LED lights: A Few Facts

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

Any experienced user willing to update this sticky? Much of the OP's advice still seems solid, but LED and battery efficiencies have improved and prices have dropped a LOT since the OP in 2010. The ratings (in lumens) of the less expensive LED lights may be BS, but it sure seems you can get a very bright compact light for <$30 that SB members have taken to 30m on multiple dives without flooding. Is it "5000 lumen"? What's the actual lux? Most divers don't care about exact measurements if it works well for the task. Sure it would be nice to have an objective way to compare, but marketing always seems to exaggerate or emphasize less meaningful stats to make their product sound impressive. Caveat emptor, but that's why it's nice to read user experiences!

These cheaper lights may or may not be optimal for photography (hot spots, etc?), but it would be great if an experienced user could weigh in (after actually trying some of these newer lights.) I am sympathetic to manufacturers who may still be selling stock of lights based on technology that was good enough just a few years ago, but folks shouldn't have to spend 4x - 10x more unless there are real-world advantages to a diver. And as a newer diver, I'm amazed by the range of prices for apparently similar hand-held diving lights.
 
Thanks for your comprehensive report regarding to the LED light. I am from one dive lamp manufacturer and we are named Ano Industries. I would like to write more about the LED Technology for more people to review and study.

LED is the forth generation of bulb source since Edison invented the bulb. The biggest advantages of the LED is power saving, long liftspan and easy color control. The most popular LED for flash light is from
Cree, like XPG, MTG, XML series and etc; you can refer to the link for more information about Cree's LED. Some of the LED from Luminus and Osram are good for flashlights as well.

First of all, we, Ano industries never uses the copy LEDs in China for our lamps, none of them was or will be use the copy LEDS. We do buy directly from Cree or Luminus; but you may not know the price between the REAL Cree and copy LED is about 5 times difference. The LED makes difference.

All of our lights are designed with constant discharging current; but the real thing you may want to know that the driving voltage of LED is 3.0V and if the voltage is lower than 3.0V, it will be dropping inconstantly. This happens only to the single or battery pack in parallel, but not to the battery pack in series or series/parallel. The burntime of a flashlight is depending on the battery, not actually on the LED. Also, the lumen of LED is not loosing the LUMEN gradually, but maybe, it is your battery and your flashlight (inconstant driver) that makes the lumen dropping. Generally, the warranty from Cree for the lifespan of LED is 50000 hrs with proper discharging current. We, Ano Industries have the warranty for 12-18 months; but if you say the liftspan is higher than this, it is not true unless the flashlight has 3 years warranty.

I agree with you on the lumen saying. When we test the lumen, we normally do not take the full assembled lights to test the lumen because the hotspot from reflector and glass is actually causing lumen losing. The light radiation should be read from the bulb and this is more accurate; you can never read the accurate lumen from the test for a full assembled light. The color temperature is another matter that could influence the brightness as well as the light absoption from a reflector.

The switch is very important for a diving light and we have magnetic switch, turning and the mechanical pushing. All of our mechanical switch can be openned and replace the O-Ring and I think this is very important when your switch is full for sands that you can open it for easier repair. Most vendors are just using a clip at the end of the pin; saving costs, but not good for repair. For magnetic switch, the most important is the plating of magnet and if this is not protected well, the magnet is easy to get rusted. We apply Chrome plating, like the 1st stage regulator and 100 hours salt spray test was passed according to the standard of ASTM B 117-03. One more thing about the material, you may want to know that the Stainless Steel does not mean it can not be rusted, but it is depending on what kind of envirement you are going to put and what type of grade of SS steel you are going to use. We are using the same grade of Stainless Steel as the diving computer and for any of our SS material, 100 hours salt spray test was passed based on ASTM B 117-03. SS302, SS304 and even SS316 can be rusted; for scuba diving, SS316L should be the only SS material used because of the corrosion of sea water.

About the battery, everyone knows there is alkaline and lithium rechargeable battery. There are many different of rechargeable battery and I am going to take only ICR as example. If a designer is designing a light with alkaline battery, the driver may not set up the low cut-off voltage because it is one-time use. If it is used a lithium battery (protected or unprotected), the dirver is better to design with at least 2.4V cut off voltage in case you are using a unprotected lithium battery. If the lithium battery is over charged or discharged, it can ruin the battery for ever and the lifespan of a battery is depending on the protection for overcharging or overdischarging. A good designed driver is free to use the the unprotected battery. All of our lights with single battery is lasting as least 1-2 hours for high power setting and all of our lights with battery is constant discharging until the battery is used up.

