Looking for help with lighting control problem

[ixlr8]

Hi,

I have a set of lights in my shop that I control with an "InLine Link Relay", shop is about 130 ft from my house. I have an 8 button dimmer keypad in my house and a 6 button dimmer keypad in the shop. I also have a sensor on my shop overhead door to sense if it is open or closed. The lights are programmed to come on when overhead door opens and go off when the overhead door closes. This works fine. In the house on the 8 button keypad, I have a button on the keypad that is programmed to turn shop lights on when button is pressed on, and turn shop lights off when button is pressed off. This also works with no issues. In the shop I have the 6 button keypad set up the same as the 8 button keypad and programmed the same as the 8 button key pad. If I press the button on the 6 button keypad to turn on the light in the shop.. the light comes on. BUT, when I press the button to turn off the light, the light does not turn off, and I can't figure out why. If I open and close the overhead door.. the light will turn off. I tried programming different buttons in the 6 button keypad, same result. It will turn the light on, but not off.

Is there any way to see/log the traffic so I can tell what is happening? I found something that looks like it might log the traffic, but it seems to require Excel to look at results and I don't have Excel on my computer.

 

TIA,

Jim

[Illusion]

Florescent light in shop?

 

You can view the log without Excel, just say no to open with excel and you should be able to save log as a .txt file.

 

If I had to make a guess based on the factors you listed, the circuit the 6 button KPL is on is far away and on a different phase than the lights. When you turn the lights on the attached load is causing enough noise or absorption to interfere with the module hearing the off command from the 6 button KPL.

 

Try connecting an incandescent test lamp instead of the current load to the module and then see if you can turn it on and off from the 6 button KPL

[ixlr8]

The shop lights are florescent. The lights, the 6 button keypad, the overhead door are all on the same phase in the shop... there is presently only one phase, a 110v feed via long extension cord, in the shop. The 8 button keypad is in the house, no idea which phase it is on, and it works fine. I am heading out of town until Thursday so I won't have a chance to test incandescent light until then. I still wonder why the lights turn on/off properly via the garage door sensor program and not the 6 button keypad.

Thanks for the info on the .txt vs .xls viewing of log files.

[edit] After reading some other posts, I may have found the answer to my question on why overhead door sensor controls the lights properly. If I remember correctly I have the lights included in a 'scene' with the garage door sensor and the 6 button keypad is not. The 6 button keypad signal has to go 140ft to the ISY and then return, the garage door sensor signals stay in the shop. Although that doesn't explain why the 8 button keypad in the house works fine. I won't be able to verify this until I return on Thursday. Hopefully the signal logs will also tell me something. [/edit]

Thank you,

[oberkc]

Do you have a means by which imsteon can communicate between legs (or phases) of your eletrial system? This would generally be two dual-band devies, properly verified on each leg.

[ixlr8]

Do you have a means by which imsteon can communicate between legs (or phases) of your eletrial system? This would generally be two dual-band devies, properly verified on each leg.

Yes, I have two dual-band devices, verified one on each leg. I only have one leg going to the garage, but I guess I should check to see if the ISY is on that same leg, just to be sure. I am just surprised that the 6 button keypad can turn on the lights but it can't turn them off.

[Illusion]

[edit] After reading some other posts, I may have found the answer to my question on why overhead door sensor controls the lights properly. If I remember correctly I have the lights included in a 'scene' with the garage door sensor and the 6 button keypad is not. The 6 button keypad signal has to go 140ft to the ISY and then return, the garage door sensor signals stay in the shop. Although that doesn't explain why the 8 button keypad in the house works fine. I won't be able to verify this until I return on Thursday. Hopefully the signal logs will also tell me something. [/edit]

Thank you,

 

You are going to have to explain this. I had assumed that the button on the 6 button KPL related to the lights was a controller for said scene. If that is not the case, and the ISY is involved in this event directly (not just an observer) please explain why the signal has to go to the ISY and then return.

 

Further, I do not think the log is what you want. It shows status changes, but not traffic. What you want is the event viewer. See the wiki article for details. The wiki is easy to get to as there is a link in the 'Help' section of the Admin Console that links there.

 

Finally, I think while you are at the wiki, you should also investigate 'Scene Test'.

[ixlr8]

 

You are going to have to explain this. I had assumed that the button on the 6 button KPL related to the lights was a controller for said scene. If that is not the case, and the ISY is involved in this event directly (not just an observer) please explain why the signal has to go to the ISY and then return.

The 6 button KPL is not in a scene so the ISY is involved with this event directly. Only the garage door open sensor is in a scene. I tried to put both the KPL and the sensor in a scene and the Admin Console wouldn't let me. I have read the Wiki until I can't think straight trying to sort out how a 'scene' works and what the requirements are.... still have no clue. The 8 button KPL in the house is set up the exact same way as the 6 button KPL in the shop, and the 8 button works as programmed.

