DMX to Insteon bridge - I/O Linc 2450

[frustratednon-geek]

I'm having trouble with two I/O Linc Contact Closure Interfaces (2450's) I installed at my church as a DMX bridge. The stage lighting controller closes 1 of the 2 relays as part of its own scene--the desired result being that the sound booth operator only needs to hit 1 button to cue both stage and house lighting systems simultaneously.

 

Unfortunately the insteon response can be quite erratic when somehow the last command sent is from the relay that's turning off. As a workaround I can set up a cumbersome sequence via 'show control' on the DMX board which turns both relays off for about 0.5 seconds, then turns the desired one on at the initiation of the scene. Even that has failed a number of times over the last year and it's a real pain whenever we need to make lighting changes.

 

Hoping someone can chime with an elegant solution... I've got an ISY99i on the system FYI. Thanks in advance.

 

P.S. here is a link to a previous post of mine about the system when I first installed it

[Michel Kohanim]

Hi frustratednon-geek,

 

Are your IOLincs setup in momentary mode? i.e. when you turn them on, they turn off momentarily? And, if so, is that what you want done?

 

Can you give a sequence of an example that does not work (i.e. IOLinc turns on, ISY does this, DMX does that, but this should have happened but didn't)?

 

With kind regards,

Michel

[frustratednon-geek]

I'm only using the sense portion of the i/o lincs, not the relay portion so momentary is somewhat irrelevant in this case right?

 

Here's the sequence of what happens:

1. Operator pushes 'Intro' scene button on the DMX stage lighting control board

2. DMX controlled stage lighting fixtures fade, change colors, move, etc. per whatever the scene settings are, including...

3. A DMX-32 board sitting in the attic which has 32 channels of 12v faders, 2 of which are currently in use to signal the I/O lincs...

4. The 2 channels on the board power relays which create contact closures for two I/O Lincs, each assigned to Insteon scenes we wish to trigger off the DMX event.

5. In the case of 'Intro' relay 1 closes, I/O Linc 1 senses the contact closure, sends a group command to 6 or so Insteon devices and executes the house lights to the corresponding 'Intro' scene in Insteon. Works great...

Here's where the problem typically comes up:

6. Operator pushes another, 2nd scene called 'Message'.

7. Stage lighting fades to new scene, DMX-32 energizes Relay 2 , deenergizes Relay 1 almost simultaneously

8. When it works correctly, House lights pick up the group command from I/O Linc 2 and fade seemlessly to 'Message'.

When things don't work the I/O Linc 2 sends the group command for 'Message' but is eclipsed shortly thereafter by I/O Linc 1 sending a group off command as its contact closure opens. Hope this makes sense.

 

I think I need to let the ISY handle this through a program (not hard-linked) but not sure how to accomplish. About 6 months ago I tried to use a program but the ISY wasn't picking up accurate status on the two I/O lincs during transition. However I did just replace a defective PLM so maybe the erratic status reporting was an early symptom of the PLM failing.

 

How would I program? If I/O Linc 1 is ON and 2 is off, Set scene 'Message', etc.? Thanks for your help!

[Brian H]

It maybe possible. That relay two is closing and sending the proper command and relay one has not opened up yet. Causing its group command to be sent after relay two sent its message.

 

If the 12 volt signals you are using. Are ramping up and down and not instantly On or Off. Good chance the relay being turned On reaches its pull in voltage before the other relays drop out voltage is reached.

 

You maybe able to use some programming in your ISY controller to compensate for your overlapping messages.

[LeeG]

Insteon does not allow more than one Scene to run at a time, cancelling a Scene if need be. When DMX Relay 2 turns I/O Linc Sensor 2 On it sends a sequence of On commands to the devices assigned to that Sensor. Although the Responders to I/O Linc Sensor 2 normally respond to the first message sent by the I/O Linc Sensor 2, there is a backup message for each Responder which takes a few seconds to complete in case that first message is not seen by all the Responders. These messages will be cancelled when DMX board turns Off I/O Linc Sensor 1 which starts a new Scene Off sequence.

 

The normal Insteon Scene message sequence is this ….

 

Send Group Broadcast message which all Responders react to normally.

