about 10 VDC with only 20 mA draw. The other board with my transformers ran fine at 12 VDC. Here are some waveforms:

10 VDC, good waveform:
formatting link

About 10.2 VDC, peaks on transitions:

formatting link

At 11 VDC, unstable transitions and high current draw:

formatting link

Output at 10 VDC:

formatting link

Output at about 10.5 VDC, getting ugly:

formatting link

Output at about 11 VDC:

formatting link

This didn't make much sense, but then I observed that the frequency was about 22 kHz, and it is supposed to be about 65 kHz. The resistor is 10k and the capacitor should be 33 nF. According to the spec sheet, the period of oscillation is 1.453 *RC so that would be 479 uSec or 21 kHz. Apparently I neglected to change the resistor to about 3k.

I changed to 3.01k and frequency is 70.7 kHz. Works fine at 12.2V and 13.3V.

You can see the 1 uSec dead time:

formatting link

formatting link

So, found the problem, solved it, and all seems OK. Total current draw of the entire PCB at 13.3 VDC is about 10 mA. It draws about 750 mA when initiated, which is about as expected.

Paul

Reply to
P E Schoen
Loading thread data ...

I "spoke" too soon.

I just measured the currents when initiated, The board draws about 900 mA and the gate current is about 190 mA. This is puzzling. That is 10.8 watts input. The DC output from the DC-DC converter is about 11.5 VDC, so with 190 mA draw that is just 2.18 watts or 4.36 watts total. There is 6.4 watts being dissipated somewhere, and I don?t sense anything getting that hot. I?ll try for a longer time.

OK, it looks like the heat is generated in the 220 uF 35V electrolytic capacitors, and the rectifiers. I used Schottky diodes on the other board. I?ll try replacing them. BINGO! Now it draws just 550 mA when initiated. I totally removed the 220 uF capacitors. There is just a 1 uF ceramic capacitor on the output, which is 12.7 VDC. When initiated, the voltage drops to 11.36 VDC, with about 0.5 V P-P ripple at 146 kHz.

Now everything seems to be running pretty cool.

I'm still not sure if I need to add any inductance to the primary side. My split-bobbin transformers had significant (25 uH) leakage inductance, but these have only about 2 uH. Regulation is better, but actually I am using this as a constant current source for SCR gates, so a "soft" voltage may be better. I'm just using linear current regulation, so some limiting via inductance may be more efficient. Efficiency is not really a concern since the application is a 600-1200 watt AC current source, but I thought I could use a 6 watt 12 VDC switcher. I will probably use this (it's about the same size):

formatting link

Paul

Reply to
P E Schoen

of

s
190

at hot. I?ll

. I?ll

y

be

e

ld

me

n-Module-p-1045523.html

I hope you're not putting that in an industrial application!

Reply to
bloggs.fredbloggs.fred

looks like an offline switcher, I've seen many of those in little controllers. Yes, they are the s**ts but they do work for specific task!

Jamie

Reply to
M Philbrook

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.