Low vf XP-L2 and H17Fx driver

cantero · 2018-11-05T03:33:33.000Z

I’m just now getting into this hobby and toying around with an M2 host with a XP-L2 and the H17Fx driver. After doing a bit more reading I realized using a driver with so much regulated output on a low vf LED may cause some heat issues due to the extra voltage needing to be dissipated. Is this assumption correct? Should I switch to a higher vf LED for a driver like this and use a PWM DD driver for the lower vf XP-L2?

chadvone · 2018-11-05T03:43:26.000Z

I have M2 H17f driver with XP-l Hi. Will get warm, but I wouldn’t let that stop you. There are many lights on this forum that get Hot. Kinda what is done here. The H17f has really good thermal step down also.

cantero · 2018-11-05T04:09:02.000Z

I’m not worried so much about the heat of the LED, I know the light is capable of handling it. My main concern here is the heat being dissipated by the 7135 regulators themselves.

For example…
Let’s say I’ve got a fully charged 4.2v cell and I’ve got it set on a regulated mode that is drawing 3 amps

The VF of the XP-L2 is ~3.2v around that current range, and there will be some voltage drop from the battery being loaded so lets say the battery is spitting out 3.9v

0.7v difference that has to be dissipated through the 7135 chips 0.7v*3amps=2.1watts of heat which seems a bit high to me for the driver PCB to dissipate since it isn’t as thermally bonded to the head as the LED is.

I could be completely wrong on all of this, I guess my worry is can the driver/7135 safely dissipate that much heat without damage?

This may not be a problem since it is being spread over multiple 7135’s, what brought about this thought is Texas_Ace’s footnotes on the “17mm V2 Texas Commander” driver on his OSHPark page.
https://oshpark.com/profiles/Texas_Ace/page/1
But, from my understanding this driver uses a single CC IC to do the regulation, hence it may be more of a problem in that type of setup vs multiple 7135’s.

nick779 · 2018-11-05T15:12:20.000Z

cantero:

I’m not worried so much about the heat of the LED, I know the light is capable of handling it. My main concern here is the heat being dissipated by the 7135 regulators themselves.

For example…
Let’s say I’ve got a fully charged 4.2v cell and I’ve got it set on a regulated mode that is drawing 3 amps

The VF of the XP-L2 is ~3.2v around that current range, and there will be some voltage drop from the battery being loaded so lets say the battery is spitting out 3.9v

0.7v difference that has to be dissipated through the 7135 chips 0.7v*3amps=2.1watts of heat which seems a bit high to me for the driver PCB to dissipate since it isn’t as thermally bonded to the head as the LED is.

I could be completely wrong on all of this, I guess my worry is can the driver/7135 safely dissipate that much heat without damage?

This may not be a problem since it is being spread over multiple 7135’s, what brought about this thought is Texas_Ace’s footnotes on the “17mm V2 Texas Commander” driver on his OSHPark page.
OSH Park ~ Shared Projects by Texas_Ace
But, from my understanding this driver uses a single CC IC to do the regulation, hence it may be more of a problem in that type of setup vs multiple 7135’s.

In this regard, youll be fine. I have multiple nichias and even multi emitter low vF lights which regulators do not suffer. In your case I would be thinking more about the FET side using that emitter. The LED can take it, but its going to be pulling 8-9A full FET.

EasyB · 2018-11-05T15:36:34.000Z

You are right, that is a lot of heat produced in the 7135s. I don’t know if it’s too much, but it will definitely mess with the temperature regulation since the MCU (where the temp sensor is) is getting hot from the 7135 produced heat. I observed this with this driver and an XPL HI; the output would step down quicker in
the regulated mode than in FET-only mode, even though it pulled more current in the FET-only mode.

The FET on this driver is higher resistance than most FET drivers (unless it has been changed), so this will limit the current more in turbo, probably to ~6A depending on the cell you use.

nick779 · 2018-11-05T16:44:55.000Z

EasyB:

The FET on this driver is higher resistance than most FET drivers (unless it has been changed), so this will limit the current more in turbo, probably to ~6A depending on the cell you use.

Why do you say that? Its the most recent revision available on a very high quality driver. MTN even states that the X revision has an even higher performance FET on it.

