17mm 3500mA Driver for XM-L U2?

[FX-32]

Hi

I'm looking for (and can't find yet) a driver for a XM-L U2 with less than 17,0mm diameter, Input should be ~3,6V to any, and output must be real 3500mA max.
Thanks in advance!

 

[Be-Seen Triker]

Are you opposed to building the driver up to meet 3.5A? Do you need boost/buck capability to maintain constant current?

What I am proposing is a NANJG 105C driver (8x AMC7135) and adding 2x AMC7135s to it. This is one I made for 3.15A.
This is a very reliable way to achieve your objective until such time the battery voltage drops below the required Vf to maintain this current. At 3.5 amps, that will be pretty soon. I think I measured the Vdrop across the regulators at ~0.1V. I used Arctic Silver epoxy to hold the 7135 in place and share the heat. You will need a good boost driver if you need to maintain 3.5A to the emitter. I haven't found a good driver that will do that in a low profile 17mm formfactor.

 

[deadrx7conv]

Stacking those AMC7135's is cheating. Great idea. Each 7135 cranks out 350ma or about a watt.

Wondering why no one is using the AMC7140 which is good for 700ma or 2 watts? You'd only need half as many.

Or the 1.5a AMC 7150.... just don't see it too often either!

 

[moderator007]

I have tried to source some AMC7135's from china. They want more than what DX 1400ma boards are. The shipping and transfer fees was killing the deal. So I just buy DX boards and strip them off. Looks like the 7140 or the 7150 would be a good circuit to build for small AAA flashlights. Really doesn't even need a circuit board. I suspect trying to source some of these will also be more than its worth.

 

[deadrx7conv]

Most chips are readily available on ebay.

 

[FX-32]

Thank for your response!
Now I'm going to use a driver from Kaidomain (SKU:S009742) with 8xAMC7135 and I will add some extra chips. This has an ATMEL chip for modes, and I already have a programmer... so, I will edit this chip and install only modes I want (no SOS for example, and custom % for each dimming mode). Here is the info to do this mod.
Thanks again for your help, it really helped me a lot!

 

[CKOD]

Hi

I'm looking for (and can't find yet) a driver for a XM-L U2 with less than 17,0mm diameter, Input should be ~3,6V to any, and output must be real 3500mA max.
Thanks in advance!

worst spec ever

 

[E1320]

worst spec ever

I have had really good luck running the Nanjg 105c with 10 7135s ie: 3.5amps. With the U2 XML they are very bright, but they do generate some serious heat so you must be careful of that.

 

[FX-32]

worst spec ever

Why? Please explain better!
In the first post, I was thinking about a Boost+Buck driver, for that reason I detailed that input.
I'm getting a linear driver now, I know it isn't constant current and input is not much flexible, but it will work OK for my project.

 

[simplec6]

Are you opposed to building the driver up to meet 3.5A? Do you need boost/buck capability to maintain constant current?

What I am proposing is a NANJG 105C driver (8x AMC7135) and adding 2x AMC7135s to it. This is one I made for 3.15A.
This is a very reliable way to achieve your objective until such time the battery voltage drops below the required Vf to maintain this current. At 3.5 amps, that will be pretty soon. I think I measured the Vdrop across the regulators at ~0.1V. I used Arctic Silver epoxy to hold the 7135 in place and share the heat. You will need a good boost driver if you need to maintain 3.5A to the emitter. I haven't found a good driver that will do that in a low profile 17mm formfactor.

Awesome pic here, very hard to find good macro shots like this. Still trying to figure out how to do this 7135 solder business.

 

[brandocommando]

I just took these, I hope they help.

Each chip adds 350mah, just bend the 3 little tabs down and piggyback them. I have made (3) 4.2A drivers now this way.

I get my chips (and drivers) from Illumination Supply. Craig is a CPF sponsor/retailer. He sells the chips for .50 a piece.

I just ordered 30 more of them, and they have fast shipping too! I get my orders from them in 3 or 4 days usually.

The driver in the pic is a 5.6A driver that has 4 chips added on the top as well as 4 chips added on the bottom!

I also just made a 1.7A 3 mode driver for an XP-G by adding just 1 chip, it should be around 400 lumens and has some great throw.

Feel free to PM me if you have questions!

