Looking good! Can’t wait to see the final
Great job on that pill! 
Looks like it has a lot of mass.
1 Thank
Absolutely!
151 g just for the pill alone. Just for comparison - a classic aluminum D4V2 weighs 115 g, with battery!
This light is just a massive metal ingot. 
714 grams without batteries and LEDs, boards, driver plus lenses! I think, it will reach 850 g when completely assembled. Absolutely insane.
7 Thanks
Something is happened since the last update, so here we go.
The LED boards and lenses have arrived. The LEDs also arrived just in time.
Do you remember the “tool” that I had to specially design?
This was used to prepare the lenses and MCPCB (mark the dimensions) to ensure the highest possible precision when fitting the dimensions.
Here are the prepared lenses:
And the boards:
After some manual work, the boards looked like this.
They fit just fine.
The lenses were also modified.
The most important part is the design of a special frame that holds the LEDs and optics securely in place and also improves the optics in the lamp head.
The first test assembly ever! Doesn’t it already look great?
There will be more later. I hope to be able to finish the lamp next week. See you later!
12 Thanks
Ooooooh! Nicely done! Very good fitment. It’s getting exciting.
3 Thanks
Thank you!
Absolutely, I hope to get first tests with the LEDs on the boards done by tomorrow. (this is important since I have to check how high the max current can be without the risk of overheating, since stainless steel has a low thermal conductivity)
I also have an idea on how to connect the LED boards securely together without the need of using seperate cables.
2 Thanks
Absolutely amazing work. So many handy and talented people on this forum.
1 Thank
This looks great. I’ve been following the project since you started it, and I really love how it’s proceeding. Looking forward to the first beamshots! Awesome project 
1 Thank
I hope that I am able to buy a better drill in the hardware store (if it is still available there) and I have to redesign the frame for contacting the boards together, so maybe I even be able to give you some beamshots today.
2 Thanks
Heck ya man. I love the way this build is going! I had no idea Lenser ever made light with a fistful of 5mm LEDs.
2 Thanks
No picture yet, but 7000 lumens full Power confirmed!
4 Thanks
Designing a battery cage:
Due to the intentionally smaller diameter at one end it cannot be inserted in the wrong direction.
And here a first beamshot of the Ledlenser. It’s alive!
6 Thanks
That looks awesome! Nice job
1 Thank
I print a first prototype of the battery cage, mainly for testing some things out.
The batteries fit nicely, and it is very sturdy. I also already build one PCB for connecting the cage, also for testing purposes (especially with regard to the manufacturing process and final assembly).
Since I discovered that my black PETG filament is just waste (even drying it several hours does not prevent jamming of the nozzle) I printed it in white first. I ordered new black filament, but it won’t be delivered until Monday or Tuesday, so finishing the project will be delayed…
I also ordered some additional stuff to get the cage properly done. This is the reason, why testing newly built stuff in small iterations (if necessary) is always better. This is product development. My experience in this area must be worthwhile for something
3 Thanks
Agreed, it’s looking awesome!
Nice work on filing down the MCPCBs and optics.
1 Thank
A lot has happened since the last post. The Ledlenser V2 mod is finished and works as planned. (Well, almost anyway)
There have also been setbacks. And chaos. The mod was not easy.
To be honest, it was by far the most time-consuming and difficult mod I’ve ever done. Actually, looking back, it’s not even a mod anymore, but a complete rebuild of a lamp, where basically only the host and pill were left…
I will publish the construction progress in several separate posts in the coming days, as these are very extensive…
Enough preamble, here we go. In this case: with the LED board or LED array.
The LED boards were already prepared. For easier mounting, a mounting frame was printed based on the already designed frame.
The pill has already been prepared for the power supply. A suitable hole was drilled for the power supply, as well as holes for easy screwing in of the pill with tweezers/lens wrench.
Finally a reason to get a (cheap) drill press. It doesn’t wobble around that much. The cheap Lidl metal drills do that
The cheap drills were still precise enough…
Now the reflowing could begin.
