> Don Y <blockedofcourse@foo.invalid> writes:
>
>
>> I don't see a market, there. Consider: anyone who already has
>> a 4-20mA loop in place would have to put converters on both
>> ends of the (existing) wire -- one to convert the transducer
>> to digital and another to convert the digital back to 4-20mA
>> to i/f to <whatever> (PLC?).
>
> Rather a 4-20ma converter at the sensor, an existing
> STP of ~500M in length, and a new digital controller
> replacing dozens of museum-quality PID controllers.
Assuming the industry being served allows replacing kit
without also revalidating the process.
Reply by David Lesher●November 27, 20232023-11-27
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> writes:
>If it's compatible with automotive Ethernet, you can get adapters from
>Amazon. Not super cheap.
>We recently did that with a couple of lidars.
I suspect those were 10BaseT1S, suitable up to 15M distances.
--
A host is a host from coast to coast...............wb8foz@panix.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
Reply by David Lesher●November 27, 20232023-11-27
Don Y <blockedofcourse@foo.invalid> writes:
>I don't see a market, there. Consider: anyone who already has
>a 4-20mA loop in place would have to put converters on both
>ends of the (existing) wire -- one to convert the transducer
>to digital and another to convert the digital back to 4-20mA
>to i/f to <whatever> (PLC?).
Rather a 4-20ma converter at the sensor, an existing
STP of ~500M in length, and a new digital controller
replacing dozens of museum-quality PID controllers.
--
A host is a host from coast to coast...............wb8foz@panix.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
Reply by Joe Gwinn●November 27, 20232023-11-27
On Mon, 27 Nov 2023 01:08:35 -0000 (UTC), David Lesher
<wb8foz@panix.com> wrote:
>Joe Gwinn <joegwinn@comcast.net> writes:
>
>
>>Well 10BASE-T1L and 10BASE-T1S still exist, and the latest version of
>>the standard was approved in 2019, so it's still active. Lots of
>>people make chips. Google on 10BASE-T1L to find many.
>
>Many What? I'm seeking things I can put in a NewEgg or Microcenter
>or ? shopping cart.
The point is that if companies like Analog Devices are making the
chips, there necessarily are many other companies buying the chips and
incorporation them into things you can buy. And the automotive market
is likely to be very large.
But you will need to do your own research to find candidate products
and determine which ones are suitable.
>>Note that if twisted pair is not required, the fiber-optic
>>options are winning big in big factories because of their total
>>immunity to EMI.
>
>That is quite true, but the value of the installed base STP in
>plants is huge. Not everything needs GB/s.
See the "Note that if twisted pair is not required" qualifier. There
are many folk who don't have that limitation, and knowing where the
market centroid is may be useful.
Joe Gwinn
Reply by Phil Hobbs●November 26, 20232023-11-26
David Lesher <wb8foz@panix.com> wrote:
> Joe Gwinn <joegwinn@comcast.net> writes:
>
>
>> Well 10BASE-T1L and 10BASE-T1S still exist, and the latest version of
>> the standard was approved in 2019, so it's still active. Lots of
>> people make chips. Google on 10BASE-T1L to find many.
>
> Many What? I'm seeking things I can put in a NewEgg or Microcenter
> or ? shopping cart.
>
>
>> Note that if twisted pair is not required, the fiber-optic
>> options are winning big in big factories because of their total
>> immunity to EMI.
>
> That is quite true, but the value of the installed base STP in
> plants is huge. Not everything needs GB/s.
>
>
If it’s compatible with automotive Ethernet, you can get adapters from
Amazon. Not super cheap.
We recently did that with a couple of lidars.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC /
Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics
Reply by David Lesher●November 26, 20232023-11-26
Joe Gwinn <joegwinn@comcast.net> writes:
>Well 10BASE-T1L and 10BASE-T1S still exist, and the latest version of
>the standard was approved in 2019, so it's still active. Lots of
>people make chips. Google on 10BASE-T1L to find many.
Many What? I'm seeking things I can put in a NewEgg or Microcenter
or ? shopping cart.
