After swapping the UV emitter in my S2+ with a higher-powered one, I noticed that the ZWB2 filter gets untouchably hot in seconds, which made me suspect that it absorbs a substantial amount of 365nm UV, not just visible light.
I ran a crude fluorescence-based comparison with and without a filter, and arrived at the conclusion that the 2mm-thick filter has a transmission efficiency of only 60%, i.e., it loses almost half of the UV going in.
This conclusion is further supported by 2 sources:
Test of 1mm ZWB2 filter indicating around 78% transmission efficiency at 365nm. For a 2mm filter, the transmission is thus (78%)^(2/1) = 61%.
First review under this 1.5mm filter listing includes a test result that shows 68% transmission. For 2mm, the transmission would thus be (68%)^(2/1.5) = 60%.
So, using a 2mm ZWB2 filter cuts the UV output almost by half. While this tradeoff is justifiable for mineral detection and such, it feels very wasteful, and unwelcome for applications such as resin curing or charging glow items.
Does anyone have a solution that allows visible light to be cut without such high losses? I think a ZWB2 filter with a lower pigment concentration would be ideal, but I don’t know if anyone makes such a thing.
Been a while since I’ve looked into this, Hoya used to be the “go-to” for 365nm pass filters, but I believe now discontinued(?).
There was a person selling UV lights (reddit?) with dichroic filters a while back, I scoured AliExpress looking for suitable material but came up with nothing.
The main attraction with the ZWB filters is the low cost- the Hoya etc material is (at least) an order of magnitude more expensive.
There is a chance your led is emitting different wavelength than the old one, even if specs say otherwise, those black uv filters are not all the same there are zwb-2 and there are zwb-3 that are comparable with different wavelength.
My test already accounts for this possibility by measuring the transmission difference between (1) one filter and (2) 2 filters, rather than bare LED versus 1 filter, so any spectral differences between the LEDs would make no difference in this test. I realize that I said
which is an oversimplification. My apologies.
That is a possibility. I would hope for consistency given that they are all from Convoy and listed as a ZWB2 filter.
Thank you very much for this info! The cost difference is pretty crazy. I was hoping that someone could just dilute the ZWB2 with clear glass–you get a tiny bit more visible light, nowhere near enough to contaminate the beam, but a whole lot more UV.
My worry is that a film doesn’t have a lot of volume to sink the heat, and could burn up or melt quickly. But maybe someone could make it heat-resistant like cinema gel filters.
If your application is to detect weak fluorescence, a filter is still necessary. But for detecting strong fluorescence or for curing UV resin, going without a filter might be the way.
An interesting hypothetical option might be using very thin pieces of the ZWB series of filters in combination with a standard glass lens for mechanical protection…
After reading this thread I decided to test my UV drop in that I got from Lumencraft, and came with zwb2 glass, after running it for 5 min, the glass was still cool to the touch, while I could feel the heat when I put my hand in front of the glass. Then I thought, maybe it was not powerful enough, so I did the same with hank’s 40W uv mule light, also with zwb2 filter, similar result. lots of heat felt in front of the light, but glass was not hot, while the light itself got pretty hot. I know it is not a scientific way, but it is something.
That would be a nice idea! What is accomplishes is essentially equivalent to diluting the ZWB2 filter. Ideally one wants to attach the filter to the glass to mechanically couple them, though I couldn’t think of a good adhesive at the moment.
Thanks for the data, very interesting! How thick is your ZWB2 glass on the Hank?
During a reddit discussion, someone pointed out the possibility that different ZWB2 filters may have different formulations and different performance. It appears that you won the lottery in this regard.
If you happen to have a Convoy filter on hand, could you run a similar test?
Wellp, you only want filters to strip out visible light when viewing stuff like minerals, etc.
I’d never use filters when charging gitd doodads or curing resins or anything. If anything, stray wavelengths might actually contribute and be helpful.
That is a definitely a valid question. Though multiple tests/datasheets indicate that a ZWB2 has an absorption loss of around 20% per mm, which is in line with my observations.
If your Hank filter doesn’t heat up at all despite the 40W of UV power, chances are that it’s not ZWB2 but some better formulation.
I had a feeling something was off with my so called test results, cuz it seemed too good to be true, so I decided to redo my experiment a bit differently, and I got different results, I charged batteries, and i used thermal camera this time, cuz " feelings" is a very subjective thing, different materials with same temp, would feel different, like steel and wood, feel very different as far as temp.
So zwb2 filters do get hot, not crazy hot but they do, maglite was coldest, but imo it is due to lower power and filter being farther from the led, other lights are hanks 40w mule and 9.3 with uv. All pics were taken after about 4-5 min of them running, temp in the room as well as all lights before I turned them on was 29-30c that is a starting point,
Thank you very much for this re-test! 55C is still lower than I expected, but as you pointed out, feelings of temperature are subjective and reflect not temperature but simply rate of heat conduction.
This is a great point. Most UV LEDs have a nontrivial amount of emission past 375nm, where the ZWB2 filter’s transmission falls below 50% per 2mm. In this sense, ZWB2 filters and 365nm LEDs are really not optimized for each other.
There is also considerable variance in spectral width among UV emitters. Seoul CUN66A1G’s datasheet looks pretty tight, but Luminus SST-08-UV-H has a much wider spectrum with a heavy right tail, which will be impacted more.
