larryllix

This is a program I created to stabilise inputs from more than one source. Any "wild" sensor inputs get cut out of the calculation. Operation the last known working average is kept for comparison purposes during the next iteration. a constant $cAVERAGE.DEV.MAX variable is set to the allowable deviation limits for a sensor to be considered "bad". I have set mine to 2.0 degrees for this purpose each input is compared to the last known average and the absolute value compared to the allowable deviation factor. good sensor readings are tallied and used to calculate the new average if all values are outside the maximum deviation then a completely new average is created from all sensors. Benefits gives a more stable resultant detects and isolates any sensors gone bad or "out-of-range of the crowd" readings. gives a contributor count that can be used for other purposes. eg: Temperature variance detection for circulation. fan operation from indoor temperature sensors I have a temperature sensor mounted up under a PV array and the oven effect gets disallowed every late morning with this program. this can use as many sensors as available to further stabilise the resultant output. Just add more lines of program code. ISY is fast. No delays can be noticed and lots of repeats allows other processing to occur during the calculations. Negatives based on slow changing sensor values like temperature readings from thermostats and probes. Sudden value jumps may disrupt the process. all three sensors outside the past average results in including all three sensors to start. This may not be the best algorithm for some cases. we sure could use an inline If/Then construct in v5. ----------------------------------------- Sync.outTemp.average - [iD 00FE][Parent 0101] If $sTag3.temp < 40 Or 'Gathering Room / GathRm VenStat' Sensor < 45.0° Or $sWC8.outTemp.raw < 450 Then $Average.sum = 0 $Average.contrib.cnt = 0 // sum inputs within dev.max of past average and keep count $Average.deviation = $sWC8.outTemp.raw <--WC8 board ships to ISY x 10 integers only $Average.deviation /= 10 $Average.deviation -= $sHouse.outTemp Repeat While $Average.deviation < 0 <----- calculate the absolute value 2 lines $Average.deviation *= -1 <----- make a negative, positive Repeat While $Average.deviation <= $cAVERAGE.DEV.MAX $Average.sum += $sWC8.outTemp.raw $Average.sum /= 10 $Average.contrib.cnt += 1 $Average.deviation = 999 <-----terminate the Repeat loop Repeat 1 times $Average.deviation = 'Gathering Room / GathRm VenStat' Sensor ° $Average.deviation -= $sHouse.outTemp Repeat While $Average.deviation < 0 $Average.deviation *= -1 Repeat While $Average.deviation <= $cAVERAGE.DEV.MAX $Average.sum += 'Gathering Room / GathRm VenStat' Sensor ° $Average.contrib.cnt += 1 $Average.deviation = 999 Repeat 1 times $Average.deviation = $sTag3.temp $Average.deviation -= $sHouse.outTemp Repeat While $Average.deviation < 0 $Average.deviation *= -1 Repeat While $Average.deviation <= $cAVERAGE.DEV.MAX $Average.sum += $sTag3.temp $Average.contrib.cnt += 1 $Average.deviation = 999 // none worked, get all new Repeat While $Average.contrib.cnt is 0 $Average.sum = $sWC8.outTemp.raw $Average.sum /= 10 $Average.sum += $sTag3.temp $Average.sum += 'Gathering Room / GathRm VenStat' Sensor ° $Average.contrib.cnt = 3 Repeat 1 times // finish calcs with what we got $Average.sum /= $Average.contrib.cnt $sHouse.outTemp Init To $Average.sum $sHouse.outTemp = $Average.sum Else - No Actions - (To add one, press 'Action')

When replacing variables or devices I always add and XX suffix to the end of the old device's name. When doing the replace it is easy to get the correct device and not make mistakes Replace kitchenSwitch XX with kitchenSwitch. Don't forget that scenes have to have devices removed and replaced with the new, as LeeG pointed out, in case that wasn't clear.

If v5++ is being used, I wrote some accumulated time functions that work well by calcs with the ISY "seconds since midnight" feature. http://forum.universal-devices.com/topic/17366-v5-tracking-fan-cycle-runtime/ This could also give you a daily report of the, length of run, from the sun.