Just buy extras, if you mess up you will only be out $7.00 or $8.00.

 

[Justin Case]

At 3.5A drive, an XM-L could have a Vf of 3.4V. Add in the 0.12V overhead for the 7135 and you might need a min of ~3.5V to reach full regulation. Realistically, you will need more than that because your flashlight (I assume that this is a flashlight application) will have parasitic resistances (e.g., from spring contacts, battery/battery contacts, thread contacts, switch contacts). At 3.5A, if the parasitic resistance is even 0.1 ohms, you need another 0.35V, for a total of ~3.85V. Even if you use 1xIMR26500, your run time in full regulation is going to be woefully short.

If you are powering the XM-L with a bench supply that allows you to dial in a specific input, then you're all set. If you can feed the driver 4xD NiMH cells, then you should have plenty of voltage to stay in regulation (at the expense of relatively poor efficiency and thus that much more waste heat generation).

But with 1xLi-ion, if you are expecting or wanting decent run time in regulation, you probably won't get it. Maybe if you paralleled a bunch of IMR26500s....

 

[simplec6]

I just took these, I hope they help.

Each chip adds 350mah, just bend the 3 little tabs down and piggyback them. I have made 2 4.2A drivers now this way.

I get my chips (and drivers) from Illumination Supply. Craig is a CPF sponsor/retailer. He sells the chips for .50 a piece.

I just ordered 30 more of them, and they have fast shipping too! I get my orders from them in 3 or 4 days usually.

The driver in the pic is a 5.6A driver that has 4 chips added on the top as well as 4 chips added on the bottom!

I am going to make a 1.7A driver for an XP-G next, it should be almost 500 lumens with great throw.

Feel free to PM me if you have questions!


Just buy extras, if you mess up you will only be out $7.00 or $8.00.

Thanks so much for the awesome pics man!!

At 3.5A drive, an XM-L could have a Vf of 3.4V. Add in the 0.12V overhead for the 7135 and you might need a min of ~3.5V to reach full regulation. Realistically, you will need more than that because your flashlight (I assume that this is a flashlight application) will have parasitic resistances (e.g., from spring contacts, battery/battery contacts, thread contacts, switch contacts). At 3.5A, if the parasitic resistance is even 0.1 ohms, you need another 0.35V, for a total of ~3.85V. Even if you use 1xIMR26500, your run time in full regulation is going to be woefully short.

If you are powering the XM-L with a bench supply that allows you to dial in a specific input, then you're all set. If you can feed the driver 4xD NiMH cells, then you should have plenty of voltage to stay in regulation (at the expense of relatively poor efficiency and thus that much more waste heat generation).

But with 1xLi-ion, if you are expecting or wanting decent run time in regulation, you probably won't get it. Maybe if you paralleled a bunch of IMR26500s....

Wish I could figure out how to do two 26650's or two 32650's in parallel in a D Mag.

 

[gt40]

I am curious about this; The higher levels seem to be at odd with the vf I am seeing on my xm-l's. I have a battery space 10000 mah M cell from an earlier project. It is only 3.3v lifepo4. hooking it up to a xm-l t6, I was hitting 4.5 amps measured with my clamp meter and climbing before I took it off. Maybe the vf is really lower or something. Anyway, I just received a bunch of 7135 chips. Thanks for this thread!

At 3.5A drive, an XM-L could have a Vf of 3.4V. Add in the 0.12V overhead for the 7135 and you might need a min of ~3.5V to reach full regulation. Realistically, you will need more than that because your flashlight (I assume that this is a flashlight application) will have parasitic resistances (e.g., from spring contacts, battery/battery contacts, thread contacts, switch contacts). At 3.5A, if the parasitic resistance is even 0.1 ohms, you need another 0.35V, for a total of ~3.85V. Even if you use 1xIMR26500, your run time in full regulation is going to be woefully short.

If you are powering the XM-L with a bench supply that allows you to dial in a specific input, then you're all set. If you can feed the driver 4xD NiMH cells, then you should have plenty of voltage to stay in regulation (at the expense of relatively poor efficiency and thus that much more waste heat generation).

But with 1xLi-ion, if you are expecting or wanting decent run time in regulation, you probably won't get it. Maybe if you paralleled a bunch of IMR26500s....