• The forum software that supports hummy.tv has been upgraded to XenForo 2.3!

    Please bear with us as we continue to tweak things, and feel free to post any questions, issues or suggestions in the upgrade thread.

LED lighting

  • Thread starter Thread starter Deleted member 473
  • Start date Start date
Which part of reducing the size of a number do you not understand? This can be done by subtraction of a positive number, the addition of a negative number, division by a number greater than one or multiplication by a number smaller than one. It was you who first mentioned division in post#94 and repeated in #98 Now FFS stop nit picking
 
Absolutely not. I think if I were doing it, I would illuminate a piece of white paper with the compound light source and set the white balance to that. That's what I do when taking photos for my stuff on Ebay. Doing that generally produces a reasonably good colour rendition regardless of whether the light source is daylight or fluorescent. As the piece of paper in BH's shadow images is substantially grey, surely the colours of the lampshades and shadows are substantially reasonable.
But that's not the point I was trying to make.
 
I now have the fitting with three of the Aldi LED bulbs running (temporarily) on a dimmer switch rated 60W minimum. I can confirm that it works, although when turning on the lamps stay off until the control gets to about half brightness; they can then be dimmed from there down to about one tenth (estimated), any further and they fade out completely.

With a 60W GLS in one of the positions, at full brightness I can hardly tell the difference between the GLS and the LEDs. However, when the dimmer is turned down the GLS fades out much faster than the LEDs - with the GLS at about half the LEDs are still around 90%. The turn-on/turn-off characteristics of the LEDs is exactly the same as without the GLS "dummy load".

I don't know how much of this behaviour is due to the lamps themselves and how much is a consequence of the dimmer I happen to have to hand. I need to look into what dimmers are available that are specifically designed to accommodate LED lamps.
 
All the articles have read say the Graphene / LED Light Bulb will use 10% less power, 10% less than what? a GLS bulb, or 10% less than a non Graphene LED bulb. If it's only 10% less than a GLS bulb I would have thought it is a non starter
 
I think we can assume it's 10% less than a conventional LED lamp, but I don't understand where the saving comes from. The graphene provides a highly effective heat conduction path, which should result in the LED chip running cooler and having a longer service life or to have a higher output for the same service life, but other than that I don't see the point.
 
No, I don't get it either, longer life - yes, brighter - yes, but only when feeding it with more power, not the same amount. There's obviously something they're not telling us. They don't explain why it will be cheaper than a non graphene version either, cheaper heatsink?
 
Last edited:
They have spent millions trying to find a use for this 'earth shattering' discovery called Graphene.

What do they come up with?

A light bulb.

:):):):):);)
 
Had been looking at them myself, just hadn't had a chance to drop down to the store to get some. At that price they are worth a try.
At that wattage, they ought to at least match an old 60W, how do they look?
 
They pass all the tests I can do apart from the obvious one, (how long they last), They out perform new CFL bulbs in terms of light output, I can't measure the 806 Lumens, as I don't have a calibrated meter, but all the light meters I have show them to be considerably brighter than new CFLs of the same wattage. They are the colour I wanted (2700K = warm) and they don't make a noise (yet), which is one of the negatives I had read about them

IMG_5028s.png IMG_5029s.png IMG_5030s.png

Three photos at different exposures, all with a white balance set to daylight (5200K),
Left = 13Watt CFL, centre = 8.8Watt LED, right = 97Watt Tungsten
 
Last edited:
Electronics Weekly 14th October 2015 said:
Don’t write off incandescent bulb

LED bulbs are fast encroaching on incandescent light sources because of their far higher efficiency when producing visible light.

But Junichi Takahara, director of photonics at the Advanced Research Center (PARC) of Osaka University, points out that incandescent bulbs are actually very efficient electron-to-photon converters if you take both visible and infra-red emissions into account.

And he said they would be efficient visible light emitters if the surface of the filament could be nano-structured to suppress the emission of infra-red. This is an idea he credits to US researcher John Waymouth.

In 2012, Takahara and semiconductor process engineer Kazunari Kimino decided to spend two years developing such a filament using nano-imprint technology – which is suitable for mass production. They did it in 18 months.

Start-up Metalumina was founded in June 2014, with the aim of creating light sources with tuned spectra to suit various situations.
Click to expand...
 
I'm sure that incandescent bulbs can be made more efficient, in much the same way that a modern steam engine would be more efficient than one designed 100 years ago, but, I still doubt these guys can get enough visible, usable, light per Watt from a modified filament to compete with CFLs and LEDs, it will always be heat source that emits light rather than the other way around
 
Quantum physics is weird. I'm not prepared to say it can't work, and a filament doesn't have a band gap to overcome.

If a large percentage of the energy can be captured and re-radiated in the visible spectrum, I see no reason such a bulb can't compete with LEDs - and assuming the filament is "easy" to manufacture, it's a lot simpler than making a LED lamp run on mains and be bright enough.
 
According to the article below, current white LEDs aren't white enough - the brighteners in washing powder make whites look whiter by reacting with ultra-violet light to fluoresce, and white LED lamps don't emit any UV (which, actually, would be a waste of energy in terms of lumens per watt). So this lot are adding some.

Electronics Weekly said:
The science behind white LEDs

Steve Bush looks into Lumileds’ explanations of how it has developed LEDs that produce warm, but also crisp white light

How do you get that ‘whiter than white’ look in clothing?

You add fluorescent dyes to the cloth that absorb short wavelength sunlight – violet and ultra-violet – and re-emit it as blue.

Visually, this extra blue works against any yellowing of white cloth to give a whiter-than-white look.

This same fluorescence is what makes some clothes glow under ‘black light’ (ultra-violet) in clubs.

According to Californian LED maker Lumileds, for it to work, the colour point in CIE u’, v’ space is shifted downwards in v’ by six to eight points.

“Perceptual studies done at Lumileds determined that this is the ideal blue shift to give a perception of crisp white without shifting too far and giving a noticeable blue tint,” said the firm.

Getting whiter-than-white to work under artificial light, in clothing shops for example, requires a light source with that includes wavelengths shorter than (more violet than) 420nm.

Halogen lamps have some of this, as do ceramic metal halide lamps, while conventional white LEDs have none – being based on blue die with emission typically peaking at 455nm (middle peak in diagram, Figure 1).

To stimulate optical brighteners, Lumileds developed LEDs that include a second (left) peak between 400nm and 415nm alongside the standard one at 455nm and the phosphor-converted yellow-green and red emissions.

Launched a while ago and branded ‘CrispWhite’, Lumileds has now revealed some of the thinking behind it.

The colour temperature is 3,000K. “This gives a warm inviting environment similar to traditional halogen lamps,” said the firm.

Specifically, emission is designed to be below the Plankian curve, to ensure all luminaires have a warm tint, while staying inside ANSI limits.

With treated cloth, additional blue shift from the stimulated optical brighteners pushes the objects down towards the psychologically ‘whitest’ point.

“This colour point is very similar to ceramic metal halide lamps that are adopted in retail environments,” said Lumileds.

The 400nm-415nm emission range was chosen because it is both short enough to stimulate brighteners, as well as being close enough to the edge of the eye’s sensitivity curve to have little effect on the perceived colour of untreated items.

Ultra-violet light can cause colours to fade, and has been linked to the development of cataracts.

“CrispWhite uses only LEDs with dominant wavelengths above 400nm, which don’t cause fading and other damage to merchandise,” said Lumileds,

“Light with wavelengths below 400nm is classified as ultra-violet and also carries increased risk of causing cataracts with extended exposure,” said the company.
Click to expand...
 
You must log in or register to reply here.
Back
Top