• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 1810 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.

1646 line driver

Bruce Egnater
Prodigy 150 points
Other Parts Discussed in Thread: DRV134

Planned to use the THAT1646 ahead of a high gain audio circuit (guitar amplifier) in place of a transformer for ground loop elimination. The 6dB of gain appears to be a problem as it causes excessive "hiss". Is there a way to reduce the gain of the 1646 to be unity without upsetting its differential/balanced output capability?

  • 0
    TI__Guru**** 189399 points

    Hi, Bruce,

    Our equivalent device to the THAT1646 is the DRV134. But, it also has a fixed 6 dB of gain.

    How much gain do you have after the line driver? These both seems to have fairly low noise.

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 150 points

    Thanks Don. Dealing with noise in high gain guitar amplifiers is an awful thing. You could have as much as 60dB or more of gain after the line driver, which I know is insane but that is what musicians do. We typically use a transformer which, of course, does not add "noise" gain per se, but has other issues. I was trying to get away from the transformers.

    Thanks

    Bruce

  • 0 in reply to Bruce Egnater
    Intellectual 415 points

    Bruce - Let me explain something about the THAT1646, DRV134 and for that matter the SSM2142. I have experience with all three parts and, for the record, do occasionally consult for THAT Corporation.


    I suspect that the noise you've experienced  with the THAT1646 may be due to the fact that if it is used as a line driver feeding an unbalanced load the opposing output is required to be grounded to maintain minimum noise. The noise that you are experiencing may not be due to the 6 dB gain. A THAT1646 feeding an unbalanced load with the other output floating will have a noise level that is typically in the -60dBu vicinity rather than the datasheet spec of -100dB. The -60 dB level clearly too high. If the opposing output is grounded (adding 6 dB gain) the noise drops to around -100 dBu or better. The results are counter intuitive. The reason for this is that the THAT1646, DRV134 and SSM2142 all have high common mode noise gain. When the opposing output is left floating, some of the high common mode noise gain gets converted to differential and the noise floor degrades by 40 dB or more. The TI DRV134 datasheet mentions the grounding requirement but doesn't say why, the SSM2142 data sheet used to also recommend grounding without saying why and the THAT1646 datasheet should mention the requirement but does not. I've tested all three parts and they all do this.

    There is also another option available to drive loads that are single-ended without grounding the opposing output though it defeats the transformer-like operation of the part. If the output of the device is a 1/4" two conductor connector (and thus can never be balanced) the THAT1646, DRV134 or SSM2142 can have the sense lines opened to defeat, in the case of the THAT1646 OutSmarts (r) or in the cases of the DRV134 and SSM2142 the cross-coupling that provides floating transformer-like operation. For what it's worth a floating transformer output with an open leg won't produce any output at all: The active approaches in this configuration work just well enough to make you think they're broken. To make a long story short always ground the unused output. That's how these parts were designed to work...