Stenlight S7 in Turbo mode - Looking for Lux behavior

[wasBlinded]

Anyone with this headlamp with unmodified emitters and a Luxmeter able to help me out? I'm interested in what happens to light output in the first 2 minutes or so in Turbo mode, i.e. is a lux drop of about 25% in that period inherent to the design? I can tell this is due to die heating in mine because there is no significant current limiting yet occuring in that time period.

Thanks

 

[cy]

never noticed it with mine, but usually never run in brightest mode for long.

where in OK are you? I'm in Tulsa.

 

[wasBlinded]

Hi Cy -

I had not noticed the sag in lumens either. To the eye, it looks like nothing has changed, but the Luxmeter says differently. I'm just trying to convince myself that this is normal Stenlight behavior, and not something I need to fix on my own now that I've upgraded the emitters. Looking at the thermal path, I suspect that all Stenlights do it to some extent. On High, that drop is much smaller, and it is non-existant on Medium and Low.

Yes, I live in Tulsa.

 

[jtice]

I cant see it either, with my naked eye, never tried a meter on Turbo.
Here is my runtime on High, if that is of any use to you.

Slight drop there, but that happens with pretty much any light.
You dont have to worry about the Sten overheating though, it has builtin thermal limiting. I have not seen mine do this yet though, and have ran it on high for long periods of time.
David (from Sten) says it only happens on turbo usually, and even then it has to be in a fairly warm environment.

ah, BTW, what did/are you modding yours to?
I have been thinking of slapping a couple Seouls in mine.

~John

 

[cy]

got two SS P4 ubin headed in mine

 

[wasBlinded]

The drop is really only on Turbo, and it is not due to current limiting for temperature - it is occuring too early for that, and watching the current supplied to the light shows that the current isn't changing significantly at that point.

On high, there is very little drop on mine. I've upgraded it with two Seoul P4 USW0H LEDs.

 

[wasBlinded]

Looking at the Seoul datasheet, it looks like you have to raise die temperature 85 deg C to get a 25% drop in luminous output with the P4 white. That implies that I have a very poor thermal path in my Stenlight, I just don't know for sure if that is the nature of the S7 or the way I have installed my emitters.

 

[jtice]

How did you install your emitters?

Stock, the Sten uses sticky thermal pads that attach the LEDs to the PCB,
which is supposed to have a copper backing I beleive.
I tried to convince them to attach the LEDs directly to the back of the Stens AL body,
but I think there were issues with alignment, and assembly.
Plus I think thats how they were monitoring the led temps.

Did you reuse the pads? or replace them with thermal epoxy?

~John

 

[wasBlinded]

I replaced the pads with thermal expoxy (I talked about it here: http://candlepowerforums.com/vb/showthread.php?t=155973)

What I failed to do was check on the light's behavior in Turbo before replacing the emitters. If an unmodified Stenlight doesn't show that 25% fade in the first minute or two, then that tells me I should rethink the mounting of my emitters. Too bad they aren't mounted on a pedestal with metal to metal connection to the body, though that would certainly increase cost of manufacture, and this is only an issue in Turbo mode.

 

[cy]

this tells me I should reuse sticky pad if possible.

was the pads reusable?

 

[jtice]

Odd, I really wouldnt think the pads would be better then the thermal epoxy.

From your description, it seems that the beam pattern turned out nice, with an increase in output.
I am still thinking whether to swap my LEDs, I have TWOJs in mine now.

~John

 

[wasBlinded]

this tells me I should reuse sticky pad if possible.

was the pads reusable?

Yes, the pads were reusable, and make positioning the LED a lot easier. I didn't think they would be as good as a thin layer of thermal epoxy either. I still don't know, since I didn't test for an output drop on Turbo when I used the pads with the UVOJ Luxeons.

 

[niemidc]

It is conceivable that thermal limiting could begin to kick in within 2-3 minutes. In Turbo mode it begins to have a subtle effect immediately, and if you don't have good airflow over the lamp it will eventually have a fairly big effect (up to 40%).

The thermal limiting sensor is right next to one of the LEDs, so it will see the heat minutes before the case feels warm. To tell whether it's die heating or thermal limiting, try your test while the aluminum part of the lamp is sitting in a bowl of water, face up (if this is practica) or find some other way to keep the lamp body very cool.

There are two thermal paths that could be at issue: LEDs to circuit board, and circuit board to body. If the first one is degraded, the LEDs will heat up more than desired. If the second, thermal limiting will be more limiting than usual and the lamp will not get as warm as it should even after 10-15 minutes. If this is the case, try to measure the total power the lamp draws after 20 minutes at room temperature in still air, and it should be around 4 watts.

 

[wasBlinded]

The initial major drop in output I am seeing comes well before a reduction in current draw from the battery, so I'm pretty sure it is due to LED die heating. I'll have to get a couple of the thermal pads from you and go back to that technique.