Relay Extension Alternative?

[Rob_in]

While the 16A rating for the Loxone relay extension seems pretty nice, for use with valve actuators or 240v LED light circuits it also seems a bit like overkill.

The retail price (in the UK) for the relay extension is 399GBP. Which gives you 14 relays. That's 28.50 for each 'output'. I'm wondering if there is any cheaper way to control smaller loads?

Today I came across the Teracom TCW181B-CM device. 8 x 3A relays mounted on the DIN rail & controlled by ethernet protocols (eg. HTTP API which Loxone claim is easy to interface with). This device can be had for under 100GBP. That's around 12.50 an 'output' or under half the price of a Loxone Extension Relay.

Then I found a device called smartDEN LAN Ethernet 16 Relay Module - DIN Rail BOX. 16 x 15A (at 120v) relays for only a few pounds more. Similar control mechanism and getting on for a quarter of the price of Loxone outputs.

I'm currently evaluating Loxone, so have to ask, are devices like this worth considering when you have a lot of outputs?

[Simon Still]

The simpler option is to use DMX relays.  The DMX extension is £240 but will run up to 128 DMX channels.  

DMX relay boards can be picked up easily at £4 a relay or less 

eg

http://www.ebay.co.uk/itm/12CH-Relay-switch-dmx512-Controller-relay-output-DMX512-relay-control-12-way-12V-/381401613627?hash=item58cd4ec53b:g:fE8AAOSwLqFV9qV5

[Rob_in]

On Monday, 1 February 2016 12:59:48 UTC+1, Simon Still wrote:

The simpler option is to use DMX relays.  The DMX extension is £240 but will run up to 128 DMX channels.  

DMX relay boards can be picked up easily at £4 a relay or less 

That that DMX extension is the killer though.

For just bare boards I've found SainSmart make 16 relay boards for just $15.27USD. Bundled with an ethernet control module (turn things on and off with simple HTTP requests) for $42.99USD. That's around 30GBP or £1.87 a relay! They are rated at AC250V 10A.

[Simon Still]

I think ultimately it depends how 'standard' you want your install to be.  You can buy a RPi for £25, install OpenHab and cut out the miniserver.....

 

[Rob_in]

Funnily enough I have been looking at the Raspberry pi options. Getting good inputs is hard, but the UniPi device looks good.

TBH, I don't see Loxone as any more of a standard than anyone else. You still have to code/configure a Loxone installation. My dealings with their support and partners - looking for information on compatible heating systems for example - has been rather frustrating to say the least :(

[TomM]

Hi Rob,

DMX is definitely your most robust, minimal configuration option here for extra relays.  I have lots of outputs (lights and relays) running DMX (approx 100 channels) and they have never failed and this includes all of my heating system.   I have several devices running over HTTP including digital input boards, milight lamps, amps, TVs etc and these are only 90% reliable.  From experience, I definitely wouldn't entrust key infrastructure such as heating to HTTP requests.

With regards to Loxone support - I've been using Loxone for about 18 months now and always found their UK support team very helpful, knowledgeable and responsive.  If you're in the UK you should take a trip to see them in Reading, very worthwhile.

Re compatible heating?  Are you still struggling with this?  I too found this unclear at the start mainly because I didn't understand how heating systems worked.  As it turns out, traditional heating systems in the UK are very basic and comprised of valves, pumps and boilers.  Each of which can be activated as required using whatever output is appropriate  (normally a relay) to control the system in very much the same way as a traditional programmer would but just with a more intelligent controller at the heart.  I now have a heating system with approx 20 moving parts all individually controlled and it works faultlessly.  It could also be reverted to a traditional setup within a couple of hours if necessary.

[Simon Still]

On Sunday, 31 January 2016 20:15:43 UTC, Rob_in wrote:

I'm currently evaluating Loxone, so have to ask, are devices like this worth considering when you have a lot of outputs?

Rob - what's the scope of your project?  Always interesting to hear what people have planned and helps with feedback or suggestions.  

New build or retrofit?  How much of it are you intending to cable (vs wireless).  What are you looking to control/automate?  Budget? 

