An RS-485 board designed with Ethermania

[Dario Di Maio]

Hi,

we are back on an previous project started by Marco Signorini from Ethermania, is the design of RS-485 boards. We are working with him in the general design of features and components, and working on a collision avoidance driver for vNet, inspired by uLAN.

We are not planning to change the design, because we are taking things simpler as we can. But gives here your comment, so that will inspire the next release.

http://www.souliss.net/2013/09/usart-driver-for-vnet.html#more

Regards,
Dario.

[Jbllea]

Hello

The idea is great, but it would not be better to create a separate relay board? so the RS-485 would decrease in size and above could command a lot more relays.

As for the inputs (provided that they are not expert in electronics) just make an entrance, or if we start from a voltage of 12 volts which is sent to all buttons, we place 1 resistor at each exit of the buttons by doing so the first button will come out a voltage of 11.8 volts per second 11.6 etc., so we can recognize the button and activate the corresponding relay, the real problem would be if they press 2 at the same time! but this is just an idea I do not know whether it is possible or not! !

and above all, sorry for my bad English

[Di Maio, Dario]

The main idea is to have a small device that can serve a single room that could be easily used, there is a third board that will be used only for the inputs, maybe we can place there a connector for expansion (like external relay boards).

The inputs are powred type, you have IN+ and IN- and shorting these two the input is recognized, you don't need external resistor for pull up and down the pins.

The boards has a 12V to 24V switching regulator, but the relay one can be supplyed only at 24V due to relay coils voltage.

Regards,
Dario.

[Jbllea]

ok, then I stop my previous comment to "The idea is great" :-),

has a big potential! keep us updated

[Dario Di Maio]

This is the board PCB designed by Marco, that is almost the final one, has an Atmega328 with header for an external USB programmer. The power supply use a DC/DC switching regulator (that's why use a lot of space) and the communication use a standard RS485 transceiver.

There are 12 MOSFET in open-drain configuration with a protective fuse for a maximum rate of 2.5A per channel and 3 digital inputs that can be connected directly to a standard wall swtich without any additional external component.

Enjoy and give us feedbacks.

Regards,
Dario.

[Flavio P.]

W

​hat are the size (W,L,D) of the board?

Flavio​

2013/9/15 Dario Di Maio <dario....@souliss.net>

--
You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

--

P ********* Consideriamo l'ambiente prima di stampare questa e-mail *** Think about environment before printing *********

Il contenuto di questo messaggio e-mail è privato e riservato al solo destinatario. L'utilizzo non autorizzato può costituire reato. Se ricevuto erroneamente siete pregati di cancellarlo e comunicarci via e-mail l'errata ricezione.

[Di Maio, Dario]

Are approximately same as UNO, with 3x2.1 inches the relay and LED boards will not have exactly the same lenght.

Regards,
Dario.

[Flavio P.]

I think there are some difficult to insert the board on a 503, and you?

From Mobile Nexus

--

[Di Maio, Dario]

It will fit, but there will be really few space for cables. The PCB looks pretty full, I guess that there were no other way for Marco while sizing it.

What do you think about specs?

Thanks,
Dario.

[Flavio P.]

Why do you use rs485? and power 24v??? They require one or two cable.... I think you can use a switching with input 220v AC and 12 or 5v DC for output in the same dimension....

From Mobile Nexus

--

[Di Maio, Dario]

The board is designed to run with Souliss but also to be used easily with other libraries, there were basically two options for the bus:

1) RS-485

2) CAN

I like a lot CAN, but is less common used than RS485 in the Arduino world and is most expensive. That's why there is an RS-485 transceiver. Likely this makes the life harder while using it in vNet, because electrically you cannot have collision detection, so we are working on a collision avoidance software solution.

There is also a nice way to use RS-485 driver to build a network that is similar to CAN but doesn't allow to get longer distances, these two links gives more details about
http://en.wikipedia.org/wiki/J1708
http://ww1.microchip.com/downloads/en/AppNotes/01230A.pdf

Using RS485 will give a way to use the boards also out of Souliss, that is really important for Marco, because we don't have yet enough visibility to drive users over an hardware solution.

The power supply use a DC/DC switching converter, so is in the range 12V - 24V. For the relay board, are used 24V relays (coils) so the board can work only at that value, but basically just changing the relays you can get a 12V version. The LED board should run at 12V due to LED strips, but using two strips in series you can power the board at 24V.

