Scenes vs Programs question

[aviatordave]

Hi folks,

 

I haven’t been on in a while because everything has been running smoothly! Love that!

 

Just recently, however, I tackled a new project that I’m pretty happy with. It’s a pond filling system. We have, what you’d call a water feature, behind our house. It’s composed of an upper and lower basin, connected by a little creek that allows the upper basin to flow down into the lower. The upper basin isn’t that large but the lower basin is probably about 1000 gallons of water. There are a couple of water falls and a pump box in the system to keep the water moving. Because of splash from the waterfalls and evaporation in general, the system loses water over time.

The Insteon system I set up has two stainless float switches that control a high and low water point. Each of those switches are wired to an Insteon I/O Linc. There is also an Insteon outdoor outlet in the system that when on, opens a solenoid valve to allow a garden hose to start filling the pond. All the Insteon stuff is housed in a weather tight project box. The I/O Lincs turn the outdoor outlet on and off. Because the float switches each have an open and closed state, I have to make sure the solenoid valve only opens when the lower float switch is closed and shuts off when the upper float switch is open. (Both float switch states are closed with the float in the lower position and open in the upper position.)

Initially, with this configuration, I tried setting up a scene where the IO Lincs both controlled the outlet. That didn’t work because when either switch state was reached it would affect the outlet switch. In other words, when the water got low enough to let the lower float switch close and turn on the outlet, it also turned off the outlet when the lower float rose with the water level and opened the switch, well before the water got to the desired upper float level.

Easy fix. Just write a couple of short programs instead of using a scene. If lower float IO linc is On (float switch closed) them turn on the outlet/water. Second program, IF upper float IO Linc is off (float switch open), then turn outlet/water off.

Both programs send me a text message alerting me to the change in states.

Works great.

 

So finally, my question: I noticed when using a scene and physically manipulating the float switches to test it, the system responded practically immediately. When using the programs, there is a noticeable delay of about 1-2 seconds before the system reacted to the float switch changes.

Is it possible, WITH A SCENE, to have two IO Lincs control one outlet but only with a single state change from the float switches? By this, I mean can the lower float IO Linc have the water turn on when the lower float switch is in the closed position but IGNORE the lower float switch when the water rises a little and shifts the float to the upper, open position and continue to fill the pond until the upper float switch is pushed to the upper/open position, turning the valve outlet/water flow, off?

 

With the reverse being true as well, can the upper float switch IO Linc ignore when the upper float drops to it’s lower closed position and not turn the water on until the water drops enough to allow the lower float switch to handle the turn on the water state?

 

This is obviously child’s play for a program to accomplish, but I didn’t see anywhere in the ISY system that allowed for that kind of detailed control in a scene. (Probably why programs are needed. The only reason I ask is because I noticed how much faster things happened with scenes vs programs. I don’t need my pond watering system to react that fast but I can think of other situations where near instantaneous reactions would be desirable.

 

I look forward to your thoughts on the matter!

-Dave

 

P.S. For those of you asking why bother having two float switches when one would do the job? I tried this originally. The float switch is just a vertical, linear setup with a reed switch inside a stem that the float with a magnet rides up and down on. Even though the float switch has about 1/4” of travel between it’s upper and lower stops along the stem, there’s a specific point where only a hair’s width of travel one way will open or close the reed switch. It took so little movement to change open/closed states with the reed switch that the system changed states about 433 times in a 24 hour period. It would cycle on and the hose would only dump water into the pond for about 10 seconds before the float switch would open, shutting the water off. Couple minutes later, it would repeat. This had the desirable effect of keeping the pond water level amazingly consistent, but cycled the solenoid valve WAY too much. The manufacturer rates the solenoid valve for 200,000 cycles. At the rate it was cycling, the valve life would be about 1.5 years. I’m currently letting the water level in the pond drop about an inch between filling cycles. Now the valve only cycles about once every two days. Muuuuch better valve life!!

 

 

Sent from my iPhone using Tapatalk

[aviatordave]

Here’s some pics of the setup if anybody is curious:

 

 

Sent from my iPhone using Tapatalk

[Whitehambone]

Nice set up! I see a few leaves on the ground, I assume you live in a climate with all 4 seasons.  Does that system require winterizing?  

 

I never tried to configure a Scene they way your describing.  I noticed there is an Adjust Scene to Ignore in the Program Action, but I never used it.  Perhaps that is a way of disabling the Scene after it triggers quickly.  Then maybe use two Scenes, one enabled and the other set to ignore?  Just thinking out loud (or in writing).  