And now the time has come to reveal the secret of the LEDs used: 9x Luminus SFT-25R G1 BC from Kaidomain are used. These are specified with 6500 K and the highest flux bin, which was also confirmed in a quality assurance test carried out in advance.
So, here we go:
And done:
The lenses fit perfectly on the boards and the frame:
To connect the LED boards together and later connect them to the driver, I thought back and forth about the best way to do this. At first I thought about connecting all the boards separately to the driver, but that would have been a nightmare with the mess of connecting wires, apart from the fact that you had to be able to screw the pill shut somehow.
So in the end I switched to the method with rigid copper strips and just one connection to the driver:
Another advantage is that the LED boards are rigidly connected to each other, which greatly simplifies subsequent handling.
The LED board array work directly:
With the lenses fitted, the lens pattern looks like this.
It looks somehow futuristic and unusual. So absolutely perfect for this lamp.
I’ll show you how build the battery cage and the driver in a post in the next few days.
15 Thanks
I like the copper connecting strips; buss bars I call them.
5 Thanks
Now we continue with the construction documentation.
I had already shown pictures where I had made a test print of the battery cage and tested a few things on it. That this was not planned and almost forced me to abandon the whole project… that will be described later…
Based on the final design, I printed the cage with new black filament (which miraculously worked straight away without any change in temperature).
Connecting the battery cage required a few parts, including small brass pills:
They were joined by two more pills and the PCBs.
I really need a new 3D printer. The printing took forever (over 7 hours!!!) and in the end the result was rather so… meh. Well, with a lot of rework…
… the result was at least presentable. Here with PCB and brass pill already installed.
With the four PCBs and springs:
Thin copper wire (rigid) was used to connect the positive and negative sides of both batteries.
Here with color-matching shrink tubing, for the optics.
Finished!
The battery cage holds two 21700s in 2P (parallel) connection. The lamp delivers the same luminous flux with one battery in both slots.
Samsung INR21700-50S are used. (Here in the Keeppower Wrap because battery parts had no more raw cells in stock - but the batteries fit optically: all black!)
They fit 21700 with a length of about 72 mm, so that the Vapcell F60 also fit. The 50S are a bit longer than the 40T, for example.
A copper washer was machined and fitted with a spring to connect the negative pole (directly to the lamp housing).
The washer holds perfectly thanks to the precise press-fit.
5 Thanks
Now comes the report on the most annoying part of the whole conversion… the driver. (And the assembly, but more on that later).
I didn’t take many pictures during this phase of the build. I simply didn’t think to do so, especially since I was under time pressure anyway due to an upcoming vacation and a subsequent forum meeting in Cologne - where I wanted to present the lamp - and simply spent as much time as possible on the project with the fastest possible construction progress.
My plan was to use a Convoy 20.5 A buck driver with 6-8.4 V input voltage. The main advantage was the simple construction, because the batteries could simply be plugged into the lamp (only with button top, the only disadvantage), and that the regulation is better. If only I had informed myself more thoroughly…
So I set to work. I had all the parts to hand, I thought, so what could go wrong when assembling the lamp?
I installed a small copper plate with a spring on it in the switch box to ensure contact with the positive pole. To do this, I ran the AWG18 wire under the switch.
The wires on the switch were also connected.After sealing, only the three wires were visible.
I considered various methods for connecting the negative pole to the driver. My first plan was to screw or solder a wire to the pill.However, due to the lack of a small M3 tap, I decided against this. I also rejected the method of ensuring a connection to the housing via the switch box. In the end, I decided to make contact with the housing on both sides using extended copper fins.
This also worked well. Unfortunately, a piece of the inductance package was broken off, although the driver should still work. A general note on these drivers: always handle these with care! The housings of the inductors are very fragile. In my case, even getting stuck with a fingernail was enough to break off another piece. It looked as if these housings were made of some kind of epoxy mixed with iron filings, which would explain the brittle material.
