apostolakisl

You should look at it like this. 1) You run a recirc pump and keep your lines hot continuously for extended stretches without specific demand. 2) You run a recirc pump on demand Situation 1) Advantages. Hot water is damn near instantly at the faucet while it is running and you never wast water/time waiting Disadvantage. You paid for a holding tank and you pay extra energy costs as you lose heat into the house . .. only a problem if you are cooling your house . . . during months when the furnace is on it just means the hot water tank is heating the house instead of the furnace. Situation 2) Advantage. No need for a holding tank. Quick hot water but not instant, unless you have a way to trigger the system a minute or so before you need it (ie bathroom light turns on, you go pee, then need hot water). Uses less energy, except if you are regularly triggering the system when you don't actually need hot water. (ie go into the kitchen to get a beer). Disadvantage. Hot water is not instant if you flip on the light and go straight to the faucet. You seem to be trying to implement situation 2 controls into a setup designed for situation 1. I think you need to go with one or the other as a hybrid isn't likely to work well. Personally, I would go with a higher gpm pump and scrap the electric tank and go with situation 2. You need to look at the on-demand heaters spec sheet temp rise at the pumps gpm, and your typical incoming water temp to figure out the highest gpm you can run and still have the water get hot. EDIT: Acutally, you need to look at temp rise versus the temp of your house, not incoming water temp. The rercirc pump will be pumping water from the pipes at house temp back into the heater, not incoming water. So if your tankless can do 50 degrees of rise at 5gpm, that will give you 120 degree water . . .perfect.

If you aren't going to run the recirc pump 24/7 (or at least a lot of the time), you really don't need the electric tank at all. The only reason the plumbers would put that electric tank in there is to prevent the tankless from constantly going on/off and short cycling it to death with a constantly running recirc pump. If you use the program I wrote and take the electric tank out of the question, you should get a very reasonable result. With 2.5 gallons in your recirc system, a 5gpm pump should clear the lines in 30 seconds. Add to that a few seconds to kick up the heat on the on-demand tank and other imponderables. . . so maybe run the pump for 60 seconds. Then let it sit for however long the pipes stay relatively hot. Then let it re-eval for occupants in the water usage rooms and kick it on again if need be.

The thing is, the logic of running the pumps is wrong (not your logic, I know). There is no reason to run the pumps for 20 minutes. As soon as the pipes get warm, which probably is going to take a minute or so depending on the gpm and the number of gallons in the pipes, there is no reason to keep running the pumps. Presumably, if someone goes into the bathroom, they are going to start using the water in less time than it takes for the pipes to cool off. And once you start using the water, there is no reason for the pump to run at all, the pipes will stay warm. The program I wrote and then slightly modified should work quite nicely. You might need to adjust the on time and off time a little bit depending on the details of how many gallons in the pipes and the pump gpm, and how quickly the pipes cool based on where they run and how well insulated they are. the overall jist of my programs is 1) When someone enters a room where they might use water, the pipes get flushed of cold water and warm water replaces it, ready to be used. 2) If after a period of time, that the pipes might have cooled, and someone is still in the room, the pump runs again to freshen up the hot water. Of course this is only necessary if they weren't using the hot water, since if they were using it, the pipes would stay warm on their own. But there is no way to detect that aside from putting a thermometer on the pipes. If you really wanted to do this right, you would put a thermometer on the pipes in each of those 3 rooms and only run the pumps if a person is in the room and the pipes are cold.

Here is a better way: IF status guest bath light is on or master bath motion is turned on or status kitchen light is on Then call program 2 if clause Program 2 If Elk is not armed away Then disable program 1 turn pump on wait 2 minutes turn pump off wait 15 minutes enable program 1 run if clause program 1 No need for a folder. Program 2 needs to be disabled as well so that it only runs when called from program 1

The locked on condition is always a problem contingency with disabling programs. Also, if I am using a motion sensor to turn on lights I like to put a delay in the non-priority program in an attempt to keep the priority lights response program faster. I am not sure how valid this is but it makes sense to me. Adding the Elk status into the conditions shouldn't be needed as the short pump cycle could just time off or the else could be used for cleanup from any condition to terminate the in progress cycle immediately. This would also prevent your burglars from getting any preheated water convenience. IF ( status guest bath light is on or master bath motion is turned on or status kitchen light is on ) And Elk armed away is false Then wait 2 seconds repeat 4 times . turn pump on . wait 2 minutes . turn pump off . wait 15 minutes Else turn pump off The above won't work because of re-triggers. You have to separate the if/then and disable the if program while the "then" program is running.

