XP-G R5 Tripple headlamp NOW WITH BEAM SHOTS

CampingLED

Enlightened
Joined
Nov 13, 2007
Messages
612
Location
South Africa
Have been thinking of this light for some time now. Finally some of the components are here and it is time to start sharing. A lot of work remaining though before it will be finished.

Decided to make this a modular design to make future upgrades easier than any of my other lights. Three SS bolts will be used to hold the parts together and no glue will be used. Optics or emitter swaps will only require the loosening of the bolts and some soldering. Heat conductive past will be used between all metal parts and the star to body coupling. The lens will press against the optics that will in turn keep pressure on the Triple R5 star. All other components will be added externally to make future upgrades easy.

The driver circuit will be similar to the ones that StefanFS uses for his P7 Mag mods (8 x 7135 multi-mode driver). I have enough 7135 driver boards, but still waiting for delivery of the multi-mode one that will piggy back to the other board. Therefore three mode with Max 2800mA. Three XP-Gs will be wired in parallel. Will either use one or two 18650 batteries.

Still need to work on the coupling between the head and the headband of this light as well as the other external components.

Here is a picture of the main components that I have right now:
XPG_HL_1.jpg


I asked a friend of mine to machine the aluminum parts for me before he left his company. He was so busy and left it with a colleague. Took a long time before he made it and he decided not to charge me. The machining is not as perfect as I wanted it, but with some extra work it will be fine.

Partial dry fit
XPG_HL_2.jpg


Dry fit
XPG_HL_3.jpg


If this works well I may just make a few more.
XPG_HL_4.jpg
 
Last edited:
Re: XP-G R5 Tripple headlamp IN THE MAKING

A test assembly of the electronics. One multi-mode driver (added an extra 7135 chip) wired to a multi 7135 chip board (was originally 3 x 7135, now 4 x 7135). The preferred programmed mode is L-M-H. Current draw on the battery side was 0.14A on Low, 0.7A on Med and 2.3A on High. On High the bare star heated up within 3 seconds. The "only" 2.3A on High could be due to a few reasons, Battery cannot give more current, fried one 7135 chip when it was soldered, internal resistance of my 20 year old Fluke 77 multi meter.

HL_Electr1.jpg


Ready for a fitment of the head components. Added some heat transferable gell to all metal parts where they made contact to each other (complete back disk, bottom of tripple star, between besel and centre protion).
HL_Dryfit1.jpg


Headlamp head unit fitted together. Used SS screws to assemble all parts together. Future optics and LED star changes will take a few minutes only.
HL_head1.jpg



HERE ARE SOME BEAM SHOTS.
Camera was set to 1 second, F5.6 & daylight white balance. Distance about 6-7m from plants. Reference light is my Petzl MYO XP headlamp modded with a SSC P4 U-bin LED.

MYO XP on high without diffuser (Spot mode)
MyoXP_H.jpg


New light on Low
XPG_L.jpg


New light on Medium
XPG_M.jpg


New light on High
XPG_H.jpg
 
Last edited:
Re: XP-G R5 Tripple headlamp IN THE MAKING

Getting ready to assemble the battery pack. Springs on + & -. Insulation tape and heat shrink around the parts. This will now be treated as a single battery. Will make a dummy battery with a female connector that will go into my charger to charge the pack.

HL_BatteryPack.jpg



The driver and switch box. Will be fitted above the battery pack. Still need to secure the power in and power out connectors to the box. The switch hapened to be a forward clicky and works just fine. My intention is to connect a male and female connector to banana plugs to measure current. If I connect it on the input/battery side I can measure current drawn. If I connect it to the output/light side I can measure current to the 3 x emitters. Will then get an idea of actual current to emitters and the efficiency.

HL_DriverBox.jpg



The main parts. Decided to take a chance on the DX headband. At max setting it feels like I need to extend 1cm more and my head is average size. This headband should be at least 5cm longer. It may stretch a bit over time and shoud be fine then. The battery pack is now complete.

HL_components.jpg



Another test with the units connected together. Lamp in low mode and on. Looking at the background the LEDs fooled my camera's white balance setting (sunlight used).

HL_connected.jpg



Getting ready to mod the head to connect to the headband and to connect the battery pack and driver box to the back of the headband. Almost there.
 
Last edited:
Re: XP-G R5 Tripple headlamp IN THE MAKING

Looks like a good start. Whats the plan for the driver? I can not see a spcae for it in the back if the big Alu block so presumably you will have to have it seperate? I had this with my small MCE HL mounted it to the headband on the right/back side of the head the 18650 went on the left back balancing things up a bit.

