The LED Fresnel Shootout

[Brian]

The product under review is the Elation ELED Fresnel kindly supplied for review by CPC.

 

CPC Product Page

 

Since the unit was first announced the price has dropped from £325 excl VAT to £265 excl VAT.

 

 

As the manufacturer claims that their product is '500W Halogen Equivalent' I've taken that claim literally and compared their product with a 500W Minuette Fresnel made by CCT and used in venues all over the UK.

 

CONSTRUCTION AND OPERATION

 

1) It's bigger than I thought it would be. Mentally, I had it as something Minuette/Prelude sized. Something with a 100mm lens. It actually has a 150mm lens and is therefore closer in size to a 1k unit.

 

2) Construction is extruded corner sections with steel panels. A bit like many other fresnels. The yoke is folded steel and is a decent thickness with a single locking knob. Access to the inside is via a single, non-captive, thumbwheel which releases the sliding top panel.

 

3) The back panel contains a fused IEC inlet, a 4 digit display with 4 menu buttons and, a nice touch, both 3 and 5 pin XLRs for in and loop. There is a handle and a spot/flood focus wheel.

 

 

4) Focus is with a leadscrew in the base of the unit, again like many other fresnels.

 

5) The lightsource is mounted on a sliding carriage and consists of a PCB containing 9 LEDs, arranged in a square, with heat control by a heatsink and cooling fan. This is the only fan in the unit and overall the noise level is similar to a desktop PC. Under test I measured a maximum heatsink temperature of 47 degrees which bodes well for lamp life.

 

 

6) The other main items inside the case are a universal input, 100-250v, switch mode supply marked as delivering 24 volts at 2.5A and mounted on one side panel, and a control/display PCB at the rear.

 

 

7) The unit can be operated from the rear panel which allows you to set the dimmer level and save it so that next time the unit powers on it returns to its previous operating condition.

 

8) The dimming is smooth with no flickering at the bottom of the range.

 

 

BEAM QUALITY

 

Due to the arrangement of the LEDs in a square, the beam takes on, under some focus positions, a similar shape.

 

 

Likewise, at extreme spot focus, you can resolves the individual LED blocks.

 

 

There is some visibility of the fresnel lens rings in the beam.

 

 

The overall effect is that the beam is a bit ragged and is nowhere as smooth as a conventional fresnel.

 

 

For comparison the conventional fresnel set to a similar sized beam.

 

 

All these photos were taken with the units at 2m from the wall which will exaggerate any beam 'quirks' and in real life you're usually not concerned with beam visibility, just what it makes visible. However, in some of the applications I could see for this unit, ie display work, such artefacts could be a problem. Also, the large nature of the light source will cause problems with poorly defined barndoor cuts.

 

 

COLOUR

 

Normally, to compare light output, you would set the beam sizes to be the same by measuring the light levels and finding the point at which the beam drops to a specified percentage, usually 10%, of the centre intensity. Because of the beam blemishes and different shapes identified above this proved difficult to do and so they were set by eye to match each other.

 

Each unit was then fitted with filters and the remaining light level measured. Filters used, and their quoted transmission values, were Red = L106 = 9.3% transmission; Green = L139 = 11.9% transmission; Blue = L119 = 3.1% transmission.

 

Conventional Unit...

 

Open = 1670 Lux

R = 14.3% of open white (250 Lux)

G = 12.6% of open white (220 Lux)

B = 3.9% of open white (75 Lux)

 

LED Unit...

 

Open = 670 Lux

R = 7.5% of open white (50 Lux)

G = 13.5% of open white (90 Lux)

B = 3.7% of open white (25 Lux)

 

 

As an experiment I then attempted to match the colour of each unit to the other one. This proved very difficult to do but the following results were obtained...

 

1) Matching the conventional unit to the LED unit was partially achieved with L201. This gave a remaining light level of 37% of open white and an absolute level similar to that of the LED unit.