Add a program outside the folder that turns the pump off when the system is armed.

This isn't really going to work out very well. You have 3 different things all controlling the same thing and doing so differently. They will compete with each other. First thing, put all the programs in a folder and set the folder condition such that when the alarm is armed away the entire folder is inactive. That takes care of that. Otherwise, I would suggest simply putting all of your conditions in a single if folder and having the pump turn on for a couple minutes (however long it takes to warm up the pipes) when any of those things is activated. Split that into 2 programs and have the second program disable the first "if" program for something like 15 minutes. You only need the circ pump to turn on at first. Once people are using the water, you don't need the circ pump. And, once the water gets circulated, it stays warm in the pipes for a while even if no one is using it. So IF status guest bath light is on or master bath motion is turned on or status kitchen light is on Then call program 2 then clause Program 2 If - - Then disable program 1 turn pump on wait 2 minutes turn pump off wait 15 minutes enable program 1 run if clause program 1 This setup should keep the pipes warm if the lights are on in the kitchen, guest bath, or if motion is detected in the master.

The Java gui needs to be fixed in my opinion. They should put a line through it or something until you hit save. I made the same mistake myself at first.

Is it possible you have a program on your ISY run amok? A program that sends a beep command to the switch caught in a loop could do this.

So in other words, things are going to be exactly the same except Apple is going to extort some money from manufacturers to say "Apple Approved"

Here is the key quote: "Some may not even be manufactured by Apple. The company is hashing out a deal with other device makers whose products will be certified to work with Apple’s system and sold in its retail stores, the Financial Times said." Basically this is what we have already. It is just that Apple is going to do some tweaking of some already existing products and re-brand them. There will I'm sure be advantages, but as is always the case with Apple, you play their way or no way.

Then effectively it is && since I put an actual "&" and the "amp;" Maybe I could delete one or the other and have it still work. I did not check the "encoded url" button and it works so maybe it is seeing the actual "&" and ignoring the "amp;". But, it works, so all is well . I think the key was putting the credentials into the path box instead of using the authentication header, along with extending the timeout. Thanks to all.

OK, thanks. It works! I also had to increase to 1500 ms the time out. . . at least to not have ISY show an error. Where did you get this syntax "&user=" etc? I have not seen the "amp;" in any of the foscam docs.

I had already tried both with and without the colon. I tried increasing time out to 1500ms but no go. Purdueguy PM'd me that he thinks the new Foscam firmware is requiring encrypted passwords. I don't know that this is possible with ISY. The ISY and the Foscam are on the same LAN, so I'm not terribly worried about encryption.

I don't manage the presets via ISY, I do those via the standard camera webpage. I then use HTTP calls from ISY to set the camera to certain presets. For example, to set my camera to Preset2, I use the command: http://CAMERA_IP:PORT/decoder_control.cgi?command=33 To enable the automatic IR: http://CAMERA_IP:PORT/decoder_control.cgi?command=95 When doing this from ISY Network Resource: Protocol Information: - HTTP - GET - Host = IP address of Camera - Port = Port of camera (default is 80) - Path = /decoder_control.cgi?command=XX (replace XX with your command number) - Details -> Headers -> Click Add, choose "Authorization" and fill in the name/password for an "Admin" or "Operator" user (a "Visitor" user shouldn't be able to move the camera, etc) More info here: http://www.zoneminder.com/wiki/index.php/Foscam Trying to have ISY turn my camera alarm on/off. I am getting error "tcp client request failed net module rule 134:401", which I believe is th user/password failing. I included the authorization header with the correct user name (admin) and password. ISY is translating that to base 64 which I reverse translated and it shows admin: mypassword I tested my syntax in IE and it works (turns the alarm on/off) Any ideas on what I am doing wrong? Thanks

