HID warm-up graphs

LEDBE
S

Superfonarik.ru

Newly Enlightened
Joined
May 6, 2008
Messages
30
Location
Moscow, Russia
Hello everybody!

I've seen no similar reviews yet, so I guess that we are first to measure startup graphs of popular HID flashlights. We believed previously that all HIDs warm-up the same way. Some ballasts are faster, some are slower, lamps differ, etc.
But we suddenly discovered that Polarion PF50 ballast does a very crazy job to boost HID output to around 10000 lumen (so getting to nominal 5200lm in just 3 seconds, cold start), then slowly settling down to 5200lm.
That's really amazing flashlight!

The PSS-01 is a generic 3500-lumen HID made of automotive HID lamp and similar ballast. The raised area on red graph between 0-8sec comes from xenon burning (and giving some instant light) while main metal-halide mixture is heating up.

HID-startup.png


Lights to go: Surefire Beast, Microfire Warrior, and maybe some top-cost cars HIDs :)
 
Great graph for the PH50! I've found the same results with my own light and I suspect all the others operate the same way. I tested mine with a light meter so I've only got raw lux data.

How were you able to measure lumens....do you have an Integrating Sphere?
 
We are building our Sphere currently, but that work is far away from end :)

Actually I've adjusted graphs to manufacturer-declared output, which seems to be quite accurate. At least, all our indirect luminous flux-evaluating methods show more or less the same.
 
Excellent idea for a thread. Members will find this very useful when deciding which HID light would suit them best.
 
I would be interested in seeing what the startup graph for the Stanley HID would be, since it also "overcharges" the bulb for the first few seconds.
 
Mjolnir said:
I would be interested in seeing what the startup graph for the Stanley HID would be, since it also "overcharges" the bulb for the first few seconds.
Click to expand...


I think it's been done already, just not in graph form.
 
Now that I go back and look at my data, my PH50 peaked at 530 lux and settled to 368 lux. So, my light didn't show nearly twice the output at 8 seconds as yours did but still shows a large difference between peak and nominal. I may retest now that you've peaked my curiosity again.


Please do post the raw lux data, that would be great.

Thanks again. :)
 
Superfonarik.ru said:
I'll retest also. Did you measure it from "cold" state with fresh-off-charger battery? What is the total runtime of your item?
Click to expand...


The light was "charged" but not hot off the charger. In other words it was probably charged two months prior and placed in the safe. Since the PH50 is regulated and because li-ion cells have very low self discharge rates I'm not sure if this is going to make a much, if any difference but I've been surprised by unexpected results in the past. It's certainly easy enough to top off charge to battery and conduct my informal test again. This isn't to say that I don't trust your test results, since I fully do. Instead, I just believe we're seeing the difference rooted in sample variation.

Total run-time for my light is 58 minutes iirc. I only ever officially measured the run-time once when it was new. I will say that my particular PH50 seems very bright. I ran a series of comparison tests and beamshots between it and my PH40 at the time and the performance difference was striking. Much more than what I thought a 10W difference would account for. My guess was that either my PH40 was slightly under average or my PH50 was a bit above average. I didn't own a light meter back then so I'm purely speculating.
 
Patriot, it seems like testing conditions are similar. I think that low charge (say, 20-30%) or extensive wear (some thousands hours run time) can make the graph differ much, but our differences are indeed most likely sample-to-sample variations. We'll also take graphs for some different samples we have.

I have also noticed that striking difference between PF40 and PF50, it's easily about 30% brighter, as seen by naked eye. Taking into account the logarithmic sensivity of an eye, the difference in output looks like twice.
We don't have 40s in stock now, but I'll certainly measure that once we get them from Polarion.

Or maybe Ken Good does have both and can take that lux readings for lights when there is some free time? :)

Anton.
 
Ok, I retested with a topped off battery. The position of the light and light meter were not exactly the same but all I'm looking for is the relative difference between peak and nominal output.

Peak during start was 522
nominal run after 2 min. 320

Slightly more spread than the 530 and 368 that I got before. It appears that my light either isn't peaking as highly or settling as low as yours. Still, if it's producing 5200 lumens nominally it's peaking to around 9000 lumens which is amazing to me.

Equally exciting and impressive is the 5200 lumens at 3 seconds!!!
 
Is it possible that the color temperature affect the measurement? I find it hard to believe that the measurement is true. As the eyes of the
light clears the first 3 seconds. My measurements show the same as for the other. I am still skeptical;)
 
vee73, can you take lux readings for first 30 seconds of warm-up of your Polarion? All our samples show more or less the same peak-and-stabilize behaviour.

I think it's not the case where CCT can affect the results. Anyway, color change during Polarion warm-up is much shorter and is much less noticeable than for other HIDs.
 
Vee73, I'm not sure that I understood your post. Were you able to test your lights with a meter or were you just saying that to the naked eye it didn't appear to be nearly double the output?

I must say that I never would have guessed that my light was 35-40% brighter at 7 seconds that it was at 30 seconds but the eyes and brain do a very poor job of tracking light differences over time.

Regarding the light meter favoring a certain spot in the spectrum, I'm sure that's a tiny portion of that is reflected in the test but the range from start to nominal run is pretty much centered in the visible spectrum. For this reason I don't think the meter itself is a cause of any major variation. I would guess far less than 5% and possibly even less than 1% in the case of quality light meters.
 
Last edited:
You must log in or register to reply here.

Similar threads

PhotonWrangler
Stanley HID retrofit LED?
Replies
0
Views
310
PhotonWrangler
PhotonWrangler
S
Vector HID
Replies
2
Views
1K
alpg88
alpg88
G
Source for 100 Watt HID Bulbs
Replies
0
Views
808
Goofy_Goober
G
I
Chinese made 35 watt HID replacement bulb
Replies
2
Views
2K
Monocrom
Monocrom
Monocrom
Anyone still use their Sam's Club HID?
Replies
11
Views
3K
Monocrom
Monocrom

Latest posts

Top