[Rob_in]

On Wednesday, 3 February 2016 09:39:43 UTC+1, TomM wrote:

DMX is definitely your most robust, minimal configuration option here for extra relays.  I have lots of outputs (lights and relays) running DMX (approx 100 channels) and they have never failed and this includes all of my heating system.   I have several devices running over HTTP including digital input boards, milight lamps, amps, TVs etc and these are only 90% reliable.  From experience, I definitely wouldn't entrust key infrastructure such as heating to HTTP requests.

Really? On a LAN dedicated to control? It's not UDP you know ;) When you can write a request to a device and then read back that status to make sure the write worked I don't see how anything can be more robust. Sure, the link could fail, but same could be said for any other hardware. Sure, the software on your controller could fail, but again, same could be said of any system. I doubt the Loxone stuff is coded in ADA with multiple redundancy ;)

In your 'only 90% reliable' TCP control system, why, exactly, was it failing?

 

With regards to Loxone support - I've been using Loxone for about 18 months now and always found their UK support team very helpful, knowledgeable and responsive.  If you're in the UK you should take a trip to see them in Reading, very worthwhile.

Well, I wrote directly to Loxone and explained what I wanted to do. I asked about compatible hardware (heating systems for example) and got the answer that they didn't keep lists like that. I asked about what projects that might be similar had been implemented and no details emerged. I asked about a knowledge base or other support forum and, well, ended up here (no offence, but it's nothing official is it). Hardly a good experience.

 

Re compatible heating?  Are you still struggling with this?  I too found this unclear at the start mainly because I didn't understand how heating systems worked.  As it turns out, traditional heating systems in the UK are very basic and comprised of valves, pumps and boilers.  Each of which can be activated as required using whatever output is appropriate  (normally a relay) to control the system in very much the same way as a traditional programmer would but just with a more intelligent controller at the heart.  I now have a heating system with approx 20 moving parts all individually controlled and it works faultlessly.  It could also be reverted to a traditional setup within a couple of hours if necessary.

We will have a heat pump for water and driving underfloor heating and possibly some low temperature radiators on another floor. Most of the heat pump systems seem to have fancy electronic control systems. I imagined being able to interface with them using some kind of protocol and communicating directly with their electronics. I have been reading documentation from the manufacturers and for heating think I see how it will be possible to control all this with relays. This isn't a very hi-tech approach though!

[Rob_in]

On Wednesday, 3 February 2016 11:07:27 UTC+1, Simon Still wrote:

Rob - what's the scope of your project?  Always interesting to hear what people have planned and helps with feedback or suggestions. 

Well, this is a little OT, but I guess it's relevant.

It's a new build project. I'm a software developer (started out in process control if that's relevant) so want to do some home automation. Just for fun really, although I do see some practical advantages. For example:

Our home will have lots of South facing windows. We will need to keep the heat out in summer and let it in in winter. I see automatic blind/heating control as a big thing.

Example scenario: it's early morning and the temperature has just dropped below the minimum 'comfortable' range, but not yet at the absolute minimum we've set. A smart system could handle this like:

- It's the middle of winter, the weather forecast is for rain all day and a temperature well below our minimum requirement. Clearly the sun isn't going to come out today so turn on the heating as it's the only way to warm up.

- It's the winter but a sunny day is forecast and sunrise is only 30 minutes away. Wait for the sun to hit the house then open all the blinds on the South side. After a while, turn on the UFH pump (but not the actual heating system) to circulate water from the now hopefully warm South to the colder North.

- It's spring or summer. The weather forecast is for sun and outside temperature will soon be at or above our absolute minimum for the house. No action required.

You get the idea. The house will also have a mechanical ventilation system which can be automated to prevent excessive exhaust of air from rooms that are at comfortable levels.

This is the main reason, but while we're at it, would also do the lights.

It's a new build so everything will be wired.

I do like the look of Loxone but the cost of their hardware, particularly cost per I/O is *much* higher than other options (see above). Right now I'm not seeing any compelling reason to justify that.

[smartbusinesstools.be]

You could go for a KNX relay module, like this one http://www.smartbusinesstools.be/shop/product/switch-actuator-16-fold-8su-mdrc-16a-230vac-317?category=53#scrollTop=119

Costs 369 EUR for 16 x 16A relais, which is 18GBP per output, plus it has a number of advantages:

• buttons and leds on the module to monitor/operate relays manually, which is very handy while installing or in case of problems
• KNX bus for signalling is even easier to install and extend than DMX (line, star, long stubs, sub-stubs) and many more devices use KNX (e.g. push buttons with leds so you don't need to use any inputs/outputs)
• KNX bus and devices is a standard. The devices can work with many other centralized building automation systems, or just on their own in a decentralized way.

