MTN Electronics: LEDs - Batteries - Lights - Chargers - Hosts - Drivers - Components - 1-Stop-US Source

[RMM]

@will34

I’ve been chasing these 9mm round XP centering rings for a while but I don’t feel like placing an order on Simon’s store just for them. Would be nice to see them for sale at MTN

I have a bunch of them, but they are not very good; the spacers are too thick and are often fairly loose on the LED. For a non-generic spacer I expected more. I will list them with that note.

[Tom_E]

@RMM

10mm is too small for the tool; you would want the 12mm driver in 0.8mm thickness. I have some 0.8mm 2 oz. copper, I will get them listed too all in one go. ....

All sounds good, Thanks! I'll wait til you get it listed, maybe the 25 version. If you fit the regular 25 w/pins, I got plenty of those and some nice custom firmware for it. I like the luxdrv type of short cycle memory, and got a pretty nice version for the 25 w/blinking out the batt voltage, etc.

I just ordered the XPL spacers from BG here earlier this morn. Maybe for the height they can be filed down, but the sloppiness sounds like a problem.

[RMM]

@wight

@RMM

I would recommend cutting the maximum PWM duty cycle even lower than half. I tried going down to PWM=100 (~40% duty cycle) even, and the customers who had them had the 7135s die after a while. I would only use the 7135 for moonlight and maybe very low duty cycles of 10% or less on 2S lights.

RMM, does this apply exclusively to FET+1 / 7+1 configs or have you also seen this on your Zener-modded qlites? IIRC you sold quite a few of those in the past, so I’m very interested in what you’ve seen.

Yes, the zener mod qlites will also overheat badly if you run only 8x7135 chips and good cells (high drain 18650; 26650), but they do fine at 3A with 18350s. With 18650s I would always recommend running at least 12x7135, and 26650 16x7135 if you are planning on running it on high for long. Generally, the chips are not damaged, but the efficiency and output really stinks if you have too much voltage overhead.

Remember, the 7135 are linear voltage regulators, so your power dissipated increases drastically as the difference between vIN and the vF of your emitter @ the set current increases. As you increase the chip count, you not only spread the heat, but you effectively lower the differential between vIN and vF. The batteries will sag more, and vF is higher at a higher current. This is also why smaller batteries can run more efficiently at lower currents (they sag more under load).

With a single 7135, the vIN + vF delta is is very big for a 2S emitter, and there is only a single 7135 to take the heat. It seems like even the 7135's internal throttling is not enough to save the chip from damage in this scenario.

[RMM]

@Tom E

Thanks! I'll wait til you get it listed, maybe the 25 version. If you fit the regular 25 w/pins, I got plenty of those and some nice custom firmware for it. I like the luxdrv type of short cycle memory, and got a pretty nice version for the 25 w/blinking out the batt

Lol, are you kidding? That won't come close to fitting! We're talking about the MU version, which is still significantly larger than the MMU.

[wight]

@RMM

Yes, the zener mod qlites will also overheat badly if you run only 8x7135 chips and good cells (high drain 18650; 26650), but they do fine at 3A with 18350s. With 18650s I would always recommend running at least 12x7135, and 26650 16x7135 if you are planning on running it on high for long. Generally, the chips are not damaged, but the efficiency and output really stinks if you have too much voltage overhead.
[snip]
With a single 7135, the vIN + vF delta is is very big for a 2S emitter, and there is only a single 7135 to take the heat. It seems like even the 7135’s internal throttling is not enough to save the chip from damage in this scenario.

Fair enough, and in fact I think I recall you making that specific recommendation (going from 8x to 12x or more for MT-G2 applications).

I’m a little surprised that 1x7135’s internal protection couldn’t quite handle it, but there must be a limit to everything. Thanks for the info.

[Halo]

@RMM

I have a attiny25 version as well, but it still needs some tweaks. It runs OK, but really it needs a 7135 or similar too, so that’s also what I’m working on (hard to make that all fit!)

When using the tiny25 you could gain some room by losing the two voltage divider resistors. Attiny25/45/85 can measure battery voltage without them. See my post here: Flashlight Firmware Repository - #790 by Halo

[RMM]

@Halo...

@RMM

I have a attiny25 version as well, but it still needs some tweaks. It runs OK, but really it needs a 7135 or similar too, so that’s also what I’m working on (hard to make that all fit!)

When using the tiny25 you could gain some room by losing the two voltage divider resistors. Attiny25/45/85 can measure battery voltage without them. See my post here: Flashlight Firmware Repository - #790 by Halo

Yes, but it isn't as precise, and ToyKeeper would be very upset with me if I reduced the voltage readout accuracy!!!

I can make it fit. 0402 aren't that small.

[Halo]

ToyKeeper commented that there should still be enough accuracy since we have resolution to spare currently.