Mounting to the headband is the ticky bit. It would go very easaly onto Silvas big plastic headband with a couple of holes tapped. Onto the headbands I have been using recently I would wrap a piece of 10mm flat bar over the top probobly 270 degrees around then bend it down to have two straight bits going down to the headband bolt. I would have a second long bolt up high just below the bends down to tighten the circle at the top and hold it in place with no bolts to the main body. This has the advantage that you can have it wrapped around the middle rather than the back of the heatsink thus moving the weight back centraly over the main bolt to the headband thus helping avoid bounce issues.
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

The driver and switch will most likely be fitted close to the battery. I may even fit it just behind the head. :thinking:

I am also thinking of a hinged coupling to the head. The screws that will hold the main parts together will also hold the hinge. In the interim I may just fit it to a cap with all the parts in front.

The more I think about the solutions, the better the end product becomes.

Open to suggestions :candle: :whistle:
 
Re: XP-G R5 triple headlamp IN THE MAKING

I really like the looks of this project. :twothumbs Coincidently I've just started on a similar project.

Are you going to bevel the front of the housing so it doesn't cut off the light from the LEDs? What's the outside diameter of the housing? Do you have any beam shots? Got any plans for a waterproof battery case?

Heat conductive past will be used between all metal parts and the star to body coupling.
Paste? Makes a difference, eh. :naughty:

The "only" 2.3A on High could be due to a few reasons, Battery cannot give more current, fried one 7135 chip when it was soldered, internal resistance of my 20 year old Fluke 77 multi meter.
Linear regulator, so it wouldn't be just your meter resistance. Anyway, 2.3 A for seven AMC7135s sounds about right to me; just under 330 mA each. There are different models of AMC7135 that can go as high as 380 mA.
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

Looking good. Interested to see the final product and how you end up mounting it.

I finished my light weight double 3 weeks back and am very pleased with it. I am not using a 7135 based driver but a homemade linear regulator. with a single 18650 cell (blue Trustfire 2500 from DX) I don't stay in regulation very long at 2000ma. 1500ma stays well in regulation for 90% of the battery capacity. With the single 14500 cell I am only well regulated up to 1000ma. I don't know what the variability in internal resistance is but picking good cells clearly makes quite a big differance.
 
Re: XP-G R5 triple headlamp IN THE MAKING

I really like the looks of this project. :twothumbs Coincidently I've just started on a similar project.

Are you going to bevel the front of the housing so it doesn't cut off the light from the LEDs? What's the outside diameter of the housing? Do you have any beam shots? Got any plans for a waterproof battery case?

Will compare the beam pattern tonight now that it is in the housing. My feeling is that it may not lose too much light in one direction (the other two LEDs will most likely make up in that direction. Beam shots will follow soon. This thing is amasing (when tested for a few seconds at a time without the housing). Totally smokes my modded Petzl Myo XP (will use as comparison headlamp in beam shots). Raw material started as a 35mm solid aluminum bar (used parts of it before in Mag mods).

The battery case will most likely contain two 18650s in parallel. May not be waterproof, but definately rain/splash proof. Still open to suggestions on the rest of the light. Also looking for a nice switch. May use an on-off switch in series with a n/c push button type switch (to change modes). I plan to use connectors between batteries and driver/switch assembly and between driver and light. This will make future mods very easy.


Paste? Makes a difference, eh. :naughty:

:oops: Is that what they call it. This also helps for waterproofing between the aluminum parts.

Linear regulator, so it wouldn't be just your meter resistance. Anyway, 2.3 A for seven AMC7135s sounds about right to me; just under 330 mA each. There are different models of AMC7135 that can go as high as 380 mA.

If the Vf of the LEDs are similar I most likely get about 770mA per LED. Output is great and I do not want more heat generated for the slightly higher output. What I want to say is that I can definately live with the 2.3A and do not need more. This light will most likely be used in Low and Med mode 95% of the time. The other 5% in High mode will be mostly to boost my ego and show-off. :wow::lolsign: That makes me think, high mode will initally be used 80% of the time.
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

Looking good. Interested to see the final product and how you end up mounting it.

I have some flat aluminum and friends who can weld it for me. This may form the basis for the mount. My main thoughts are around making it swivel with a nice clicking feeling.

I finished my light weight double 3 weeks back and am very pleased with it. I am not using a 7135 based driver but a homemade linear regulator. with a single 18650 cell (blue Trustfire 2500 from DX) I don't stay in regulation very long at 2000ma. 1500ma stays well in regulation for 90% of the battery capacity. With the single 14500 cell I am only well regulated up to 1000ma. I don't know what the variability in internal resistance is but picking good cells clearly makes quite a big differance.