>Note that if twisted pair is not required, the fiber-optic
>options are winning big in big factories because of their total
>immunity to EMI.
That is quite true, but the value of the installed base STP in
plants is huge. Not everything needs GB/s.
--
A host is a host from coast to coast...............wb8foz@panix.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
Reply by Don Y●November 26, 20232023-11-26
On 11/26/2023 4:11 PM, David Lesher wrote:
>>> Is 10BaseT1L real or just another nothingburger???
>
>> But that's not the question you asked! You were looking
>> for a media converter: something-base-something to 10BaseT1L.
>
> Sorry, I thought you meant 4-20mA converters.
I don't see a market, there. Consider: anyone who already has
a 4-20mA loop in place would have to put converters on both
ends of the (existing) wire -- one to convert the transducer
to digital and another to convert the digital back to 4-20mA
to i/f to <whatever> (PLC?).
There *might* be a market for devices that convert N (>> 1)
devices to digital for transport over the link and a
similar "demultiplexor" to convert them back to 4-20mA to
interface to existing i/f's.
But, the smarter move is to deploy this on newer installations
so there's no real economic incentive, going backwards.
> Yes, I seek 10BaseT1L to 10/100/1000BaseT1L
>
>> My point was that this would likely not be an initial
>> offering for a new market -- because there aren't
>> something-base-something drops in use in that application
>> domain, now.
>
> But given the extensive 100BaseT/1000BaseT deployment, a
> converter to 10BaseT1 is a starting point towards wider
> deployment.
I see a USB-based "adapter" to be more likely to find a
market. It lets it be deployed to PCs/MACs as well as
embedded systems ("If you can speak USB, you can get
this hardware/protocol capability for free!")
>> Instead, look for something with a *computer* (host) on
>> one side and a network interface (of your desired flavor)
>> on the other side.
>
> Won't fit this need. Must put printers in places with only
> STP cable to them.
Why wouldn't a USB dongle do the trick?
Why isn't 10/100/1000BaseT(X) useable (are there only two
EXISTING conductors in place that must be leveraged?)
>> As to how real? A quick search shows lots of hits for
>> devices targeting that market. But, (in my initial
>> search), more at the component level. E.g., TI, AD and
>> Microchip stand out.
>
> Yep. That was my starting point, not components but
> deployable products.
"PCI 10BaseT1L", "USB 10BaseT1L", etc. search terms?
>> You might check inventory (and lead time) for some of the
>> devices to see if it's all hype or if foundries are
>> actually committed to making silicon.
>
> That's true.
The dilemma, there, is "lack of stock" can indicate NO
demand just as easily as HIGH demand (insufficient supply).
Reply by David Lesher●November 26, 20232023-11-26
Don Y <blockedofcourse@foo.invalid> writes:
>On 11/26/2023 2:47 PM, David Lesher wrote:
>> Don Y <blockedofcourse@foo.invalid> writes:
>>
>>
>>>> 10Base1L {& S} is 100% Ethernet. But while I've read lots of
>>>> hype about it, I also did for other earth-shaking inventions
>>>> such as bubble memory, ATM, Token Ring, etc.
>>>>
>>>> So what, if anything, is actually available for 10BaseT1L
>>>> {Or for that matter, 10BaseT1S..}?
>>
>>> But why would you imagine a media converter as the solution?
>>
>> It is not, for me. I was explaining where 10BaseT1L came from.
>>
>> I'll try again:
>>
>> 1) I need Ethernet, 10baseT is fast enough.
>> 2) I must use an embedded 1 twisted pair cable that is in place.
>> 3) The solution would be 10BaseT1L, if in fact it is anything but vaporware.
>>
>> So:
>>
>> Is 10BaseT1L real or just another nothingburger???
>But that's not the question you asked! You were looking
>for a media converter: something-base-something to 10BaseT1L.
Sorry, I thought you meant 4-20mA converters.
Yes, I seek 10BaseT1L to 10/100/1000BaseT1L
>My point was that this would likely not be an initial
>offering for a new market -- because there aren't
>something-base-something drops in use in that application
>domain, now.
