Speakerplans.com Homepage
Forum Home Forum Home > Other Chat > Lighting Talk
  New Posts New Posts RSS Feed - UV Cannon, 400watt vs 100watt COB
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

UV Cannon, 400watt vs 100watt COB

 Post Reply Post Reply
Author
Message
BJtheDJ View Drop Down
Young Croc
Young Croc
Avatar

Joined: 28 November 2012
Location: Bristol
Status: Offline
Points: 884
Post Options Post Options   Thanks (0) Thanks(0)   Quote BJtheDJ Quote  Post ReplyReply Direct Link To This Post Topic: UV Cannon, 400watt vs 100watt COB
    Posted: 21 August 2017 at 8:30pm
I have a few 400 watt UV cannons and I've been very happy with them over the years.  Yep, they're delicate and subject to breakage if you look at them in the wrong way andf remember not o stick them on top of the bass bins without a strap to hold them on.

Been seeing lots of adverts for the new 100watt CoB UV cannons - at considerably more expensive prices than my current lights have cost cost me.

I realise that breakages will be a thing of the past (by and large) if I invest and the LEDs will be unlikely to fail while I'm still alive.

But, I already have 8 of the giant things with their 400 watt consumption and propensity to warm up a 16amp cable if you have 4 of them on the same outlet - but they can keep you warm on a cold night in a marquee LOL

Any opinions on light output comparison between the two ??  Or is it just near-UV like a lot of the led-uv luminaires??

cheers  BJ
If your vaccinations were mandatory and nobody in your class had measles,
then you know that you're not a Millenial kid.
</f
Back to Top
Hemisphere View Drop Down
Old Croc
Old Croc


Joined: 21 April 2008
Status: Offline
Points: 2272
Post Options Post Options   Thanks (0) Thanks(0)   Quote Hemisphere Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 10:20am
Unless anything has changed significantly in the past 5 years or so, it won't quite match the performance of 'true' UV lights but if you're lucky it'll come pretty damned close.

Last I heard there were some new high powered 1w and 5w LEDs on the market but they were very expensive, like 15 quid for the 5 watt ones, and they extended a lot further into the UV spectrum than the smaller, cheaper LEDs they use to make the cheapo flat panels. (those use 0.2w LEDs)

It's likely these are using those newer type of UV LEDs or something similar, and that seems to be reflected in the price.


Looks like it's probably using 100x 1w LEDs

Obviously just because theoretically you can get LEDs that reach that part of the UV spectrum is no guarantee this manufacturer is actually using them, but they do exist:

https://www.amazon.com/world-365nm-Ultraviolet-Light-Power/dp/B00R17ME7C

It's the 365nm you need. That's the wavelength UV cannons usually operate on.

Well they specify it as '365-370nm' - it's more likely to be 370 but they stick 365 on there because they know that's what people are actually looking for.

Surprisingly even that 5nm can make a slight difference to some colours, but not much. It's close enough.

The cheapo flat panel LEDs are usually 390-395nm.

One option you have would be to cut out the middleman and make your own. Might work out cheaper and could be a fun project:

http://www.topledlight.com/100w-100-watt-ultra-violet-uv-365nm370nm-high-power-led-9000mw12000mw_p1903.html

$200 for a 100w LED. 50 grams LOL

Imagine sticking two or more of them in one DIY mega-cannon! Actually they're so small that with a bit of bodging you could maybe retrofit them into an old style cannon which needs a new bulb.

Edit: So, I just 'analysed' the published chart from the $200 LED I just linked to, the one rated at 365-370nm. The chart is only divided into 100's of nm but it didn't look like the peak was centered on 365 or 370. Turns out it's centered on 377nm. 