[Tom_E]

@RMM

@Tom E

Thanks! I'll wait til you get it listed, maybe the 25 version. If you fit the regular 25 w/pins, I got plenty of those and some nice custom firmware for it. I like the luxdrv type of short cycle memory, and got a pretty nice version for the 25 w/blinking out the batt

Lol, are you kidding? That won't come close to fitting! We're talking about the MU version, which is still significantly larger than the MMU.

I was think'n (maybe dreaming?) the 12mm board would have all this extra space over a 10mm board, hhhmmm, well a little more?

[wight]

@Tom E

I was think’n (maybe dreaming?) the 12mm board would have all this extra space over a 10mm board, hhhmmm, well a little more?

On anything smaller than a 15mm board an SO8 MCU isn’t very practical for single-sided applications. Take a look at my 10mm/12mm/15mm “DD-L” drivers for comparison if you like.

[LSX]

Well this was good timing for me to read this thread. I just pulled apart my Cu Tool to begin modding it with a 12mm FET driver. But with a purpose built one being released I will save this driver for something else.

Richard, this has probably been answered plenty of times before, but can an XHP35 be run on a FET driver from a 3S carrier?

[DB_Custom (Dale)]

The LD-2 can be modified to run the XHP-35 from 4 cells, 3 might not be enough overhead for the 12V emitter.

[WillyD]

Ok, I feel like this has to have already been asked, but I did a ctrl-F for ’S2’ and went several pages back. So, is there any estimate on when you might be getting any red or blue hosts along with the shortened tubes I have a pretty large order to place and wanted to do it all in a single order. If it’s going to be a while on the S2 hosts I’ll just place the order for everything else. And, thank you.

[jhalb]

@DB Custom

The LD-2 can be modified to run the XHP-35 from 4 cells, 3 might not be enough overhead for the 12V emitter.

How would you wire this? Series? Parallel?

[thijsco19]

The cells has to be in series.
The ld-2 is not a boost driver.

[99-sts]

@RMM

I am going to try my hand at selling a few Lumintop Tool AAA lights. I got a few of these a few months ago, and I really like them. Even in stock form, they are very nice, but they are also pretty sweet modified. I have one with an XP-L HI + FET driver and another with a Nichia 219C + FET driver. I have a driver that has been specifically designed for these lights, but I won’t be able to build any until next week.

Tool Promotion

I am, however, going to run a promotion on these lights: everyone who buys one of the first 20 lights will automatically be entered into a random drawing to receive a modified Tool and battery (Nichia 219C, FET driver, and a 10440 Efest). If you buy two lights, you get two numbers. The drawing will occur after the first 20 are sold. Open to international customers as well, as long as I can ship the light to you (if your country does not allow batteries, then you will not receive a 10440 with the light).

Ordered my lumintop tool, hope I’m the lucky one.

[saypat]

Ordered my lumintop tool, hope I’m the lucky one.
[/quote]

buy all 20 and you’ll be the lucky one

[RMM]

@WillyD

Ok, I feel like this has to have already been asked, but I did a ctrl-F for ‘S2+’ and went several pages back. So, is there any estimate on when you might be getting any red or blue hosts along with the shortened tubes I have a pretty large order to place and wanted to do it all in a single order. If it’s going to be a while on the S2+ hosts I’ll just place the order for everything else. And, thank you.

As soon as Convoy gets them to me. They have been really dragging lately and do not communicate well, so I am a bit in the dark myself.

[Rufusbduck]

@RMM

@ToyKeeper

@pscal

It may be possible to use a 7135+FET driver for a 6V LED without additional heatsinking if the firmware limits the current through the 7135 chip to an acceptable level, but I am not even sure if that is possible (code is not my forte).

Some people already do this and it seems to work. Double the voltage, cut the maximum PWM duty cycle in half, same overall wattage and heat. Still provides nice low/med modes.

I would recommend cutting the maximum PWM duty cycle even lower than half. I tried going down to PWM=100 (~40% duty cycle) even, and the customers who had them had the 7135s die after a while. I would only use the 7135 for moonlight and maybe very low duty cycles of 10% or less on 2S lights.

Would it make sense to stack two and cut the duty cycle more? Moon wouldn’t be quite as low but the wattage would be split between the two chips.

[RMM]

@Rufusbduck

@RMM

@ToyKeeper

@pscal

It may be possible to use a 7135+FET driver for a 6V LED without additional heatsinking if the firmware limits the current through the 7135 chip to an acceptable level, but I am not even sure if that is possible (code is not my forte).

Some people already do this and it seems to work. Double the voltage, cut the maximum PWM duty cycle in half, same overall wattage and heat. Still provides nice low/med modes.

I would recommend cutting the maximum PWM duty cycle even lower than half. I tried going down to PWM=100 (~40% duty cycle) even, and the customers who had them had the 7135s die after a while. I would only use the 7135 for moonlight and maybe very low duty cycles of 10% or less on 2S lights.

Would it make sense to stack two and cut the duty cycle more? Moon wouldn’t be quite as low but the wattage would be split between the two chips.

Yes it will help a bit, but not too much. I would just use the 7135 for moonlight and maybe 2% levels, then let the FET do all of the others.