Re: 1-wire to USB converter that can use 1820s directly

["Rod Speed" <rod.speed.aaa@xxxxxxxxx>]

Dave Houston <nobody@xxxxxxxxxxxx> wrote
> Rod Speed <rod.speed.aaa@xxxxxxxxx> wrote

>> Yes, kludges are quite common. Still makes a lot more sense to
>> use a bog standard USB/1wire commercial converter/adapter if it
>> will drive the string of 1820s fine, and it appears that that will
>> be fine. At worst it might need one of the spare wires on the RJ11
>> cables used to drive the 1820s in non parsitic mode and doing that
>> is a hell of lot simpler than using a IOM #135E or IOM #136 Module
>> and a FT232BM USB - RS232 adaptor packaging wise alone.

> The CP2102 is 5mm x 5mm and requires only 2-3 capacitors. It's
> capable of very high bus speed and a very high serial baud rate.

> I agree that a standard USB/1-wire adapter is simpler
> but it doesn't appear there are any that are self-powered.

They are all powered from the USB. No reason why
not, plenty of 5V power available on the USB port.

And parasitic power of the 1820s should work fine, thats
what Pete Anderson uses with his strings of 1820s.

> It should be easy enough to modify your Elektor adapter to use a separate
> power supply. +5V regulated switching supplies are readily available.

Yep, but a commercial USB/1-wire converter/adapter
is a lot cleaner on packaging alone.

> I've posted here about one that I've tested to be trouble free as far as
> causing X-10 problems. However, you may have problems with voltage
> drops if sending it very far over a wire in a telephone cable, especially
> one that passes through a modular connector or two at each node.

Cant see why, the 1820 doesnt take much and
I'd doing them in a star config, not a daisychain.

Happy to use CAT5 cable to the stack of RJ11 sockets for each cluster of
1820s.