I have a 12x .380 7135 slave board connected to a 1.4A board. Together they should provide around 4.56A + 1.4A = 5.96A. However, I am only getting around 4.6A out of them. I direct drove the LED’s with the same battery I tested the boards with and got 6.2A, so Vf of the LED’s isn’t the issue and battery selection isn’t the issue. I tested the 1.4A board at 1.42A. The slave board is wired using 20ga wire. Why am I so far off from the desired current?
When you direct drove the LEDs, you didn’t have the voltage sag of your driver constellation in your circuit… I think that’s the reason you’re seeing less current.
It sounds to me like the 7135 chips on the driver aren’t working, but the slave board is. Can’t think of why that would happen though.
No, direct driven was the battery directly to the LED’s. As for the driver, it works perfectly fine without the slave board at 1.42A. With the slave, the modes on the driver work, just the current isn’t there. I don’t know why or how it could bypass the 7135’s on the drive and only use the 7135’s on the slave. Seems impossible since the slave is connected to a ground and pwm on one of the driver’s 7135’s.
I’ve got a similar set up, a triple XP-L with a 12x .380 7135 slave board connected to a 3.04A board. Running on Samsung nr18650-25R I get about 7.4 amps. So in theory it can work. Check all those solder joints, foil traces and wires.
No matter what you will get some losses through the driver traces and vias, along with the 7135s themselves. These are mostly useful for multiple LED solutions since the effective forward voltage requirement is so low.
Put a jumper across one of the 7135 ground and output pins then run it and see what you get. Those are the losses from the driver traces, spring, contacts, and wiring. You then have to add the losses from the 7135s themselves on top of that.
Even an awesome FET driver will lose some power compared to true direct drive. You have a 1-6 mOhm FET (when cold and battery full), a few mOhms through the traces and vias, extra wire, etc. LEDs are so sensitive to voltage that even a few mOhms can make a large difference in some setups.
Richard, I know that there are some losses, I’ve just never experienced 1.5A loss. That seems pretty significant. I stacked the same amount of 7135’s on a different driver and didn’t have that much loss.
Sounds silly but do you have led - connected to both boards?
^ Good question RBD.
This doesn't sound like your problem, but just in case. 7135's can get hot real fast if they are on a slave board when there is a good gap between the cell voltage and the Vf of the load. 7135's have a thermal shutdown protection feature. I've seen it kick in real fast on slave boards Like in seconds. Maybe some of the chips are flowing more of the current early and shutting down fast. Seems doubtful, but you may want so hook up just the slave board and watch the current closely right out the gate. The solution is to heat sink the slave board.
7135's eat like .1 volt I believe.
Not silly. I’m a moron. I’ve been up for 18 hours. Plus this is my first time to wire up this slave board I’ve had for 6 months. In other words it’s very likely I have it wired wrong, and the more I think about it the more I think it is wrong.
I have the pwm, ground, led , and led - from the 1.4A driver to the slave board and then led and led - from slave to the led’s.
So is that wiring wrong?
As long as you have LED- connected from the slave to the driver, that answers RBD’s question.
It sounds like you’ve covered everything
Not wrong just covering the obvious. How hot is the cell?
When I tested it, the slave wasn’t hot and the cell wasn’t hot. It was only on long enough to check amperage and that’s it.
Sorry, I meant the charge.
3.09 ohms in there somewhere. But where? Sounds like alot when your taking small megohms from the7135. Are the led’s in parallel? Heat doesn’t help out either. I’m wanting to learn more about modding.
Yes, 3x XP-E2 Red LED’s in parallel, so low Vf with the LED’s.
Have you tried shorting the ground and output pin on the 7135s yet?
I have not. It will either be tomorrow morning or Monday morning. I work weekend nights in the ER, so technically I should be sleeping right now but the doorbell rang and woke me up. I’ll probably play with it tomorrow morning when I get off work.
I'd do that just to bypass the 7135s and eliminate that variable. There will be some drop going through the 7135s regardless, but if it is still way low then you'll know to look elsewhere. If is still way low bypassed, then the next step is to drill a hole and run the positive wire directly to the battery side. Most of those drivers only have a few small vias for the positive end and you can end up with a lot of resistance there at high amps, especially as that copper starts to heat up.