Further, I do not think the log is what you want. It shows status changes, but not traffic. What you want is the event viewer. See the wiki article for details. The wiki is easy to get to as there is a link in the 'Help' section of the Admin Console that links there.

 

Finally, I think while you are at the wiki, you should also investigate 'Scene Test'.

I will spend some more time with Wiki reviewing these, thanks for the suggestions

[Illusion]

Sounds like maybe the ISY cannot hear the 6 button KPL when the lights are on.

 

Open Admin Console>Event Viewer

 

Turn lights on, try to turn lights off with 6 button KPL. See if the ISY can hear the KPL by looking for entry in the 'Event Viewer'

 

I had a lamp in the living room that I could always turn on, but not off. It was a flo that was going bad. Switched to LEDs and have not had a failure since.

[ixlr8]

I am home again and have tried a few things, I found that the 8 button KPL in the house also no longer works properly. I tried using Event Viewer, but all the code stuff makes no sense to me. But I did get a popup window saying that ISY couldn't communicate with the IOLink relay when I tried to turn the light off. After more playing, it does seem that with the lights on, the signal does not make it reliably from the house to the shop. I spent some more time reading about 'scenes', I still don't understand them... but I was able to create one for the shop lights with the shop 6 KPL and now lights come on quicker, and now they shut off. But then they come on again and sometimes shut off again. I suspect that the 'scene' is turning them on/off, but the program I had in place is also turning them on/off and that is why they are turning off/on/off when I command them to turn off. If I wait 15 secs or more before I turn the light off, it stays off!!??Which leads me to the next question, how to I remove the existing program from the keypad. I tried 'deleting' it from the 'program' tab in the console... that worked, but how do I delete/remove it from the keypad?? I searched the Wiki and I couldn't find anything on deleting/removing programs... only creating them. Or did I just delete the program from the console and it needs to be deleted from the PLM.. I wish I understood this system better.. but thanks to your help I am getting there. I need to get this working with the florescent bulbs.. I am putting in 12- 8' 4 bulb fixtures to light the place and I am just dealing with the first 3 at this point.

TIA

[LeeG]

ISY Programs reside in the ISY only. They are not in the PLM and not in any device.

[ergodic]

Just to amplify on what Illusion said: you are describing a classic case of Insteon powerline interference.

 

When the fluorescent lights are on, the ballasts (I'm betting electronic ones, yes?), will generate a truly astonishing level of HF noise. That noise feeds back into the powerline. That is why things work when the lights are off, but unreliable when they're on. I wouldn't spend a lot of time trying to analyze this past that - there are too many variables and "X" factors.

 

In my experience even dual-band devices are no guarantee of immunity from this problem when the line noise level is high enough.

 

What I finally had to do in my garage to fix this same problem was wire an outlet J-box in the ceiling where the wiring to the fluorescents begins. I then plugged in a Filterlinc to that and wired up a little pigtail 120V plug to the fluorescent wiring chain. That's plugged into the Filterlinc and that setup took care of it. The Filterlinc keeps the ballast noise off the powerline -- at least most of it.

 

I had to do a similar thing in my kitchen remodel which uses undercabinet fluorescents with an electronic driver, though the plug was already present for that so it was simple.

 

As a plus, taking all that noise junk off your powerline might well improve your overall Insteon reliability, even though it's 100+ feet away from the house and the noise should be reduced a lot over that distance.

 

I'll also mention in first attempting to solve this, I did try a couple of different X10 inline filters. None of them were successful.

 

Also, not sure exactly what topology you were describing, but my experience is that the dual-band devices do not provide the line signal level that the true Insteon access points do. And 140 ft is a long way to run the signal anyway. So if you don't have true access points on each leg, I'd consider installing them.

[ixlr8]

I created a scene and properly deleted the program I had written from the ISY and now it seems to be working properly, and responding quicker. The lights are medium quality and do have electronic ballasts. I am learning that the cheaper the fixture, the noisier the ballasts tend to be. I will also try a Filterlink and see if that helps. If I am installing 12 of these lights... I am going to have a ton of noise to deal with. Thank you to all for their help and suggestions, online and off.

 

Jim

[ixlr8]

I rewired my lights from hardwired to plug in. The Filterlinks I ordered finally arrived, they fixed the issue. Now I can control the shop lights from in the house as well as in the shop. Just disappointed that it is going to cost me an extra $40 per lighting string, $160 minimum, to do the full shop lighting for the Filterlinks, plugs, socket and electrical box to make this work. Thank you all for your help and suggestions.

[Illusion]

Or you could put that $160 toward LED bulbs instead of Flos. Great if it gets cold in the shop, and much more energy efficient. No filters, plugs, etc required...

