HOME › Forums › Input/Output › EZIOxx › EZIO2x4 not fast enough?
-
AuthorPosts
-
Anonymous
InactiveJuly 10, 2007 at 6:19 amPost count: 4I finally got around to replacing my X10 IOLinc with an EZIO2x4 to monitor my doorbell. This was one of the few remaining X10 devices in my setup (the only remaining now are a subset of my KeypadLincs that I need to be able to turn individual button status lights on), so I was happy.
Unfortunately, I have erratic success. Basically, the thing doesn’t always tell indigo when an input turned on, even if it tells it when the input turns back off. Is anyone else having this problem?
An example of a success:
Jul 9, 2007 10:01:09 AM
Received INSTEON “EZIO2x4” inputs status changed: 1, 0, 0, 0
Trigger Action Doorbell (EZIO)
Received INSTEON “EZIO2x4” inputs status changed: 0, 0, 0, 0An example of a failure:
Jul 9, 2007 8:37:20 PM
Received INSTEON “EZIO2x4” inputs status changed: 0, 0, 0, 0With the IOLinc I got 100% reliable signalling. Once I pressed my doorbell button, the doorbell detector circuit triggered the contact input, and the ‘on’ was sent. A few seconds later an ‘off’ was sent – I guess the IOLinc had some buffering in it.
InactiveJuly 11, 2007 at 2:47 pmPost count: 256Detection of the switch closure should not be an issue for the EZIO devices since the inputs are sampled several times per second. Even when “debouncing” is on, the effective sampling is over 10 times per second. The generated message is a broadcast and we’d like to know where it is missed. Unfortunately we don’t have Indigo set up but we’ll attempt to investigate. Meanwhile tell us:
– What version of firmware is reported by the EZIO2X4?
– Does debouncing the input make a difference in missing events?
– Is the input linked to another Insteon device and thus sending a group message on activation?InactiveJuly 11, 2007 at 2:58 pmPost count: 18@simplehome wrote:
Detection of the switch closure should not be an issue for the EZIO devices since the inputs are sampled several times per second. Even when “debouncing” is on, the effective sampling is over 10 times per second. The generated message is a broadcast and we’d like to know where it is missed. Unfortunately we don’t have Indigo set up but we’ll attempt to investigate. Meanwhile tell us:
– What version of firmware is reported by the EZIO2X4?
– Does debouncing the input make a difference in missing events?
– Is the input linked to another Insteon device and thus sending a group message on activation?This is the same problem that I encountered (and reported to Al) a while back; I have firmware v. 1.7.
>> Does debouncing the input make a difference in missing events?
The Indigo UI only allows you to set input debounce for analog inputs; In retrospect, that doesn’t make a lot of sense…
>> Is the input linked to another Insteon device and thus sending a >>group message on activation?
No — just to the PowerLinc.
Let me know if you need more details…
Mike
InactiveJuly 12, 2007 at 12:39 amPost count: 4I have the same setup. My doorbell detector is connected to I1 in the ‘non-isolated’ mode, so I don’t have an option to debounce the input (other than building some additional circuitry to do it, which I’m not willing to do since my SmartHome IOLinc worked fine with the same detector). The EZIO is linked to the PowerLink, and nothing else.
My EZIO just came in like a week or so ago, and has v1.7 of the firmware.
InactiveJuly 12, 2007 at 10:10 pmPost count: 256The debouncing algorithm within the EZIO applies to the digital inputs, not the analog (which would not make sense). Perhaps there is a misnomer in Indigo. In any event, this is available such that the user doesn’t have to worry about debouncing switches externally.
InactiveJuly 12, 2007 at 11:41 pmPost count: 18@simplehome wrote:
Perhaps there is a misnomer in Indigo.
I have already pinged Matt.
InactiveJuly 16, 2007 at 4:00 amPost count: 3I was having the same problem so I tried linking back to the PowerLinc. I started with input 3. It worked like charm, reporting input 3 as “Button 3”. Then I linked buttons 1,2 and 4. Now all four inputs always show up as “Button 3”. I tried linking 1, 2 and 4 again but same problem. I can’t figure out what I am doing wrong. I want to unlink input ‘button’ 3 to see if I have the same problem on other inputs but I can’t figure out how to unlink from the powerlinc. I figured it would be the same as linking to the powerlinc but starting with 2x 10 sec presses of SET but that doesn’t appear to be working and I can’t find any unlinking proceedure in the EZIO manual. HELP!
InactiveJuly 21, 2007 at 3:51 amPost count: 4One of the other users has reported that manually setting the buffer bit (I assume he used the Windows configuration tool) did not solve the problem.