I would like to take 1 example for our design for the canister light, CL2300. We stated 2300 lumen and even you use a lumen tester to test, it is higher than 2300 (the actual is about 2700 lumen, maybe we should call 2700 lumen because the consumer likes to hear this), but we are using 6 x ICR18650 battery pack (3 hours burn time at HI power mode) and a lot of lamp manufacturers are using the tube with 2 or 3 ICR26650 and they stated the longer burntime than us. I do not know how they can do it. The proper designed battery pack is battery than un-matched singles cells.

I hope everyone can understand more about the LED lamp and if anyone has knowledge about other type of lamps, please talk like this that more people can learn.

Welcome to contact us on Skype: anowolf for any questions you have on the LED lamps.


If a light manufacturer is being more realistic they will advertise runtimes down to 1/2 of the max lumen rating not all the way until it actually quits.

It a light doesn't have a constant current driver (in which case it will be at max brightness until it just suddenly quits) it will only be at it's max advertised lumen output for a short while. As the battery is used the output will decrease. It may be advertised as 200 lumen for instance but after running for 1/2 hour it may be 170.

After using it for several dives it may now only be 100 lumens. When it gets so the battery is only able to produce 1/2 max lumen this should be considered its effective runtime. Maybe that's 6 hours (depends on the batteries).

If a light is advertised as running for days that's somewhat misleading. Sure it is still burning but it may not be any brighter than a candle. It's also something any nonrechargeable battery will do when powering led's. Led's are very efficient at lower power so it's a long, long time before the light actually doesn't come on at all.

However, if you buy a light that is advertised at 200 lumens and that is claimed to run for days and days you might think that means that it's running at 200 lumen for days and days...it's not.

Another fact is that there are only so many led's out there. Most lights, regardless of the sales prices, are using some of the same led's as other much cheaper lights. The actual retail cost of most led's is in the $15 range so the difference between a $700 light and a $50 light isn't the led.

It may be the build quality (higher production/test/development costs) or it may be looks or status or marketing but it's not the parts. You can find an expensive light that isn't well made and you can find a cheap light that is. Of course your chances are probably improved with a more expensive light but there is no direct correlation.

One other fact not always appreciated is that two lights with differing lumen outputs can appear to be the same at first glance. You may think that the higher lumen light must be the one that appears to be brighter. That's not necessarily the case.

The higher lumen model is producing more light but it may not be any brighter at the hotspot than a lesser lumen model with a tighter hotspot and no spill. Lumen is just describing the output. Lux is the measurement at a particular spot.

A 100 w bulb in the ceiling of the living room of your house may produce more lumen than the 60 w bulb that is by your chair that you use as a reading lamp. The light falling on your book from that 60 w bulb is brighter than the light falling on that book from the bulb in your ceiling.

Lights that need to be twisted to work underwater may be preferred as they are simple with no complicated parts involved or they may not be preferred because they can come on by themselves at increasing pressure. They also rely on o-rings to handle the dynamic motion of being twisted at pressure.

Magnetic switches have the advantage of not penetrating the body at all (and can't be a cause of leakage) and don't put any dynamic stress on sealing o-rings. They are more complicated and need to be well designed to keep magnets from falling out or if not sealed by degrading in saltwater. The circuitry internal to the light is more complicated and expensive although once properly designed is not so much of a problem unless the light leaks.

Non-rechargeable batteries have the advantage of lasting longer and rechargeable batteries have the advantage of saving money, being able to produce higher power for shorter periods of time with the right drivers and therefore allow for matching power to anticipated dive times.

Also, there is a limit to the use of non-rechargeables for higher powered lights. Or rather there is a useful limit. If a 1500 lumen light can only burn for 30 minutes that's not a useful dive light. If it can produce 1500 lumen for only a minute or two but burn for 2 hours at a rapidly declining level of output then that isn't really a 1500 lumen light either.

Led's are more efficient as power is reduced so with multiple power settings on a dive light battery power can be extended by reducing power on those dives when it's not needed. Since led's are more efficient as power is reduced this also means that a 50% reduction in power may only reduce the brightness by 35% so this can be a real benefit.

