diyAB Amp The "Honey Badger" build thread

Hi UncleMud,
The very best oscilloscope you could pick is an analogue model. The best news is that they are only a couple hundred for really excellent scopes that were over $1K new. A digital scope has toys, but they can lie to you. They are also very noisy (the trace). For a digital, you need 4x to 5x the bandwidth as an analogue. So what that means is that my 1 GHz digital is about the same as a 200 MHz analogue scope. I kept my analogue scopes.
 
My Honeybadger is seriously ill and I am looking for a cure. After about four years of perfect performance one channel developed a warbling/scratching background-noise that starts on powering-up and stays constant in content and volume independent from temperature, input and output loading. Both channels use a common power supply. The noise is not that loud and masked at a medium sound level. It feels like a transistor gone rogue.
Last weekend me and the badger attended a meeting of enthusiasts, some much more knowledgeable regarding electronics than myself. After spending a couple of hours with checking the PCB / soldering points and replacing everything semiconductive from Q1 to Q13 even the experts ran out of ideas. The badger is still warbling. Anyone here with a clue?

Right now the only promising solution seems to build one completely new board from scratch ...
 
You say that you have run it for four years before it started to misbehave recently, right? If nothing else helps, I have a pair of untouched spare boards, purchased about four years ago, which I do not plan to use. Happy to send them over from CH, if needed. Good luck with the repair, though!
 
Hello all,
many thanks for your input. I just tried the proposed swap of R19to 15k and replaced the 1.3W Zener with a 500mW type - no change.This does not really sound like oszillation, it is a scratching/crackling/warbling noise.

How to attach an (existing) MPEG4-recording? The "attach files" option does not show the file.

The board and the solder joints were checked by various (experienced) guys and I did some "reheating" wherever quality seemed in any way questionable.

By the way, one of my ground rules: NEVER ebay or aliexpress for electronic components or boards !

Thank you "mcsontos", but I make the boards myself (one in blue would look strange) ...
 
This is the patient (initially without the zeners):

board.JPG
HB4.JPG
 
Tried to modify HB, don't know about Ltspice accurate or not or cirquit is wrong, hope it's works ,
 

Attachments

  • Screenshot_20241012-215912.jpg
    Screenshot_20241012-215912.jpg
    270.5 KB · Views: 96
  • Screenshot_20241012-220106.jpg
    Screenshot_20241012-220106.jpg
    171.1 KB · Views: 87
  • Screenshot_20241012-220441.jpg
    Screenshot_20241012-220441.jpg
    160.5 KB · Views: 84
  • Screenshot_20241012-220013.jpg
    Screenshot_20241012-220013.jpg
    173.5 KB · Views: 84
  • Screenshot_20241012-215628.jpg
    Screenshot_20241012-215628.jpg
    270.4 KB · Views: 91
  • Screenshot_20241012-221009.jpg
    Screenshot_20241012-221009.jpg
    208 KB · Views: 101
manic mechanic,

pleasing layout and finish, indeed! Optimized board placements, short and spatially well separated air-wires, no mess, etc. Just beautiful. 🙂

I had a coffee-break chat with my reliability colleagues who hunt such things 7/24. Assuming that you swapped everything in the environment (left-right cables, speakers, source cables, source itself, protection boards etc) and checked for GND faults, their further suspects are:

  • aging of (potentially overheated or mechanically strained) film capacitors
  • (aged) solder resin residuals between nearest neighbor solder pads
  • soft dielectric breakdown of the pcb insulator between closely spaced, large potential difference lines/pads

Since it is not a matter of lawsuit (in which case they routinely do high-res thermal maps, X-rays, electron microscopy etc), they also recommend to simply redo the whole board from the scratch.