PLM with 277v lighting?

[Jay M]

I want to control the lighting in a new conference room at work.  The lighting is all 277v.  We will be using LED fixtures that can be controlled by the Insteon 2475DA2 0-10v ballast dimmer.  Does the ISY support the 2475DA2?

 

The Insteon paddle switches can run at 277 volts, but the PLM is only 120 volts.

 

What is the most reliable way to have the ISY talk to the 277 volt lighting system?  I assume it will only be through RF? Is that reliable enough?

 

thanks,

~Jay

[Brian H]

You are still going to need a PLM. As it will be needed to send the Insteon  RF commands along with Insteon power line commands.

Maybe a small 277 volt to 120 volt step down transformer can power it. Even if the transformer doesn't pass the power line signals correctly.

[Jay M]

Thanks Brian,

 

We have 120v outlets in the room, but the lighting is 277v.  

 

The question is: will the RF only connection between the PLM and Lights be reliable?

 

The other question is if the 2475DA2 is supports in the ISY.

 

thanks,

~Jay

[mwester]

Nobody can tell you if the RF connection will be adequate.  That's a function of too many factors unique in your environment.

 

IF your building has no other sources of RF, particularly around the 900 MHz band, and the fixtures into which you'll mount the Insteon dimmer units is not metal, and you have line-of-sight between each Insteon dimmer and the PLM, then it's reasonable to expect that it will work.

 

However, all it takes is for some industrial machine operating in the 900MHz unlicensed range to power up, or for someone to take some other 900MHz device (baby monitor?  Walkie-talkie?  Who knows what?) and put it in exactly the wrong place (and nobody can tell you where that might be either), and suddenly you'll loose the connection.  It's largely trial and error, so you'll just have to try it, find the issues, move things or add a few range extenders, etc, until ti works.

 

I'm also a little concerned about the LED lighting power supplies -- they're undoubtedly switching supplies, which generate so much noise that they tend to render the Insteon power-line communications useless.  That means that your mesh may have to rely entirely on RF, rather than the ideal case where the PLM need only have comms to one of the dimmers, and it would relay over the powerlines to the others.

 

Or even worse (and again, you have to try it, nobody can tell if this will happen), if the noise at the zero-crossing point of the AC power is bad enough, Insteon devices cannot process the RF signal either -- so basically, if you have bad power, the RF may also become useless.

 

Buy a couple units, make sure you can return them, and test the comms before you go any further with this project.

[Jay M]

Thanks, that's very helpful.  The whole building is 16 years old and so are all the florescent ballasts that are on the 277v lighting circuits.  We are not a hospital, but we do have our share of wireless technology in the building.

 

The one thing we do have going for us is that it's all in one room that is not very big, and I'm pretty sure the new LED lighting will be on it's own circuit.  I suppose we could just run a separate 120v circuit just for the Insteon controllers.  In fact, the controllers could be located in one place together and then just run the low voltage to each fixture.

 

~Jay

[stusviews]

It seems that you have a 480/277VAC wye 3-phase supply. Although there are transformers that can drop the voltage to 12VAC, the transformer itself will probably block the Insteon power line signal. Depending exclusively on RF is iffy, at best. The only way to determine if it will work is to try it.

[Teken]

Jay M,

 

I would appreciate it if and when you decide to move forward with this endeavor to reply back with your experience with the 2475DA2 0-10v ballast dimmer. I ask because there haven't been very many people or threads about its use and experiences.

[stusviews]

Jay M,

 

I would appreciate it if and when you decide to move forward with this endeavor to reply back with your experience with the 2475DA2 0-10v ballast dimmer. I ask because there haven't been very many people or threads about its use and experiences.

 

I agree. But I'm not sure that you electric supply is a good test.

 

Paper weight lifter. With effort.

[Brian H]

I did not see the 2475DA2 (Insteon Category 01, Sub Category 25) 10V Ballast Dimmer in the drop down list of modules in (RC2) 4.5.3 Firmware. I don't use the 5.0.? Alpha firmware so it may or may not be in it.