 

2) Matching the LED unit to the conventional unit was partially achieved with L204. This gave a remaining light level of 58% of open white and an absolute level only 25% of that of the conventional unit.

 

For a bit of fun I then tried a piece of L181 in both units. The LED unit had, visually, a reasonable amount of light left.

 

Overall the light has a very 'crisp' blue quality to it. Colour rendering is what you'd expect from white LEDs.

 

 

LIGHT OUTPUT

 

The question is 'Is this unit, as the manufacturer implies in their claim, a direct replacement for a 500W TH lamped unit?'

 

Some maths...

 

The T18 in my 500W fresnel gives out 11,000 lumens. The LED block in the LED unit is quoted as being 50W. If the LEDs were to give out 11,000 lumens they would need an output of 11,000/50 = 220 lumens per watt. Which would be very impressive.

 

The reality...

 

My 500W units has a quoted beam angle of 18 degrees to 57 degrees. In the spot position I measured a peak of 4300 Lux and at flood a peak of 800 Lux. The LED unit is quoted as being adjustable from 10 to 50 degrees; in spot I measured 2600 Lux and 400 Lux when in flood.

 

When set to visually similar spot sizes (see photos above) I measured 1670 Lux for the 500W unit and 670 Lux for the LED unit.

 

 

COMMENTS AND THOUGHTS

 

1) Heat. Even a 500W conventional unit gets hot; very hot. For speed I was hanging the gels over the front with a bulldog clip as a weight laying on the top of the unit. In the 30 seconds or so it took to take a measurement, write it down and pick up the bulldog clip it was getting too hot to touch. In any application where heat is an issue then LEDs, even with their other faults, should be seriously considered.

 

2) Installation. The LED units comes complete with built-in dimming. Never underestimate how much money that will save you. I was able to test the LED unit with nothing more than an IEC lead.

 

3) Lamp Life. Putting aside extravagant claims of 100,000 hour LED life, even if you only get 30,000 hours that still, on average, is 40 or 50 times the life of a TH lamp. Or even more if you judder the TH lamp whilst focussing it!

 

4) Mixing LEDs in a conventional rig. Let's be honest, LEDs aren't as bright, have an 'odd' colour and 'different' beams. So don't think of them as a direct replacement; find new ways to use them.

 

 

NOTE

 

I've avoided too many absolute measurements as, to be honest, they don't really give a useful picture. Any measurements are best used as a comparison between the two units.

[Guest lightnix]

A very interesting post - thank you very much for putting in the time and effort :up:

 

I wonder what the "official response" will be? :D

 

3) Lamp Life. Putting aside extravagant claims of 100,000 hour LED life, even if you only get 30,000 hours...

TBH I'm not sure you'd even get that, at a running temp of 47°C

 

Let us not forget, that an LED's "working life" is often defined as the time it takes for the lumen output to drop to 50% of the original brightness (although some big name manufacturers now quote 70%). There's a difference between "working life" and "usable life" and I can't help but wonder what the latter is, in the case of this unit.

 

I notice that, on the CPC site, it says, "50W 6,000k white LED module", but also, "Power Consumption: 75W" (my bold). Am I missing something here? Are the fan and electronics really pulling 25W?

 

Sorry to ask you to go back, Brian - but is there any chance you could investigate just what is drawing what in there?

 

What would be good, would be to see how it looked after 6 months of 6 hours on, 6 hours off, 24/7 and how much variation there was between the LEDs.

[Just Some Bloke]

Thanks Brian for the hard work. Much appreciated.

 

Edited highlights:

 

the manufacturer claims that their product is '500W Halogen Equivalent'

[snip]

When set to visually similar spot sizes I measured 1670 Lux for the 500W unit and 670 Lux for the LED unit.

 

Does it strike anyone that the manufacturer might be telling porkies?

[scjb]

I know "me too" posts are frowned upon, but just wanted to also say thanks for the time and effort put in. Much appreciated, and results to ponder on.