[larryllix]

With scenes you cannot have any conditional logic. They are not as dependent on ISY, for sure, but when the power is off the pumps don't work anyway.

Scenes can be modified, on the fly to create conditional logic, but that still is dependent on ISY then and scenes can make your logic obscure and hard to troubleshoot, or understand, a few years later when you need to update something.

That many operations per day, something will give up the ghost as contacts do not like breaking motor load circuits that much. Your ISY programming is obviously smoothing things out there. If exact levels are important your ISY program timers could be tweaked somewhat, I would think. 1-2 cycles per day could be a good compromise.

[tazman]

I would think either turn the pump on X amount of time after the float calls for water or only have the program active twice a day.

[aviatordave]

Whitehambone, we most definitely live in a 4 season environment. I will have to winterize the setup. I’ll use some compressed air to blow the lines clean. I’m debating whether I should unplug the IO Lincs and bring them inside. I can’t think of an immediate reason that they would be harmed by the cold since they’re in a weatherproof enclosure and wouldn’t see any contact with snow . . . The power would be off. But I don’t know if there are any components to the circuitry inside them that might not like being frozen. (Capacitors?) I might be more concerned with ice damaging the float switches . . . This exercise in thought is making me wonder about the extent of winterization this system will need. Thank you for bringing up this point!

 

I suspected that Scenes weren’t that robust, but larryllix, you make a good point about potential obscurity. It’s is much easier to see what’s going on with program statements. Scenes just seem to be a bunch of statements of responder/controller relationships. And even those can be confusing sometimes. I agree that several years removed, it might be difficult to interpret. I don’t need the water level in the pond to be that precise. If I need to adjust it, I’ll do it with changes to the levels of the two float switches. I made the brackets for them, specifically to allow for sliding up and down to refine positions after installation. I was just curious about whether or not Scenes could could be manipulated and to what degree.

 

I sure love the creativity allowed by IO Lincs!

 

Thanks for the feedback, folks!

 

 

Sent from my iPhone using Tapatalk

[aviatordave]

Hi tazman,

I used to have a simple garden hose timer turning on a couple times a day for X minutes. I could never seem to dial in the exact times needed. What I’ve learned about water features is that they lose water at different rates based on temperature and humidity. Also, if it rains a lot, then there’s the problem of having the pond overfill and now I’m needlessly dumping water over edges into the area around the pond.

That, and most importantly, I like to tinker so I can’t just leave well enough alone. ; -)


Sent from my iPhone using Tapatalk

[larryllix]

37 minutes ago, aviatordave said:

Whitehambone, we most definitely live in a 4 season environment. I will have to winterize the setup. I’ll use some compressed air to blow the lines clean. I’m debating whether I should unplug the IO Lincs and bring them inside. I can’t think of an immediate reason that they would be harmed by the cold since they’re in a weatherproof enclosure and wouldn’t see any contact with snow . . . The power would be off. But I don’t know if there are any components to the circuitry inside them that might not like being frozen. (Capacitors?) I might be more concerned with ice damaging the float switches . . . This exercise in thought is making me wonder about the extent of winterization this system will need. Thank you for bringing up this point!

I suspected that Scenes weren’t that robust, but larryllix, you make a good point about potential obscurity. It’s is much easier to see what’s going on with program statements. Scenes just seem to be a bunch of statements of responder/controller relationships. And even those can be confusing sometimes. I agree that several years removed, it might be difficult to interpret. I don’t need the water level the pond to be that precise. If I need to adjust it, I’ll do it with changes to the levels of the two float switches. I made the brackets for them, specifically to allow for sliding up and down to refine positions after installation. I was just curious about whether or not Scenes could could be manipulated and to what degree.

I sure love the creativity allowed by IO Lincs!

Thanks for the feedback, folks!


Sent from my iPhone using Tapatalk

I would unplug the pumps and leave the i/OLincs powered up for the winter. You wouldn't want a frozen pump to be powered up by some HA error. The minor electronics heat will keep the Rh down inside the case by raising the temperature inside the units, lowering the Rh.

[lilyoyo1]

1 hour ago, aviatordave said:

I suspected that Scenes weren’t that robust, but larryllix, you make a good point about potential obscurity. It’s is much easier to see what’s going on with program statements. Scenes just seem to be a bunch of statements of responder/controller relationships. And even those can be confusing sometimes. I agree that several years removed, it might be difficult to interpret. I don’t need the water level in the pond to be that precise. If I need to adjust it, I’ll do it with changes to the levels of the two float switches. I made the brackets for them, specifically to allow for sliding up and down to refine positions after installation. I was just curious about whether or not Scenes could could be manipulated and to what degree.