If you're looking for high reliability, this is cheaper and better quality than Loxone extensions, and future safe. You can do much cheaper with DMX and IP controlled devices but that adds complexity and risk, e.g. if you ever sell your house, how is the new owner ever going to manage such a complex non-standard setup?  

[Simon Still]

On 4 Feb 2016, at 08:01, Rob_in <rain...@gmail.com> wrote:

Well, I wrote directly to Loxone and explained what I wanted to do. I asked about compatible hardware (heating systems for example) and got the answer that they didn't keep lists like that. I asked about what projects that might be similar had been implemented and no details emerged. I asked about a knowledge base or other support forum and, well, ended up here (no offence, but it's nothing official is it). Hardly a good experience. 

I see the lack of their own knowledgebase/support forum is a sign of them moving from low margin/high support individual installs to an approved installer model.   They’ll still support individuals but it doesn’t make business sense to drive a lot of that type of business as it doesn’t scale for them.   The profit margin on a single install can’t be that great - and they’ll be even less interested if you’re planning on using the minimum amount of their kit and interfacing with commodity stuff. 

We will have a heat pump for water and driving underfloor heating and possibly some low temperature radiators on another floor. Most of the heat pump systems seem to have fancy electronic control systems. I imagined being able to interface with them using some kind of protocol and communicating directly with their electronics. I have been reading documentation from the manufacturers and for heating think I see how it will be possible to control all this with relays. This isn't a very hi-tech approach though!

Loxone will theoretically interface to all sorts of things - there are RS232, RS485, Modbus extensions as well as the inbuilt KNX and HTTP.  More of an issue seems to be that the heating system manufactuers are most interested in selling you their own solution so support for interface to others is limited. 

I’ve looked at linking my Viessmann boiler but you need £1000 of their own kit to interface it (and even then it's far from simple).  Given that their own weather compensation controller works so well I've concluded there is t any point. 

[TomM]

Rob, based on your responses I'm not sure I can coherently argue the case for DMX to control outputs such as relays except to say that in my real life experience from my own install I have found it more robust.  Maybe my IP network isn't up to scratch, but either way I'm not willing to spend too much time coding control over IP when I know I can get a perfectly robust solution from DMX which is about as 'out of the box' as you'll get in this area.  That said, Filip proposes a good solution in KNX which is industry standard of course.

I would also echo Filip's wise words re complexity - whilst it's lovely to have all these cool gadgets automating and controlling the house you shouldn't forget that other people will live in your house too (current family & future purchasers) and I can tell you from experience that its not very pleasant to get a call from the wife when you're abroad to tell you the heating has failed over night and then you find remote access to the network is down so you can't just remote in to see whats going on and calling any old plumber is unthinkable because you haven't had time to finish the docs for the install so they won't know what's going on!!  As a developer myself I completely understand the urge to increase complexity to gain more granular control (and thus automation) but I now find that a very selfish notion as it creates complexity and risk for my family that they wouldn't be able to deal with without me - therefore it's something I decided to shelve a long time ago.

In Loxone I believe I have a decent compromise which enables quick deployment of updates & fixes, (almost) zero configuration for the interface, robust infrastructure for controlling key components and a potential backup in Loxone support.  Loxone components aren't exactly cheap but if push came to shove and you really needed to buy they're not extortionate either, especially when you compare to proprietary control systems for things like UFH from, say, Heatmiser.

Having spent about 2 years researching systems (and an interest in HA for nearly 10 years) I chose Loxone for my house and, despite a few minor issues, i'm very, very happy with it.  Even the wife is happy with it, which is quite a feat considering the botched systems I'd tried out before in our old house!  So good luck to you, i'm sure you'll pick the right system for you but just make sure its the right system for the house and the other occupants too

[Duncan]

loxone can be made to work with almost any heating system to give end-user functionality and with regards to interfacing with heating systems, i think in most cases you dont actually need to at all - loxone works extremely well controlling the zones/outputs, usually replacing the controls that would be supplied with a multi-zone heating system where you replace the programmable thermostats and central stats with temperature sensors and switched outputs for controlling manifold valvies, zone valves and boiler pumps.