The use of the switching is due to lower heat dissipation, a typical linear regulator at 12V is pretty hot and cannot go over. Probably is easier found an AC/DC at 12V rather than 24V I agree, but we can move from 24 to 12 just using proper relays, without any change in the PCB.

It will require two cables, one for power supply and one for data. There are two options, run a power cable around the home or have a local power supply near the board. In the first case you cannot run at 12V because the voltage drop will create troubles after few tens of meters.

Does I have answered your questions?

Dario.

[Flavio P.]

And why you want to use 328p again?

2013/9/15 Di Maio, Dario <dario....@souliss.net>

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

I didn't got the question. For the scope the 328 has enough RAM, in case of CAN there are microcontroller with a build in transceiver for CAN.

Is nice to see that if you have CAN, you haven't I2C. Both protocols are electrically quite similar, they works on a pulled line and drawn current to transmit data using an open-drain/collector device.

Dario.

[Flavio P.]

I'm thinking for a microcontroller with built-in rf transceiver

2013/9/15 Di Maio, Dario <dario....@souliss.net>

I didn't got the question. For the scope the 328 has enough RAM, in case of CAN there are microcontroller with a build in transceiver for CAN.

Is nice to see that if you have CAN, you haven't I2C. Both protocols are electrically quite similar, they works on a pulled line and drawn current to transmit data using an open-drain/collector device.

Dario.

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Flavio P.]

L​

ike ​ATmega256RFR2 for example.....

An innovative board!

2013/9/15 Di Maio, Dario <dario....@souliss.net>

I didn't got the question. For the scope the 328 has enough RAM, in case of CAN there are microcontroller with a build in transceiver for CAN.

Is nice to see that if you have CAN, you haven't I2C. Both protocols are electrically quite similar, they works on a pulled line and drawn current to transmit data using an open-drain/collector device.

Dario.

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

Yes, the wireless is also a way that can be taken in account, Atmel has a series of micro with built in 0.9 or 2.4 Ghz radio.

Some months ago I've discovered also uracoli that is a library for such micro, I've never used but it con be integrated as alternative to chibi one.

Regards,
Dario.

From mobile.

[mc bit]

You have never seen this?

2013/9/16 Di Maio, Dario <dario....@souliss.net>

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

I'm still waiting for the free samples that they promised me... arghhh :)

I've a couple of Zigduino for testing the stack that use the same family of microcontroller.

Dario.

From mobile.

[Flavio P.]

And you have tested? 

The main advantage are the dimension and not need bus cable.(i like the cable,but the community seems not like too)

What do you think?

Flavio 

From Mobile Nexus

2013/9/16 Di Maio, Dario <dario....@souliss.net>

I'm still waiting for the free samples that they promised me... arghhh :)

Dario.

From mobile.

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

Actually we support chibiduino, that is same of Zigduino but with the radio not embedded into the micro.

I've not yet tested the Zigduino due to lack of time, but after the last bugfix that we have done on the stack, chibiduino is pretty reliable.

Me too I like wireless but some people are scared about.

Dario

Dario.

From mobile.

[Flavio P.]

The rs485 board,have the same installation difficult of most other board or domotic system.

The winner idea, for me, and i repeat, for me :), is to think an hardware like zigduino, with power output, mosfet or relay, but in a little dimension. This is the better hardware for souliss node.

I don't think that are useful to have 12 channel mosfet on board!

And a board with only input for the beginning i would avoid.

I would focus on the size and wireless, to keep her on a 503 comfortably, on the rapid start-up and on the cost.

One mainboard, to be used as input (without mounting the relays), as input-output (with relays), and an additional mini-board with 3 mosfet (but 12A not 2,5A) to use it as RGB LEDs, high power LED,motor control, solenoid control, connected to the mainboard only by a little cable(like 4 pin strip).

This is my thought :)

2013/9/16 Di Maio, Dario <dario....@souliss.net>

Actually we support chibiduino, that is same of Zigduino but with the radio not embedded into the micro.

Dario.

From mobile.

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

I like wireless, but RS485 is cheaper and doesn't scare people about EMI. In the starting idea of Marco there was a bus wired connection, so we moved on that way, but a wireless board could be of interest.

For the MOSFET, have multiple channels with lower rate allow to have multiple light points without have to wire all togheter to get a big series LED of tens of ampere. But I've to admit that you board is also interesting.

Regards,
Dario.

[Flavio P.]

