Dive-Rite LED Lux to custom SST-90 Drop in conversion

fenix-store
K

kd5ahl

Newly Enlightened
Joined
Sep 26, 2006
Messages
59
Moderators, I was torn between starting this thread here, or in the diving light forum. If it belongs there, feel free to transplant me!

I'm in the middle of designing a drop-in replacement for the electronics and LED's that come standard with the Dive Rite LED Lux canister light.

I like to take underwater video, and the LED lights have a nice white color to them.

The original led's were triple luxeon k2's with focussing optics to give a nice tight beam. After removing the optics from both of the light heads, the spread was nice and wide, but had 3 rings of light, one from each emitter which really annoyed me.

I decided I wanted more lumens and a more even light pattern, so I have decided to replace the guts of each light head.

I have a bunch of photos, and some drawings of the planned upgrade slug that I will post later after I get some sleep.

I'm hoping that I can learn from any responses in this thread and that my foray into modding lights can help someone else out as well.

I have been poting in seversal of the other forums here at CPF, as I try to suss out all the design elements, and since I have actually started the modification process, I felt it was a good idea to make a single build thread so everyone that wants to, can follow along.
 
First pics:

IMG_3296-vi.jpg


Original Light Pattern

IMG_3297-vi.jpg


After Optics Removed

How to get the optics out:

IMG_3308-vi.jpg

carefully remove the o-ring that holds the lens in

IMG_3309-vi.jpg


The component parts

IMG_3311-vi.jpg


The triple Luxeon K2's

Here's a comparison between the head with optics removed (left side) and a modified sunray lamp I did for a friend with an SST-50 driven with a single shark buck. You can clearly see the 3 rings of light from the original Dive-Rite LED head.

IMG_5182-vi.jpg
 
Last edited:
Do you have any pictures of the LED driver in this light?

The reason I ask is they claim to have 940 lumens of output power
http://www.diverite.com/products/catalog/lights/lt6250led-lux

Yet when you read the technical specs, the numbers don't add up. The Luxeon K2 has a max of 300 lumens at 1500mA.
http://www.philipslumileds.com/pdfs/DS60.pdf

Adjusting for a little cooling, you might get another 5% out of the LED so say 315 lumens per LED. This will give 945 max lumens assuming 100% optical efficiency...wich won't be the case.

The datasheet says 6 hours of run time of 6 hours on a 12V 4.5Ah battery
http://www.diverite.com/downloads/diverite_146.pdf

How is this possible? 4.5/6 = 750mA....which means one of the specs is wayyy off. The light output or the run time.

Any comment on this?
 
I have photos of the driver somewhere. If I can't find them, I'll pull the light apart and take some tonight.

I too was skeptical, and I'll be honest, I have never used the light for 6 hours straight without recharging it. I typically recharge after 2 1+ hour dives.

I do recall during the burn-in of the batteries recommended by Dive Rite taking an extraordinary amount of time. I did not time it though.
 
Hey that would be great. Do you have any idea how much current the LEDs are running at? Do you have an ammeter you can put in line with the battery when it is on?
 
I have an ammeter I can place in line, yup.

I dont have any driver pics, so I'll shoot those as well as hook up and measure the draw from the battery. ~700mA sounds like what I remember when I hooked it up a while back.

I suspect the driver is converting the extra volts from the battery to extra current to drive the LED's at their maximum. I'll admit I'm not sure how the driver circuit works.

maybe someone from the flaslight electronics forum will read this and chime in.
 
Don't worry, you're talking to the right guy about electronics. I work as an EE for a company that makes LED drivers.

If we can find out what driver they are using, the rest will be readily apparent. I am wondering if they are using direct drive or a buck or boost converter. THat's why my interest in the drive electronics.

Depending on the drive current, the Vf of the LEDs will change the power equation.

So we need to know:
(1) What is the LED drive circuitry
(2) What is the LED drive current? Is it constant current?
(3) What is the VF of the LEDs (this will verify our LED assumption)

Then we can figure out the rest. I emailed dynazign to get the datasheet on their MR-11 light engine. Based on your picture, I am assuming they are the manufacturers.

Thanks for your help!
 
One thing I forgot to add: we actually need two current measurements: Battery current and LED drive current (at the output of the driver).

You may not be able to get the second measurement without de-soldering or using a clamp meter. I think we can derive it based on the drive electronics as long as the specs for the circuit are available and we have the total LED VF (voltage across all three LEDs). I am assuming it is an OTS part.
 
Last edited:
You can see the traces going from LED to LED, so they have to be driving those luxes in series.

That almost guarantees a buck circuit right?

I have a clamp type meter, will it read DC as well? I always used it for work on AC Panels.

In either case I am mot worried about unsoldering a wire from the driver to the LED's to get a current reading - The warranty was willfully violated when I put one of these in the light head:

DSC00241-vi.jpg


DSC00238-vi.jpg


that's a SST-90 driven at 9A by 3 shark bucks running at 3A each.

