I have been thinking a bit about creating a stand-alone HMI component. Basically a terminal with a touch display and optional keyboard, mouse and customized input equipment. The key principle would be that we use a Ethernet/RSX as bacbone for data traffic, while the HMI takes care of graphics, input etc.
This is just some random GIF I found on the net, but an HMI app could consist of a set of standard components where layout is controlled by XML files that are uploaded from the device. Communication could be Ethernet/RSX. Equiped with a Plain VM this could be an interesting concept where HMI logic is executed on the HMI unit and only events/data transferred to/from the rest of the system.
Making the app above might seem complicated, but it’s nothing I have not done before from scratch in C/C++. I am thinking more and more about creating a standard HMI app in Qt since that can run on Windows, Linux, Android, OsX etc. I need to check around for options.
Since we are talking about a network of stand-alone units we could create a full control center as well.
This picture is from CubeMX, just a few clicks and I have assigned 6 x UART’s, 3xSPI ports, crystal, SWD + added USB – the IO capabilities on the F405 never stop to amaze me + you gotta love CubeMX. Be a bit on your watch because not every possibility is available through CubeMX. Another tool I like is the 3D Paint version in Windows.
This is for the RS-X Switch.
I still got a few components to make\upgrade before I have a full Home Automation network. The CCM and RSX module allows me to create a central, but I need sensors and I need to feed the sensors power. One solution to this is to create an active RSX Switch with power output on each network.
The RS-X Switch as drawn here will take 2 x Ethernet or 2 x RS-X (M1 & M2) input and provide 4 x RS-X output with power. Each of RS-X E1,E2,E3 & E4 can be switched on/off separately and be used for networks or 1:1 with sensors/actuators.
The functionality in the switch is simple message switching. The idea of having 2 x Ethernet and 2 x RS-X Input is redundancy, but I realise this might be a bit tight to achieve so will see what pins & space I have available.
This module will also be a great RS485 Hub used stand-alone.
3D of a 3-port RS485 GW. This actually have 5 RS485 (or RS-X). 2 on the backbone bus running high speed, and 3 isolated with lower speeds in front. These can be used Standard RS485, RSX or even Profibus/485.
The MCU is actually capable of running 3 port Profibus using UART’s saving the need for specialised hardware.
This module do not have the Raspberry PI Zero W add-on option. The reason is because the connector would conflict with the isolation area.
Added 2 coils on the 16 x Servo Module as well. On this I have plenty of space. I should have moved the MCU a bit down, but I will leave it for now.
This module provide output only, meaning it can not do generic IO as my previous 32xServo/IO could. The limitator is the opto couplers providing a full galvanic isolation. It can still do digital output and low frequency PWM (< 100Khz) in addition to Servo signals.
Servo PSU is selectable from 5, 12 & 24V by jumper. Signal is also selectable from 5, 12 & 24V by Jumpers. The MCU 5V is separate from the Signal/Servo 5V etc.
It is also has a 4 pin HMI – the same as on the CCM board. I simply did not see any point in removing it + might be cool for demo/testing.
Doing 16 channel Servo on a STM32F405RG is well, the reason I use a M4 is the speed of the backbone bus.
Not very exiting – just added the 3 extra 15uH coil options I mentioned earlier on the back. As mentioned these are added because I suspect I need them to separate 5V PSU to different high frequency modules. The issue, as we learned earlier is that these freuencies interact. The coils here simply let 5V through but stop higher frequencies – it cost me little to just add the space for these on the back side.
I have a weird package error on the 25 Mhz crystal that I need to check, otherwise this is ready to go.
Also starting to use “CCM” (Communication Central Module) as name on this because it contains Ethernet, GSM, GPRS, GPS, WIFI, Bluetooth, USB++
I updated the module list again – don’t worry – this will not be the last change here.
CCM (Communication Central Module) is the new name for the combined Ethernet, Wifi, Bluetooth, GSM/GPRS, GPS, RPI Module. I packed so much into this that it covered 3 planned modules.
Servo Controller Module was illustrated earlier.
I need to work on PSU/Battery.
3 x RS-X Module comping up.
8 x Analogue Input Module coming up.
PWM Output Module coming up.
Digital Input Module coming up.
DC/Stepper Controlles – we probably talk several modules here.
Sound I/O – now this becomes interesting. We have so far talked about a PLC that control robotics, but we could easily talk about a system that also does sound processing… I am very, very thin on analogue electronics, but some interesting options here.
These two diagrams show drafts of analogue data acquisition modules. The first is a 8 channel 24bit @30Ksps while 2nd is a 16 channel 12bit at 2,5Msps.
The ADS1256 is available for ca 5.- USD and provide 8 channels with 24 bit resolution at a sampling speed of ca 30ksps. This is very good for an ADC with this high resolution.
The faster ADC technique is achieved by using the ADC’s integrated in STM32 directly. These are 12 bits, but have a much higher sampling rate capability (2,4Msps).
I am not sure if I want to make both boards + I need to dig in a bit on analogue scaling & calibration options. I also need opto isolators that have limitations of their own. I also have the issue that the higher frequency of the 12 bit is of little usage + 12 bit resolution is a bit low for sufficient accuracy. Assuming I have space I could actually extend the 24 bit board with a few faster 12 bit channels as well.
This is a block diagram of the All-In-One Home Central. This is basically a merge between two PLC cards to create a smaller, mobile bread & butter node. I am not going to do anything on this yet – I simply want the MicroPLC modules up running first since this is re-use of the same technology.
This also provide a full wireless RS-X switch as an alternative to the low cost wired one. The RPI module have sufficient juice for a secure wireless connection.
IoT is all about wireless connections these days. The main challenge is that providing sufficient security becomes a challenge becausse you need some computing power for secure encryptions – which easily drive cost and size. Myself I prefer to use secure wireless links to cover distances between buildings and floors, while I use more classic wired networks locally.
To enable this I need two new components to my Home Automation system:
This is a basic PLC module to provide 3 x External RS-X networks. I will create a all-in-one with these integrated later, but we can stick to the PLC backbone for now. Isolating everything will drive cost and size, but I want these 3 connections isolated because they will support long distance wiring exposing the entire system for lightning or pulses from other quipment. I will only get 3 networks on each card, but I can stack loads of cards if I need to.
The second component is an Active RS-X Switch. This will bridge between an isolated network to a short distance non-isolated with 12V power added. With 4 connectors we can connect 4 devices or add passive switches to create entire sub-networks. We can also add another Active Switch by chaining these up to create larger networks.
The main advavtage is that we add power locally and don’t attempt to send DC power over any distance. Power is best transported as AC. We also create barriers where a lightning strike or other spikes will only have limited impact while we still can add low cost sensors in numbers. I created sensors and passive switch earlier that I will make upgrade revisions of.