Very sorry to hear of your loss. The first thing to try would be to pull the power disconnect tab to reset it. This is a small piece of clear plastic at the bottom middle of the keypad immediately below the lowest button(s). Use a butter knife or small screw driver to pop it out. The tend to be a little tough to pull out, but it should come out 1/8 inch or so and stay out when you let go. Let it sit out for 10 seconds or so then pop it back in by just pushing on it. Don't push and hold it, just push it in and as soon as it snaps in let go. If you hold it for too long, it will do a factory reset. Hopefully that is all it takes. If not, we can take you further down the road. Again, sorry to hear about your husband. We will miss him around here as well.

A cai webcontrol board can accept the one wire temp input and post directly to ISY variables.

I'm thinking bad plm. If your regular, non-isy, scenes are all working, then a noise issue is low probability. I doubt it is the ISY.

And then restore PLM. I agree it should work, but it is expensive. And you would have to keep both ISY/PLM's up and running waiting for the possibility of being called into action. Since it seems that PLM's have a ticking time bomb in them that explodes just after 2 years, you might find that the backup PLM is also no good when it comes time to switch to it. That is why I would suggest not having the backup PLM powered up until it is needed.

As Lee said that would be very difficult. If you are concerned about a PLM failure, you could run 2 plm's and use a relay to switch the 120v power between the 2. You would also need a splitter for the rj45 so that both PLM's would be connected at the same time to the ISY, but since only one is powered on, the other would be invisible to ISY. You would need a controller for that relay that could be operated via a network connection. If the PLM failed, you could switch to the second PLM, reboot your ISY remotely and then do a restore PLM remotely.

Glad to hear you have it going in the right direction. I would be curious to know what device ends up being the problem.

So you are saying that querying the links table from a device is yielding only 1 link, despite the fact that the device currently functions properly as a member of one or more scenes? That I don't quite understand. It seems that the links must be there or it wouldn't work in a scene. But how a query from ISY is only showing one link I don't know. It seems to me that it should either show all of the links or fail to show any. I assume you have installed the new PLM in the same spot as the old one? This would make me think that a comm issue is not the problem. However, you might try putting the PLM onto a power splitter (just a plain splitter, no other surge suppressors as they can create comm issues) behind an Insteon noise filter and check the communication and ability to write links between the plm and that device. If that all works, it would seem that some sort of comm issue is the problem. You might also just have a bad PLM. ELA mentioned in a thread that he had an RF device that was jamming his system. It doesn't seem like that is what you have, but it would be worth reading.

It sounds like you did things correctly And my modem also died just after the warranty. I have read this same thing many many times. Based on what you are saying, it sounds like your devices do not have the correct address for the PLM written into them and the plm doesn't have the right device addresses written in it. When you did restore modem, it should have written the link between each device and the PLM. So each device should have the PLM address and the PLM should have each device. The process of restoring the PLM can take a long time depending on how many devices/scenes you have because it is not only writing the links in the PLM, it is also erasing and rewriting the links in the device that reference the PLM. If it didn't take a while, then that would indicate a problem. Mine took I think like 2 hours. If you can manage to bring up the links tables on your individual devices, you should see your plm address in there. Also, you can bring up your PLM links table and you should see the addresses of all your devices in there plus a whole bunch more relating to all the scenes. For example, I have 366 links in my PLM and about 60 devices. You can check your "tools" "diagnostics" "plm info/status" and see if it lists the PLM as connected and also what the address is.

Yeah I just realized myself that might be the problem. So if I understand correctly, programs that are disabled won't re-evaluation the "IF" clause while they are being executed. So I'll need to change the dimmer program to be event based... Programs that are disabled never evaluate the if clause unless directed to do so by another program (or manually).

I think I see the issue. Your last program is disabled. It won't trigger from the status of those 2 programs changing condition. That last program must be called by another program.