Dimmable LED Landscape Lighting: Question

[ergodic]

I'm starting a back patio remodel. The lighting will be connected to the ISY. Part of it will include LED RGB strip accent lighting in various places.

 

To test this, I've obtained one of the Foster 60W dimmable 12V LED transformers, and connected it to 40W of LED strips - a little more than the max I would intend to put on any one transformer. It dims nicely and evenly with an Insteon SL dimmer, remarkably down to about 5-10%. So far, great.

 

Now, these transformers are officially spec'ed for standard triac dimmers only with symmetric forward phase (ie Lutron.) I rather doubt the Insteon dimmers have symmetric control circuitry.

 

But does anyone know? Has anyone actually checked the DC component on the Switchlinc dimmers? I've been using them for several years on my incandescent landscape lighting (running every night at 66%) with no issue other than a minor buzz from the transformer.

 

I'm thinking if I just 1A fuse the 120V side for safety that should be enough. I've already run the Foster for several hours at 50% dim without any meaningful temperature rise showing on my little IR gun.

 

Anyone else tried this?

[IndyMike]

Hello Ergodic,

 

The following shows a pretty healthy DC content from a LampLinc. Don't remember what the Dim level was. This was likely from my "old" V1.0 LampLinc that I typically use for test purposes. It's possible that later units or SWL's preform better (since the SWL's are sometimes used for fan control).

 

What type/version device are you using.

 

[ergodic]

I'll be getting new inlinlincs for this.

[IndyMike]

In that case I'll try the output of a newer SL dimmer. I've just put one on a long term power consumption test. These newer units appear completely different than my older models. This particular model isn't even registering on my Kill A Watt (previous units drew ~1W).

 

A quick look at the components shows some significant changes as well. The old models had some rather large electrolytic and ceramic caps. These are gone on the newer models. The inductor appears to have at least 2x the turns on the newer units.

 

I'm hoping the output performance is improved on these units. I'll try to get you some plots by the weekend.

[ergodic]

Thanks.

 

You motivated me to hook up a V2 (V28) Lamplinc and scope it.

 

The first image is on the output at 50% into a pure resistive load. Almost no DC bias with this unit.

 

 

The second image is the 120V side of the 60W Foster LED transformer, again at about 50% dim with the same Lamplinc. A lot of overshoot, harmonics, and general bouncing around as I'd expect into an inductive load. But still nothing at all worrisome - in fact no discernable shift at all between AC and DC coupling. All things considered a pretty clean waveform.

 

 

I'm pretty comfortable with using it for this at this point, especially as it will only be at about 20% load.

 

In fact, from what I see I'm not entirely sure why SH doesn't just label for MLV. But I guess that's a complex issue involving NEC 110, UL approvals, etc. etc. and all sorts of other stuff I know little of. Still, it would be interesting to know if there is any kind of phase balancing in the newer units. Looks possible.

[IndyMike]

Hello ergodic,

 

Thank you for the plots. I haven't tried a dimmer output into an inductive load.

 

As I was staring at my ongoing test setup this weekend, I came to two conclusions:

 

1) My plots above were generated with a 0.1u capacitive load.

2) I didn't have the proper load for your application

3) You've already performed the "acid test" when you hooked up your transformer and monitored the temperature.

 

Sorry, I should have posted back earlier.

[ergodic]

Thanks.

 

I should say I'm assuming it's an inductive load from the transformer block diagram I have - that waveform looks more like a capacitive load characteristic to me. But the unit is labeled as triac-dimmable (as long as the input is balanced), and so far that seems true.

 

The Foster claims to have some sort of internal protection to handle 15 days of a direct short. And aside from that, the thing's totally potted and about the size and weight of a 200W landscape transformer so I imagine there's lots of margin with a 60W rating.

 

Not a cheap unit either and so the cost of doing RGB with LEDs this way isn't exactly economical, but - stupidly - I just have to see if it's possible.

[IndyMike]

In fact, from what I see I'm not entirely sure why SH doesn't just label for MLV. But I guess that's a complex issue involving NEC 110, UL approvals, etc. etc. and all sorts of other stuff I know little of. Still, it would be interesting to know if there is any kind of phase balancing in the newer units. Looks possible.

 

I thought I remembered where the dimmers were rated for magnetic LV loads - I can't find any evidence of that now. I may be remembering the X10 line.

 

One comment on the overshoot near the zero crossing - this is substantial and I have not seen this with capacitive loads. I'd be interested in how this is referred to the line input of the switch. It would be occurring in the area (time frame) where Insteon messages are located.