audio-talk

I've recently been playing with LM2596 converters as an alternative to 7805 and 7809 regs.
they are cheap as chips on ebay on the small packaged boards.
so far so good and the benefit of minimum losses when dropping lots of volts seems a good plus point.

however I'm having a senior moment and I can't quite work out how to get rid of a 5mv spike at precisely 52.36 khz on a 5v output with no load.
I don't think it's going to affect anything in use, it's just irking me.

no contributions, so I'm guessing not much interest in this one.
To have one last attempt at provoking some input:

I chickened out of the spikey smps dc-dc supply and went back to 7805 for the platform mentioned, It didn't work and the reg got toasty. I'm guessing there was a voltage sag when the modem asked for a momentary current rise. This platform wants something like 2 amps for about 200ms every minute or so.

I went back to a previous platform where the processor and modem are seperate and this worked perfectly with normal regs, but I did use two regs, one for the processor and one for the modem.

I'm now going to put back the lm2596 and suffer the spikes....

if anybody has any ideas about filtering this then please share

No ideas.. but in may I bought some 2 amp adjustable dc to DC converters. Did not work on the mofo, but great as a fan controller with light currents
I since bought a couple of 5 amp boards to see how they fare, but I'm a few weeks off trying those Ed.

I don't have an answer for you other than something I looked at when I was concerned about the HF artifacts coming out of my SMPS. Murata have application notes on suppression of diff and common mode noise which might be useful. I chose to ignore the problem as I couldn't quantify it with the equipment I have. From what you say though, have you got an issue affecting the equipment or something you just don't like the look of?

steve s wrote: ↑Fri Jul 03, 2020 9:39 am No ideas.. but in may I bought some 2 amp adjustable dc to DC converters. Did not work on the mofo, but great as a fan controller with light currents
I since bought a couple of 5 amp boards to see how they fare, but I'm a few weeks off trying those Ed.

Many SMPS don't like surge currents, shut down and then do not start up. The Meanwell supplies I've tried shut down on overload and then try again slowly to start up so I found them good for filaments. They have a term for this behaviour/spec I can't remember just now. You then might deal with the HF artifacts if you feel the need.

My very cheap heavy duty laptop supplies cut in and out when overloaded too Steve. My next amp will use the meanwells I'll see how they fare

They have a term for this behaviour/spec I can't remember just now.

Maybe Foldback limiting?

This is the Murata stuff I mentioned

https://www.murata.com/en-eu/products/e ... asic#part1

Nick wrote: ↑Fri Jul 03, 2020 12:27 pm Maybe Foldback limiting?

Yes that, as well as constant current limiting as well as the thing I couldn't remember which was hiccup current limiting.

I could have spent many hours thinking about this one......I'm going to leave it for another day.....

I've used ld1085 out in the field with a few caps for good measure.......it works so far.

ed wrote: ↑Fri Jul 10, 2020 3:48 pm I could have spent many hours thinking about this one......I'm going to leave it for another day.....

I've used ld1085 out in the field with a few caps for good measure.......it works so far.

All my kits are LT1083, cheapo Chinese ones, but they seem to work well where I have used them.

ed wrote: ↑Fri Jul 03, 2020 9:29 am if anybody has any ideas about filtering this then please share

I've had good results by following with a common mode rejection filter. Basically two coupled coils on a ferrite toroid followed by a cap, 100 - 220 nF sort of thing, like what you often see on mains inlets. This kind of thing:

Google search - rn212 common mode filter