[Ynot]

Yup - thanks Brian - as usual a most detailed and logical report.

 

I can't say I'm surprised at the actual outcome - nor indeed that the manufacturer's claims are all but proven false - but then again, how often do (cheaper end) mfrs do just that and bull up their product?

There are, of course, lesser 500W fresnels out there....

I would imagine had Brian used a patt 45 instead of a minuette......

NO contest there, I suspect!!

 

I still think there may be a valid use for these LED fixtures. It strikes me that the output is maybe brighter than the budget LED pars on the shelves, and they're certainly appear to be more controllable....

Don't think I'd rush out and buy any JUST yet, but will be watching the market with interest.

After all, I may very well be able to source some funding for a more energy efficient/carbon efficient set of lanterns where I wouldn't for a set of incandescents....

:D

[Brian]

Are the fan and electronics really pulling 25W?

 

The internal SMPSU is a 24V, 2.5A unit = 60W. It's a standard generic part and similar units have an efficiency around the 80% mark. 60W/0.8 = 75W so I suspect that the 75W figure is a maximum theoretical draw figure based on the PSU.

 

Actual measured consumptions...

 

LEDs off = 8W

LEDs full on = 63W

[david.elsbury]

Is the dimming curve steppy at low levels, like some cheap LED fittings can be?

[Brian]

Is the dimming curve steppy at low levels, like some cheap LED fittings can be?

 

I used the inbuilt control to test as it has an auto-step function if you hold the 'UP' or 'DOWN' buttons in. Stepping 'DOWN' from DMX value 255 the steps became noticeable at around DMX value 30 and very noticeable at 20.

[Chris Beesley]

Just for reference on this one - we (as always) quote the manufacturer's spec directly in our catalogue/web text.

 

http://www.americandj.eu/product_info.php?products_id=1995

[Ynot]

Curious - the spec list gives 1 or 2 ch DMX control...

 

What's the second channel?

Strobe?

[Chris Beesley]

Single channel mode is dim 0-100% or two channel is CH1 dim 0-100% CH2 Strobe.

[Ynot]

Single channel mode is dim 0-100% or two channel is CH1 dim 0-100% CH2 Strobe.

A logical guess then...

:D

[karl]

Looking at the photos it seems the design is basically the same as a standard fresnel with the cluster of LEDs placed roughly were the lamp would normally be. This could explain the shortcomings in terms of beam shape and focus. Maybe a more radical design may be required.

 

As mentioned in the previous thread it would seem sensible to have the electronics mounted in the rear of the case so the case itself can act as a heat sink. This would reduce the need for fan cooling. Instead of mounting the light source on a movable platform the whole case could telescope/concertina to provide focusing.

 

Would it not be possible to use light pipes to gather the light and consolidate it into a smaller area? On our servers at work LEDs mounted inside have the light routed to the front panel using plastic light tubes. These do 'leak' light but I'm sure a better quality ones could be produced. If they were made of a material with good thermal conduction (like glass) they could also act as a heat sink for the LEDs.

 

I have no idea if this is practical and there may be even better solutions. I just think that maybe there are better solutions than doing a direct swap of one light source for another.

[Guest lightnix]

Just for reference on this one - we (as always) quote the manufacturer's spec directly in our catalogue/web text.

But if you are the importers of this equipment, do you not (under European Law) assume the legal responsibilities of a manufacturer - including those relating to trade descriptions?

 

If the manufacturers claims are shown to be nonsense, then how can you - in all honesty - continue to display them on your site?

[Chris Beesley]

True .. very true Lightnix. However for this product we are not the importer - that is American DJ Europe/UK, CPC acts as one of many distributors in the UK. I will however when im back in the office (im on hols at present) review the text on the basis of the results here - and also make American DJ EU/UK aware of the test report. Dont get me wrong, while we quote the specification given by the manufacturers we are not in the business of tricking people into buying products that do not perform their claimed specification.