 

From the FCC Database

The module has the small RF Daughter Board in it. With the pig tail antenna.

If the module is mounted inside a metal fixture. I doubt Insteon RF would reach it reliably.

[Jay M]

This is all very helpful!

 

Today we tested one of them.  It's not a complete test, but I'll tell you what we did so far.

 

We don't have a ISY here yet.

 

We attached the low voltage wires to the fixture, and left 2475DA2 dangling from the ceiling.

We attached relay 1 wire from the 2475DA2 to the LED fixture.

We installed an Insteon dimmer in the wall to the 277v box that switched the old lights. We capped off the load line.

 

I pressed the set button on the switch, and then the 2475DA2 to link them.

 

It worked quite well.  The fixture got very dim, and then at the bottom the relay finally clicked off.

Double tapping for fast on clicked the relay on and the LEDs went to 100% instantly.

 

The dimmer feel was pretty good... Not as good a incandescent, but sadly, that's no longer an option.

 

What we still need to test is how well it will communicate via RF if the dimmer is in a metal box.

We also need to know if the ISY supports it. 

 

 

~Jay

[Teken]

Jay,

 

That is great feedback so far but can you let us know if the dimming was smooth in a gradual manner?

 

I know the rest is highly dependent on the LED being used / ballast. But if you could let us know make and model along with did you notice any flicker, hum, noise, etc.

 

 

=========================

 

The highest calling in life is to serve ones country faithfully - Teach others what can be. Do what is right and not what is popular.

[Jay M]

I'll see if I can find more specifics for you.  I'll try to post that later today.

 

I called universal devices and they told me that the 2475DA2 is definitely supported.

 

~Jay

[Teken]

I'll see if I can find more specifics for you.  I'll try to post that later today.

 

I called universal devices and they told me that the 2475DA2 is definitely supported.

 

~Jay

 

Awesome & thanks.

[Jay M]

I'll try to describe how well it dims.

 

It seems to dim a little faster than incandescent.  Also there is a slight delay before it begins to dim or brighten.

 

The delay is helpful in one instance.  If you want the fixture to dim as low as possible, it will go to its lowest point, and then pause for a second before the relay shuts off power.

 

For $100 more we could have purchased a lighting fixture that dims to 0.  With that fixture, I don't think we would need to wire in the relay, we could just use the 0-10v wires.

 

The fixture we used is Acuity Brands Lighting 2GTL2.

 

~Jay

[Teken]

I'll try to describe how well it dims.

 

It seems to dim a little faster than incandescent.  Also there is a slight delay before it begins to dim or brighten.

 

The delay is helpful in one instance.  If you want the fixture to dim as low as possible, it will go to its lowest point, and then pause for a second before the relay shuts off power.

 

For $100 more we could have purchased a lighting fixture that dims to 0.  With that fixture, I don't think we would need to wire in the relay, we could just use the 0-10v wires.

 

The fixture we used is Acuity Brands Lighting 2GTL2.

 

~Jay

 

Jay,

 

Specifically using that brand of LED was it smooth or did it have a stepped dimming?

 

As always thank you!

[jtara92101]

Specifically using that brand of LED was it smooth or did it have a stepped dimming?

 

 

 

I doubt the brand of LED (or driver - I presume these lights have a built-in driver) is going to make a difference in terms of "stepped dimming". If it takes a 0-10V analog input, I think it's unlikely it will have a ADC (if it even uses an ADC, brightness control may well be analog) with fewer steps than the 2475DA2.

 

The Insteon 2475DA2 specs say it has 32 steps. It goes from 0-10 volts in steps of 0.3125 volts. 

 

Looking at 2447D, regular Insteon SwitchLinc dimmer, the spec says it has 32 steps "locally", but "increments of 1% with software".

 

Seems a bit strange that 2475D would only have 32 steps, since it has no "local" means of control. (No physical switch.)

 

Has anybody ever noticed that SwitchLincs are more smoothly controlled "with software" (whatever that means!) than they are using local control from the switch paddle?

