• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 48 subscribers
  • Views 697 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM675 heat generation problem

Thomas Dawson
Prodigy 10 points
Other Parts Discussed in Thread: LM675, LM1875

Hi - 

I'm running a pair of LM675's as a differential output driver at 70 khz at up to +/- 20V peak sinewave output from  +/- 24V supplies.  Each is running at about 23db gain closed loop so there wouldn't seem to be a stability issue although my circuit is somewhat different than the simple example circuits in the spec sheet for the part.

The problem is that each of them are dissipating about an amp internally under these conditions and the customer is rejecting this lack of efficiency.

This extremely high current remains constant even with no load which appears extremely strange and wasteful for an amplifier claimed to have a 70 Khz power bandwidth, which is how I am operating it.  

The heat problem appears to get linearly worse as frequency increases above 10 khz, and the amplitude exceeds about 20%.  When with no output level or at much lower frequencies, the LM675 appears to be in spec current wise.

I can supply a schematic to show exactly how I'm operating them.

Is the LM675 actually this inefficient?  It appears to be hugely wasteful to waste 24 watts as heat  @ 70 khz below clipping. with no load at the output.

Btw, it appears likely that the 1 ohm, 0.22 uf stabilization network is the major culprit here.  In the lack of margin of stability characteristics in the product spec, what might TI suggest as an alternative stabilization approach to allow the LM675 to meet its claimed performance as a ''70 khz full power bandwidth amplifier' while maintaining a reasonable level of efficiency up to 3A?

You can reach me at tom.dawson@siliconengines.net or by phone at 224-795-5372.

Thanks Tom Dawson

  • 0
    TI__Mastermind 49906 points

    Hello Tom,

    Normal Class AB amplifiers are around 70% efficient, and less efficient as you approach half output swing.

    According to the standard power full power bandwith formula, the max slew-rate limited frequency is about 31KHz.

    The LM675 is a tighter DC spec version of the LM1875 audio power amplifier, so it is targeted for power apps in the audio range (DC-30Khz).

    The example circuit is for audio-band up to 30kHz - and you may have figured out the issue....

    The circuit is not designed for continuous 70KHz signals. The "snubber" circuit is designed to kill high frequency gain to prevent oscillations in the output stage.

    I think you have found your major power dissipation problem - the snubber network. At 70KHz, the impedance of the 0.22uF cap is 10 ohms! So that is a 11 ohm load on the amplifier in addition to your applied load. You may notice that cap getting warm, too.

    Try removing the snubber cap and see if your efficiency get's better....it should. You also need to look at the output waveform with a scope (preferably analog) to watch for oscillations around the zero cross and peaks. You may have to change the caps to 0.047uF - which will still be about 50 ohms. Since the oscillations are highly dependent on the loads (Resistive? Capacitive? Inductive?), you may need to tweak the R/C values.

    Regards,