Run air conditioning diagnostic software remotely over the internet.

Hi,

I am new to this forum and have no experience as yet with Arduino. I have a connectivity issue I would like to solve and am hoping that this forum about general project advice and feasibility is the right place to ask.

I am an air conditioning engineer and regularly use diagnostic software to look at air con systems and fix problems. The manufacturer of the systems I work on is Daikin and I use a machine called a Daikin Service Checker to do this. It is basically a box with a pcb in it and various connection ports and terminals on the outside. The software is on my windows laptop and the service checker acts as the interface between the laptop and the air con unit to be worked on. The service checker connects to the laptop with a lead from an RS232 serial port which goes to a usb port on the laptop. The service checker is also connected to the air con unit via a simple two core flex. This comes from two screw terminals on the service checker onto two screw terminals on the main pcb of the air con unit. Once connected up I can control the system and monitor many different values from it such as temperature sensors, pressure transducers, fan and compressor revolutions and solenoid switches.

Here is a link to a sketch on my dropbox of how it normally connects up:

What I would like to be able to do is run my laptop and service checker from a remote location, such as home or the office, and have someone else on a site with a means to manually connect to an air con system over the internet. My idea is that we have some sort box on the site (Arduino based?) that connects to the air con unit with a two core flex and is also connected to the internet via a mobile phone or one of those mobile broadband boxes with a sim. Then at the my end, in the office, I have another box that communicates across the internet with the other one to send data back and forth between my laptop/service checker and the air con unit on site.

Here is a link to a sketch of what I'm visualising:

Daikin call their communications system that runs between their ac units and controllers "D III Net". From what I've managed to gather from the internet it runs on a protocol called BACnet, which is an HVAC industry standard protocol that most air con manufacturers use as this can be integrated into many common Building Management Services such as Trend for remote monitoring and control. It can also communicate with another protocol called LONworks I believe.

That exhausts my knowledge so far on protocols and how they actually work. What I'm wondering is would my magic boxes have to be programed with the protocol that the air con unit runs on ie BACnet or could they litterally just act as connectors passing data backwards and forwards between the air con unit and my laptop/service checker (that obviously already speaks the language/protocol of the system)?

Any help or advice with this would be really great.

Thanks,

Markie.

Is your primary objective to get remote access to the AC units or is reducing the number of Daikin Service Checkers also a major objective?

My initial reaction would be to attach a GSM modem to the Daikin Service Cheker on-site. The laptop in your office would also have a modem. You would call the modem on the site then use the normal software.

AD---DaikinServiceChecker----GSMModem------#------Modem------Laptop

Since you were already thinking of using a Smart Phone this solution is not too different.
It does not use an Arduino though.

Of course these things are never quite as easy to setup as they sound, but it should not be too expensive to make a test setup.
I would be very surprised if Daikin do not already have a similar setup.

A big advantage of going the SmartPhone or Modem route would be avoiding having to go through the customer's firewalls. However you need to be sure that nobody can dial into the AC unit and play with the controls.

ardly:
Is your primary objective to get remote access to the AC units or is reducing the number of Daikin Service Checkers also a major objective?

Hi, thanks for the reply.

Yes I guess I should have made that clearer initially. The idea is definitely to have the service checker with me. I have a few other people working with me for the same company who are not trained in using them and they are expensive so we are stuck with one for now. The thing is if we need a diagnostic I have to physically go there and do it whereas if I can get this idea to work I can give a box to other engineers to keep with them. When they need a service check run I could get them to wire the box to the unit, connect/tether to their phone and then I could then connect to that across the internet.

I realise that this complicates the whole process but if I can get it to work then my life at work would become so much easier!

Markie.

Here is another possible angle;
http://www.xdimax.com/cool/cool.html

However it is not going to run with the Daiken Service Checker so you would have to evaluate if it does what you need.
If you don't use this product or something similar then you are going to have develop the equivalent yourself.

The link states

In other words CoolMaster hides complexity of internal Daikin Communication line (including very unique physical layer) and instead gives simple, comprehensive, well documented protocol running on popular RS232 or USB connection.

This suggests that Daikin have made their internal communications protocol and even their physical connection proprietary so it may be quite a task to develop something.

Hi,

Yes I'm aware of the Coolmaster product as I've done a bit of searching online regarding this before I posted here. This has actually been superceded by their latest version CoolMasterNet:

It seems like a good bit of kit developed by third parties to provide remote access, control and linking of Daikin VRV units to BMS systems, an alternative to Daikin's own and more expensive version. It's not what I'm after in that, as you say, it does virtually the same job as a service checker but needs installing and setting up on site and is actually even more costly.

I'm looking at this from a layman's point of view but it occurs to me that I already have the necessary means to control and interrogate the AC system i.e. the service checker and the service checker software program installed and running on the laptop. What I want to do is simply replace the means of connecting this apparatus to the AC system i.e. replace the 2 core hard-wired connection from the service checker to the AC's PCB with a connection that goes via the internet. The service checker and it's associated software communicates with the AC system with the Daikin protocol, why would I need to create something else that can do this?

The whole Daikin AC communication network between all the interconnected indoor and outdoor units and remote controllers is run on two wires daisy chained from unit to unit and from the indoor units to controllers. It is a 16 volts DC circuit that sends pulses at a certain frequency which, from my understanding, is the language or protocol that it works on and the service checker speaks this language. I just need to create somethig that will pass these communications back and forth between the service checker and the AC system. Does that means of bridgeing across the internet need to even understand or process in any way the signals it is passing? Can it not simply act as a conduit that replaces the usual piece of wire that connects the service checker to the AC?

One result of doing this I can see would be that the speed of signal exchange would be slowed down by passing it over the web rather than direct hard-wiring. However, I know that the Daikin comms system has been designed so that it can operate through two core unshielded flex that can be run for hundreds of meters through buildings that contain all sorts of electrical interference such as lighting balasts etc and hence it must have a fairly robust operating system that can cope with a fair bit of environmental disturbance I would think. Probably why it runs at 16 vdc and not 5vdc?

If I can get some help with how to make up these "bridge boxes" I'd really like to give it a try to see if it would work.

Markie.

It might be possible to build "bridge boxes" but obviously that would involve developing some electronics particularly as the physical layer is described as "very unique". You would have something like;

16VDC------Bridge box--------# internet #-------Bridge box-------16VDC

It would be good to have more detail of the 16VDC network e.g. what is the data rate, what is special about the physical layer, how is the data physically encoded onto the network, how does each device know when it can 'talk', ideally what are the packet sizes and format. Maybe you could hook up an oscilloscope and get an idea of how the signalling works.

Perhaps you could try this as a first pass;
Sample 16VDC network 'A' with analogue to digital converter and output using digital to analogue converter onto 16VDC network 'B'.

Sample 16VDC network 'B' with analogue to digital converter and output using digital to analogue converter onto 16VDC network 'A'.

You will need to sample at least twice as fast as the data rate.

If you can get a Daikin Service Checker to communicate properly with an air conditioning unit with a microprocessor in the way doing doing all this AtoD and DtoA conversion then you may be well on the way to cracking the problem.

The second step would be to turn the digitised data into packets and pass the packets between two microprocessors.
You should include security. Whether this step works or not will be even more dependent on data rates and whether or not communications delays are tolerated.