1-wire serial protocol and expensive sensor

My interpretation for this was for what they call the "Communication Hardware", the DX80 Sensor-Node that they recommend connecting to the Sure Cross® U-GAGE K50U Ultrasonic Sensor

They write 1-Wire, why would they bother mentioning it's a wire and not say that for Ground (1-Wire ground ?)

if that was not the One-Wire ™ interface invented by Dallas (subsequently acquired by Maxim and then AD)

(the concepts of 1-Wire serial communications and 1-Wire serial device select are related to One-Wire ™ interface)

but sure, it's very possible

Hmm ... maybe it's not:
image
For the "compatible Node" to be able to request data, it might need to be 2-way - so half-duplex...

There are many interface that use just one wire, and even call themselves "1-wire" - but they are not the same as the Dallas 1-WireTM.

Half-duplex UART would be such an example.

The interface to the well-known WS2812 LEDs is another 1-wire interface.

Added:

Amother example:

fair indeed

in the doc they mention

So it seems the device can transmit and receive

they mention as well that there is a Sensor Configuration Tool which can configure the device, so there is definitely a way to send information to it. The tool runs on windows and you need a specific USB cable to connect to the device

so now they mention RS-232 and 1-Wire :slight_smile:

--

coming back to this

I've read in the past about "sinking input to sensing device" in the context of 1-Wire communication (A "sinking" input implies that the device expects a connected wire or circuit to provide a path to ground (i.e., it "sinks" current). so may be it's just an enable pin that you drive LOW to activate the component ?

1 Like

They also call it 1-Wire Bus as multiple DS18B20 sensors and the like could be connected in parallel with the same bi-directiional signal wire.

To me, 1-Wire Serial Interface clearly refers to a UART Port equipped with TXD or RXD line only.

Dang... y'all boys are good -- REAL GOOD! This is the info I was dreaming of getting, but doubted anyone would care enough to give... THANKS - this is great stuff.

Just using logic, is there a chance this thing might constantly be sending (TX) an 8-bit reading, but using 1-bit "start" and "stop" signals so the receiver would know when to start reading a particular 8-bit data-point? as in, it sends a singe "1", then delays 50ms and sends "10101010," then delays 50ms and sends "0" and delays 50ms -- repeat, repeat, forever...

That would eliminate the need for the sensor to receive any commands. Because the sensor only does ONE thing, it can just constantly do it and when you ask for the "reading," it already forthcoming.

if it's a Serial 9600 bauds 8N1 communication, chances are there is a protocol to send the information

for example, the A02YYUW distance sensor has two lines Rx and Tx

When "RX" floats or is HIGH, the module outputs processed value, the data is more steady but the response time can be 300ms. When Rx is LOW, the module outputs real-time value at 10Hz with no processing.

the frame sent has 4 bytes

image

so you need a small parser to extract the distance.

I'd say that's extremely unlikely.

It sounds like a standard half-duplex RS232 (UART) interface, so there would have to be some protocol that determines when it sends, and when it listens.

Possibly this is where the 'device select' line comes in ... ?

MODBUS is also mentioned in that datasheet ...

EDIT

According to the datasheet, the Sensor sends its data when requested over the 1-wire serial interface::
image

So it seems it's a "polled" operation

Sure there is a chance but if you buy one for $300 and it does not work that way then what?

...AND THAT is why I'm asking people with much more expertise than I possess!! LOL!

I'm going to look into the "MaxBotix" you mentioned, but still want to pursue using Banner sensors

You really need to be asking Banner Engineering - it's their product, so they're the ones that will know.

You need to ask them for the interface specification for this "1-wire serial interface".

1 Like

indeed - may be it's all proprietary and you need to buy one of their nodes as well

that certainly seems to be the way they want you to play it!

Thanks to all y'all who helped on this, but here's THE OFFICIAL RESPONSE FROM BANNER:

Hello,

The 1-wire interface on our devices is the proprietary Banner interface. Our sensors with 1-wire interface only work with our wireless radios with 1-wire serial interface.

If you have any other questions, you can reply to this email, chat through the web site, or call 800-450-8144 and ask for me. If I'm unavailable, please leave me a voice mail, or if you need immediate assistance, any one of our other Application Engineers can help you out.

Thanks,
Nha Phan

OK that does not mean you can't hack it, but it's a different undertaking...

True, But at $300, it's an expensive hack. I think some of Y'ALL might be able to do it, but the risk of failure is to high for me.

I'm assuming these "reputable" sensor manufacturers are aware of our hacking skills and have built the newer models to not work with Arduino/Pi/ESP -- lot's of lost profits! Have you ever met anyone at the "national chain" electrical supply businesses that knows anything?? NOPE! This way only the "Banner Engineers" can help you and they lead you right to another "Banner product" -- The "salesmen" are just there to place the order.

Have you checked out Maxbotix

the cost of hacking the device is probably higher than buying the node anyway :slight_smile:
it would make sense only if you needed lots of sensors and of course as the protocol is proprietary it could break at every update.

They did answer the question you asked - which was, "is this the AD/Maxim/Dallas 1-WireTM ?"

As suspected, the answer is, "No."

I would ask the follow-up question of whether they would provide the specification(s) for their proprietary Banner interface, so that you can implement it on your own system.

They may well say no - but there's no harm in asking ...

Probably not.

More likely, they don't want the hassle of having to publish & maintain public documentation, and to provide support for people asking them questions about it. Much simpler (ie, cheaper) to just keep it all in-house.

I suspect that they also feel that there's going to be very little demand from the Arduino community for a sensor that costs $300!