Isolated N to N volts

[graynomad]

Can anyone tell me what you call a "DC/DC converter" that will take any DC voltage (well say 9-30v) on the input and produce the same voltage (a small loss is OK) on the output that is isolated from the input?

Please don't say a transformer , I don't want to dick around changing from DC/AC/DC. 

I though a quick Google search would find something, and there are no end of DC/DC converters but they all have fixed OPs as far as I can see.

I'd like a PCB mount device, good for around 1-2A.

Maybe it can't be done because you have no reference across the isolation barrier, although a VCO could do that and I'm sure the Murata (or whoever) guys would be up to the job.

EDIT: Actually thinking about it 2A at 30V might be a bit of an ask. Still any ideas welcome.
______
Rob

[johnwasser]

I think that to supply isolation but transfer power the two sides would have to be either magnetically coupled (sorry I have to say it but "transformer") or capacitively coupled (capacitors).  Either way you are talking AC.

Have you thought of separate batteries?

[graynomad]

I guess it pretty much has to be a transformer, it's just that I don't want to do it, I'd like a little black box filled with clever stuff by someone who knows about this type of electronics

I may decide that the far side has to have it's own PSU/battery, that's certainly an option but I didn't want to force that as it might be a remote sensor(s).

I can also stipulate that one side will be for example 12v, and the source > 12v (or whatever headroom is required). In this case there are a 1000 options.

_____
Rob

[Grumpy_Mike]

Isolated DC to DC converters do this. Do you want something that when you put a varying voltage on the input you get a varying DC output. You can do that even simpler using a one one transformer and simple rectification on the output side. On the input side you put an oscillator going to the rail.
However you do get the peak value output for the DC input, not the RMS.

[graynomad]

Thanks Grumpy, I have some idea of the basics but this sort of hardware is a little out of my comfort zone and I was hoping for a black box off the shelf.

Isolated DC to DC converters do this.

I can't find one that follows the input voltage though, they all seem to have fixed outputs.

______
Rob

[RuggedCircuits]

I can't find one that follows the input voltage though, they all seem to have fixed outputs.

Sometimes, when you can't find a part that does something it is a red flag that you're looking for the wrong thing or solving the wrong problem.

Backing up, why do you need to transfer power across an isolation barrier while also keeping the voltage the same?  Maybe there's a better answer out there for the big picture goal.

--
The Aussie Shield: breakout all 28 pins to quick-connect terminals

[graynomad]

Yes I should have stated the app.

I have a network with RS485 + power (9-30v) on the wires. I'd like to have a small box to plug my RJ11 connectors in that isolates both the signal (that should be easy) and the power. The theory being that if I need a single node or maybe a few nodes to be isolated I just plug in this box.

As I said before I'm sort of happy to have a reduced V on the output, as long as it's about 9v or more the nodes will be happy, but that does place a limitation on the input as it has to have enough headroom over the 9v. However as the nominal system power will be 12 or 24 volts this wouldn't be a problem.

If I go this route then I can use any DC/DC converter.

______
Rob

[RuggedCircuits]

I think you may be worrying about the wrong thing. Isolated DC/DC converters don't really need headroom. I'd settle for 12V DC/DC converter (12V in, 12V out) and just design everything for that. Lots of choices for these modules.

--
The DIN Rail Mount kit for Arduino: quickly attach your Arduino to standard DIN rail

[graynomad]

I think you're right, I can't use 12v->12v as such because the input is from batteries that go through the normal charge/discharge range of volts.

But most of those converters have a large input range so that's not a problem.

Thanks everyone, I design by myself with no one to run ideas passed and it often makes things clearer having to spell the problems out.

_______
Rob

[jackrae]

It's might be even easier to solve than you think.  An in-line capacitor will block the DC supply but permit the data signal to pass through.
Might not be the exact answer to your question but may add to the thought process
jack

[graynomad]

Trouble is I want the DC power supply to be isolated as well.

______
Rob