 

I sure love the creativity allowed by IO Lincs!

 

Thanks for the feedback, folks!

 

 

Sent from my iPhone using Tapatalk

I'm opposite of larryllix in that I use scenes predominantly and use programs to add capability to them. I prefer scenes mainly due to the fact that I have manual control over my devices regardless of the isy. For me, it's paramount that my system works outside of the Isy. 

If the isy dies how do you operate things? Unless you have one laying around your system is down until you get a new one. At least by setting scenes, you can operate the pumps from home with a kpl as needed until then. 

Edited July 26, 2020 by lilyoyo1

[tazman]

1 hour ago, aviatordave said:

Hi tazman,

I used to have a simple garden hose timer turning on a couple times a day for X minutes. I could never seem to dial in the exact times needed. What I’ve learned about water features is that they lose water at different rates based on temperature and humidity. Also, if it rains a lot, then there’s the problem of having the pond overfill and now I’m needlessly dumping water over edges into the area around the pond.

That, and most importantly, I like to tinker so I can’t just leave well enough alone. ; -)


Sent from my iPhone using Tapatalk

I would still leave the float to turn it off you can do 2 different timers I'm thinking something like this

Float activates program

Wait 5 minutes 

Run pump

Float rises and runs another program

Wait 2 minutes 

Turn off pump

That should stop the on off that you are seeing and if that runs to frequently you can also have the program only active twice a day or something like that

[simplextech]

2 minutes ago, lilyoyo1 said:

For me, it's paramount that my system works outside of the Isy. 

I'll second this sentiment.  Systems should be able to "function" through manual control as well as automations (programs).  The programs add spice and "smarts" to the overall system but "things" happen and it defeats the idea of a smart home when a dumb switch doesn't work because the "system is down".

[aviatordave]

Tazman, I’ve already stopped the problem of too frequent of cycling with the addition of the second float. The system only runs once every couple of days now. It’s great.

Lilyoyo1, simplextech, I couldn’t agree more. That’s one reason I posted this question. (Which, I’ll admit, is hard to keep track of since I have a small novel’s worth of rambling to explain why I’m asking.)

Do you know of a way to set up a scene so that two IO Lincs can control one outlet in the manner that I describe in the OP? I want certain states of the float switches to trigger the outlet that controls the water valve, and other states to be ignored.

To simplify, I want one IO Linc to only turn on the outlet (open the water valve) and the other IO Linc to only turn off the outlet. Lower float switch closed, upper float switch open.
I want both IO Lincs to ignore the float switches when the lower one is open and the upper one is closed.
So far, what actually happened with scenes was that each IO Linc controlled the outlet when either float switch changed states, regardless of open/close status. Which means the water level around one float basically takes control and ignores the second float, which puts me back at cycling too much.
I solved this with programs but wondered if it is possible with scenes.


Sent from my iPhone using Tapatalk

[larryllix]

39 minutes ago, aviatordave said:

Tazman, I’ve already stopped the problem of too frequent of cycling with the addition of the second float. The system only runs once every couple of days now. It’s great.

Lilyoyo1, simplextech, I couldn’t agree more. That’s one reason I posted this question. (Which, I’ll admit, is hard to keep track of since I have a small novel’s worth of rambling to explain why I’m asking.)

Do you know of a way to set up a scene so that two IO Lincs can control one outlet in the manner that I describe in the OP? I want certain states of the float switches to trigger the outlet that controls the water valve, and other states to be ignored.

To simplify, I want one IO Linc to only turn on the outlet (open the water valve) and the other IO Linc to only turn off the outlet. Lower float switch closed, upper float switch open.
I want both IO Lincs to ignore the float switches when the lower one is open and the upper one is closed.
So far, what actually happened with scenes was that each IO Linc controlled the outlet when either float switch changed states, regardless of open/close status. Which means the water level around one float basically takes control and ignores the second float, which puts me back at cycling too much.
I solved this with programs but wondered if it is possible with scenes.


Sent from my iPhone using Tapatalk

The basic part should be easy.

• create a scene with the upper limit/float as the controller. Set the pump receptacle as a device 'on'
• create a scene with the lower limit/float as the controller. Set the pump receptacle as a device 'off'

There you have all the non-ISY control you will get

• create an ISY program to monitor the upper float and set a fail-safe timer to turn off the pump
• create and ISY program to monitor the lower float and the upper float and set a fail-safe timer to query both IO/Lincs.
•  
• create more ISY programs to overview/monitor things, as required and thought of, to attempt to make the system safer than just the two floats. Monitor IOLinc failure, Insteon comms missed, etc etc... Your cabinet is not metal so you should have no problems with RF standing waves inside the "resonator can" fro dual-band Insteon devices, should any be added or needed later. (new dual-band i/oLinc comes out in 2029)
   

Edited July 26, 2020 by larryllix

[Whitehambone]

7 hours ago, larryllix said:

Scenes can be modified, on the fly to create conditional logic, but that still is dependent on ISY then and scenes can make your logic obscure and hard to troubleshoot, or understand, a few years later when you need to update something.