 

http://store.earthled.com/collections/l ... I1O5r_Npyc

[ergodic]

ixlr8, good to hear it worked out. Also, it may not help much, but take a look at a Crestron or Control4 to see what a home automation control system can really cost. It's no fun to spend $140. But it beats having to hire somebody to even do it.

 

I can second the LED suggestion. The dimmable LED bulbs are getting very close.

 

I've just installed the FEIT floods (sold at Costco) in all my outside soffits. The warm white looks very nice outside, and they use less than 1/5 the power of the incandescent floods I had in there. Given that they're on about 50% all evening, they'll pay for themselves in a few months at my marginal energy price here (about 32 cents/kwh). They also dim down almost to off, and no detectable Insteon interference so far though I'm still checking that.

 

I think these would work well in a shop environment, and I'm thinking about replacing the fluorescents in my garage with these too. There's also a FEIT narrow-angle spot (the light from that is noticeably bluer though). And of course LEDs don't give off that life-sucking illumination of fluorescents or CFLs. But at $20+/bulb, this route isn't exactly cheap either, though of course you benefit from the lower energy cost if you use them a lot.

 

I still haven't found a decent LED replacement for my inside 12V halogens MR16s, though I keep looking. Most of the ones I've tried have stupid little fans inside for one thing. They also don't dim well at the bottom end. And the color temp doesn't change with dimming which doesn't come across all that nice for inside light. Plus the LV halogens are pretty efficient on their own so the cost benefit isn't as big.

 

But I'll also admit I'm fussy and I expect to see better in a couple of years. The stuff is still very much a work in progress. I'd be very surprised if anyone is even installing fluorescent fixtures five years from now.

 

I was talking to a contractor working in the local Target parking lot a few weeks ago. They were replacing all the arc floods in the lot with LEDs. (And man, are these things BRIGHT!) What he told me was that the bulbs cost about $600 apiece, but they pay for themselves fast because the cost of bringing in a crew and cherry picker to replace bulbs is so expensive. He said they're booked up for months, including hotels with big atriums etc. That demand should ramp up the technology and bend the cost curve.

[Illusion]

ergodic,

 

I know this is off topic, but I thought I would mention it here. Philips has developed a LED product line called Dimtone that warms up as it dims to match incandescent. I do not know if they have an MR16 yet, but might be worth investigation.

[apostolakisl]

I just bought a new building for my office, gutted it, and used the CREE ceiling troffers instead of fluorescent in the new build-out. http://crseries.creeledlighting.com/

 

I primarily used the CR22 model. They are very good. The light they make is in no way a compromise. It has great color rendering, it is bright, and it is soft. They cost about double a conventional ceiling troffer but use about 1/2 the electricity and last 50,000 hours to 80% original lumen (of course I'm going on their word with that). So they should be maintenance free for the life of my office. My electric bills are half what the previous occupants paid (I also replaced the HVAC units and put in spray foam insulation). I did get back about 20% of what I spent on them in rebates from my local electric provider so the ended up costing about 1.5x what I would have paid for normal ones. Of course normal ones also cost you about $40/ea every 5 years or so for re-lamping in addition to double the electric use (and double the heat production which is important where I live).

[ixlr8]

Or you could put that $160 toward LED bulbs instead of Flos. Great if it gets cold in the shop, and much more energy efficient. No filters, plugs, etc required...

 

http://store.earthled.com/collections/l ... I1O5r_Npyc

Interesting, I didn't know about these, but to do my whole shop with them would cost about $3100 for the bulbs!!!

Filterlinks suddenly look cheap.

[IndyMike]

Hello ixlr8,

 

I'm jealous - that must be a sizable shop to fit 12 - 8' fixtures. I have 9 - 4' (4 tube) fixtures in mine. When I fire them up in the morning, it will make you squint for awhile.

 

You are fighting both signal attenuation over the 130' distance from the house and noise at both ends.

 

1) I saw where you are currently running off an extension cord. I have no numbers for signal attenuation over an extension cord (stranded / xx? gauge). It sounds like you have a sizable load in the shop. You may want to put in the final service drop before you spend money trying to fix communication problems. Things may get better or worse - but they will likely change.

2) You are currently running some distance of wiring within the house (distance between the nearest Insteon device and the extension cord connection). Try minimizing this by placing an Insteon device near the connection point (minimize losses). When you put in your "real" service drop, it will be at a panel. Make sure you have an insteon device close by (same phase) to relay the signal to the shop.