What I'd ---<REALLY>--- like is a 0-9A Buck Driver in a single module for driving single high power LED's.:cool:
 
Nice modification :)

The voltage measurement across the drive circuit output will tell us for sure if they are in series or parallel. I am assuming they are in series, it makes the most sense. They might be using a buck or direct drive. The Vf per LED will be around 3.85V @ 1.5A and 3.5V @ 700mA. It's so close to 12V that direct drive might be what they are using.

Can you confirm what the battery pack specs were on the battery that came with it? Also, how do you know that this is the 940 lumen LED Lux and not the 700 lumen or 500 lumen version? I just want to make sure we have the correct model for comparison.

Thanks again, look forward to hearing your findings later.
 
I returned the 700 lumen for this model. Unless Dive Rite screwed me, they sent me the LUX, it's specifically what I asked for.


I seem to recall the pack saying 4.5AH NiMH on the side. I'll check that tonite too.
 
allrighty.

photos incoming.

DSC00820-vi.jpg


This is the slug out of the light head taken as I was disassembling it. The red connector is NOT standard. I added that to facilitate the swapping of the head for the SST-90 slug.

DSC00821-vi.jpg


This is the underside of the driver board. The hot-glue that I added keeps the wires from breaking off at the circuit board.

DSC00822-vi.jpg


This is the top of the driver board.

DSC00827-vi.jpg


This is the draw at the battery.

I will remove a wire from the driver to test the draw at the LED as well. I will post the results of voltage and current measurements a little later, I need to grab some batteries for my meter :eek:

EDIT: and the bright light from the side in the last photo is not the flash from my camera, its the light from the head!!!


UPDATE:

Voltage measured at the driver board 10.2V
Current Measured between the driver and the LED 722mA. I removed the negative lead from the board and inserted another wire there to make the measurement.

I guess I really need to ask myself if the LED's are really K2's

are there any LED's that look like the ones in the image in my first post that are brighter than k2'S?
 
Last edited:
Great shots! Thanks.

Any chance you could read off what is on the 20 pin SOIC chip on the circuit board? Or take a close up of the text? This will tell us what the max current the driver could handle is.

I verified that these are indeed K2s and all operated in series.

I am guessing that the light output will only be ~500 lumens, no where near the 940-1000 lumens I keep on hearing about.
 
kd5ahl,

you rock!

Well, after a call into Diverite, they claim that the method in which they calculate the lumen output is proprietary! I can't believe that, it's an industry standard test.

I know for a fact they are using Luxeon K2s, so the maximum output they could ever get at full current (assuming a perfect optic) is 900 lumens at 1.5A of current. I am willing to bet that the actual light output is no where near that.

knowing that the LEDs are being driven by ~722mA (to get the run time they claim), the light output assuming an optical efficiency of 90% (really good for such small lenses) is surely closer to:

190 Lumens per LED* 3 LEDs * 0.9 Optical Efficiency =
513 Lumens!

aha! we have wayyyy overstated spec sheet.
 
kd5ahl said:
allrighty.

UPDATE:

Voltage measured at the driver board 10.2V
Current Measured between the driver and the LED 722mA. I removed the negative lead from the board and inserted another wire there to make the measurement.

I guess I really need to ask myself if the LED's are really K2's

are there any LED's that look like the ones in the image in my first post that are brighter than k2'S?
Click to expand...

I verified that they are FOR SURE K2s.
 
Hi,

Firstly thanks for all the info on the DiveRite LUX unit. I'm a complete noob to this LED and electronics business so it has helped quite a bit.

I'm wondering if there are parts that I can replace in the head that will produce more light from the head. Specifically I'm looking for a tighter more intense beam. I'd be happy for this to come at the sacrifice of burn time, say down to 4 hours minimum. I've read about the XPG LEDs and am thinking that a new MR11 size board and optic will do the trick, though pairing it with the right driver and calculating burn time are things I'm still looking into.

thanks in advance, damien
 
Dive Rite LED 700 mod?

I am in the process of upgrading our dive lockers collection of halogen and HID lights to LED. Among the collection of lights to be upgraded is a Dive Rite LED 700 that I would like to upgrade from to a single or possibly triple Cree XM-L. Unfortunately I can't figure how to access the back side of the light head. Has anyone here modded one of these or familiar with them enough to help me out a touch?
 
You must log in or register to reply here.

Similar threads

D
APLOS AP70C dive light
Replies
4
Views
3K
ABTOMAT
A
S
Advice for refurbing a dive light
Replies
1
Views
4K
donn_
D
lucca brassi
some new build at divelight forum ?
Replies
0
Views
2K
lucca brassi
lucca brassi
L
Piezo in canister lid / hex nut vs epoxy
Replies
2
Views
4K
DIWdiver
DIWdiver
K
Dive Rite Rx2 diagnosis - help needed
Replies
1
Views
7K
lucca brassi
lucca brassi

Latest posts

Top