Send Group Cleanup Direct message to Responder 1

Receive ACK from Responder 1 - if this ACK is not received the I/O Linc Sensor sends another Group Cleanup Direct to Responder 1 - this will occur a few times before the I/O Linc stops with this Responder

Send Group Cleanup Direct message to Responder 2

Receive ACK from Responder 2

Send Group Cleanup Direct message to Responder 3

Receive ACK from Responder 3

 

and so on until all Responders receive a Group Cleanup Direct message which is ACKed. This sequence occurs each time an I/O Linc Sensor turns On or Off. When the next Scene starts before all these messages have been sent and ACKed, these follow-up messages will be terminated. This means that some Responders will not see the Group Cleanup Direct. This is not unique to I/O Lincs. Pressing a button/paddle on two Insteon devices at near the same time will have the same result.

 

It is rare to have two Insteon devices pressed On/Off at the same time. In this situation two I/O Linc Sensors are being activated at near the same time all the time which will have some bad results.

 

The DMX board can delay the change to the second I/O Linc Sensor for a few seconds to allow all the messages from Sensor 1 to complete before starting Sensor 2. An alternative would be to eliminate the Responders being linked directly to the I/O Linc Sensors, having ISY Programs issue the Scene commands. This may make things worse if Insteon comm is poor as an ISY Program turning a Scene On/Off only sends the Group Broadcast message.

[frustratednon-geek]

If the 12 volt signals you are using are ramping up and down and not instantly On or Off. Good chance the relay being turned On reaches its pull in voltage before the other relays drop out voltage is reached.

 

Ironically the system works best with the associated DMX channels set to 'Crossfade' (as opposed to 'Snap before fade' or Snap after fade'). The relay being turned on reaches its pull voltage after the other relay drops out... that is it usually does. Maybe 9/10. Good enough for home but that 1/10 times when the house goes completely dark after a cue makes me very very angry

 

Also, the relays whine and chatter during the fade which just doesn't sit well with me... Seems like there would be an electronic workaround (resistor? Flux Capacitor?) on the 12v side but that's above my paygrade.

[frustratednon-geek]

The DMX board can delay the change to the second I/O Linc Sensor for a few seconds to allow all the messages from Sensor 1 to complete before starting Sensor 2.

This is what we've been doing for the past year or so but:

• [*:iij6kqni]Instead of programming a scene, you have to program a 2-step 'show.' Step 1 - a special 'Relay Blackout Scene' for 0.5sec, then Step 2 - desired scene.

• [*:iij6kqni]During the 0.5sec relay blackout the lights fade down slightly before going up--kinda cheesy.

• [*:iij6kqni]Too complex for the average user to program lighting board.

An alternative would be to eliminate the Responders being linked directly to the I/O Linc Sensors, having ISY Programs issue the Scene commands. This may make things worse if Insteon comm is poor as an ISY Program turning a Scene On/Off only sends the Group Broadcast message.

We've actually got pretty good Insteon Comm and most if not all devices are dual-band. I'm pretty sure I tried this about 6 months ago and the chatter was confusing the ISY as well.

 

What if I used a program to query the devices again after a wait. The trigger being a change in state of either I/O Linc:

If: either I/O lincs are switched on or off

Wait: 0.5 secs

Query both I/O lincs again

If: I/O 1 is on, set scene 'message' etc.??...

[Xathros]

How about replacing the relays with OptoIsolators? That should give you a clean on/off point without the chatter. You will still need to time things to avoid Insteon traffic jams but it sounds like you have a few options for that already.

 

-Xathros

[frustratednon-geek]

How about replacing the relays with OptoIsolators? That should give you a clean on/off point without the chatter. You will still need to time things to avoid Insteon traffic jams but it sounds like you have a few options for that already.

-Xathros

So that's what OptoIsolators do then? Where would I find them? Is this something I would just wire inline with the sensor inputs?

 

Quick update: after some tinkering with the length of 'crossfade' in the DMX control board, I seem to have the system working as intended. There's a fairly loud whine as the relays fade in and out but it works. The opening relay triggers an 'off' command just prior to the closing relay triggering the 'on'. They don't step on each other any more and I don't have to program a blackout... Will keep you posted. Googling 'optoisolator' now...

[frustratednon-geek]

Would this work?

 

http://dx.com/p/4-channel-12v-opto-isolator-relay-module-w-high-level-trigger-blue-233833?tc=USD&gclid=CNjL47zxprsCFQqGfgodOTcALA

 

Something with adjustable set points would be perfect.

[Xathros]

That looks to me like it would do what you want. You won't be able to adjust the trip point but it should be very consistent allowing you to work out the timing with the DMX board or ISY programming.

 

-Xathros