• You must be a Supporting Member to participate in the Candle Power Forums Marketplace.

    You can become a Supporting Member.

Sold/Expired FS: empty LED tower for Surefire M3T/M4/M6/KT1/KT2/KT4 head(accept 17mm driver)

I saw that in your other thread...awesome :thumbsup:

Steve Ku limited the drive level for his towers to 1A in the later runs (the first run had either 1.2 or 1.5 as an option I think)...I presume that was due to thermal limitations. But his were brass; copper might do a little better. And using a 14mm board would allow for a bit more mass,as well as the (probably) more efficient driver.

As far as widening the reflector opening...I'm sure it's possible, but much like the turbohead tower, I'd prefer to keep the original parts stock myself. If you need to bore out the head, I think you end up competing with Steve's current project, which is a complete replacement head for the e-series in XP-G and XM-L flavors. But I'm sure there are still a lot of people (like me) who want a drop-in tower option but weren't able to get it from Steve before he moved on to the complete replacement head. I was (and still am) in the market for between 6-10 to use with the high CRI XP-G, myself.

edited to add: taking a closer look at one of my steve ku modules, the XP-G is actually mounted on a MCPCB...but the PCB has been ground down so that it's just a touch wider than the LED itself. Really, all that's left are solder pads, and those don't seem to be much larger than the pads on the LED. Not having the skill to assemble one of these in any fashion anyway, the relative merits of taking the time to grind down the PCB (and what I think must be obligate degradation of the thermal path) compared to attaching the LED directly to the post and soldering to the tiny tabs on the LED package, are beyond me.
 
Last edited:
I really do appreciate your input. An e series tower really caught me by surprise, because until just recently all my work was on the 1" tube diameter lights.

I know that some development work has been done on the head for the e series. My collection includes a McG PRT turbo head that is still impressive today. It seems like most users do not want a head that large.
 
There are at least two Steve Ku products that can work on E-series lights. The first is probably the most straightforward -- the Veleno Designs drop-in for the stock SF E-series head. The opening is about 1/4" diam (or about 6.4mm). The diagonal for a Cree XP-G on its substrate board is about 4.9mm, so there is some "reasonable" amount of leeway. You still have to grind down an 8mm or 10mm MCPCB but you'll have some amount of solder pad left that shouldn't be overly difficult to solder hookup wires. The Veleno Designs E-series tower uses a very thin gauge hookup wire with a very thin insulation. However, by eye, it looks like one may be able to use the fairly common 26 gauge Teflon wire. And if that's too big, 28 gauge looks almost certain to fit.

Another issue with the E-series tower is that there are apparently two styles of SF E-series heads and they use a different profile at the bottom face of the reflector opening. The old, original E heads have a completely flat surface inside of the head. The newer E heads with the hexagonal flats have a raised center hollow metal cylinder. A tower made with the same geometry as an incan MN02 lamp would work in both types of E heads (but with less contact area to the new style head), but a tower machined the way the Veleno Designs towers are made will fit only in the new-style heads.

The other tower is the M3 tower. This one is more challenging because the reflector opening is only about 0.2" diam (5.1mm). Now you have to grind down the MCPCB to basically the same size as the XP-G's footprint. That leaves very little solder pad left to connect the hookup wires. Also, you need an E2M adapter to attach the Z46 head to the E-series body.
 
what I'm interested in is the towers for the e-series heads. I have one of each style of steve's towers (for the teardrop & new heads), but personally only have the new style heads. I have the teardrop-designed tower in a new head and it seems to work fine, but I don't know how much thermal management suffers.

I guess it would make sense to use the one design that fits both (the one for the teardrop head), so long as thermal management is sufficient for the XP-G in either head. I'm also perfectly satisfied with the 1A drive current limitation, myself. But if designing for the new head meant an XP-G could be safely driven at 1.5A, that might sell better overall (even though you lose the teardrop head market).

I'm willing to loan a new style head and both towers for measurements, whenever you're ready.
 
Last edited:
There are probably a lot more E series lights out there, but there are/will be at least a couple of LED replacement heads available.

Veleno Designs has been selling E series towers fairly steadily for quite a while, though is now shifting to complete heads, motivated, at least in part, by trying to optimize results with modern LEDs without the compromises introduced by using heads designed for incandescent lamp assemblies. The towers show up fairly often in the Marketplace. Lumens Factory also makes LED heads.

LED towers for Surefire M series Turbo heads have been available sporadically, or on a custom basis, from a variety of makers. The towers typically don't compete with the highest output incandescent Surefires, but offer better runtime and often good throw. They infrequently show up in the marketplace.

Given probable pricing, M series owners might be more likely to buy a premium tower than E series owners (highly speculative). Surefire's discontinuation of most incandescent lights might be added motivation for experimentation with LED towers, especially if a highly reliable (moderately) high-output version can be brought to market for an attractive price, with tint options that will appeal to incandescent users making the shift to LED.

Die-hard incandescent M series users can switch to bi-pin lamps, with various options for powering.

So... for my needs, the M series towers would address an under-served market, and would thus be of greater interest.


- Syncytial.
 
So... for my needs, the M series towers would address an under-served market, and would thus be of greater interest.


- Syncytial.

I agree, the M Series towers are more 'necessary'

I think the Veleno sells so well because it is a drop-in. If only kits are available sales may not be large, although I'm certain someone would assemble a run or two. But it's tiny work that most people wouldn't attempt, just my conjecture -
 
totally agree, M-series turbo towers should come first. I'd just like to see the e-series towers available eventually. I never meant to supersede the M-series towers, just wanted to plant the idea on the e-series 😉
I want both :thumbsup:

coordinating with someone interested in doing assembly work might prove lucrative for both projects in the long run; one name in particular does come to mind.
 
