1. Home
  2. Electronics Design
  3. Trouble-shooting an old circuit

Trouble-shooting an old circuit

A couple of decades after last using it, I dusted off the Surf Simulator I built in the late 90s and have been trying in vain to get it working again. I'll probably build a much simpler version for Arduino, as I'm now discussing here:

formatting link
1

However I'm still hoping to get the legacy stuff working again, mainly as a challenge. It *was* still working very well, the only flaw being a faint but still audible 'click' within the crashing breaker stage, not good for a 'sleep inducer'!

But I then made the stupid mistake of fiddling with several of its nine presets and physically replacing some wiring... and after a couple of days I still have not recovered that '95% OK' status. And I'm darned if I can still fully follow my own design from the limited stuff I documented in late 2000. A big ask I know, but I'm hoping for advice on a few specifics. Here's a 23 year old block diagram:

formatting link
Q1: I've got the 3340/380 attenuator/amplifier section working (it's on a separate veroboard strip). But I dispensed with the dual supply that I'd apparently used originally. Works fine from a single supply of around

12V. Does that look as if it was used anywhere else?

Q2: Assuming I can identify it amongst the very crowded main veroboard (see below), I wonder if I really needed two 'Rumblers'? As I now see it, the sound has the following elements, with no 'mid rumbler'.

Gentle surf slowly increasing Breaking with a sudden but rapidly decreasing crash Deeper 'rumble' as it recedes on a pebble shore, fading to virtual silence. (All with some randomness over the cycle's duration.) Q3: Any other ideas or advice welcomed.

formatting link
Terry, UK

Reply to
Terry Pinnell
Loading thread data ...

It's hard to help you debug the circuit, because you didn't provide a schematic. Do you have a schematic??? Or is that lost to the ages? I can only recommend that you start at one end (or both) and work toward the other end, or the middle. You should be able to probe using an input to an amp and speaker, to hear what the different signals sound like. I would isolate with a small value cap, maybe 0.1 uF or 1 uF.

When I read your posts about using the UNO, you talk as if it will only be providing a very slow fade and turn the unit off. Why not implement the entire design in the UNO? If you actually understand what each block in the diagram is doing, it should be a simple matter to implement each of those in software. The UNO should be able to keep up with this amount of signal processing over the audio range. But since you ask if the UNO can even implement the fade, I'm guessing that software is not something you are very familiar with.

Reply to
Ricky

snip

I tried to find Q4, because the block diagram shows a line from it leading to -Ve which presumably is the original negative supply. See if you can find that connection on the board and physically trace it to wherever it goes. That might answer Q1.

Do NOT NOT NOT eliminate ANYTHING until you have diagnosed the problem! Fix it first, then consider eliminating whatever. Your problem is hard enought as it is, without introducing any other variable(s).

This won't help, but trust me, I feel your pain. Many of us (all of us?) have been there, done that and have the tee shirt to prove it.

Ed

Reply to
ehsjr

Thanks Ed. Although I abhor unsolved puzzles like this, I'm admitting defeat!

I was already adopting Ricky's suggestions of stage-by-stage testing, mainly with my digital 'scope. At 18 test points, 8 of which I've freshly soldered in for easier access. But apart from the original noise and its immediate successors, including the 4017 'randomized astable', all the rest before the LM380 amp (rumble filters etc) appear to have gone AWOL.

The interactions of the 4016 (or 4066) electronic switches with the various voltage levels set by nine presets plus a couple of pots, (duration to auto cut-off and overall volume respectively), add to the nightmare.

The intended -ve voltage currently shows zero but I didn't get as far as clearly establishing its source. FWIW (curiosity value?) I've included below a couple of two more of the regrettably few things I documented in

2000. The 'stages chart' was a .doc file but I've converted it here to an image for convenience.

My assumption back then was presumably that all would be perfectly obvious to me later from the few key documents. Maybe true a few months on - but after 23 years definitely not! Afraid the same applies to much of my stuff from around 1980 - 2001. Probably still true now to some extent ;-)

formatting link
formatting link
Terry

Reply to
Terry Pinnell

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.