This is what the light reproduction of a classic 400w UV bulb looks like:
Image result for uv cannon 400w wavelength
This is what you get with the LED, even the recent high powered lower wavelength ones (the ones you're looking at in the Cob cannons, almost certaianly)


So it's much more of a sharp peak than a scattered range (edit: that might be because it's a simulation rather than a measurement, but if the UV bulb cannon was shown as a simulation it would be focussed on 365 rather than 377. It seems likely that LEDs are more focussed in their range than metal halide bulbs but that's just speculation). By 365nm it's 23% intensity, but the 400w cannons are 90% intensity at 365nm, and they also reproduce a good dose of 340-360nm which is no doubt useful.

So I was a little hasty in my first analysis. The UV LEDs have still got a long way to go.


Edited by Hemisphere - 22 August 2017 at 11:44am
Back to Top
Hemisphere View Drop Down
Old Croc
Old Croc


Joined: 21 April 2008
Status: Offline
Points: 2272
Post Options Post Options   Thanks (0) Thanks(0)   Quote Hemisphere Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 11:12am
Just to go a little more into the science as to why you really want that range at - and below 365nm for best possible results:

Fluorescent materials have what's called an excitation spectrum and an emission spectrum.

Unfortunately I can't find them for most fluorescent colours but I did find one for the bog standard, fluorescent blue-ish colour, the one that they add to washing powder and the one that makes white t-shirts glow under the blacklight.


That's one of the easiest colours to fluoresce under a blacklight, and if you've ever bought one of those cheapo panels you'll know that, while it doesn't do much for lighting up a full colour painting, it always does a fine job of making people's t-shirts glow. That's about all the cheap LEDs seem to do well.

You can see from the chart that's because they're already absorbing light at near to maximum intensity at around 390-400nm.

But a lot of the fluorescent colours' excitation spectrums begin much lower, so much that they only catch the tail end of the cheap blacklight's range.

Only the classic bulb cannons will excite the whole of the range. The new higher powered LEDs will catch most of it, enough to reproduce all the colours brightly, but the relative intensity between colours won't be equal, with colours focused towards the higher end of the UV excitation spectrum appearing brighter than those at the lower end, and most of the colours feeling less rich and full bodied than they otherwise might be. Very bright and pronounced, but still lacking compared to the 'real deal'.

Edit: I wouldn't count on UV LEDs 'never breaking' by the way. I had a couple of LEDs die on a flat panel. They're not invincible! Certainly way, way longer lifecycle than the big bulbs though.


Edited by Hemisphere - 22 August 2017 at 11:30am
Back to Top
BJtheDJ View Drop Down
Young Croc
Young Croc
Avatar

Joined: 28 November 2012
Location: Bristol
Status: Offline
Points: 884
Post Options Post Options   Thanks (0) Thanks(0)   Quote BJtheDJ Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 12:24pm
Hemisphere,  Many many thanx for your (very) in-depth reply.

I shall have to keep a close eye on the <365 output specs before biting the bullet and investing.
If your vaccinations were mandatory and nobody in your class had measles,
then you know that you're not a Millenial kid.
</f
Back to Top
Hemisphere View Drop Down
Old Croc
Old Croc


Joined: 21 April 2008
Status: Offline
Points: 2272
Post Options Post Options   Thanks (0) Thanks(0)   Quote Hemisphere Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 1:03pm
I don't think you'll find a better output in an LED fixture than the one I linked to. That's $200 just for an LED on it's own, from a no-brand Chinese import shop. If it was installed to a commercial lighting product it'd be almost double that. That unit will be more or less what these 100w COB fixtures will be using.

If the wavelength went significantly lower it would be significantly more expensive. It's a basic limitation of the technology. I remember about 5 years ago, maybe 6 or 7, when these 377nm LEDs started appearing on the market. You couldn't buy any commercial panels at all with LEDs that were this good at the time, the 5w was 15 quid and the 1w was about 5 pounds, and you couldn't get these big high powered clusters. So I think these are really quite a new development. State of the art as far as the consumer market goes.

It's similar to why blue LEDs are more expensive than the other colours (or at least they used to be for a very long time). The lower the wavelength, the more expensive materials and processes they need to use to reproduce it, but sometimes the field advances and one type of process previously reserved for specialist applications, edges over into market viability.

