2
\$\begingroup\$

datasheet: L6203 DMOS FULL BRIDGE DRIVER

I was testing this H bridge on wire wound high power resistors. I have noticed a high drop in the output voltage. The measurements taken are as follow:

Vin=13V; Vout=9V; Iout=2.7A =>Rds=1.48 Ohms

Vin=12.8V; Vout=8.7V; Iout=2.46 =>Rds=1.6 Ohms

Vin=15.2V; Vout=10.8V; Iout=3A =>Rds=1.46 Ohms

Vin=17V; Vout=11V; Iout=3.3A =>Rds=1.8 Ohms

Vin=20V; Vout-19V; Iout=0.56A =>Rds=1.78 Ohms

Ofcource some of the test where done where the Driver was really Hot (near 140 C as measured by an infrared Thermometer) so here is the temperature Derating:

Rds on as a function of temperature

Also There is another curve that describes the variation of the supply voltage :

Rds on as a function of supply voltage

The curves implies that Rds on should never reach higher 0.8 Ohms (suppose derating due to temperature = 2 @ 150C)

Tests I also tried:

  • Use of a demo board that uses the same chip but SMD footprint. Almost identical results
  • Use of 99% duty cycle signal instead of a constant DC signal at the PWM inputs of the driver (maybe charge pump needs that). Same results
  • Use of 5V logic instead of 3.3V logic for PWM and enable signal... Same results.
  • Trying to drive a motor instead of series wound resistors. Same results.
  • Removed snubber/schottcky diodes at output. Same results
  • Buy drivers from a different supplier to make sure that the driver isn't a counterfeit component(I didn't trust the supplier).Same results.

All measurements are taken using a true RMS multimeter. Vin is the voltage between GND and VS. Vout is between OUT1 and OUT2. I am measuring exactly at the pins so no drops are present because of the PCB/wiring.

circuit:

Schematic

Question is obvious: WHAT THE HELL IS GOING ON? The efficiency is crap and the driver is overheating quickly. I'm using a very large heatsink, still, the driver is unable to drive more than 2A continuous!! What am I doing wrong??

\$\endgroup\$
  • 2
    \$\begingroup\$ What is the voltage? \$\endgroup\$ – Gregory Kornblum Nov 8 '16 at 18:50
  • 3
    \$\begingroup\$ By the way, the input capacitor is very small. Should be tens or hundreds of microfarads \$\endgroup\$ – Gregory Kornblum Nov 8 '16 at 18:53
  • 2
    \$\begingroup\$ What Cboot capacitors are you using? \$\endgroup\$ – Marko Buršič Nov 8 '16 at 18:58
  • 1
    \$\begingroup\$ voltage is Vin of each experiment. That is 12 .8V,13V,15.2V...etc \$\endgroup\$ – fhlb Nov 9 '16 at 7:57
  • 1
    \$\begingroup\$ I placed 200uF-50V very close to the IC (not seen in schematic) along with a 220uF at the input of the switching regulator I am using to power the Logic and microcontroller. so total capacitance is 420uF. I used aluminum caps \$\endgroup\$ – fhlb Nov 9 '16 at 8:00
1
\$\begingroup\$

I have a feeling that Rds-on here is (per leg) so you need to multiply it by 2 to compare it with delta(V)/I. Many integrated H-bridge IC specify the Rds-on for the high side FET and for the low side FET. Take for example this datasheet.

Returning to the definition for Rds-on, it is easy to see why and to expect this.

\$\endgroup\$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.