In a typical gas boiler install, you use loxone to control the manifolds or actuators on radiators, and receive the zone temperatures, but leave a modern modulating boiler to just be triggered by loxone using the single traditional stat input, delivering the currently required 'boiler lockout' for building regulations.
obviously if you get the chance for a new install, then you wouldnt buy any controls from the ufh supplier, just the hardware and manifold actuators, then you are not throwing anything away. in a recent design, the heating controls for a large ufh building with 4 manifolds and 38 actuators would have cost over £2k for the traditional zone controls - this was more than the cost of a loxone miniserver, 1-wire sensor for temperature sensors, dmx extension and dmx relays to control all the actuators, and so all the fancy features available through loxone including lighting, audio, etc came for free as part of the heating install.

if you are using ufh then even a boilers own temperature compensation wont do anything as the manifolds maintain their own temperature indepenently of the boiler output temperature

You can use loxone to do temperature compensation of the manifold temperature to even out room heating times with different outside temperatures- that was a sudden cold night will cause the manifolds to be hotter and the IRCs startup time will still be approximately correct, resulting in rooms at the right temperature even with big variations of the outside temperature. In this situation you need to replace the manifold blender valve with an electronically controllable proportional actuator - eg a 2,3 or 4 port valve with a 0-10v actuator and a manifold temperature sensor so loxone can manage the manifold temperature

if you have a heat pump system, it probably has temperature compensation and a heat store in its own control system , so you leave the system with its own control system, and use loxone again to drive the output/demand via actuators/pumps etc but there isnt much point trying to replace the heat pump control systems or interfacing with them.

[Dpin]

HI Simon
tks for your post, I got a question:
when you attach a DMX Relay board to the Loxone DMX, that relay board gets a channel address, right, how is the control of each relay output?, I think my question goes straight to the Loxone configuration, would appreciate if you have any information

Thanks
Dp

[Duncan]

in the loxone config, you add a dmx device and ensure thats its dmx address matches the physical dmx address set on the relay hardware

some dmx relays switch on with any dmx value above 0, some above 128 and probably many other options, so in the config you use the 'correction' parameters with input value 0 = 0, target value 1 = 100, and that means any positive value on the output drives the dmx output to 100% or dmx value 255 and ensures the relay switches on.

[Rob_in]

On Monday, 1 February 2016 12:59:48 UTC+1, Simon Still wrote:

The simpler option is to use DMX relays.  The DMX extension is £240 but will run up to 128 DMX channels.  

DMX relay boards can be picked up easily at £4 a relay or less 

Well... time to revisit this topic as our project moves closer to implementation.

Interestingly, since researching and thinking more I have discovered that if one uses DMX for all the lighting with 24v LEDs then this is actually cheaper than trying to implement 240v lighting with relays. And that's even if one uses the Loxone RGBW dimmers (I'm sure there are cheaper options), which I'd be using to drive 4 single colour ('white') channels each.

Given this, yes, I think what Simon said about using DMX for relays way back in February is a good idea. BTW, the only relays that we need are for blinds (roller shutters).

Whenever I try and search for DMX modules online am inundated with cheap Chinese gear which I am rather wary of. The board Simon originally mentioned back is no longer available too.

Can anyone recommend a known supplier for DMX relays and/or RGBW dimmer modules that are known to work perfectly Loxone? Maybe I should ask that in a new thread?

Cheers,

Rob

[seb303]

For DMX lighting drivers I'd highly recommend the range from eldoLED:
https://www.eldoled.com/

Seem to be of high quality, reliable in my experience (I have very many of these), excellent to-zero dimming, and a good variety of models: constant current / constant voltage / high/low power, etc.

For their constant voltage drivers, you generally need to provide 24V power - I used some Meanwell power supplies, shared across a few drivers, and switched off by the Loxone when not in use to minimise power consumption.

[Andrew B]

Just toss in another idea: I'm using the Elexol IO72 for input and output. It has 9 x 8-pin ports, and a variety of input and output and combo solutions. Have been running many input ports for my wall switches and one now uses their darlington transistors to control a few din rail mounted relays in my mechanical room.