For a long time blue LEDs weren't even possible for any price. Each wavelength of LED uses a fixed chemistry, certain types of metals and gases that work to produce certain wavelengths in the LED. The current most market viable chemistry centres around 377nm, and the second most viable is around 395nm. There might be small fluctuations but I don't think there will be any hidden gems producing true 365nm in a consumer product.

More details about the science. The rabbit hole goes really deep on this stuff!
The most relevant point among the rest though is "Near-UV emitters at wavelengths around 375–395 nm are already cheap and often encountered, ". It's just a Wiki page and hardly a definitive authority, but it does seem to suggest 375nm is the low end of the widely available range.

But anyway there's a very good chance you'll be satisfied with what they can do. Maybe just buy one from a shop with a good returns policy, carefully open and test it and send it back under your legal right for a 14 day cooling off period.


Edit: Here's another example of a COB LED that claims 365nm, https://th-led.en.alibaba.com/product/60444580173-800307320/Tianhui_High_Power_COB_UV_LED_Module_120W_365nm.html

Their published chart looks superficially as though it might be 365nm, but again on careful analysis (copy chart into Photoshop and zoom in), it's actually centred on 372.5nm.

I'm pretty sure they're all using the same chemistry and there's a hard limit they hit up against around 375nm, but they massage and misrepresent the data in every way they can to make it seem like 365nm, because they know that's the number people actually want. Again the issue of a much steeper roll-off than metal halide is apparent, down to 10% intensity by 360nm.


Edited by Hemisphere - 22 August 2017 at 1:32pm
Back to Top
Jasonstry View Drop Down
Young Croc
Young Croc
Avatar

Joined: 11 September 2007
Location: Cannock. J11 M6
Status: Offline
Points: 1367
Post Options Post Options   Thanks (0) Thanks(0)   Quote Jasonstry Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 1:18pm
There was a similar thread on the Blue Room forum that might help.



Down with signatures!
Back to Top
norty303 View Drop Down
Old Croc
Old Croc
Avatar

Joined: 18 August 2004
Location: Eastbourne
Status: Offline
Points: 8806
Post Options Post Options   Thanks (0) Thanks(0)   Quote norty303 Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 1:26pm
The 100W COB have dramatically more output than a 400W cannon.
They give a horrible visible purple light that you don't get nearly as much of with a decent lamp like an EYE Iwasaki, but in terms of coverage and throw the COB is just a beast.

You can buy just the COB chips direct from China and retrofit into a suitable LED COB floodlight fixture (as in outdoor garden floodlight type light) of same rating, and have the bonus of them being IP rated as well.

Lower wavelengths are available through binning of diodes. they are more expensive because they are rarer.

Binning is how we get our longer wavelength blue laser diodes like 462nm


https://www.aliexpress.com/store/product/High-power-LED-chip-UV-COB-Purple-Light-360Nm-370Nm-380Nm-390Nm-400Nm-410Nm-3W-5W/508860_32804292618.html

Edited by norty303 - 22 August 2017 at 1:27pm
My laser stuff: Frikkin Lasers
Back to Top
studio45 View Drop Down
Old Croc
Old Croc
Avatar

Joined: 16 October 2007
Location: United Kingdom
Status: Offline
Points: 3863
Post Options Post Options   Thanks (0) Thanks(0)   Quote studio45 Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 3:15pm
So....if a broader range of wavelengths like the mercury lamps produce works better for illuminating multi-coloured UV art, and the LED production process gives chips with a range of actual output wavelengths which are then sorted - why don't they just throw a random mixture of chips straight out of the machine on there, and end up with the desired multi-wavelength output?
Studio45 - Repairs & Building Commotion Soundsystem -Mobile PA
Back to Top
Hemisphere View Drop Down
Old Croc
Old Croc


Joined: 21 April 2008
Status: Offline
Points: 2272
Post Options Post Options   Thanks (0) Thanks(0)   Quote Hemisphere Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 3:54pm
How do you know that's not what the (seemingly most widely available) 375nm chips have done?

