Testing a Cree XHP50.2 J4 3A led

Yes, this is possible and something I have been wanting to get worked into Bistro from awhile now.

Although at this point I think we might end up skipping that rather crude method and going to something better unless someone gets it implemented before we make the move to the newer MCU’s.

I just finished up a build this past weekend; an XHP50.2 J4 3A 5000k in a single 26650 tube, using a H1-A booster driver all from KD. It has a much nicer hotspot with no X in it vs the original 50; the corona tint shift is still there like the XPL2s I’ve built, but doesn’t seem as pronounced. Compared to an L2 with a MT-G2 direct driven I built, the max on this 50.2 light boosted to 6v seems around 2500 lumens, or about 50% power on my MT-G2 L2. I don’t have a light meter to check the lumens or candela though.

as texas Ace said the max. Setting of the driver can be limited in the firmware

instead DD a 80% or so cycle as max. setting can be set in the firmware on your TA driver

Can an expert do this please?
Thanks

Low Vf on a DD driver with FET
higher current might blow the LED because it can get too much

on a linear driver like 7135
more heat on the driver less on the LED

on a Buck/Boost driver
less power consumption of the lioght, more runtime, less heat

I got some LiitoKala 26650’s in yesterday and charged a pair up, tested them against the Basen’s in my DBC-05 Triple 50.2. The new cell gave more amperage at the tail end but less actual lumens. Weird when I see that. “Only” 10,833 lumens pulling 22.44A on the LiitoKala’s.

TomE is being an outstanding guy and sending me his pair of Sanyo NCR20700A to test in this light against the other contenders. It’ll be interesting to see how they compare…

Lexel, is the below correct?

This can be controlled through carefully battery selection or programming the driver to run at a lower duty cycle on the highest level.
.
.
.

Emitter will see same amp draw as a non-low Vf emitter.
.
.
.

You get the same amp draw to the emitter, just with better efficiency.

this is correct

I want to ask if someone tested the nano MCPCBs from fasttech (https://www.fasttech.com/products/1609/10031707/7156107)? What i can find on the net if i search for nano MCPCB sounds promising (http://www.camnano.com/nanotherm-lc/?gclid=CMuynYPo2NMCFaoK0wodHMUOzw) but i do not know if the pcbs are from this supplier or have the same specs.

Noctigon or SinkPAD XM-L mcpcbs, tried and true.

The issue with LED lights is more about heat then anything. We are pretty much always heat limited and it appears that you would be as well in those lights.

They should handle 1.5A @ 6V fine (about 10W of power) fine from how those lights appear to be made.

You can drive these LED’s up to 10A+ if you can keep them cool and they will make ~5000 lumens each at that level if you do.

Although I am not sure what you mean by 100% efficiency. There is no such thing with LED’s. If you mean the max efficiency then you would only want to run them at like 100ma. efficiency drops as power goes up, see some of the tests we have posted on here for examples.

They will still be pretty efficient up until around ~6A though, so that is not a problem in your case, the issue is once again keeping the heat in check.

That link is just advertising. There are no actual details or specs on it from which to derive any useful data.

It appears they use a DTP copper mcpcb which is great but also no different then any other copper DTP mcpcb on the market. Also they say 100% power, not efficiency. Big difference. 100% rated power for the xhp50 from cree is like 2.8A. There is no such thing as 100% efficiency as the lower you driver it, the more efficient it will be.

We run the xhp50 up to 8-10A in flashlights but that will quickly overheat in that light you have. 1.5A should be about right I figure.

You can run the LED in 12V mode but better off in 6V mode, that will ensure it stays in regulation and that you can use a cheap XM-L DTP copper star.

Are those lights you linked to already LED? It might be easier to either just swap the LED’s in them or mod the driver for a bit more power. Got a link to those lights?

Hi stormrider88, Welcome to BLF!!

The BD "Mad Scientist" guy sure is funny. Talks like he discovered or invented copper for LED thermal management, something we've known and practiced for years. Also several of the major LED high performance flashlight manufacturers have been doing copper for years as well, like Fenix and ThruNite.

That BD site though says 100% power, not 100% efficiency. I think that's mis-leading what's goin on there.

It's a shame because he's dis'ing a lot of us out there, including Richard at MtnE who sells light bars here: http://www.mtnlitebar.com/, who I'm sure is using copper in the LED MCPCB's.

The "Thermal Gap Pad" thing is interesting though. Wonder how efficient that is...

Edit: Ooops cross posted with TA.

+1 with TA!

I can help with that last part.

The xhp50 and xhp70 are available in both 6 volt and 12 volt versions. Each emitter has 4 tiny dies which are 3V each.

So the 12 volt version have all 4 dies wired in series.

The 6 volt versions have the dies wired 2 in parallel and those 2 in series.

So 3 amps at 6 volts equals 18 watts.

1.5 amp at 12 volts also equals 18 watts.

Same power levels.

Watts law states volts X amps = watts

Make sense?

That thermal “gap pad” is nothing more then a silicone thermal pad, aka those thermal cubes you can get from fasttech, except in a sheet. It is used in place of thermal paste in assembly as it is a lot easier to apply. It doesn’t work quite as good as thermal paste but good enough for most things, particularly if the surfaces are not perfect.

I didn’t watch the video before, ditto all you said, it is all marketing and I would take a Mthlitebar over those anyday. At least he gives us actual data on voltage, power, lumens, throw ect. Plus I trust his lights to actually do what he says because he does give us all that data.

If a manufacture is too ashamed to show that data, then there must be a reason for it.

That said, the BD lights look better then a lot of the ebay junk light bars I see but I would not class them with the Mthlitebar.

Yes the LED will draw current proportional to the output voltage you set that module at. Regarding those DC-DC modules though, I would recommend one with Constant Current capability. There are many varieties available.

The DC-DC modules themselves do make great drivers if you get a CC version. I use many different kinds in various projects of mine, so far they hold up pretty well as long as you don’t drive them too hard.

No. Emitters will draw as much current as you give them. Hook any 12 volt led to a car battery and they will pop. Poof, burn up very fast. In order to control an leds output, you have to control the amperage you give it.

When Cree rates their leds they choose the current. They choose 1.5 amp for the 12 volt version even though it will stay alive at 4 to 5 amps (8 to 10 amps at the lower 6 volts in a flashlight).

So if that module can really supply a full 2 amps at 12 volts to the emitter, and the specs may not be fully accurate, then the 12 volt xhp50 will run at 2 amps.

To see what the lumen output is just look for a xhp50 output test by matchless or djozz. If they test at 6 volts, then look at the 4 amp numbers which is about 3,000 lumen.

I would highly recommend just spending a bit more for a better pre-built light. I have not looked at off-road lights in a long time but I have to believe that someone makes a quality light for a reasonable price.

Or heck, get one of the 12” mtnlitebars and it will blow away anything else I have come across. Something like 30,000 lumens from a 12” bar IIRC. I plan to get one of these when I can afford it.

I guess in theory you could use a flashlight driver that was a buck driver designed to give 2 amps to a 12 volt led. Or a buck driver that delivered 4 amps for a 6 volt led. Set the driver up for single mode use.

That’s not a big amp draw.

The guy that makes the mtn lite bar also runs MTN Electronics. You should send him an email and see if he has a flashlight driver that suits your needs.

We are getting pretty far off topic here, but have you verified your motorcycle has the extra capacity for these lights? Bikes usually has very marginal charging systems and most don’t have any extra output for a constant 2 or 3 amp load. If you add too much load you’ll slowly discharge your battery when the lights are on. Some bikes, like goldwings, have extra capacity built in. What model bike are you working with?