The Elexol products are quite cost effective per pin. And flexible.

I've set my system up to avoid dropped UDP packets, but in practice with a switched simple network they just don't happen.

[Rob_in]

On Monday, 1 February 2016 12:59:48 UTC+1, Simon Still wrote:

The simpler option is to use DMX relays.  The DMX extension is £240 but will run up to 128 DMX channels. 

So for something like this...

https://www.amazon.co.uk/Channel-Controller-Switch-Dimmer-Equipment/dp/B0111QP790

... has someone fitted one into a DIN Rail enclosure? I know there are many such enclosures available but imagine finding the right one is a bit hit and miss. If anyone has already done this would be nice to know.

Interestingly I can't find any DMX relay modules that come with any kind of enclosure.

Our project will also have a Modbus extension (for control of the heating) and there are a few Modbus modules around too. For example:

http://info.kmtronic.com/modbus-relays.html

That KMTronic board is €59.00 and can be controlled with Modbus or TCP/IP and they have documentation for use with Loxone. I think this could be a good option.

[Duncan]

ive used these
https://www.aliexpress.com/item/Wholesale-12CH-Relay-switch-dmx512-signal-led-Controller-relay-output-only-use-the-signal-control/32445272725.html?spm=2114.01020208.3.21.TygBk4&ws_ab_test=searchweb0_0,searchweb201602_5_10065_10068_10084_10083_10080_10082_10081_10060_10061_10062_10056_10055_10037_10054_10033_10059_10032_10099_10078_10079_10077_10073_10100_10096_10070_423_10052_10050_424_10051,searchweb201603_8&btsid=fb9de47d-d924-4d33-b808-032cfc8bf57d

rated 2 amps - i used these for some blinds which needed 3 relays per blind for up/down/stop, but they are voltage free contacts

http://www.srslight.com/index.php?index=product&product=DTS16
is a din mount 16ch dmx relay which i use for UFH manifold control - the contacts are 3A which is pretty good, but ive used a high current relay as well triggered from a channel to drive water pumps, just in case

[Julien Larios]

Hi Rob

I built also a hybrid solution to be cost efficient, and my choice was this "smartDEN LAN Ethernet 16 Relay Module - DIN Rail BOX".

I'm running a KNX bus for switches/dimmers/recessed actuators (blinds), a loxone miniserver and a smartden over http.

So far it's running fine, but we are not living in the house yet. Move is planned before end of year.

I would be happy to share somes insights about this setup if you are still considering it.

-Julien

[Rob_in]

Hi,

Yes please. I would be very interested in how you implemented this.

Am curious why you chose KNX and where you sourced all you switches, dimmers, etc. Are you able to share links to the suppliers?

Regards,

Robin

[Duncan]

ive also used knx switches and dimmers where necessary, but only in certain areas:

the right choice of knx switch can give you a compact elegant solution with multiple buttons, feedback indicators, background illumination, icons, text etc all without using any loxone inputs or outputs - i tend to use them for 'public' rooms such as kitchens, entrance hallways, lounges etc, and normal buttons elsewhere

i tend to use mdt glass buttons - the version 1 you make your own lables/icons, and version 2 uses an lcd screen where you can program your own icons and send information back to the button
the version 2 is a single switch size and can be up to 12 buttons and feedback leds
http://www.mdt.de/EN_Glas_Push_Buttons_Smart.html

theben make an 8 channel knx dimmer which does 200w per channel of led lighting in a very compact format - 8 channels in a block 6 din spaces wide, and it works out around £60 per channel
https://www.theben.de/en/Products/Home-and-Building-control/KNX/Actuators/DM-8-2-T-KNX

[Rob_in]

Nice. I did a little hunting around after you mentioned KNX and it certainly looks like a more 'up market' product!

[Julien Larios]

Hi,

The initial idea in my project was to not be locked into one single technology/supplier. Loxone is great "brain" for the system so I've put it at the center of the project, but I didn't wanted to go "full loxone" on all and every aspects.

- KNX was my choice for switches for the same reason than Duncan mentioned : "compact elegant solution with multiple buttons, feedback indicators, background illumination, icons, text [+temperature senors]", but also because of the ease of wiring. The bus network reduces the planning phase from my point of view. There are several suppliers that let you find the perfect design/budget tradeof that works for you. We chose the very standard brand JUNG, sourced from at my local electrician. We have a mix of 4072.02 and A 5093. I have also planned emply wallboxes for motion detectors (3180-1 A + 2070 U), but will come later.

