0
\$\begingroup\$

There is a primitive DIY battery charge monitor which blinks a LED after input voltage drops lower a certain level. Let's call it battery gauge. Also, there is an electric guitar with active electronics whose battery holder this 'device' is fitted in. When plugged into amplifier, the whole thing seem like working, except when in "low-bat" condition, battery gauge kicks in and along with light indication one may hear clicks to come out of loudspeaker.

I was trying to wrap my head around this the other day. But no luck. The big picture of what I have arrived to with extremely loose understanding is like:

The Big Picture

Battery gauge schematic is this (Here you are the MC33161 datasheet):Battery Gauge Sch

Guitar wiring \w battery gauge device: Gt Wiring Sch

My guess is: because an instrument cable is... say it 'imperfect' (it has some internal resistance and capacitance, right?) the guitar ends up sitting not the same ground the amp does. This way, during the C1 capacitor's charge-discharge events the guitar's ground level is shifted back and forth a little which is 'seen' by an amplifier. Here's the unwanted clicks.

By the way this is what I get on scope when connecting the probes into signal cable break (an area marked blue on The Big Picture image).

Scope Measurements

Some additional facts:

  • When added a tiny (.01uF) capacitor between the signal and ground on the guitar side, clicks are getting weaker drastically (affects instrument tone);
  • Clicks go away completely if guitar is powered with a lab power supply instead of battery (sure, there's a solid ground on PSU).

Okay, drawing the bottom line, I'd like someone to help me figure out what is really happening here, please. And how it can be fixed, given the current design, if it can be at all?

\$\endgroup\$

1 Answer 1

0
\$\begingroup\$

If that schematic of the battery gauge is accurate, its only connections are to GCOM and V+. Just put two capacitors across the gauge, i.e., across GCOM and V+: a small (perhaps 0.1 µFd) ceramic cap to eliminate higher-frequencies, and a larger, polarized electrolytic (perhaps 100 µFd) to eliminate the "bass" note of the pulse. Unless the gauge creates a large inductive spike that might show up on a parallel wire, that should eliminate the ticking.

The capacitors should be rated for 16 VDC or higher, and the + of the electrolytic goes to V+.

On the other hand, if this does not work, consider playing:

\$\endgroup\$

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