 

The tear-sheet for the lamp isn't very specific. Just says "0-10V dimming".

 

https://www.1000bulbs.com/pdf/lithonia-2gtl2-lp835-brochure.pdf

 

http://www.acuitybrandslighting.com/library/LL/documents/specsheets/LED%20TROFFERS.pdf

 

It seems to dim a little faster than incandescent.

 

 

 

 

Can you clarify just what you mean by that? I presume that, like any Insteon dimmer, you can program the ramp rate? No? maybe the default ramp is different?

[Teken]

What I have found in my environment is incandescent bulbs dim very smoothly. I've tested many type and brands of LED lights in the past and have noticed some brands reflect a stepped dimming. Whereas other LED bulbs mimicked almost exactly what a standard incandescent bulb does.

 

I was more curious to know if the *Visual* dimming transition from 0-100% was similar as I stated up above. The 32 steps you indicated are the positions the system can render. That has nothing to do with how smooth the bulbs will dim per say. 

[jtara92101]

What I have found in my environment is incandescent bulbs dim very smoothly. I've tested many type and brands of LED lights in the past and have noticed some brands reflect a stepped dimming.

 

 

Oh, oh, oh....

 

That hot tungsten filament will take a while to heat up/cool down, and will smooth the steps!

 

So, a really good LED bulb might try to smooth the transitions when brightness is changed, using essentially (either an analog or digital) filter.

 

And to do it right, they will have to determine the ramp rate and adapt.

[Jay M]

I would say that the LEDs I tested are as smooth as the best LEDs I've seen.  Because the don't warm up as they dim like the Phillips Warm-Glow bulbs, dimmer is a very special occasion.  What makes incadescent so beautiful full is how smoothly they go from black to 100% and back to black.  They way the filament  naturally dissolves is not only pleasing to the eye, it's also what we have been accustomed to seeing for most of our lives.

 

That said, these LEDs as well as the cheap Philips Warm-Glow bulbs are very smooth and don't appear to be stepped.

 

When I refereed to the dimming speed, I wasn't very clear at all.  What I meant to say was that an incandescent bulb directly attached to a switchlink dimmer will ramp up or down at a specific speed when you hold the paddle.  When that same Switchlink is a controller for the ballast dimmer that is dimming these particular LED fixtures, the speed at witch it dims or brightens is noticeably faster.  It's not so fast as to be difficult to get where you want without overshooting, it's just difficult when you are so used to a particular speed. 

 

I think that creating scenes with these lights, and knowing their limitations, it will be a very pleasant transition from one scene to the next.  The only exception, will be when I need to turn off, and then back on, like for a video presentation or something.  It's not going to be nice.  But if I can program the scene with the right ramp rate, then I think the result will still be much smoother than manually moving a more traditional dimmer.

 

We won't receive the rest of the lights until mid-October, but we are taking a leap of faith that Insteon + ISY on a 277v system will work.  If not, it's not a big deal to rewire for 120v.  Our fixture can handle both voltages.  So at the end of October, I'll try to remember to give an update.  I'm sure I'll be here frequently seeking programming advice!

 

~Jay

[jtara92101]

I have some Commercial Electric screw-in eyeball conversions for 4" and 6" cans. They have phosphor on the face of the central "bulb" part, and have a very significant afterglow when turned off, similar to the Philips A19 phosphor buibs. I've been assuming this is purely the response time of the phosphor. I'd think that all high-CRI buibs will have to use phosphor, and so would naturally filter steps.

 

It may be, though, that that afterglow is too weak to really act to filter steps.

 

I noticed your lamps are CRI 82, not particularly high, but also not awful. But high enough that I'd imagine they need to use phosphor.

 

Did you set the ramp rate of your scene that controls the lamp(s)? The scene will have default ramp rate unless you change it. The ramp rate you set on the controlling Switchlinc will affect the ramp rate of it's (I presume, non-existent) load.

 

To avoid cognitive dissonance, set the ramp rates the same for the SwitchLinc and for the scene, so that the LEDs on the SwitchLinc will follow the dimming of the scene.