Help with interesting control problem

An area for more general discussion by the user community for topics not covered elsewhere
Post Reply
chill
Posts: 5
Joined: Sun Jul 05, 2020 4:04 pm

Help with interesting control problem

Post by chill »

Hi folks, I'm looking for direction on creating an Arduino sketch/project to help solve a unique problem.

What I have are 8 separate IR Emitter signals coming in from 8 classrooms over HDMI baluns to an equipment rack. All 8 of these are pointed to one IR Receiver. This IR receiver is communicating it's received signals to a home automation processor. The home automation processor controls 6 devices located in the equipment rack via IP, RS232 or IR depending on the device.

Currently this works only if one remote control is operating the equipment at a time.

What I'm trying to build is a way to assign each of the 8 modulated emitter signals to individual pins on an Arduino, analyze the IR signal, filtering out any unnecessary signals such as TV/Projector commands, then buffer the rest of the data on those pins until IR signal transmission is complete. Once IR signal transmission is complete, fire only one emitter at the receiver. In theory this should allow each of the 8 classrooms to be controlled without crosstalk from any of the other rooms.

I do understand this will cause delays in IR signal processing, however that will not be an issue for this solution.

I get the gist of what needs to be done, but looking for advice from IR experts and enthusiasts.
User avatar
AnalysIR
Site Admin
Posts: 776
Joined: Sat Aug 31, 2013 3:51 pm
Location: Dublin, Ireland
Contact:

Re: Help with interesting control problem

Post by AnalysIR »

Several possible approaches....depends on how hardware is configured etc.

So first post a hardware connection diagram/layout, showing:
- how is the IR transferred over HDMI. (Modulated, demodulated, serial data or similar). Does is come out in the rack as IR (or ?)
- physical connections to Arduino(s)
- Why do you need everything pointing to 1 IR receiver.
- Summary of project/background info???
- max run length of HDMI cables & distance from end of HDMI cables to rack inputs.
chill
Posts: 5
Joined: Sun Jul 05, 2020 4:04 pm

Re: Help with interesting control problem

Post by chill »

AnalysIR wrote: Sun Jul 05, 2020 4:54 pm Several possible approaches....depends on how hardware is configured etc.

So first post a hardware connection diagram/layout, showing:
- how is the IR transferred over HDMI. (Modulated, demodulated, serial data or similar). Does is come out in the rack as IR (or ?)
- physical connections to Arduino(s)
- Why do you need everything pointing to 1 IR receiver.
- Summary of project/background info???
- max run length of HDMI cables & distance from end of HDMI cables to rack inputs.
Ill get a diagram together. I'll answer the questions above, but to start with a brief summary.

I was called in to solve control issues in a private school where they have 8 classrooms with either smart boards, or projectors in the room. They are controlling, in the equipment rack, 4 shared video/audio sources, and 2 "Matrix" switches to distribute audio and video to the rooms.

These are:
1. 10x10 HDMI Matrix Switch
2. 8 Channel Distributed Audio System
3. Dish Network Receiver
4. Samsung Blu-Ray Player
5. Apple TV
6. Intel NUC/Windows 10

They are using Harmony remotes in each room to control watch and listen "activities". The automation processor is in place to simplify the IR command macros by use of "Trigger Codes". When the automation processor receives a "Trigger Code" it runs either a series of IP commands to the video matrix switch or RS232 commands to set the distributed audio system to set the desired source. "Trigger Codes" are utilized to processes volume commands to the distributed audio system via RS232. The processor also "passes through" any IR commands received that it does not utilize as a "Trigger", for control of the AppleTV and Blu-Ray.

Now this system works great when you are controlling one room at a time. Unfortunately the school is not in a budgetary position to replace what is there and upgrade to a much better control system. I have to work in this box I have been put in.

- how is the IR transferred over HDMI. - The signal at the rack is modulated IR.
- physical connections to Arduino(s) - At this time none. Plan is to "wire" the IR signals directly to pins, eliminating all but 1 emitter that sends the buffered commands.
- Why do you need everything pointing to 1 IR receiver. - The home automation processor provides only 1 receiver input to process IR Trigger commands.
- max run length of HDMI cables & distance from end of HDMI cables to rack inputs. - less than 50 meters, HDMI runs. They are utilizing HDMI over Cat6 baluns. The receiving side of the balun (in the classrooms) has the demodulated IR receiver input. The transmit side of the balun (at the rack) has the modulated IR emitter output. The baluns installed are only capable of 1 way IR transmission. At the rack, the distance from the emitter output on the transmit balun to the processor IR receiver is about 1 meter total.

Attached is a basic layout of the control of the system. I only drew 2 rooms, and omitted HDMI and Speaker cabling to simplify the chart.

Colors depicting control topology on the layout are:
Red: IR Signal
Orange: RS232
Blue: Ethernet

I hope this helps.
CRBasic.png
You do not have the required permissions to view the files attached to this post.
User avatar
AnalysIR
Site Admin
Posts: 776
Joined: Sat Aug 31, 2013 3:51 pm
Location: Dublin, Ireland
Contact:

Re: Help with interesting control problem

Post by AnalysIR »

Some points for consideration:

- The IR signal at the rack, is not likely a full swing 0->5V (or 3V3)..as its used to drive an IR LED with a swing of ~1.8V. Also, it won't have GND( OR 5V), unless GND can also be provided by the HDMI box. Will probably need some additional electronics to get the signal into a GPIO.
- Essentially, you are looking for something to store & forward IR signals at the rack, when more than 1 signal source is active simultaneously.
- Although it would be feasible to code this on a standard Arduino...you could run into an SRAM issue (e.g. if everyone kept VOL+ pressed at the same time. So I would suggest using DUE or ESP32 or Teensy 4.0 for the solution. (Teensy 4.0 looks like best value by far with ridiculous amounts of horsepower)
- Essentially, you need to pass the first signal that arrives thru transparently. If another arrives within a certain time after the last you need to store it and send it after the active one is complete (+ a gap).
- Of course you need to write some firmware to do all this signal capturing (non-trivial / not for beginners as you need to simultaneously capture the carrier frequency for up to 8 signals in parallel. Do not underestimate this task).
- Is the HDMI connection a 2-pin 3.5mm Jack?