Great light you have there. :thumbsup: I hope the 2x batteries that I plan to use keeps it in regulation for longer in High mode. This is the first time that I use the DX 7612 driver (lost one in the post before) and I am about to order more. Like the L-M-H mode without strobes. Soldering these tiny things with my 2mm soldering tip is a challenge though (wonder when the 1mm tip will arive that I ordered two months ago).
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

I have some flat aluminum and friends who can weld it for me. This may form the basis for the mount. My main thoughts are around making it swivel with a nice clicking feeling.
Welded Aluminium would be nice. I bought some stuff that was meant to let you do it but gave up trying to get it to work. Don't think I was getting enough heat out of my blowtorch. Just stuck with the glue, dose not look good but works OK and I can repare as nessasary....


Great light you have there. :thumbsup: I hope the 2x batteries that I plan to use keeps it in regulation for longer in High mode. This is the first time that I use the DX 7612 driver (lost one in the post before) and I am about to order more. Like the L-M-H mode without strobes. Soldering these tiny things with my 2mm soldering tip is a challenge though (wonder when the 1mm tip will arive that I ordered two months ago).
Yes that small soldering stuff is hard I often find it easier to flood things so I end up with everything shorted then you can use solder wick to clean up the excess and things work out OK.

Here is the discharge graph for various curents with 1 and 2 18650 cells. Next time I get batteries I am going to get some lipoly ones and see how they perform with the far lower internal resistance but lower basic voltage as well....
mce_hl_runtime_2.jpg
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

Thank you for the graph. It seems that in all cases the 2x batteries stay at the same output about twice as long (makes sense). Pitty that at the higher currents both seem to drop very quickly. The next question will be at what voltage the drop starts to happen (Vf of LEDs at that current + inefficiency Vdrop of driver)? Nice thing about the XP-Gs are their low Vf.

Just looked at the friction mechanism of my windows and I think I should use a piece of plastic bolted between two metal strips for the swivel part of the mount. The light can swivel at the bottom and at the top the friction plates can be mounted. One to the light and one to the bracket with plastic in between.:thinking:

The simplest way will be a hole from side to side through the base of the light (it is 10mm thick and there is enough space). I can then bolt a U-shaped aluminum bracket to it with small fibre washers in between. Tightness of the bolt determines the force required to swivel. :thinking:
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

Yes with a linear driver the driver, the actual vf for the leds the internal resistance of your cells and any wiring loses are all critical for good regulation at higher curents. You can not be sure till you have built a few how well your setup will work... Idealy you would get multiple xpgs, and cells and test them all then pick the best as there is definatly enough variation to make a quite a big differance....

The simplest way will be a hole from side to side through the base of the light (it is 10mm thick and there is enough space). I can then bolt a U-shaped aluminum bracket to it with small fibre washers in between. Tightness of the bolt determines the force required to swivel. :thinking:

That sounds similar to what I have. I am mounting to a cheap lights strap, there is a long bolt about 20mm through two plastic bits molded to the plastic which gose against the forehead. One of these plastic struts is shaped to hold the nut for the bolt secure. My alu downstruts go inside the plastic bits. There is a seperate plastic spacer between my alu downstruts, this used to give you the clicky action onto the original plastic lighthead. I don't have the clicky action but with suitable washers I just tighten it all up together and it holds nice and firm but can still be adjusted. I have had no problems so far but the real test will be after more extended use.

P1050319.JPG
 
Re: XP-G R5 Tripple headlamp IN THE MAKING

Thank you for sharing ifor powell.

I have taken a few beam shots. Added to post #2. I think my previous headlamp has just become obsolete. :naughty:
 
Re: XP-G R5 triple headlamp IN THE MAKING

Those beam shots are great! I can imagine you're pretty happy with it. The beam profile looks nice and smooth.

Thanks for the link to your headlamp, IP.
 
This is a very rewarding project and I am very pleased with the results. Need to sort out the rest of the light soon.

Also waiting for some specific heat sinks to arive to complete this one.
The emitters are currently glued to the 2mm thick copper plate. These optics should give a tighter beam and brighter hot spot. Also 3 x XP-G R5 emitters.
TrippleXPG2.jpg
 
Those optics are tighter than the little ones I have used in the 2 up light. I have used 2 of them in my curent bike helmet light with xpe's, I think they will be a good bit wider with XPGs though.

If you want a tighter spot from the xpg you need somthing big. The ledil Iris is the best I have seen by far see here.
 
Been lurking around in the MTBR forum before (not a member there), but missed that post. I agree, the Iris is fantastic. From the datasheet it should be ±5.5° (i.e. 11 deg) with MCE. Should be even less for the XP-G then. Unfortunately the size is still too big for something like a P60 drop-in. The optics designers have a very tuff job at the moment. With the smaller LED footprints the users need smaller optics, but smaller is less efficient for the optics.
 
This year started very busy and I did not get enough time to make progress as I wanted to.

Updated post 3 with some progress pics. Almost there. :candle:
 
Top