In our first home I filled it with X10 devices. I removed some of the wallswiitches but just couldn't live without a few of them so we listed and showed the house with the X10 switches still in the wall. Mostly they were plug-in modules. After the sale, during the long closing period, I wanted the devices to go with me so I found some of the old small rocker switches to replace the old and similar X10 pushbutton types. They were obsolete by that time and hard to find. When closing day came, I was in a panic as they had bargained in a final inspection, the last day. This was a new thing just starting up in our area. The new owner woman came in the hallway, and went right to the X10 switch that wasn't there, and asked what happened to the switch. I explained it was part of the security system (BS - well sort of) and if they wanted it I could replace it, but they wouldn't know how to operate it anyway. She came back with something like "Good! I was having nightmares for the last four months about those strange electrical gadgets in my house and would have had to hire an electrician to take them all out!" YMMV

To clarify further (maybe). It the indented lines after the Repeat 1 time that gets executed once. "Repeat" is not syntactically correct. Read as "Do 1 times" Then Set Scene 'Night Time Mood Dim' On Repeat 1 times Set Scene 'Night Time Mood Dim' On

Post the programs. You may have a quirk in your negative logic that stops the program process. Right click on program and "copy to clipboard"

Are we sure the lights were on in the first place? Could be more Insteon comm problems.

Tube amps with 2-5% distrotion were very common. They were not all that great of a quality. When the solidstate amps came into being even with low 0.1 - 0.5% distortion figures people didn't like the sound of them. The crossover distortion from the class AB or class B amplifiers switching from push to puil annoys human ears and left engineers wondering why. Tubes have smother crossover in a gentler linear switching fashion. Their 2-5% distortion figures are mostly harmonic distortion and smotther to human ears. Think winding lead guitars with the smooth distortion eg: Carlos Santana, Eric Clapton etc.. It just isn't as annoying and simulates the harmonics naturally found in many instruments. I think the crossover distortion is also consatnt with transistor switching so that very soft levels of music leave crossover distortion figures into the 10-20% distortion at low levels. Of course they always rate them at full output where the distortion is a low percentage. Harmonic distortion is typically a fixed ratio to the volume. Many claim CDs have a terrible sound and the sampling frequency (44.1kHz?) math proves they do, at high frequencies. This sounds like your expensive amps use Class A outputs and no transistor crossover switching distortion is heard simulating the old tube smoothness. I bet it gets hot when it works though..

Talk about bug detector radar!!!!!!

They must be running full class A push-pull outputs (think EasyBake) to get that good a sound with the technology back then. I know FETs can do it better but the initial decades of solidstate transistors had a nasty crossover distortion. The tube amps, cherished so much, had much higher levels of distortion but peoplr liked it 'cause it was "good" distortion and not harsh to the human ear.

Yeah, the floppies were a typo. should have been $100 and hardsectored which lost out due to MS taking over and I had to drive to Toronto to get them, not found beside the kiddie candy at the grocery checkout as in later years. Always wanted an 8" drive at home because they could handle 77 tracks and over 100K of storage. Makes ya' laugh now.

LOL. With all that audiophile hype I thought it would at least be a tube amp and the tubes would be worth that much. Those amps look really familiar from the good ole' days. These were the days when power ratings had no "peak Watts", or "RMS Watts" or "Average Watts" rating BS, just real power and heated our houses too.

I should dig my old 64K byte statuic memory board out of the dump then. The memory quality should kms (US=miles) better than the massive GigaByte memories today for only $20, as I paid almost $1000 for it, populated to 32K with 16 x 2K rams, but it was only 2MHz access speed? Buyers anyone? Wait!... I paid almost $100 for a box of 10 floppies too... DIT: The floopies were not $1000

You need to edit the setup in your router, adding an IP address reservation for the ISY's mac address. This way ISY always gets the same IP address. Then you need to enable DHCP in your ISY Now in your router you need to configure a port forward ttable entry. This is a conversion from an outside your LAN port address, to an inside your LAN IP address and port address that your ISY will always be found at. So when a remote app, like MobiLin,c calls your home IP address (router external) with the port that you have forwarded, you router will now convert that IP and port address to ISY's internal address on your LAN. Now you are talking through your router with the firewall hole you just punched with the port forwarding feature in your router? Clear as mud? So if your ISP's DHCP server has issued your router an IP address of say 187.128.56.19 and you ISY has been assigned the IP address of say 192.168.0.101 by your router's DHCP server, and you haven't changed the default port of the ISY from 80. You would set up a router port forward of say port external port 443 to 192.168.0.101, port 80 internal. (your ISP 187.128.58.19 is assumed or it wouldn't get to your router) Now when somebody anywhere in thw worlds uses this URL https://187.128.56.19:443 (note "s"), you router will convert it to 192.168.0.101:80 on your LAN and your ISY will answer. Username and password required still.

Well If I knew exactly what I was doing and had worked it out to always work, maybe I could instruct you to be able to do it yourself. Having said all that, I can't. It would still be in the experimental stage for a boy that's been doing this kind of thing for many decades. At this point, usually somebody with different angle of experience pops in and says....why don't you just use this gadget? ...and that is what the forum is all about.

