Not getting expected "tailcap" results from BLF17DD driver

I recently built two, almost identical T08s, one with a BLF17DD driver the other with a Q-Lite + 8. Tailcap amp readings have me a bit confused.

Other than driver choice builds are identical, removed all components from the original drivers and used them as contact boards. 20ga silicone wire from the contact boards to the 17DD/Q-Lite, drivers potted with Fujic. 20ga silicone wire to identical, Dedomed XM-L2s on home made DTP MCPCBs.

All modes work properly on both light, including the step down from Turbo on the 17DD.

Here’s where I’m stumped:
I’m using a Fluke 88 to measure tailcap amperage, a set of stock Fluke leaks cut down to 10” with a 1” piece of 12ga copper home wire soldered to the - lead to bridge the two batteries.

With the Q-Lite light I get consistent 6 amps regardless of batteries used.
With the BLF17DD:
Samsung 25R, 4.95a
Samsung 20R, 4.90a
AW IMR, 4.8a
All batteries fresh off the charger

I realize I’m using the stock Fluke leads but they’re rated for 10a and a fraction of their original length. It doesn’t really matter anyway because I’m using the exact same setup with the Q-Lite.

I’m not sure what I was expecting to see from the FET driver but thought it would be higher than the Q-Lite. The only thing I can think of is a high resistance connection somewhere between the contact board and LED. Kinda wish I hadn’t potted everything.

For what it’s worth, both lights are very nice and I can see no appreciable difference in output. In fact, the FET light measured a corrected 159kcd* to the Q-Lite’s 122kcd* though the FET light may be better focused.

  • My output numbers are only relevant my output numbers…

So, anyone want to venture a guess at what I did wrong? As I said, I’m stumped.


One of the things that keeps me at this hobby is that with every success or failure, I learn something.
This time?
Despite a full amp difference I can’t see it, outside, side by side.
Despite more than more than 35kcd* difference I can’t see it.
Throw away the ammeter and lux meter, go outside and turn it on!
Always test, every step of the way (I though I had learned this a while ago, the combination of two batteries, contact board, driver, LED and only two hands led me to only check for shorts while building).

The first thing that comes to mind is to check that the BLF driver is grounded solidly.

In my own experience, a Qlite with 8 added 7135’s won’t do a consistent 6A with much of any cell, much less all tested. I wasn’t getting 6A and more until I started building the FET drivers. And I have Qlites with 22 7135s.

The only times I’ve had the FET driver show a weak performance was when the ground was insufficient. The 20R should be showing in the 6.6A range or higher.

By consistent I mean 6.0a with a fresh cell. In my experience a 16 x 7135 Nanjc or KD V2 will do 6a until heat or sag become a limiting factor. In my case I’m measuring on start up, maybe 15 or 20 seconds at most, heat and sag (particularly with 2P) are not the issue.

Interesting that you’ve seen reduced performance with a high resistance ground, I would think the same could also happen with a high resistance source side. Looks like I’ll have to decide it it’s worth digging into the potting to find out.

Drivers that don't get hot don't need to be potted. Doing so will only run the risk of making them hotter than they would otherwise be, as they pick up heat from the pill.

Dedomed LEDs are pretty notorious for being inconsistent, something about messin' with them alters the Vf, and the altered Vf also affects the current they'll pass in a direct drive setup where Vin is already so close to Vf. Do you have a light meter to do even a basic ceiling bounce comparison, just to double check the current readings? It should show up on the meter even if you can't see it by eye.

Also, bypass the drivers in both lights (just temporarily short LED- to GND) and see if there's still a difference in the tailcap numbers. That will give you an idea of what the LEDs are doing.