- In the old house we are renovating, I needed recessed actuators for blinds to reduce the amount of cables. As the KNX bus was already running around I chose the entry level from Hager TXB202A. The program inside is quite simple and there is no digital input, but with Loxone behind it's all fine. I got them on voltus.de

- For the dimmers, I didn't wanted to be locked on loxone technology. I've looked into DMX but to avoid one more protocol to integrate I'd preferred to stick to KNX and selected the DIMinBOX 2CH from Zennio. In bonus they provide digital inputs but I'm not using it for the moment. I'm dimming the 2 lights in living room, 1 in bedroom and 1 in office.

- I also bought 2 presence sensors "ESYLUX PD-C360i/8 KNX Mini" but not installed yet. We changed our mind and no more false ceiling will be installed in living and kitchen, so now I'm figuring out what to do with them ... ;) fallback will probably be bathroom and bedroom

- I found the smartden product while randomly reading forums as an alternative to the IPX800 v3 http://gce-electronics.com/en/home/57-web-relay-board-control-ipx800-v3.html . The IPX800 is a great box but more as a cheap alternative to the Loxone, not as extension board. The smartden is quite easy to integrate with loxone, I can share all the http parameters if required. You declare a http virtual output and you're all good - well ... to be honest when I say it's easy, it took me several hours... Downsides you have to be aware of :
* It's really difficult to get a "status feedback". Meaning that if you turn on light in loxone, but the http request to smartden gets lost somewhere, it will appear ON in loxone but your light is still off. It might frustrating if you don't "know" what happens. I'm confident that I'll find a way via scripting or a webservice trick, but out-of-the-box you just don't have it.
* Depending on how you configure your controllers in loxone, it will send to smartden the http request in sequence of each relay. Because of the low processing power of the smartden, the side effect is a delay in each relay switch. So you'll not have an instant switch for all lights, but one after the other spaced by roughly 250ms
For all these reasons I decided to use the smartden outputs for "non critical" things: cellar lights, outdoor light, facade, outdoor power outlet ...

- I'm using loxone outputs for the garage door and the most used lights : kitchen, corridors, bedrooms, bathroom . Loxone inputs are used for classic switches in the cellar, all wired up with cat6 cable.

- I planned some rgbw lighting in the living, and I'll use the KNX MDT AKD-0424R.01 module. It was for me the best tradeof between cable management, cost, robustness, power management (4A / channel) and integration efforts. For info, I also have a Philips HUE bridge from my previous flat, and thanks to Loxone everything will be seamlessly integrated.

- Lastly I have all windows and corridors wired to receive alarm sensors (opening and movement) in the future. I don't know yet what I'll do with them, a dedicated alarm system or just another input extension board, but what is sure is that I want to enjoy these inputs in loxone engine for scenarios.

- I have also a multiroom audio plan, but still struggling with the technology choice... It's offtopic but we can talk also in another thread.

For the moment I'm pretty satisfied with my setup, but again we are not yet living there yet. In 2 months I'll maybe come back here and say that I swaped the smartden for a knx relay board ;)

I also want to emphasis the security aspects of the smartden. The box is really vulnerable - no https, weak password policy, multiple ports wide open (snmp, http...). Depending on your network skills, it might become a potential breach as this box will most probably run on your regular LAN with all your computers, NAS, smartphones, guest wifi and so on. I will mitigate this via different VLAN, routing and firewalls but it's not for everyone.

[Rob_in]

I cannot fault your logic.

Sure, the SmartDEN and other solutions I have mentioned so far are clearly more effort to integrate and setup and look like more economic options. Thank you for the information about the SmartDEN speed - that is a point against this unit. I guess quality/speed/ease of use is the trade off.

For output, when considering KNX one can get a Zennio MAXinBOX 16 Plus (16A x 16 channel KNX relay with NO/C/NC contacts) for around 400EUR. The Loxone relay extension is not only more expensive, but it has 2 less channels, doesn't have the NO contact and will only ever work with a Miniserver. Why would I choose Loxone's Relay extension over the Zennio unit? You have to question the Loxone pricing strategy in this case because in the case of this KNX Zennio unit I imagine there is no such quality disadvantage like the other 'economy' options I mentioned before.