My guess it that these sensors just trigger a 12 vdc signal at the GDO that could take a sensitive realy to isolate and interface to a I/OLinc input. Some electronics knowledge would be necessary to do the measurements and trial and error necessary to make this work. A solid state opto-isolator type or even a diode, capacitor and a few voltage dropping resistors could make it work but polarities and more expertise would be necessary. An open collector buffer transistor may work better yet. I am looking to do the same thing on my furnace fan with 6-7 input speed controls at 24vac each going into an I/OLinc sensing.

Remove the LED from all other scenes and create a new scene for each LED. Drop the KPL LED into the scene and adjust to 'ON. (Scene On = LED On) This program is about as simple as ISY gets. This gives a 1 second delay to avoid Insteon traffic collisions but on an LED indication response it shouldn't matter. This does not change any other ISY logic that controls the door, just the LED indication Note: My contact is closed when the door is closed. Either contact can work with the 'reverse' option. Garage Door Open.LED - [iD 0097][Parent 0109] If 'Garage / Garage Door.closed' Status is Off <----I changed from 'On' to get status Off = door open = LED On Then Wait 1 second Set Scene 'Mudroom / MudRm kpl.LED.a' On Else Wait 1 second Set Scene 'Mudroom / MudRm kpl.LED.a' Off

I would just switch the scene command in the program lines to reverse the LED indication. Contacts do NOT need to be changed or switched from NO/NC. A better contact is not going to change the look of your LED. This is what having a PLC HA = ISY is all about... not having to change hardware. Coincidentally I just did this for my garage door last night. It took about 15 seconds after seeing the LED results and I decided the same thing. I want the LED to look like an unusual/alarm situation when it is On. Paul has a point about scenes vs. programs and that takes a certain type of contact but if your LED doesn't work while you are away will you notice? Don't use the "reverse" option in the ISY option box. It will cause you troubles, as mentioned above by others. To do this you need to remove the LEDs from the I/OLink scenes and create a new and separate scene for each LED. Then you need to write a program for each LED to operate that scene. If Garage door is open Then set LED scene On Else set LED scene Off

"...only pennies per day", has added up, through my lifetime, to be almost $100K per year. Most of us want ..finite and absolute... and "cloud' is becoming a dirty word, a way to disguise taking money from our wallets, a little at a time by dazzling our eyes with technology we haven't seen before.. Doesn't "cloud" suggest "confusion", "haze", "can't see", "unresolved", "unknown"? Probably a really bad nickname for something we are not what to call it in a "one-size-fits-all" naming attempt. Clouds blow away and the sky may eventually become blue and clear..

Your Elk zone is probably timing out, sending an Off to ISY. Your pogram sees the Off and runs the Else section, stopping the Then section from finishing Either make a disabled program and call it from the first one, shorten your Wait 10 minutes, or lengthen your MS time off delay so it survives the Wait 10 minutes timer. If From 9:00:00PM To Sunrise (next day) <----these two lines can cause Else to run and Then to stop And Elk Zone 'Upstairs MS' is Violated <---- Then Run Program2 Else ---- Program 2 (disabled) If -- Then In Scene 'Stairs Light' Set 'Stairs Light - Top of Stairs' 35% (On Level) Wait 10 minutes Set Scene 'Stairs Light' Off Else - No Actions - (To add one, press 'Action')

If you are referring to the Amin Console and you have your full colour option on for the icons, you will see a solid green icon when the program runs, and a partial green icon once the program is done running. Your program would run so fast that you would not likely ever see the solid colour. The green colour is a remnant that will stay to indicate your program last ran True = Then section. This is normal behaviour. You could add a wait 5 seconds fr testing so you can see the different states.

Ditto what Techman reported. Also there are many temperatures sensors CAO Tags are very accurate but cloud dependant to report. Autelis with 1 wire sensors for totally LAN CAI WC8 with 1-wire sensors for totally LAN. Several Venstar, Honeywell, ecobee3 with NodeLink running on an RPi but the Venstar is the only one not cloud dependant and they all take AC power. I use 2441ZTH stats for probes and one for anti-freeze heater in my utility room.

Is this the box you are talking about? also... If so, do you have an IP address installed in the list as shown in the screen shot?

What does ISY Finder say?

When the ISY Finder comes up, click "Add" and put in your ISY URL like http:///192.168.x.xxx. If you only make one entry it will boot up the admin console automatically. If you have more than one you may have to double click on the desired URL each time.