I see some misconceptions sometimes regarding led lights and that's the reason for this post. It's long and maybe isn't effective because of that but maybe it will help someone.

Anyone knowledgeable regarding led lighting feel free to clear up other misconceptions. Too much of a buyers knowledge regarding led lights frequently comes only from manufacturers' ads. This can be misleading.
 
To get 2700 lumen with 3 LED, so per LED is 900 lumen. You are drive the Cree XML at just below 3A, right? What is the 18650 capacity that gives you 3 hour of run time?

1. If the 3 XML is in parallel, the total current drawn will be 9A. 3 hours runtime meaning total capacity of 6 18650 is 27000mAh, each will be ~4500mAh.
2. If the 3 XML is in series, you have to run 3 18650 in series in bank, two bank in parallel. Current draw is still 3A here. Per bank needs to prive 1.5A for 3 hours. Each 18650 still needs to be ~4500mAh

None of these estimation include the typical lost of efficiency due to linear current regulator, head lost due to battery internal resistance (kind of substantial at that kind of current drawn). At least from this simple math, 2700, 3 hours of run time, 6x 18650 is a very optimistic number if possible at all. Most 18650 is at 3000mAh today. Taking off parasitic lost at 3A, you will get 70% of the capacity only. I am thinking you will have 1.5-2hr of runtime if you are really driving at 3A.

But I also think driven XML at 3A is also a very optimistic design. The forward junction voltage needs to be 3.3-3.4V. A typical ICR will run out of head room long before using it full capacity.
 
Thanks for the question. You may consider 3S2P for battery pack and the LEDs are in series? You know how to caculate this?

This is why I said to optimize the design is very important, not only on the driver, but also on all details like the LED linear and battery pack. I do not know why you are losing 30% power from the battery pack, what a wastage for this and I believe no manufacturers now have the same loss on power like what you said. If you check our canister and light, you will not say this after you test it. We, Ano Industries just stated 2300 lumen on our manual with 6 x ICR18650, not like others that exaggerate the lumen to get more attention.

Maxium driving current is not recommanded because this can shorten the lifespan of LED to half. Thinking about the driving current, 3.0A, how can you make this last for 2 or 3 hours by 3 ICR26650? Do you think this possible no matter by Parallel or Series? This is not possible. The capacity of one ICR26650 is normally 4Ah. If the batteries are installed into a small tube, like the housing of a backup lights, the batteries are in series. If it is in series, we can get below data:

1. Max. capacity from the canister is equal to one single battery, means the capacity is 4.0Ah
2. The voltage of the canister could be 12.6V
3. The LEDS can be in series or paraell; if in parael, the total driving current will be 9A and if in series, the driving current will be 3A, but the voltage will be 12.6V.
4. The burntime will be 1 hour 20 mins or even shorter with single battery; the single cell needs to be selected to match for closer


To get 2700 lumen with 3 LED, so per LED is 900 lumen. You are drive the Cree XML at just below 3A, right? What is the 18650 capacity that gives you 3 hour of run time?

1. If the 3 XML is in parallel, the total current drawn will be 9A. 3 hours runtime meaning total capacity of 6 18650 is 27000mAh, each will be ~4500mAh.
2. If the 3 XML is in series, you have to run 3 18650 in series in bank, two bank in parallel. Current draw is still 3A here. Per bank needs to prive 1.5A for 3 hours. Each 18650 still needs to be ~4500mAh

None of these estimation include the typical lost of efficiency due to linear current regulator, head lost due to battery internal resistance (kind of substantial at that kind of current drawn). At least from this simple math, 2700, 3 hours of run time, 6x 18650 is a very optimistic number if possible at all. Most 18650 is at 3000mAh today. Taking off parasitic lost at 3A, you will get 70% of the capacity only. I am thinking you will have 1.5-2hr of runtime if you are really driving at 3A.

But I also think driven XML at 3A is also a very optimistic design. The forward junction voltage needs to be 3.3-3.4V. A typical ICR will run out of head room long before using it full capacity.
 