Alternative approach
- As you seem to have networks available???
- Using WiFi, you could use an ESP32 (or similar) to capture all the signals locally in each class room (obviously 1 at a time) and send them over WiFi to another ESP32 @ the rack. The unit at the rack would just send out each signal as it arrived (Essentially, using the network as the store and forward/queing mechanism)
- However, using this approach there would be a small but noticeable lag on EVERY signal sent.
chill
Posts: 5
Joined: Sun Jul 05, 2020 4:04 pm

Re: Help with interesting control problem

Post by chill »

Thank you, I was concerned about the IR emitter voltages, as you say, they are less than 3V. Someone mentioned that if I had the emitters "taped" to 8 receivers, then I could parallel the receivers into the processor.

I would Imagine the same problem would exist and would still need to capture data in parallel.
User avatar
AnalysIR
Site Admin
Posts: 776
Joined: Sat Aug 31, 2013 3:51 pm
Location: Dublin, Ireland
Contact:

Re: Help with interesting control problem

Post by AnalysIR »

I was concerned about the IR emitter voltages, as you say, they are less than 3V. Someone mentioned that if I had the emitters "taped" to 8 receivers, then I could parallel the receivers into the processor.
Yes, using an IR receiver would address that problem. However, every time a signal goes thru a standard receiver it gets distorted by up to +/- 200uSecs per mark or space.

If you used a demodulating (standard) receiver, you would lose the carrier frequency, which could impact performance when it is finally delivered to the target device. That is you would have to re-transmit every signal at a fixed frequency, say 38kHz. It may work, but I suspect with some performance issues on current or future devices.
If you used an IR Learner then this issue would be addressed in the best way possible thus retaining the frequency.
they are less than 3V
They should be around 1.8V or less. but not necessarily reference from GND or 5V/Vcc. However, nothing s transistor or 2 couldn't correct for reliable input into a GPIO port.

Using an IR Learner (TSMP58000) would be the best & most reliable approach, which will require some good firmware to be written and the use of a capable MCU.
chill
Posts: 5
Joined: Sun Jul 05, 2020 4:04 pm

Re: Help with interesting control problem

Post by chill »

Maybe there is an easier solution surrounding firmware writing... after thinking this through, it may just be possible to ignore emitter output when the receiver is busy. The user of a second, third, forth, etc... remote would simply press the command button on the harmony again. Clunky but may just do the trick.

EDIT: I would guess that I would have 8 emitters pointing to 8 receivers then "ignore" the receiver side of the circuit.
User avatar
AnalysIR
Site Admin
Posts: 776
Joined: Sat Aug 31, 2013 3:51 pm
Location: Dublin, Ireland
Contact:

Re: Help with interesting control problem

Post by AnalysIR »

If you block any simultaneous signal that arrives while the 1st is active, then that makes it a LOT simpler.
EDIT: I would guess that I would have 8 emitters pointing to 8 receivers then "ignore" the receiver side of the circuit.
Yes just wrap the receiver & emitter each in a dedicated enclosure (or possibly shrink wrap). You must isolate each emitter from each receiver ecept the one is is meant for.

You should be able to do this with a UNO or Nano.

Interestingly, with external logic gates you could simplify things even more (would require a bit of thought & design).

I just checked and we have an IR repeater Module (not in our shop, but in stock) which could work well with this, As it has an IO interface that can both detect a signal & disable a signal. So you could wire up 8 in parallel and connect up 16 GPIOs from the areduino for detection/enabling/disabling. It receives any signal from 20->60 kHz and transmits it via an onboard IR emitter. So you would have 8 emitter of which only 1 would.could be active at any time.

The modules do all the repeating stuff for you & are PCB modules 25x25mm (WxL).

This way the code would be relatively simple & you would have to wire up the 8 modules & isolate in enclosures or otherwise.

This would be expandable from the 8 to n channels in future.

You can check out the RepeatIR modules here https://www.analysir.com/blog/2014/10/2 ... -repeater/
I don't have pricing off-hand, but they are less than $20 each, I will have to check.
chill
Posts: 5
Joined: Sun Jul 05, 2020 4:04 pm

Re: Help with interesting control problem

Post by chill »

Yes just wrap the receiver & emitter each in a dedicated enclosure (or possibly shrink wrap). You must isolate each emitter from each receiver ecept the one is is meant for.
Perfect for my un-punched RJ-45 connector shields....
Interestingly, with external logic gates you could simplify things even more (would require a bit of thought & design)
PM me on this....
I just checked and we have an IR repeater Module (not in our shop, but in stock) which could work well with this, As it has an IO interface that can both detect a signal & disable a signal.
PM me on this too...
User avatar
AnalysIR
Site Admin
Posts: 776
Joined: Sat Aug 31, 2013 3:51 pm
Location: Dublin, Ireland
Contact:

Re: Help with interesting control problem

Post by AnalysIR »

un-punched RJ-45 connector shields
..not sure what these are...

Our RepeatIR modules are US$15 each + shipping (US$5.75).


I will send you an email offline...
Post Reply