It still probably results in a distribution similar to the one shown in the graphs (especially when you get up to clusters of 100 or more).

To accurately emulate the performance of a metal halide bulb, you'd need a fair few LEDs centered around 340-350nm, and that's pretty much out of the question. It seems like 360nm is the practical outer limit.

This discussion would be a lot more useful with the actual excitation and emission spectrums of the most commonly used fluorescent materials. I swear there used to be retailers of pigments that listed the wavelengths, or there may be a webpage somewhere which details the minerals most commonly used to create fluorescent products, along with those wavelengths.

A good rule of thumb is 365nm though. There are loads of minerals on or around 365, and the manufacturers develop their products around thate standard. I doubt they're all dead on the mark though. Some colours always seem dimmer than others even under a really good quality light.

Another useful tidbit:
Quote A 365 nm LED is only about 20% efficient as a 405 nm LED.  Individual 365 nm LED's are also relatively expensive.  Therefore, manufacturers need a massive number of expensive LEDs to create a decent LED Blacklight.  Furthermore, light from true 365 nm LED’s is not visible and therefore uneducated consumers are not happy with the 'light output'.

You can see on the page Norty linked to, that their most expensive LED is actually legitimately 365nm. And that's all the confirmation you need that a commercial lighting fixture isn't centered on 365. Metal halide bulbs will show a tint of visible purple only because their range is so broad.

The other risk with these no-name LED imports is you might simply not get the advertised spec, as with the listing claiming 365 with a chart that shows 377.

Tell you what though, with a bit of R&D you could probably put together a premium quality LED blacklight, by studying the precise ranges of all the most common fluorescent materials and precision matching a set of UV LEDs of varying wavelengths in optimised proportions, to make every colour of paint light up to exactly the same brightness, or at least as close as it's possible to get. It'd probably work out 3 or 4 times the price of a regular one though..and high quality paints probably do a fair job of that already.


Edited by Hemisphere - 22 August 2017 at 4:00pm
Back to Top
Andylaser View Drop Down
Registered User
Registered User


Joined: 16 April 2010
Location: Southampton
Status: Offline
Points: 300
Post Options Post Options   Thanks (0) Thanks(0)   Quote Andylaser Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 10:35pm
I still think that LED UV sources produce far too much ambient light in the upper purple and blue spectrum. Might be worthwhile experiment using a filter to try and reduce this, but I always preferred the 400w cannons.
"music so loud, that if we move in next door to you; your lawn will die" - Lemmy
Back to Top
Hemisphere View Drop Down
Old Croc
Old Croc


Joined: 21 April 2008
Status: Offline
Points: 2272
Post Options Post Options   Thanks (0) Thanks(0)   Quote Hemisphere Quote  Post ReplyReply Direct Link To This Post Posted: 22 August 2017 at 11:07pm

No ambient light here though. Seems even less than the 400w cannons.


Edited by Hemisphere - 22 August 2017 at 11:08pm
Back to Top
norty303 View Drop Down
Old Croc
Old Croc
Avatar

Joined: 18 August 2004
Location: Eastbourne
Status: Offline
Points: 8806
Post Options Post Options   Thanks (0) Thanks(0)   Quote norty303 Quote  Post ReplyReply Direct Link To This Post Posted: 23 August 2017 at 12:27am
You can see from my link that they are clearly selling the same chip that has been binned. The chance of getting the shorter wavelengths is lower so they fetch more money. As the wavelength increases they get cheaper presumably because they're more prevalent. We have the same situation with laser diodes. The 462nm diodes are just binned 445. They typically have less output for the same current too, so you trade efficiency for more desirable wavelengths.

Andy's comment about too much visible light was true a few years ago, but paying decent money for the more powerful lower wavelengths gives some very good results (coming from a man who was more than happy to sell his bulky power hungry cannons)
My laser stuff: Frikkin Lasers
Back to Top
 Post Reply Post Reply

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.06
Copyright ©2001-2023 Web Wiz Ltd.

This page was generated in 0.109 seconds.