However, for input I actually think the Loxone Tree Touch is a good option. It is much cheaper than KNX switches. Even if does not have the aesthetic variety you might need. Same for Loxone's DMX extension & LED control. But let us not discuss that - this thread is about outputs.

My thought process continues... ;)

[BartVB]

On Thursday, November 24, 2016 at 12:53:53 PM UTC+1, Rob_in wrote:

For output, when considering KNX one can get a Zennio MAXinBOX 16 Plus (16A x 16 channel KNX relay with NO/C/NC contacts) for around 400EUR. 

If you are considering the KNX route then this is also a nice contender:

http://www.mdt.de/EN_Switch_Actuators_AMS_AMI.html

The MDT AMI/AMS actuators. The AMS-1216.02 can be had for approx €300,- for 12 relays and the big plus is that they integrate 12 fold current measurement.

[Rob_in]

Ahhhh... well I see MDT do a model number JAL-0810.02. This has control for 8 blinds (16 relays) and can be had for 230EUR.

Given my requirement is to control blinds then at this price it's getting to the point of not bothering with the other 'home brew' options.

So just to be clear (because I didn't consider KNX before)... all these KNX modules can connect directly to the Miniserver and will work just fine? There's no gotcha or oddity in the way KNX is implemented in the Miniserver?

[Duncan]

the only thing you need to use knx with loxone is a knx power supply

i cant say about any specific compatibility with loxone, partly because of the massive range of products.

the biggest issue with knx is getting the devices programmed to suit your install requirements

[Rob_in]

Ahhhh... and here come the gotchas I was talking about. The Loxone site states:

"For the KNX bus a separate and product specific power supply needs to be used (the KNX/EIB power supply)."

OK - cheapest one I can find is 87EUR.

"For addressing and configuration of the sensors/actuators a full ETS programming software license is required."

Oh dear - this sounds expensive! And voila! A full (ie. professional) ETS license is 1000EUR (see https://www.knx.org/no/software/ets/prices/index.php). Ouch!

That said, there is a 'demo' version they say is capable of programming up to 5 devices. So this might do. Who knows, because Loxone say you need the 'full' license.

For my project, assuming the demo version is good, devices could be:

1. Loxone Miniserver

2. KNX gateway for heating system (if we're using KNX then no need to use the Modbus version)

3. MDT JAL-0810.02 #1

4. MDT JAL-0810.02 #2

5. Oooh... a slot for future use!

I've searched around a bit looking for 'Loxone KNX howto', 'Loxone KNX demo', etc. with little luck. There is a video of a KNX test rig which I have to say has less than impressing lighting response times (noticeable delay between the button being pressed and light coming on). Technical doco on Loxone/KNX integrations seems thin on the ground.

So let me get this straight... all the guys here talking about using KNX switching, etc. Do you all have a full KNX license or have you paid an installer who happens to have one because it's their job?

Robin

[Duncan]

you can use the demo licence, but certain devices such as the mdt glass button smart wont have all the programming features avaiable. The lite licence for ets works with all devices, and allows upto 20 devices per (knx) project - there is nothing stopping you break up your knx devices into groups of 20 (or 5 for the demo) if you go that big, particularly if you are using knx simply as inputs/outputs rather than knx to knx then each one is effectively programmed separately anyway and its inputs/outputs set up in your loxone config file

the course is free and takes less than an hour on line, then you pay around 60 euro for a hardware usb dongle to enable the features in ets

its an odd sort of program and will take a little bit of playing to get the hang of it, but at least you can then reconfigure your devices or add new ones whenever you like

[smartbusinesstools.be]

Rob_in

We sell all MDT KNX devices pre-programmed for use with Loxone to our Loxone customers so they can simply be connected to the KNX bus and used in Loxone. So the customer does not have to bother installing the ETS software, finding/downloading the programs for the devices, configuring the devices in ETS, assigning groups addresses to every object, downloading the configuration to the device. We keep records of the assigned addresses per customer so they never overlap when they order additional devices. We keep the ETS project file so we can provide support or make changes remotely if needed, or give the file to the customer.