The rest of the assembly was then simple and corresponds to what we know from triple conversions of Convoy lamps. Insert the front glass and O-ring, solder the wires for the connection to the board, screw in the pill with the frame, the board and the lenses.
Before screwing the lamp head shut (which takes four turns), I also twisted the connecting cables from the LED board four turns in the opposite direction to the direction of rotation so that they would balance each other out.
And then, hopefully connected to the lab power supply and pressed the switch…
The lamp did not work. Well - well, it did work…
I was also able to switch through the light levels and activate the special functions (strobe etc.).
The heat development of the lamp head was pleasantly bearable, only after about 45 seconds did the head start to warm up noticeably.
But I couldn’t switch the lamp off. At first I thought it was a faulty assembly or a problem with the driver itself. But a check revealed nothing conspicuous. When I then researched the driver to check the UI, I realized that I had made an absolutely stupid rookie mistake…
The driver needs a mechanical rear switch. You can’t turn it off with the side switch. Like with the TN12 or the EC35 (to be fair, the article description on the Convoylight site didn’t mention anything about this - you can only find this information if you check the description of the Convoy L7, for which the driver is actually intended).
I don’t like this lamp format with side and rear switches anyway, because it still doesn’t make sense to me. And now I was beginning to really hate this operating concept… why not just make it possible to use both side and rear switches optionally, depending on preference, adjustable if you like? No, just not. Everything has to be predetermined and unchangeable…
What did this setback mean?
Well, starting from (almost) the beginning.
I thought about various ways of getting around the problem. The first idea was to simulate a kind of tail switch using a MOSFET, which would be switched by a small converted 105c driver, which in turn would be controlled by the actual side switch. However, this would have meant a complete reprogramming of the firmware, as I would have needed a two-channel output, and it would not have been safe to place such a PCB with a powerful MOSFET on it in the lamp head or switch box.
Another idea would have been to build a tailcap. I could theoretically have printed it, but how ugly a beautiful stainless steel lamp with a cheap plastic tailcap would look…
So I decided to change the concept completely. The new plan called for a classic FET-based driver for a 3 V power supply. This also meant building a battery cage, as I really wanted to keep using two 21700s to avoid wasting space and potential capacity in the battery tube. In the end, this is the reason why I had to design and build a battery cage as I described in the previous post.
The new driver with 25 A maximum current had a different outer diameter (32 mm). This required a redesign of the driver adapter in the lamp head, but fortunately it was done very quickly.
The connection was made in exactly the same way as with the previously used driver, with copper fins on the driver.
I also wanted to change the way the LED board was connected to avoid twisting and possibly destroying the connection cables.
In the end I decided to use sliding contacts to connect the LEDs. I’m not normally a fan of this, simply because of the contact resistance, especially at higher currents. However, since the lamp is not constantly used at the highest level, this should be ok.
The sliding contacts are mounted on a specially manufactured PCB with an insulating washer between them, as the PCB is plated through. The sliding contacts are connected to the board with thin copper wires.
A first test of the assembly was successful. I then carried out various tests with regard to temperature behavior, light levels, LED tint, etc.
At first I was surprised that the lamp occasionally switched on at the highest level after assembly with the sliding contacts. This did not always occur, mostly after slight side impacts. It was then no longer possible to change modes on the light. This was due to a short circuit between the pill and LED minus, which put the LEDs into direct drive mode. You can’t see it well in the picture, but due to the limited space in the area of the copper wires, the edge of the cut-out in the pill cut into the shrink tubing, which was enough to cause this issue:
With an adapted design of the insulating washer, which also protects the area of the hole in the pill with a sleeve, there was then peace of mind:
An official presentation of the lamp will follow today, and with that the project is officially completed!
9 Thanks
Wow! A relatively simple project became much more complicated and time consuming. It does appear though, that it was worthwhile. Icm glad to see you persevered.