We have an issue with Modbus over RS-485 where the controlled
pump seems to be insisting on us padding each command with
leading/trailing filler bits.

Anyone experienced with such err features?

I am looking at
<http://www.modbus.org/docs/Modbus_Application_Protocol_V1_1b3.pdf>
and nearby. Is there a more definitive document defining
ModBus over RS-485?
--
A host is a host from coast to coast.................wb8foz@nrk.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

tirsdag den 8. juni 2021 kl. 22.05.43 UTC+2 skrev David Lesher:
> We have an issue with Modbus over RS-485 where the controlled
> pump seems to be insisting on us padding each command with
> leading/trailing filler bits.
>
> Anyone experienced with such err features?

does the bus have fail safe bias resistors?

On 8.6.21 23.05, David Lesher wrote:
> We have an issue with Modbus over RS-485 where the controlled
> pump seems to be insisting on us padding each command with
> leading/trailing filler bits.
>
> Anyone experienced with such err features?
>
>
> I am looking at
> <http://www.modbus.org/docs/Modbus_Application_Protocol_V1_1b3.pdf>
> and nearby. Is there a more definitive document defining
> ModBus over RS-485?

The Modbus/RTU framing is a good candidate for the sickest way
of defining packet boundaries.

In addition to the pull up/down resistors mentioned by Lasse,
are you sure that the interframe timings are met?

A serial interface chip with internal FIFOs will quickly mask
the timing of the octets on the line, so the software has
very limited control.

--

-TV

Tauno Voipio <tauno.voipio@notused.fi.invalid> writes:

>The Modbus/RTU framing is a good candidate for the sickest way
>of defining packet boundaries.

That's ....encouraging.....

>In addition to the pull up/down resistors mentioned by Lasse,
>are you sure that the interframe timings are met?

>A serial interface chip with internal FIFOs will quickly mask
>the timing of the octets on the line, so the software has
>very limited control.

The bothersome client uses a ADM2486. I can't recall what out end is.

On 10.6.2021 01:36 AM, David Lesher wrote:
> Tauno Voipio <tauno.voipio@notused.fi.invalid> writes:
>
>
>
>> The Modbus/RTU framing is a good candidate for the sickest way
>> of defining packet boundaries.
>
> That's ....encouraging.....
>
>> In addition to the pull up/down resistors mentioned by Lasse,
>> are you sure that the interframe timings are met?
>
>> A serial interface chip with internal FIFOs will quickly mask
>> the timing of the octets on the line, so the software has
>> very limited control.
>
> The bothersome client uses a ADM2486. I can't recall what out end is.

About a quarter century ago, I had to code a Modbus/RTU master/slave
unit with MC68332 and four 16C550 UART's.

I had to disable the internal FIFO's and harness a pair of 68332
timers per Modbus channel, to get the timings to match the
specification.

The ready unit had enough processing power to handle the increased
interrupt load, in addition of the other things to do. The unit
did obey the spec, we tested it with the customer. In the same
testing session, we noticed that there were no PC Modbus handlers
performing to the timing specifications.

--

-TV

Tauno Voipio wrote:
> On 10.6.2021 01:36 AM, David Lesher wrote:
>> Tauno Voipio <tauno.voipio@notused.fi.invalid> writes:
>>
>>
>>
>>> The Modbus/RTU framing is a good candidate for the sickest way
>>> of defining packet boundaries.
>>
>> That's ....encouraging.....
>>
>>> In addition to the pull up/down resistors mentioned by Lasse,
>>> are you sure that the interframe timings are met?
>>
>>> A serial interface chip with internal FIFOs will quickly mask
>>> the timing of the octets on the line, so the software has
>>> very limited control.
>>
>> The bothersome client uses a ADM2486. I can't recall what out end is.
>
>
> About a quarter century ago, I had to code a Modbus/RTU master/slave
> unit with MC68332 and four 16C550 UART's.
>
> I had to disable the internal FIFO's and harness a pair of 68332
> timers per Modbus channel, to get the timings to match the
> specification.
>
> The ready unit had enough processing power to handle the increased
> interrupt load, in addition of the other things to do. The unit
> did obey the spec, we tested it with the customer. In the same
> testing session, we noticed that there were no PC Modbus handlers
> performing to the timing specifications.
>

The 332 was a very nice chip for its day.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com