I like the wired connection, and for me are the best solution.....But i think all other people like wireless (no new cable), and for Souliss, the interest of the people are importart, to increase user!

For the mosfet, i think to a "normal" room, almost never have more than 3 light point, and if you have only one or two light point, with led you need about 6-8A for point. With 2,5A you power a little led, an this one cannot light your room. If user don't use led lighting, there are the relay board, for the other type of light.

Flavio 

2013/9/16 Di Maio, Dario <dario....@souliss.net>

I like wireless, but RS485 is cheaper and doesn't scare people about EMI. In the starting idea of Marco there was a bus wired connection, so we moved on that way, but a wireless board could be of interest.

Regards,
Dario.

--
You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[EtherMania]

Hi

When started this project we had in mind to propose a set of boards able to communicate each other in an easy way. We studied some protocols like, for example, the ethernet or zigbee, but at the end of our analysis the RS485 was the winner.  The main reason is that it's widely adopted on industrial automation, robust and very cheap to be implemented. Ethernet is the best protocol we can actually use in a lot of environment but, for me, it's not easy to deploy (each node requires a port in a switch and this could be very difficult to implement in old buildings). RS485 devices could be connected together by mean of a twisted cable. They could be all in parallel and there is not predefined topology (on the contrary, the ethernet has multi-star topology).

RF is a very good solution because... it's wireless. But I don't think is the only solution involved in a system: normally each node requires power supply and this is more true if you think to nodes that are used to switch lights, move blinds and so on. All these loads require 230VAC, so these type of nodes require cables at least for the power and this means that it's possible to reach them though one more cable with data. I think RF is the best solution for sensors and/or command pushbuttons but they are only part of the whole system. The big problem with RF is that is not cheap: I know that there are cheap Tx and Rx modules... but they require to be inserted in a more complex board.. and that board should be sold. This involves a lot of processes that, for RF, are very expensive. We actually can't afford these costs until we understood how big is the market. I'm not saying that we never be able to do, but I'm saying that actually we can't.

Saying that, what we now want to propose is the starting point for a complex system where multiple protocols and technologies are running together. I think the correct answer is to alway use the right tool for the right job. RS485 solves part of the targets me and Dario identified one year ago. One of that was to have a board that easily fit on a standard 503 wall mounted box. 

The relay board dimensions are 78.5x54mm. A standard 503 board is able to contain a 95x57mm so I think the fist goal has been reached (we have to check for the relay availability but I don't think there will be problems for that).

Our idea is to have more than one for each room you have to control. Each board provides three digital inputs and three outputs and they could be used, for example, to move blinds on a window and check for anti theft sensors for that window.. and you will have a spare relay for controlling something other if you need. 

The LED board is something Dario asks to have. I think having 12 channels by 2.5Amps is better than having a single 6 or 8Amps channel. This is better because you can more easy deploy cables for 2.5Amps then for 8Amps and you can freely assign how LEDs are distributed on each of the 12 channels. If, again, for a problem on a line, a channel stops working, you will temporarily loose one of 12 channels instead of 1 of three. I don't think the cost for three 8Amps board would be very different to the one needed for a 12 small channels: the cost is not only related to MOSFET used, but you have also to invest on different screw-type connectors and so on. Handling 8Amps for channel exposes to a set of electric noise and problems I don't think we can easily overcome.

The third board we are working on will contain only three input and a set of three onboard RGB LEDs (and obviously an RS485 and power supply). The idea is to provide a base block  useful for convert a standard pushbutton (like, for example, the one you can mount on BTicino or Geviss styled wall mount buttons) to a more "intelligent" bus aware buttons. This would simplify the conversion of old style plants to a bus oriented system and provides a sort of colored feedback for whatever system status you want to show.

But this is not the end of the project. At EtherMania is already available an RS485 shield that could be mounted on top of a normal Arduino or Netduino systems. This allows people to create gateways between ethernet and the RS485 busses so they can control the bus from whatever system he want.

I hope these boards would be the starting point for a big project. Me and Dario we are investing a lot of our time in order to have a set of items we can propose to help interested people to actively participate to this small domestic revolution.

Starting with something simple and versatile is the key concept.

Thank you and best regards,

Marco Signorini.

[Di Maio, Dario]

Yours are good discussion point that we can use for further release of the board.

Almost for the wireless that in my point of view is a plus.

Dario.

From mobile.

[Flavio P.]