The products on our webshop need updating for the newer products and adjusted (lower) prices. We also give discounts on the non-Loxone products to our Loxone customers. Contact us if you want a quotation.

[Naran Khetani]

Hi Tom,

I'm just about in the process of setting up my heating system based on a dmx relay and would be really greatful if you could help out a newb with the whole configuration and logic of how to get the heating system working

Thanks!!

[Ray A]

Have u thought of using a standardp din relay and using a 24v trigger from a spare output. That way u keep all loxone control. I have same issue in that i have a wc and lobby that need 2 circuits. I may just use a 230v 360 detector direct to a live circuits. £15

[Redguy]

Reply on an ol dpost but still..

I have been experimenting with Moxa io controller from their oLogik E1200 series...   They are controlled using standard modbus TCP, so no MODBUS extension is needed.  Turns out, this is easy to set up and works great ! Not dirt cheap like the modules from aliexpress and such, but still a nice and professional alternative for buying more extensions.

The 1200 series has all kinds of in/out options :

- ioLogik E1210 - 16xDI

- ioLogik E1211 - 16xDO (Sink)

- ioLogik E1212 - 8xDI 8xDIO

- ioLogik E1213 - 8xDI 4xDO 4xDIO

- ioLogik E1214 - 6xDI 6xDO Relay

- ioLogik E1240- 8xAI

- ioLogik E1241 - 4xAO

- ioLogik E1242 - 4xDI 4xDIO 4xAI

- ioLogik E1260 - 6xRTD (RTD temperature sensors)

- ioLogik E1262 8xTC (Thermo couple temperature sensors)

With prices ranging from 120 - 350 euro's i think this is an interesting option.

Another option i am still working on is using APC rack PDU's for switching heavy loads.  I bought a couple of their AP7921's for like 30 euro's on ebay.  They can be controlled using SNMP : 

snmpset -v 1 -c <write community> <ip address> 1.3.6.1.4.1.318.1.1.12.3.3.1.1.4.<outlet number> integer <command>   (2=off 3=on)

Having trouble with hooking it up to the Miniserver as a virtual output though..   The manuals specify how to define a snmp output but not how to configure the commands to send..

Simple 1U boxes (about 20 cm deep) with 8 C13 outputs.. 

[Rob_in]

Also check out the Weinzierl 570 multi IO KNX thread. Looks like a very nice module indeed.

[Techdoctor]

If you are using a Loxone miniserver (rather than the Go). You have 4 0-10V analogue outputs, you can use these to control a relay board. Google 0-10V relay control. Most of the boards have several per analogue input. Though some are single relays only. I think they work like this  a voltage of 1v turns on relay 1 and 2v turns it off. 3v turns on relay2 and 4v turns it off. 

Personally I would use the Loxone relay extension, it might cost a bit more, but think of the time you will save when it comes to programming. Plus fault finding will be easier as its all one system. 

[Hidde Beumer]

I prefer the troubleshooting capabilities of other vendors, where I simply have a LED (or in the case of the Weinzierl 570 a small LCD screen) which tells me what the state of a specific output/input port is. In loxone I must always go into the config tool for troubleshooting

Op woensdag 5 februari 2020 11:20:05 UTC+1 schreef Techdoctor:

[Andrew Brownsword]

Looks like a pretty sweet series of units.  How much does the 16 digital output model cost (for example)?

--
You received this message because you are subscribed to a topic in the Google Groups "Loxone English" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/loxone-english/VAqI0Sk72_0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to loxone-englis...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/loxone-english/9a216b9c-8928-4dc9-a22e-ffb8ff5642ec%40googlegroups.com.

[Andrew Brownsword]

Found some pricing — digital outputs it works out to around ~$12.5/output (before shipping and taxes).  That’s pretty high.  On the other hand, it allows daisy chaining on Ethernet as well and looks like a pretty solid unit.  So if your budget allows it, looks like a good choice.

I haven’t looked at the programming model, so I can’t help with that, I’m afraid.  For interacting with the Elexol stuff I use, I wrote it in PicoC for maximum flexibility, but have also done some of the simple output stuff using virtual outputs.  Their protocol is very simple.

[Skarsol]

These look really interesting, but it looks like they're too tall to fit in most of the common cabinets used for Loxone installs. The miniserver is 57mm tall while these are 84. :(