....
I would like to take 1 example for our design for the canister light, CL2300. We stated 2300 lumen and even you use a lumen tester to test, it is higher than 2300 (the actual is about 2700 lumen, maybe we should call 2700 lumen because the consumer likes to hear this), but we are using 6 x ICR18650 battery pack (3 hours burn time at HI power mode) and a lot of lamp manufacturers are using the tube with 2 or 3 ICR26650 and they stated the longer burntime than us. I do not know how they can do it. The proper designed battery pack is battery than un-matched singles cells.


Quote from your previous post. Actual 2700 lumen, 6x 18650 ICR, 3 hour burn time in Hi, 3 XML LEDs. I didn't think this combination is possible.

---------- Post added July 21st, 2015 at 12:07 AM ----------

Thanks for the question. You may consider 3S2P for battery pack and the LEDs are in series? You know how to caculate this?

This is why I said to optimize the design is very important, not only on the driver, but also on all details like the LED linear and battery pack. I do not know why you are losing 30% power from the battery pack, what a wastage for this and I believe no manufacturers now have the same loss on power like what you said. If you check our canister and light, you will not say this after you test it. We, Ano Industries just stated 2300 lumen on our manual with 6 x ICR18650, not like others that exaggerate the lumen to get more attention.

Maxium driving current is not recommanded because this can shorten the lifespan of LED to half. Thinking about the driving current, 3.0A, how can you make this last for 2 or 3 hours by 3 ICR26650? Do you think this possible no matter by Parallel or Series? This is not possible. The capacity of one ICR26650 is normally 4Ah. If the batteries are installed into a small tube, like the housing of a backup lights, the batteries are in series. If it is in series, we can get below data:

1. Max. capacity from the canister is equal to one single battery, means the capacity is 4.0Ah
2. The voltage of the canister could be 12.6V
3. The LEDS can be in series or paraell; if in parael, the total driving current will be 9A and if in series, the driving current will be 3A, but the voltage will be 12.6V.
4. The burntime will be 1 hour 20 mins or even shorter with single battery; the single cell needs to be selected to match for closer

I am not a light manufacturer, but I have enough knowledge in analog circuit, LDO design to do a rough estimation. My first questions to you is that in order to get 2700 lumen out of 3 XML, what is the current driving the LED? Is it ~3A? If LEDs draw 3A, how much is drawn from the battery pack?

- It got to be more than 3A because the driver does consume something. That is lost of efficiency #1
- 18650 internal resistant, say ~100mOhm. At high current drawn application, this lost is not small. This is lost of efficiency #2
- The driver. I assumed you use linear constant current diver previous. This type of driver is NOT know for efficiency because of the voltage between the supply (battery) and the output (LED) is wasted. The more efficiency type of regulator is switching type, but it causes flicker. A well design switching driver can be flicker free. If you use this type, the efficiency lost here is minimal

But still, at 3A kind of current drawn in such small design, efficiency up at 90% is very unlikely
 
Last edited:
Thanks for asking and I think this is good for us to consider all client's concerns into our product design. You may add our Skype and if you have new ideas/requirements for product, we can design accordingly for you and for sure, we can answer your questions timely.

The driving current can't be 3A or it is not possible to have a 3 hours burn time with current battery pack design. The battery match for each single cell is very important as well as the protection circuit board. For the canister light, it is more to consider the voltage drop than the current that the from the output of battery to the driving PCB of light, the instant voltage drop is huge. The bad cable can lose a lot of power, but the copper cable can make the voltage drop and power less waste.

We are using magnetic switch for our canister light and you can refer to this link for more details. The burn time I talked is the continuous burning. As you, the battery pack itself can be discharged and the wastage from the magetic switch is causing the power losing as well.

I am not sure about the flicker issue you mentioned above; but we do not have this issue and if the light is using the body for power connecting and the push button switch or turning switch and the cotact parts are not touched tightly; it may have the flicker issue. The tact switch and magnetic switches are free of flicker issue and if it has, the light will not turn on.

---------- Post added July 21st, 2015 at 06:03 PM ----------

By the way, we are using U2 LED, more efficient than the older generation of T6.
 
Reading yours spec from yours site looks you are still rather sloppy .....and full of marketing .... for instance

404 Not Found Zajeta slika.jpg
 
Thanks for your sharing&#65292;and I will consider the battery.Though the rechargeable is cheaper than non-rechargeable ,I still will choose non-rechargeable.As you said non-rechargeable batteries have the advantage of lasting longer....
 
https://www.shearwater.com/products/perdix-ai/

Back
Top Bottom