Hello Marco,
I'm very pleased to know you! ;)
I'm not saying you are in a wrong way! I'm just thinking to a real application of a normal user....i think that an RGB strip in any normal home can be used only for scenic effect, but i agree with you to the RGB indicator! But 2,5A are wasted in this case....
RS485 are pretty good for me, and in my home the cable are waiting....but i'm ever thinking to a normale user....for the user cable=tecnician=money=less probability to use souliss!
I say for a wireless bus only for this reason....

​For the mosfets, I tried to enlighten my room with LED 2.5 A, and should make many points of light, difficult to achieve in reality, then in my opinion we need to focus on a single point light with adequate power, and now means use 6-8A....​

​I have nothing against your work, in fact I appreciate it, and I'm here to try to help you

!

​I hope to do it!!!!!!!

Flavio​

​

From Mobile Nexus

[Di Maio, Dario]

Using Souliss you can even build a network mixed of many Rs485 links bridged wirelessly (a chiniduino and a 485 shield).

IMHO feedbacks help every time, no matter if the design in your mind doesn't mach with ours.

Is every time a treadoff :)

Dario.

From mobile.

[EtherMania]

Hi Flavio. Nice to meet you too and thank you for your answers!

I've understood your point and opinion and my answer has to be read in the sense of clarify what are the technical and feasibility reasons me and Dario based our choices.
For what is related to the RGB indicator I think there is a misunderstanding. There will be a third type board without MOSFET but with three RGB LEDS already mounted on it (or accessible through pin-headers). So there is no need to buy a 12 channel MOSFET featured for the sole purposes to have LED indicators. The board will be cheaper and smaller than the MOSFET based one, so it will easily fit on the back of a standard push buttons in a 503 sized box. I'm working these days on it so probably more details will follow in a few days/weeks.

My general idea is to start to propose something real and ready to be programmed for two reasons: the one is that I need to install some of these boards in my house... the second is because I like to do this and I need to find the funds to continue to do it.

So feedbacks are welcome and my consultancy job is to aggregate them in a something either technically then economically and time-limited feasible.

Thanks!
Marco Signorini

[Di Maio, Dario]

For my actual job we was looking for absorption of FGS horn and bells to be drived with an open drain MOSFET. As is designed the board, you can even use more channel in parallel, up to drive a mega LED at 12V and 30A :)

Dario.

[Flavio P.]

....very good! you have the connection detail for the mosfet?

Flavio

2013/9/18 Di Maio, Dario <dario....@souliss.net>

For my actual job we was looking for absorption of FGS horn and bells to be drived with an open drain MOSFET. As is designed the board, you can even use more channel in parallel, up to drive a mega LED at 12V and 30A :)

Dario.

--

You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

All MOSFET are open drain, so you just need to connect the DIODE cathode to more than one channel.

Of course, if the software goes wrong and only one channel is activated, the risk it to burn that channel out.

Dario.

[Flavio P.]

Hello Marco,

sorry for misunderstanding

ok, now i understand right the 3 RGB LED indicator board :) 

I'm waiting for your hardware, i'm very curious!!!

But what is the target price do you think to have this board?

Flavio

From Mobile Nexus

2013/9/18 EtherMania <sales.et...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "souliss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to souliss+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Di Maio, Dario]

Attached.

Dario.

[Dario Di Maio]

Just today I get from Marco the rendering of the boards, we worked on the ground plane of the LED one. Now there are two planes joined at the output contact, using faston type connector.

http://www.souliss.net/2013/09/3d-rendering-for-new-rs-485-boards.html#more

Regards,
Dario.

On Saturday, September 14, 2013 10:03:22 AM UTC+2, Dario Di Maio wrote:

Hi,

we are back on an previous project started by Marco Signorini from Ethermania, is the design of RS-485 boards. We are working with him in the general design of features and components, and working on a collision avoidance driver for vNet, inspired by uLAN.

We are not planning to change the design, because we are taking things simpler as we can. But gives here your comment, so that will inspire the next release.

http://www.souliss.net/2013/09/usart-driver-for-vnet.html#more

Regards,
Dario.

[Dario Di Maio]

Just received from Marco the rendering for the bridge between the RS485 and the Ethernet network.

[Alessandro]

awesome work. I'm sorry my knowledge in electronics is too low to comment in a propositive way

Alessandro

2013/9/30 Dario Di Maio <dario....@souliss.net>

--

[Di Maio, Dario]

At least I hope that our is enough to make these boards working :D