Electronics funny.

[Dave Plowman (News)]

Three way speaker driven by three power amplifiers all balanced input. Crossover Behringer CX3400 analogue type which sounded not bad at all.

Wanting to experiment further, got a Behringer DCX2496 which is a direct substitute, but digital, so allows a choice of crossover curves, delays to each unit, and EQ too.

First impression, setting it to the same as the analogue one, was no audible difference, which is good.

But slowly, when left on, it developed a buzz from the speaker(s) Loud enough to be annoying.

Approach the speaker and the buzz drops in level and get close enough and it goes. So you can play it with a hand rather like a Theramin. ;-)

Thinking it faulty, got another. Same thing. Swapped back to the analogue one - fine.

Mains grounds and XLR grounds all check out OK.

Difficult to tell if the buzz is from all three speaker units, as when you get your ear close enough to each one, it stops.

If you power up with no signal, no buzz (at least for over an hour). With signal, it starts after a few minutes.

Any educated guesses? I've never come across anything like this before.

[Richard Tobin]

Can other people hear it?

-- Richard

[Dave Plowman (News)]

Heh heh. Yes. It's hardly subtle.

[Graham.]

Have you tried turning off your phone, hearing aid, cardiac pacemaker, (in that order)

[Bob Minchin]

Dave, try turning off all the lights and see if it improves. I once had a job when the lights were modulating some of the semiconductors particularly glass envelope diodes and putting 100Hz side bands on the signals passing through (100Hz as this is the rate old style mains lamps turn on and off when driven with 50Hz).

[Clive Arthur]

Unlikely but...

I once had to repair a small mixer - a simple noisy pot replacement was needed - but I noticed a distinct buzz from the headphones. Taking scope probe in hand I set about trying to find the fault, but it disappeared whenever I tried to hook a probe on.

Long story short - it was a very old mixer using Germanium transistors, at least one of which had some paint missing. Every time I bent over it to hook a probe on I put the now light-sensitive transistor in shadow from the fluorescent lights. Under normal use, with the covers in place, this couldn't happen.

Cheers

[Dave Plowman (News)]

Both amp and crossover are fully enclosed in ally cases. Putting a hand near then from any direction gets rid of the buzz. I'm thinking to do with body capacitance. Somehow.

[Bill]

In message [e-mail]>, "Dave Plowman (News)" [e-mail]> writes

I'm way out of date, so unlikely to offer anything worthwhile, but

1. are you feeding this speaker from a balanced or unbalanced source?
2. Does the hum stop if you disconnect the source?
3. Where, if anywhere are mains and xLR grounds connected together?

I'm a transformer man, so servo balancing sounds like a black art or hole to me.

[Clive Arthur]

On 13/04/2019 14:53, Dave Plowman (News) wrote: <snip>

Are these cases grounded?

Cheers

[tony sayer]

In article [e-mail]>, Dave Plowman (News) [e-mail]> scribeth thus

Nah! Water don't flow uphill;!..

What are the pin connections on the 7 way XLR this unit has?..

[tony sayer]

Bin there seem that even in more recent Varactor controlled and modulated VHF transmitters;)...

[Dave Plowman (News)]

Adjacent pins for each pair. Central pin has all the screens connected to it.

It worked just fine with the analogue unit.

The analogue X/O has an analogue PS. The digital one an SMPS. Clutching at straws there. ;-)

[Dave Plowman (News)]

Yes.

[Brian Gaff]

Sounds like one of the amps may be a little unstable or on the other hand its a psu issue of some kind where the hum is being picked up from somewhere else that generates the frequency. I think I'd try the old one of attaching a length of wire to bits of the circuit like a speaker terminal, or the earth or another obvious place if its affected by proximity, sounds more like a parasitic oscillation of some sort. Brian

[Brian Gaff]

When the unit is humming does the current it takes increase?

Brian

[Brian Gaff]

Normally one can tell its the phone. Its a very obvious sound. Also some electrolytic capacitors need a normal capacitor across them to act correctly I used to find if driving sensitive electronics. I had an old Sinclair like power amp that was a wonderful radar pulse receiver and needed this mod to fix it but it was not hand sensitive it was just wide open from dc to light kind of thing. It might be interesting to drive it into power resistive loads and see if you can measure the hum on any or all of the outputs. You could do that now but its very fiddly. It might at least localise it, but if it is drawing extra current when it hums I still think of parasitic oscillations issues. You can easily blow up tweeters this way if you get a feedback loop going! Brian

[Brian Gaff]

There is an old rule in DIY electronics. Your wonderful device works fine when its a spiders web of bits joined by wire on a workbench, but as soon as you put it into a case things start to not work quite right.

At the old TV company I worked at they designed a little rf transformer wound on a bit of ferrite bead. It did what it was supposed to on paper and on the breadboard version of the circuit, However when fitted on the pcb and the case put over it, a pattern came on the screen. Eventually I happened to put my finger on the transformer when it was open while testing it and up came the patterning. It soon transpired that a stage was very slightly unstable and the capacitance of the case or of a finger moved the frequency to a resonant one and off it went. Mind you I have a bass bin here which picked up Radio China International till .1 uf capacitors were fitted to all the psu lines inside the cabinet. Its totally unscreened inside and hence all sorts of crap gets in, still some problems from switch clicks. Brian

[Brian Gaff]

Even if they are, try a capacitor from xlr ground to case. Brian

[Brian Gaff]

That was why later transistors used opaque plastic or tin cans. Brian

[Charles F]

Just wondering if all the electronics and PSU is effectively floating inside the case, on the assumption that this will be earthed by the external connections. If there isnt a hard connection already, might be worth trying a 100 ohm resistor between PSU earth and mains earth?:

Charles F