One-wire EEPROMs--any wisdom?

Passing digital fast edges down wiring can be something of a fools errand, unless both ends are impedance matched, or line tx/rx are used. Try to do it on the cheap = flaky result ime. Find a better way...

Chris

On Monday, July 3, 2017 at 6:31:01 AM UTC-7, Chris wrote: ..

..

I2C has controlled slew rates and I would't consider it fast. It is used successfully in millions of consumer HDMI devices.

Surely that would indicate that it is not a "fools errand"?

kevin

The original i2c has a max data rate of 100KHz and is usually part of the m/board, where track Z can be controlled , or be ignored because of short track length. Controlled slew rates looks like sticking a band aid on the problem and constrains device selection. The only external use i've seen is in monitor interfaces, but there are limits to cable length, fwir.

Industrial stuff has run for years using a 4-20mA loop and works well in practice, so why not use analog methods ?...

Chris

HDMI uses I2C for the DDC channel. Displayport went to differential signalling for the aux channel.

Limits greatly the number of bits. Process control is a dog's breakfast with more 'standards' than you have fingers to count them on. Everything from HART (1200 baud over 4-20mA) to things like BISS that are only standard for a couple of makers.

--sp

On Monday, July 3, 2017 at 8:21:56 AM UTC-7, Spehro Pefhany wrote: ...

... The DDC channel has similar funcitonality to Phil's requirement - a mechani sm to identify a device at the other end of a cable - i2c doesn't meet Phil 's desire to implement the signaling over one wire though.

Some PC manufacturers use Onewire devices to identify the power module to t he computer.

In the case of Dell modules I have seen the unwire device is in the module and communicates with a single wire down the same cable as the power.

I accidentally destroyed one device when I probed the female coax connector to determine the power supply voltage. They have a concentric connector w ith the outer and inner surface of the cylinder being the negative and posi tive of the supply with the pin going to the onewire device. If the probe shorts the pin to the inner surface of the coaxial connector (almost imposs ible to avoid) it applies the 19v or so to the one-wire device - there is n o protection of any kind so it is irreversibly damaged. I wasn't the only p erson to do that and found a few links on the web for how to program a new device to get the PSU working again.

If the computer can't identify the PSU it reverts to a non-charging low spe ed operation mode.

Apple also uses a OneWire device to identify the PSU and turn on the chargi ng indicator in the Magsafe connector. Apple puts the OneWire device in the connector itself so there is only an inch or so to communicate - the signa l does not travers the cable.

kevin

Obviously not, but if something else can be shared then the extra wire may appear.

Yes, I have a few of those things.

They've been successfully cloned, fortunately.