The suggestion also didn’t really make a lot of sense, since the problem wasn’t tons of spurious on/off transitions but rather the lack of the ‘on’ transition being sent if it occured less than a second before the ‘off’.
Are there any other suggestions on getting this fixed? At this time I am going to have to go back to using an X10 relay sensor so that my doorbell functions consistently.
InactiveAugust 3, 2007 at 6:33 pmPost count: 4Any more suggestions, or do I need to go add some components to make the signal longer? (at best a capacitor, at worst a latch with a 1-second clock)
Maybe there is a firmware update coming that will help?
InactiveMay 14, 2008 at 5:20 pmPost count: 8I’m considering the purchase of an EZIO2x4 but had a question that seemed directly relevant to this thread. The short version of my question is, will the EZIO2x4 detect a momentary contact closure that lasts only 10-20ms?
Now for the longer version (with a few more detailed questions)…
I have a Rainwise RAINEW 111 rain bucket (http://www.rainwise.com/products/detail.php?ID=6697&Category=Rain_Gauges:Wired&pageNum_cart=/products/index.php) and I’d like to know if I can hook that to the EZIO2x4. The rain bucket makes a momentary contact closure whenever the “rocker” device tips from one side to the other. This momentary contact closure may only last 10-20ms. As long as your EZIO2x4 is fast enough to happily detect this brief contact closure, then I believe it should work just fine. However, this thread leads me to believe it simply won’t work.
If it won’t work, I could add something like an ELK 960 to make the signal longer (as X-caiver suggests), but that ELK 960 requires 12v. I see from the pictures of the EZIO2x4 that it has a terminal labeled 12v, but I can’t find any documentation that says anything about it. Does it really provide 12v? Is there any documentation beyond the QuickStart PDF and the CommandSet PDFs? Is there a cheaper way to get the EZIO2x4 to signal short contact closures, if in fact it can’t already?
InactiveMay 14, 2008 at 11:23 pmPost count: 1001Someone else will have to handle the minimum contact time but I can address the 12V question. The EZIO2X4 has a +5V output, it does not have a +12V output. What you are seeing are the two opto-isolated inputs; I1-, I1+, I2-, I2+ (not 12+). The EZIO6I does have a +12V output, but does not have output devices (relays), like the EZIO2X4.
InactiveMay 15, 2008 at 1:42 amPost count: 8OK, I felt like a bit of an idiot for confusing an input with a power source, so I had to go back to the picture to see how obvious it was. Maybe I’m rationalizing, but the characters “eye two” really do look like a twelve, and I guess hopeful thinking on my part just wanted a twelve volt supply. So much for using an ELK 960 on this (I’ll check out the EZIO6I), so hopefully someone can tell me how often the input is sampled on this EZIO2x4.
Thank you so much for setting me straight.
InactiveMay 15, 2008 at 2:19 amPost count: 1001Easy enough to make that mistake from an internet picture. I had the advantage of having three EZIO2X4s already installed. I’ve had good success with the EZIO2X4 but all of my circuits are basically states. Door open/closed, flood light motion on/off, etc. If you don’t hear something in a few days, call Simplehomenet support. They have a good reputation for responding to questions about their hardware.
Also, from one of the posts from last year it looks like the EZIO2X4 was set up to issue Broadcast commands on state changes. The Broadcast command is not directed to any specific device so if it is not received, it is not resent. If you link your Input to something, A Group Cleanup Direct command is used, which is retried if the linked to device does not acknowledge receipt of the command.
EDIT: Corrected ACKed command name to Group Cleanup Direct
InactiveMay 15, 2008 at 2:40 amPost count: 8That’s great information about the Broadcast command. Thank you. I definitely want some reliability in the messages, since the whole reason for using this is to avoid hard-wiring the rain gauge. I want this to communicate to a Powerlinc Modem, which in turn is connected to a HomeVision home automation controller, so linking it up and checking for the broadcast cleanup should be easy enough.
InactiveMay 16, 2008 at 4:38 amPost count: 408A 2X4 samples every 10 milli seconds if debounce is disabled. This means that if a signal is less than or equal to 19 milli seconds, there is a good chance it will be missed.
My recommendation is to use the 5VDC from the 2X4, and rig up a latch, with one of the outputs from the 2X4 used to reset the latch. You would sense on the input the signal, and then create a link to the output to reset on sampling a signal (plus a broadcast to whatever device you want). This should only be a couple bucks from Digikey, however it will involve some soldering skills…
Let me know if you want more details on how to rig something like this up.
-
AuthorPosts
- You must be logged in to reply to this topic.