3) I'm not clear on your fixtures. Are these true 8 foot T12 bulbs ( 4 tubes per fixture X 75 watts = 300 watt / fixture)? ... or are they 4 foot T12 bulbs ( 4 tubes per fixture x 40 watts = 160 watts / fixture)? I am running 3 of my fixtures on a switchlinc relay (4 tubes X 40 watts X 3 fixtures = 480 W total) as an endurance test. This is a pretty high fluorescent load and has a high inrush current. I haven't had problems so far - but it's only been in place 3 months.

4) If you are running 160 watt fixtures, I would rather see you put your filter money toward better ballasts. If you are running 300 watt fixtures, this might be rather expensive. Fix your service drop first before making any improvements here.

 

For shop lighting, it's hard to beet the fluorescent tube fixtures. Each 40 watt T12 bulb puts out ~ 3000 lumens. I buy the bulbs in 10 packs at roughly $2 / bulb. I've had my fixtures in place for 11 years now. I'd estimate that roughly half of the tubes are original (half replaced).

[ixlr8]

The fixtures are 8' long with 4- 4' T8 bulbs each, they are wired in strings of 3 fixtures. So far my electrical loads are one small air compressor, one garage door opener and never more than 2 of the 4 stings of lights on at a time. 100 amp panel installed in shop, I need to replace the panel in the house before I can properly wire/feed the shop. I know that I am at the limit of my extension cord power supply to the shop.

[ELA]

hello ixlr8,

IndyMike made some excellent points that I wanted to add to.

 

His suggestion to wait until you have the service wiring installed is important. As he indicated things can get better or worse but most likely they will change. My experience with extension cord is that the wires are typically closer together and therefore exhibit a higher capacitance level ( more signal attenuation).

 

It appears that many Fluorescent ballasts cause issues with Insteon. What is often not fully understood is whether the issue is input capacitance ( signal suckers) or noise. Either way a filterlinc can often be a solution.

 

I have a few fluorescent fixtures that do not exhibit significant signal sucker nor noise issues.

Many newer fluorescent ballasts will have power factor correction and/or EMI filters.

With the right ballasts filterlincs may not be required.

 

If you have already purchased the fixtures you may want to stick with filterlincs but if not then I felt IMs suggestions were very good ones. Can you list the part number of the existing ballasts?

[ixlr8]

hello ixlr8,

 

If you have already purchased the fixtures you may want to stick with filterlincs but if not then I felt IMs suggestions were very good ones. Can you list the part number of the existing ballasts?

I already have the fixtures, I can post the ballast part numbers when I get home again in 2 weeks.

[ergodic]

ELA:

 

I can tell you in my case the fluorescent issue was definitely noise. I tried two different electronic ballasts and scoped both. One that came with the fixture and another I bought at the local wholesale electric.

 

The backfeed powerline noise from both was absolutely incredible. (I actually started laughing when I first saw it.) If you could post your ballast mfr/model/part # -- the ones that work with Insteon - it would be useful info.

 

With Insteon it is usually easy to check if there is a noise problem. Just turn the the circuit off. If the comm. problem goes away it is almost certain to be noise. With the two ballasts I tried, everything worked fine as long as they were off.

 

My guess is that a magnetic ballast would not cause comm problems - it never did in my previous house, but those are almost impossible now.

[ELA]

Hello Ergodic,

To be sure there are noise sources out there and you clearly defined yours as one using an oscilloscope. I was not doubting you.

 

My comments were more in general to the point that either a signal sucker or noise can cause issues and most people do not have the equipment required to identify which it is.

 

In particular, with regard to electronic ballasts they most certainly do produce noise. Manufacturers make efforts to suppress that noise to varying degrees.

 

Most electronic ballasts will have a rectifier input into a relatively large capacitor. This produces both a power factor issue and a possible signal sucker issue for Insteon. Adding power factor correction can help with Insteon by somewhat isolating that capacitive input with inductance.

 

Some manufacturers will include power factor correction with a side benefit of suppressing some noise.

Others specifically include an EMI filter to further reduce that noise.

 

A possible complication comes in how they suppress the noise. If a large across the line cap is used in an EMI filter on the line side you may not have a noise issue but instead a signal sucker issue.

 

I have to disagree with your comment about it being easy to decide if it is noise or not by turning a circuit off. Turning a circuit off or unplugging also serves to isolate a possible signal sucker. Turning an entire circuit off also may greatly change the overall impedance of the network.

 

Of course using an Oscope as you did will help clarify the issue.

 

As an example I can take two surge strips with 0.1uf capacitors in them and plug them in certain outlets. Turn them on and Insteon communications takes a dump. With them off all is fine. Definitely no noise involved.

 

 

I will open my fixture when I can to post the part number. Maybe you could post your ballast information as one to avoid?

 

12_11_12 update: found time to open and test ballast:

Posted results here:

http://forum.universal-devices.com/viewtopic.php?f=28&t=5923&start=180 (5th down)