I think it would be nice to have 2700K, 3500K, 4500K, and 5700K temperature options available for B35AM. That provides nice coverage of the spectrum without needing to have too many different versions available. Though, if there must be only one, 4500K is the best choice.
+1 on that suggestion.
Simon,
Any chance for the B35AR coming soon, with an appropriate 20mm MCPCB?
I would be interested in 3500k.
It also doesn’t help that the drivers used sometimes have slightly different power consumption.
For instance, I have one of the earlier Convoy S2+ (SST40, 4-mode driver). It’s supposed to be a 5 Amp driver.
When I measure tailcap current, it registers around 4.5 Amp.
I have ordered the same Convoy S2+ (SST40, 4-mode driver) several months later, I noticed in ceiling-bounce lux measurement (unfortunately, I don’t have a light-integrating measurement device…), the lux value I get is a bit brighter. Sure enough when checking tailcap current, the max is now at 4.9 Amps.
Then quite recently, I got another Convoy S2+ (SST40, 4-mode driver), now this one is brighter (not that visibly different when just viewing with my eyes, but the lux meter does pick up that it is brighter). I would guess that based on my lux measurements, this one is brighter than the earliest S2+ by around 200-300 lumens (my “guesstimate” lumens basing from lux measurement). The tailcap current measurement is now at 5.3 or 5.4 Amps.
~
In another case, I bought a couple of Convoy S21A (Osram CSLPM1, 6Amp driver)
I notice that when on Turbo mode, one of them has become a bit “purple-bluish).
I decided to check tailcap current measurement.
One of them measures almost exactly 6.0 Amps, this one is fine.
But the other one measures around 6.5 Amp on max, this one has a slightly purple-bluish tint when set to max brightness.
So definitely, there is some variance, and 0.5Amp difference of one sample to another sample will probably affect runtime measurements too…
My early batch of red ‘5A’ drivers with 12group ui pulled 4.3A. This honestly was a nice thing for W1s and SST20w as there isn’t much flux to gain above that. Haven’t checked my 6A driver readings, but the 8A buck in my L21B sft40 was only pushing 6.7A or so at the LED. This is one of the reasons my sample wasn’t hitting the throw numbers. I swapped in a PM1 since that is the sweet spot for that emitter and it’s a nicer combo at that amperage.
It would be nice to have more consistency and accuracy to the driver outputs.
Most importantly, it’s almost 4 times smaller (by package) and 2 times (by die size). Do you need a further comparison?
There is this test from Clemence:
It follows very closely the output curve of 4xE21As 5000K 9080 tested by maukka, which makes sense because it’s 4xE21A in one package, except it’s much better since the dies are packed very close and there is less phosphor around the dies giving much more throw and better beam uniformity.
In term of efficiency, B35AM (6V) vs FC40 (12V), both 5000K.
1A/0.5A : 121 vs 108.5lm/W
2A/1A : 108.6 vs 101.5lm/W
3A/1.5A : 94.9 vs 92.6 lm/W
4A/2A : 79 vs 86.6 lm/W
I’m using Djozz test with a 0.93 factor to his djozz lumens to get maukka lumens, given how close the B35A test is to 4xE21A I’m assuming it’s close to maukka lumens.
Though it’s important to note that Clemence’s test is done on his very high performance MCPCB.
Simon, what is the footprint/gasket size for Nichia B35AM? is it 5050?
3.65x3.65 mm package. IIRC dual pad, only anode and cathode.
Thank you Agro. That’s quite an odd size hahaa
Any idea when this will be available in something like the M21B or perhaps the C8+?
Is there any difference between the XHP50.2 drivers other than size?
Any spectrum analysis on these yet? :innocent:
Just the size
It is just the voltage is different.
i have drivers for B35AM,
If I order B35AR, i have to produce another driver for it .
B35AM R9080
GT FC40 R9050
the same footprint as the nichia 276A
it can be installed in M21B ,
i dont have suitable driver for C8+ with B35AM