Ain't that the truth!  I can't tell you how many hours I have spent looking at PLC programs I wrote years ago trying to figure out what I did or why.  Even with device names, comments and notes it can be a challenge.  

[aviatordave]

Larryllix,

Thank you for the info. I’ll give that a shot and let ya know how it turns out.

-Dave


Sent from my iPhone using Tapatalk

[keepersg]

Fantastic project!  I have a similar set up with two ponds, a waterfall feeding the upper pond, and a stream connecting the ponds.  I've got it set up so that the upper waterfall runs when a motion sensor (MSII which I've had all kinds of trouble with) detects someone.  The programs for that set a state variable when motion is detected, the next program disables the first, runs the pump for 10 minutes, then enables the first program.   The pump for that waterfall uses lots of watts which is why I have it set up that way.   I really like the stainless steel float sensors you have in your system.  Could you give the source for those?

[aviatordave]

Hi keepersg,

 

Your pond setup sounds pretty nice!

 

Drprm1 hit the nail on the head. Amazon is a great source for those kinds of float switches. Just search “Stainless float switch”. They have them with 90 degree bends, short stems, long ones, etc.

Be aware though, even though the float will have a travel of 1/8” to maybe 1/4”, the actual trigger point is somewhere in the middle, and it takes almost no movement of the float to toggle the switch on and off. Don’t think that the float will travel to it’s limits to make the state changes.

 

They also make double float switches. Two floats on one stem. Two switches. Set distances. You can’t adjust the range. (That’s why I went with two separate units. I can adjust the range.)

 

Other than those things you should be aware of, the switches work great.

 

Oh, one last thing: The L bracket you see on the picture of the float switch I posted does NOT come with it. I made that myself out of a small piece of metal. I cut a slot on the vertical part of the L so that I wouldn’t have to be so precise in the placement of the hole I drilled to mount it. Get the hole you drill within about a half inch or so of where you want it and make the slot in the bracket about 2ish inches long. That will give you plenty of play in getting the float height right without having to go through crazy gymnastics with a tape measure to drill in the exact spot.

 

Ok. I’ll get off my soap box. ; -)

 

 

Sent from my iPhone using Tapatalk

[aviatordave]

Hi Larryllix,

I tried the solution you suggested, one scene with the IOLinc connected to the lower float switch as a controller and the pond water valve ON as a responder. A second scene with the IOLinc connected to the upper float as a controller and the pond water valve OFF as a responder.

It’s not working.

Each individual float will turn the pond water valve on if a float reaches the down position and off when it reaches the up position.

I can’t seem to find a way for the IOLincs to ignore one position of their respective floats.

Could it be in my physical setup? Each float switch has two wires. It’s a simple completion of a circuit; Open or closed. I have one wire connected to the IOLinc ground terminal (GND) and the other to the sense (S) terminal. One float switch per IOLinc, both set up as described.

I can’t seem to click on anything in the scene setup that allows me to customize the conditions to be more specific than controller, responder, on, & off.
I’m looking for something like ‘ON only’ or ‘Off only’ but it doesn’t seem to be an option in the ISY.

Any advice?

-Dave


Sent from my iPhone using Tapatalk

[larryllix]

7 hours ago, aviatordave said:

Hi Larryllix,

I tried the solution you suggested, one scene with the IOLinc connected to the lower float switch as a controller and the pond water valve ON as a responder. A second scene with the IOLinc connected to the upper float as a controller and the pond water valve OFF as a responder.

It’s not working.

Each individual float will turn the pond water valve on if a float reaches the down position and off when it reaches the up position.

I can’t seem to find a way for the IOLincs to ignore one position of their respective floats.

Could it be in my physical setup? Each float switch has two wires. It’s a simple completion of a circuit; Open or closed. I have one wire connected to the IOLinc ground terminal (GND) and the other to the sense (S) terminal. One float switch per IOLinc, both set up as described.

I can’t seem to click on anything in the scene setup that allows me to customize the conditions to be more specific than controller, responder, on, & off.
I’m looking for something like ‘ON only’ or ‘Off only’ but it doesn’t seem to be an option in the ISY.

Any advice?

