LED Lighting, solution

[marcuslee]

Hi All,

For those who have looked down the GU10 LED lighting route, I would hazard a guess that at some point other than Megaman lamps etc, one would have come across Philips MasterLEDs.

Generally speaking my research pointed that this was probably the most mature, and the only really proven good working LED solution (unfortunate as the lamps are more commercial in nature, and therefore a bit expensive).

Going with this however, Idratek have been testing post call out to them about taking on compatibility with these bulbs (currently not all dimmers either HA or otherwise, are necessarily compatible).

I'm pleased to say though, since late last year when I first called it out, the Idratek team have looked into it and have developed a compatible firmware for the QLD dimmers.

Having now deployed on a larger scale (60-80) bulbs, initial results look to confirm that Idratek QLD dimmers can now be considered another compatible dimmer with these bulbs.

FYI for those who are considering the same.

The only tips are:
- currently the QLDs have been derated by the same amount as Danlers dimmers ie hte 7W bulbs equate to a 50W in the calcs (ie 4x bulbs per QLD channel)
- as a backup solution, you could consider using Danlers RESLOAD product on non QLD firmware upgraded dimmers, which also work (indeed Danlers RESLOADs being a product which I think might be able to bring compatibility with any non compatible dimmer)

Cheers,

Marcus

[Paul_B]

How about MR16 rather than GU10? I've been following a thread on Greenbuildingforum and they back up my own experience. A GU10 needs a switching circuit to reduce from 230v to 12v which is closer to the voltage requirement of the LED. The general experience is the small circuitry required to control voltage fails a long time before the actual LED. A MR16 offloads the voltage transformation to an external PSU which is better at doing the job and in the worst case can be swapped for a new unit.

Paul

[ludditeal]

Electricity consumption obvioulsy on everyones thought as I was recently asking Karam about LED GU10's and he mentioned that they had done some work on Philips bulbs.

Having had a couple of rooms with Idratek I have just ordered all the kit to finish the rest of the house so will be putting off the LED conversion until everything is in and settled but useful to know there are options out there.

Cheers
Allan

[marcuslee]

How about MR16 rather than GU10? I've been following a thread on Greenbuildingforum and they back up my own experience. A GU10 needs a switching circuit to reduce from 230v to 12v which is closer to the voltage requirement of the LED. The general experience is the small circuitry required to control voltage fails a long time before the actual LED. A MR16 offloads the voltage transformation to an external PSU which is better at doing the job and in the worst case can be swapped for a new unit.

Paul

Interesting point.

No particular comment on that, only thing I would have to add on that, was because of the very long length of the MasterLED bulbs, and accommodating it in tiltable light fixtures, one possible solution we broached was to indeed use a MR16 version of the bulb in the range (which as still dimable). However the only thing being was the lumens output isn't as high.

Their comment was no MR16s are (yet), but they'd expect in the coming year for the tech to get there.

[toscal]

One of our clients is currently testing our 6W GU10 LED dimmable bulbs with the Idratek system. These are almost the same size as a standard halogen GU10 bulb, and in 90% of fittings fit perfectly.
Will let you know how he gets on with the testing etc.

[Karam]

In a sense MR16s would be easier to handle but need a different driver technology (PWMing low voltage DC). The GU10s on the other hand contain within them, as Paul points out, the necessary step down and control electronics. In fact it is quite complex technology because in order to work with 'standard' mains dimmer technology the electronics has to interpret the chopped up mains signal which the dimmer creates into effectively a PWM'd low voltage drive signal at the LED end. Since the step down is a bit like a switch mode power supply it tends to have a highly capacitive front end which is not great for triac type dimmers and there is additional circuitry to try and present the right sort of load characteristics at various points in time to the dimmer - tuned to suit the mass market dimmer. At least that's how I understand it . The high derating factor is as a result of what must present as a capacitive characteristic, i.e when the dimmer triacs fire ON they see a much lower resistance than might be expected for a resistive load of the same wattage. To be honest we are for the moment being cautious and just following recommendations from others and based on this theoretical understanding. At some point we will perform more detailed tests to see if we can raise the bar with our particular set of components.

It may well be that other dimmable GU10's will work fine with the existing QLD firmware but if anyone finds they flicker then they can contact us to get and try the alternative. Firmware can be downloaded to the module in situ (over the network) so its not a difficult task.

[cliffwright]

One of our clients is currently testing our 6W GU10 LED dimmable bulbs with the Idratek system. These are almost the same size as a standard halogen GU10 bulb, and in 90% of fittings fit perfectly.
Will let you know how he gets on with the testing etc.

Do you have any update on how the client has found the use of these dimmable GU10's?

[toscal]

They dim to about 20% then switch off. But its not just Idratek dimmers that do this. I have been testing some myself with an X10 dimmer and a normal rotary dimmer and would say its the same with both of these.
It may be possible to adjust the firmware in the Idratek module to take this into account, and produce a better dimming curve. I guess Karam is the person to ask on this one.

[Karam]

The Philips bulbs seemed to dim pretty much most of the way but had other symptoms which we had to create firmware work arounds for aside from also tweaking the mapping curves. If someone wants to send us a sample of the other ones happy to investigate. But yes, arbitrary mapping curves can be uploaded to the modules. We don't have a convenient utility to generate the mapping data so its still us who have to create the necessary file.

[cliffwright]

The Philips bulbs seemed to dim pretty much most of the way but had other symptoms which we had to create firmware work arounds for aside from also tweaking the mapping curves. If someone wants to send us a sample of the other ones happy to investigate. But yes, arbitrary mapping curves can be uploaded to the modules. We don't have a convenient utility to generate the mapping data so its still us who have to create the necessary file.

Reason I asked for the update is I've just re-done our kitchen and in doing so, installed spotlights in both the Kitchen and Utility - there's 12 spots in total @ 50w each so I've got 600w at any point burning away - in the meantime I've bought 2 pairs of LED GU10's from simplyled (these; http://www.simplyled.co.uk/GU10-LED-...1CDY.aspx?nh=1) to see what the 'warm' and 'bright' whites actually look like.

In trialling these, they seem to dim to what I would guestimate to be about 40-50% brightness, which is probably sufficient.

Once I've decided whether I'm going for a full set of 'bright' or 'warm' whites, I'd be happy to lend you one of the others as a source for alternative mapping curves?

Cheers

Cliff