But given the extensive 100BaseT/1000BaseT deployment, a
converter to 10BaseT1 is a starting point towards wider
deployment.
>Instead, look for something with a *computer* (host) on
>one side and a network interface (of your desired flavor)
>on the other side.
Won't fit this need. Must put printers in places with only
STP cable to them.
>As to how real? A quick search shows lots of hits for
>devices targeting that market. But, (in my initial
>search), more at the component level. E.g., TI, AD and
>Microchip stand out.
Yep. That was my starting point, not components but
deployable products.
>You might check inventory (and lead time) for some of the
>devices to see if it's all hype or if foundries are
>actually committed to making silicon.
That's true.
--
A host is a host from coast to coast...............wb8foz@panix.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
Reply by Joe Gwinn●November 26, 20232023-11-26
On Sun, 26 Nov 2023 16:38:27 -0000 (UTC), David Lesher
<wb8foz@panix.com> wrote:
>Don Y <blockedofcourse@foo.invalid> writes:
>
>>4-20ma loop is a point-to-point technology. Why would you
>>expect to find 100BaseTX in that space?
>
>10BaseT1's driving justification was to replace 4-20 with
>Ethernet, using the existing single twisted pair cable up
>to ~1000ft. Obviously, the field end would need an A to D
>converter. But hello, it's no longer the 1950's, and process
>control ain't still vacuum tube Taylor PI analog controllers.
>
>10Base1L {& S} is 100% Ethernet. But while I've read lots of
>hype about it, I also did for other earth-shaking inventions
>such as bubble memory, ATM, Token Ring, etc.
>
>So what, if anything, is actually available for 10BaseT1L
>{Or for that matter, 10BaseT1S..}?
Well 10BASE-T1L and 10BASE-T1S still exist, and the latest version of
the standard was approved in 2019, so it's still active. Lots of
people make chips. Google on 10BASE-T1L to find many.
Here is the list of Ethernet physical media options:
.<https://en.wikipedia.org/wiki/Ethernet_physical_layer>
Note that if twisted pair is not required, the fiber-optic options are
winning big in big factories because of their total immunity to EMI.
Joe Gwinn
Reply by Don Y●November 26, 20232023-11-26
On 11/26/2023 2:47 PM, David Lesher wrote:
> Don Y <blockedofcourse@foo.invalid> writes:
>
>
>>> 10Base1L {& S} is 100% Ethernet. But while I've read lots of
>>> hype about it, I also did for other earth-shaking inventions
>>> such as bubble memory, ATM, Token Ring, etc.
>>>
>>> So what, if anything, is actually available for 10BaseT1L
>>> {Or for that matter, 10BaseT1S..}?
>
>> But why would you imagine a media converter as the solution?
>
> It is not, for me. I was explaining where 10BaseT1L came from.
>
> I'll try again:
>
> 1) I need Ethernet, 10baseT is fast enough.
> 2) I must use an embedded 1 twisted pair cable that is in place.
> 3) The solution would be 10BaseT1L, if in fact it is anything but vaporware.
>
> So:
>
> Is 10BaseT1L real or just another nothingburger???
But that's not the question you asked! You were looking
for a media converter: something-base-something to 10BaseT1L.
"I'm wondering re: media converters to existing 100BaseT
etc LANS. I've found zip so far... Anyone seem same?"
My point was that this would likely not be an initial
offering for a new market -- because there aren't
something-base-something drops in use in that application
domain, now.
Instead, look for something with a *computer* (host) on
one side and a network interface (of your desired flavor)
on the other side.
As to how real? A quick search shows lots of hits for
devices targeting that market. But, (in my initial
search), more at the component level. E.g., TI, AD and
Microchip stand out.
This makes sense, historically, for embedded devices as
you likely are talking from "inside" one such device
to other I/Os THROUGH an i/f on that device.
Consider how ubiquitous PC CAN bus interfaces are NOT.
Yet, folks routinely develop with CAN...
You might check inventory (and lead time) for some of the
devices to see if it's all hype or if foundries are
actually committed to making silicon.