Last edited:
Certainly, if he's interested. I didn't get the impression he was looking to take on anything so time consuming, though. Justin, what say you?

he obviously knows this type of project inside out. There are quite a few modders on this forum who do the machining and the electronics but it is all I can do to keep up with the machining.
 
still one up on me...best I can do is go out and earn enough to pay you for your expertise. The few projects I've tried my hand at....turned out not so well 😀
 
The few projects I've tried my hand at....turned out not so well 😀
Don't feel like you're alone, I still have the occasional fubar. It always occurs after spending hours on a project, when there are only one or two final operations needed to complete the job, and my attention lapses for just a second ...
😳

During the times that I'm working in my shop, the door is locked, my ears are covered by the best hearing protectors available (David Clark Model 27 with the optional gel filled ear seals) & the phone isn't answered unless the caller ID indicates an existing customer is calling. As long as distractions are minimized, getting from Point A to Point B usually happens without incident. Lots of CPF members don't have the luxury of locking themselves away in a temperature controlled Man Cave, making it many times more difficult to take a project from start to finish.
 
Certainly, if he's interested. I didn't get the impression he was looking to take on anything so time consuming, though. Justin, what say you?

I probably can take it on. I have a day job, but I expect towers to be moderately low volume and hopefully not excessively spiky in terms of demand.

What do you think is a fair price for the service? I see at least three possibilities in terms of what may need to be done:

1. Assemble all of the parts for the tower, no need for any further soldering of the driver itself (remains stock from The Shoppe). Work may involve reflowing the XM-L (or XP-G) to the 8mm diam board and soldering the hookup wires. An example of this case may be if SOB1500s can be ordered directly from the Shoppe. I don't know what Wayne charges for this since it isn't listed as a regular item. It may be cheaper to order blank SOBs and solder the sense resistors yourself, but those resistors are very tiny. They are 0603 resistors. The 0603 size designation means roughly 0.06" long x 0.03" wide (about 1.6mm x 0.8mm).

2. Also have to solder sense resistors to the SOB (as mentioned above) to get drive currents higher than 1A (SOB1000). Or I might have to de-solder and re-solder sense resistors if I have to mod existing SOB1000s, for example, to bump them up to SOB1500s. I'd have to check with Wayne as to what sense resistor combo that the BBNG and GD can take. Can you go to a max of BBNG1500 and GD1500?

From the datasheets for the ICs used for the BBNG and the GD, it looks like both ICs can deliver much higher drive current than 1A. So it probably depends on the ratings for the other components that Wayne uses on the boards, and also whether or not the combination of Vbatt, Vf, If, and driver efficiency results in Ibatt exceeding the switch current limit of the driver's IC (assuming the cells can deliver that level of Ibatt).

3. Tower requires also building a two-board parallel driver sandwich (and potentially also having to install sense resistors if folks want 2xSOB1500, for example). This clearly involves the most work.
 
Last edited:
when I was ordering drivers for Barry to play with, it *looked* like you could have whatever resistors you wanted soldered on for an additional 3-5 dollars total cost ($3 for the soldering service plus whatever odd cents for the resistors). SOB1500 was definitely an available configuration with this service, as well as SOB1200, SOB1333, etc. I just decided that I personally would stick with the 1A drive current for known reliability and minimal heat issues, as well as better runtime. After all, I'm not designing something to compete with the MN21 :devil:

Since I don't know how long assembly might take you, and don't have the skills to even try to do it myself, I wouldn't hazard a guess at what a reasonable fee might be. I just know I'm gonna pay it when the time comes 😀
 
Last edited:
when I was ordering drivers for Barry to play with, it *looked* like you could have whatever resistors you wanted soldered on for an additional 3-5 dollars total cost ($3 for the soldering service plus whatever odd cents for the resistors). SOB1500 was definitely an available configuration with this service, as well as SOB1200, SOB1333, etc. I just decided that I personally would stick with the 1A drive current for known reliability and minimal heat issues, as well as better runtime. After all, I'm not designing something to compete with the MN21 :devil:

Since I don't know how long assembly might take you, and don't have the skills to even try to do it myself, I wouldn't hazard a guess at what a reasonable fee might be. I just know I'm gonna pay it when the time comes 😀
if you can figure out what each minute of your time is worth and then assemble a tower or two, that should tell you about what you need to charge. Depending on whether you have 1 driver board or 2, cost may vary. The community here on CPF is more than willing to pay a fair price for high quality service.
 
when I was ordering drivers for Barry to play with, it *looked* like you could have whatever resistors you wanted soldered on for an additional 3-5 dollars total cost ($3 for the soldering service plus whatever odd cents for the resistors). SOB1500 was definitely an available configuration with this service, as well as SOB1200, SOB1333, etc. I just decided that I personally would stick with the 1A drive current for known reliability and minimal heat issues, as well as better runtime. After all, I'm not designing something to compete with the MN21 :devil:

Since I don't know how long assembly might take you, and don't have the skills to even try to do it myself, I wouldn't hazard a guess at what a reasonable fee might be. I just know I'm gonna pay it when the time comes 😀

Yes, you are right. I had forgotten that the SOB Blank Converter offers quite a good range of sense resistor choices. Definitely, you can order configurations such as SOB1185, SOB1227, SOB1278, SOB1333, SOB1416, and SOB1500.
 
Barry, one each of BBNG1000, GD1000 and SOB1000 should be on their way to you 😀

Got a small package in the mail from Wayne, and my first thought was that I didn't order anything from Wayne ... But you did 🙂

Thank you! If the emergency jobs ever stop, and maybe they will when cooler weather comes, this is high on the list. It is hard to keep up when an existing customer, and a really nice person also, brings a rush job into the shop at noon & wants it done by 4:00 ...
 
Back
Top