-Dave


Sent from my iPhone using Tapatalk

That is what I was thinking of, so it looks like you are on the right trail.

If the floats had a mechanical hysteresis that overlapped the other float...could work but OTOH the float never never reset to be triggered again.
Actually if you could introduce mechanical hysteresis you could do it with only one float. I have seen this done with a long shaft, several locking rings and a floating, sliding ring in between the two limit rings. The second float could be used in a wider allowance for extreme limits/emergency/failure  levels detector.

I have seen sump pumps with this style of float system where you can adjust either stop on the same float "stick" for the float pontoon to push.

Edited July 29, 2020 by larryllix

[aviatordave]

larryllix,

I had to chuckle when you mentioned hysteresis. I haven’t had to work with that term since my previous life in the aerospace industry. ; -) That said, I understand what you were getting at.

What I’m hearing from you though, is that what I’m trying to accomplish with the current physical float setup is probably not possible with just using scenes. As robust as Insteon devices are, I don’t believe that the IOLinc can ignore one of the two states of a switch, open/closed, through a normal Insteon linc, which is what a scene essentially creates. It just saves you from running back and forth between devices performing tap adds, ad-nauseam.

I agree with and have already contemplated a sump pump style switch. I might have to fabricate something to fit the bill though. If you go back to the pictures I uploaded, the second pic shows the space constraints that I have to work with for float placement. It’s pretty tight. Locating the switch elsewhere is not an option.

Thank you for taking an interest and engaging. I love being able to come to the forums for brainstorming! Even if it’s only to confirm suspicions. ; -)

-Dave


Sent from my iPhone using Tapatalk

[larryllix]

46 minutes ago, aviatordave said:

larryllix,

I had to chuckle when you mentioned hysteresis. I haven’t had to work with that term since my previous life in the aerospace industry. ; -) That said, I understand what you were getting at.

What I’m hearing from you though, is that what I’m trying to accomplish with the current physical float setup is probably not possible with just using scenes. As robust as Insteon devices are, I don’t believe that the IOLinc can ignore one of the two states of a switch, open/closed, through a normal Insteon linc, which is what a scene essentially creates. It just saves you from running back and forth between devices performing tap adds, ad-nauseam.

I agree with and have already contemplated a sump pump style switch. I might have to fabricate something to fit the bill though. If you go back to the pictures I uploaded, the second pic shows the space constraints that I have to work with for float placement. It’s pretty tight. Locating the switch elsewhere is not an option.

Thank you for taking an interest and engaging. I love being able to come to the forums for brainstorming! Even if it’s only to confirm suspicions. ; -)

-Dave


Sent from my iPhone using Tapatalk

The other thing that comes to mind is to put in a latching relay with each float activating Latch and Reset. Then run one I/OLink off the contact. You can get many voltage rating on P&B? KUPxxxxx relays for decent prices. I was a protection and control tech for 35 years and wired control panels for 5 years before that. I still remember some relay solutions from back then.

Edited July 29, 2020 by larryllix

[kclenden]

22 minutes ago, larryllix said:

The other thing that comes to mind is to put in a latching relay

I was just about to suggest a latch.  As I see them, the requirement looks like:

Float Position	Lower Limit Sensor	Upper Limit Sensor	Refill Valve
Both Up	OFF	OFF	Turn Off
Lower Up, Upper Down	OFF	ON	Ignore
Lower Down, Upper Up	ON	OFF	Not Possible
Both Down	ON	ON	Turn On

So if you use the Lower Limit sensor as the SET signal for the latch and invert the Upper Limit sensor signal and use it as the RESET for the latch, the output of the latch could then be used to control the IOLinc.

[larryllix]

7 hours ago, kclenden said:

I was just about to suggest a latch.  As I see them, the requirement looks like:

Float Position	Lower Limit Sensor	Upper Limit Sensor	Refill Valve
Both Up	OFF	OFF	Turn Off
Lower Up, Upper Down	OFF	ON	Ignore
Lower Down, Upper Up	ON	OFF	Not Possible
Both Down	ON	ON	Turn On

So if you use the Lower Limit sensor as the SET signal for the latch and invert the Upper Limit sensor signal and use it as the RESET for the latch, the output of the latch could then be used to control the IOLinc.

He could just run the pump off of the latching relay contact  or...

... get an ISY994 box and do it all in ISY programs.

 

[aviatordave]

You guys are too funny. ; -)


Sent from my iPhone using Tapatalk

[larryllix]

1 hour ago, aviatordave said:

You guys are too funny. ; -)


Sent from my iPhone using Tapatalk

We have gone insane from too much home automation. SkyNet has already started.