Edit: I suppose actually looking at the datasheet for the FET on mine would have been a good idea…

cantero · 2018-11-05T16:33:32.000Z

nick779:

In this regard, youll be fine. I have multiple nichias and even multi emitter low vF lights which regulators do not suffer. In your case I would be thinking more about the FET side using that emitter. The LED can take it, but its going to be pulling 8-9A full FET.

It gets hot pretty quickly, but the M2 handles it quite well. I don’t have a clamp meter or any custom leads to accurately measure the current draw but I would guess it’s close to that. It’s certainly brighter than my BLF A6.

EasyB:

You are right, that is a lot of heat produced in the 7135s. I don’t know if it’s too much, but it will definitely mess with the temperature regulation since the MCU (where the temp sensor is) is getting hot from the 7135 produced heat. I observed this with this driver and an XPL HI; the output would step down quicker in
the regulated mode than in FET-only mode, even though it pulled more current in the FET-only mode.

I haven’t noticed it doing that, but I haven’t been playing with the regulated modes for extended time periods either. Luckily it has an adjustable temperature cutoff to compensate, the only concern is at what point can the 7135’s no longer handle it.

nick779:

Why do you say that? Its the most recent revision available on a very high quality driver. MTN even states that the X revision has an even higher performance FET on it.

assets.nexperia.com

PSMN3R5-25MLD.pdf

728.29 KB

That is the FET in mine, which I just ordered last week, it’s the H17Fx with the blue solder mask, so I’m fairly sure the latest one. It’s not the best but it’s not terrible I suppose.

4.46mΩ RDSon typ. at 4.5v, it’s also got a pretty high gate resistance at 0.73Ω

I ordered a couple of SiR800DP’s to try out on it and my A6.

nick779 · 2018-11-05T16:51:44.000Z

cantero:

nick779:

In this regard, youll be fine. I have multiple nichias and even multi emitter low vF lights which regulators do not suffer. In your case I would be thinking more about the FET side using that emitter. The LED can take it, but its going to be pulling 8-9A full FET.

It gets hot pretty quickly, but the M2 handles it quite well. I don’t have a clamp meter or any custom leads to accurately measure the current draw but I would guess it’s close to that. It’s certainly brighter than my BLF A6.

EasyB:

You are right, that is a lot of heat produced in the 7135s. I don’t know if it’s too much, but it will definitely mess with the temperature regulation since the MCU (where the temp sensor is) is getting hot from the 7135 produced heat. I observed this with this driver and an XPL HI; the output would step down quicker in
the regulated mode than in FET-only mode, even though it pulled more current in the FET-only mode.

I haven’t noticed it doing that, but I haven’t been playing with the regulated modes for extended time periods either. Luckily it has an adjustable temperature cutoff to compensate, the only concern is at what point can the 7135’s no longer handle it.

nick779:

Why do you say that? Its the most recent revision available on a very high quality driver. MTN even states that the X revision has an even higher performance FET on it.

https://assets.nexperia.com/documents/data-sheet/PSMN3R5-25MLD.pdf

That is the FET in mine, which I just ordered last week, it’s the H17Fx with the blue solder mask, so I’m fairly sure the latest one. It’s not the best but it’s not terrible I suppose.

4.46mΩ RDSon typ. at 4.5v, it’s also got a pretty high gate resistance at 0.73Ω

I ordered a couple of SiR800DP’s to try out on it and my A6.

I dont know enough about how much that small change in gate resistance can affect the overall circuit in this use case, but considering the FET on the D4S has a gate resistance of 1-1.7Ω and still pulls around 20A on 100%, I dont quite see its relevance.

infineon.com

Infineon-BSC009NE2LS5-DS-v02_00-EN.pdf

1 B

EasyB · 2018-11-05T16:52:33.000Z

nick779:

EasyB:

The FET on this driver is higher resistance than most FET drivers (unless it has been changed), so this will limit the current more in turbo, probably to ~6A depending on the cell you use.

Why do you say that? Its the most recent revision available on a very high quality driver. MTN even states that the X revision has an even higher performance FET on it.

Edit: I suppose actually looking at the datasheet for the FET on mine would have been a good idea…

Nevermind about that then. The original version had a different FET.

Barkuti · 2018-11-05T16:53:58.000Z

Use a good buck driver. Problem solved.

Flashy_Mike · 2018-11-05T16:54:12.000Z

cantero:

… the only concern is at what point can the 7135’s no longer handle it.

7135 should have internal heat protection and throttle down.