thanks for the nice review
Thanks for the thorough review!
This one’s on my list of lights to buy down the road, given its unique characteristics.
Hopefully they can work out any software glitches in an update - then I’d be a lot more keen to pick it up!
The Wurkkos HD20 looks pretty good, though maybe a little too heavy for me, and it only maintains a lot of brightness for two minutes.
I think it's cheapest on AliExpress right now.
https://www.aliexpress.com/item/1005001565181379.html
Thanks for the review. 
Yeah it’s an amazing light but just a bit too heavy for long term headlamp usage, and getting it in the headstrap is rather finnicky. Works fine as a headlamp for short periods though. Produces very nice, neutral light!
Why on earth haven’t Sofirn sorted out their thermal regulation issue!? This behaves exactly the same as the Sofirn SP33 V3.
That’s an instant deal breaker for me, no customer wants to see their brightness jumping up and down.
Why was this not a con listed in the conclusion?
FIX IT
It’s a change over several minutes, I didn’t really notice it when I was using the light out on a walk. If using it up close I don’t think you would be running it so bright to hit the intense thermal regulation. I agree it could be better.
Okay, so probably not as bad as it looks. I’m looking forward to Anduril 2 being used more or just more testing before release.
I have the new SP35 on its way for review so hopefully it’s better.
The SP33 V3 that someone reviewed said it was noticeable out walking so I immediately thought this to be the same.
Lumens are a lot less than advertised on this HD20. I wish Sofirn would get some maukka lights to get a better calibration. Their old models like the SP32 were almost spot on.
Turbo
Both – 1520
Spot – 1070
Flood – 565
Advertised
Both – 2000
Spot – 1300
Flood – 700
Well, it should be according to the runtime chart Barry was so kind to send me.
I also pointed out that i.a.w. gchart's review of the SP33 V3 their new ATR (with thresholds of 45°C, 55°C and 65°C) deserves some closer investigation. I was told they have improved the new ATR to be less "jumpy". To be fair, many budget flashlights can step down but not as many can step up again (automatically) once the driver's temperature has reached a certain value. My assumption is that Wurkkos HD20's driver was in development when SP33 V3 was launched to the market, so neither Sofirn nor Wurkkos had enough feedback from flashoholics and enthusiasts to investigate this issue. There's always a certain delay between finding an issue and actually fixing it with a new generation of drivers.
Yesterday, I took a 2 hour walk using HD20 as a "neck-light", i.e. I wore it around my neck using the headband. It felt absolutely comfortable, not feeling any (over)weight of its big size. Maybe this could be an alternative way to some people who feel concerned about too much weight on the forehead? Just thinking...The floody beam in front of my feet was very useful to locate and evade piles of dog sh... or while following a narrow footpath through the woods. The spot in combination with the flood beam was bright enough to illuminate the next 50-75m in front of me. Indeed, I was impressed by the peak distance of the spot beam. The (previously unintended) triple click to cycle LED modes turned out to be really convenient.
What cell were these readings done on, and I assume fully charged? Manufacturer says they used a 4000 mAh cell, so probably a 40T.
Yeah, I used a charged 40T. Sofirn’s ratings are always off, everyone else has tested the same for their lights.
There are exceptions like their SD05 is spec’d at 2550 and I got very close to this, it’s odd
Yes, but, why aren’t they doing the testing themselves or send prototypes out for feedback. Don’t release it and leave everyone with a problem.
When they test the runtime, surely they would catch this behavior on their equipment.
I'll give them 12% because that's about the difference between maukka cal #'s and previous, past manufacturer's numbers. But this looks more like 25% which means there's something wrong.
1300 lumens out of a XPL V6 is pushing high amps, which these kind of lights just can't do, typically. Notable exception are our BLF designs like the Q8 of course 
I just got a HD20 to review, but rt. angle lights are an issue in my PVC light box thing, so I can probably use a reference source repositioned and calculate based on an adjustment. So far in some simple indoor testing, it looks pretty good overall. Sure it's big and triple clicks are somewhat awkward to switch, but it's nice having flood and some throw in one package.
Couple years ago I developed a dual LED version of Narsil for the Lumintop SD Mini II, which has side mounted aspheric flooder, but in the UI I made some sacrifices in function, but made the switching easier. I modded a FET+1 driver, so used the 0.35 amp 7135 for the side LED, and the FET for the main front LED. UI summary:
- From OFF: click & hold (1H) always goes to the side LED starting at moon level
- From OFF: click (1C) goes to main LED at last used level. If previous use was both LED's on, then both LED's come on at last used level
- From side LED ON, dbl click (2C) turns on the main LED, keeping the side LED on
- ramping works pretty much normal, but on the main LED it doesn't ramp that low (because of the FET), and flickers a little at the low ramping levels
- if you have just the main LED on, there's no way to switch to, or turn on the side LED
The idea is to make use of the flooder at low level output - see where you are walking for example, but give full FET power to the main thrower, but also allow both to be at the same time for looking down a trail, and also seeing where you are walking. The 90 degree offset of the beams helped accomplishing that.
In Narsil, I had no support for 2H (Click & hold) though - that's a simple way to switch between LED's, and as recommended and intended for the HD20.
I just tested the HD20 I got yesterday from Amazon (Prime 2 days) on a freshly charged 40T and got as high as 6.6 amp tail readings. Amps crept up the more I compressed down on the spring, using a clamp meter with the tail off. The battery fits very tight, so good spring compression on both ends.
My maukka calibrated readings was about 1470 lumens at 30 secs on just the spot LED (XPL HD) both LED's!! Initially, the reading actually goes up a little, then didn't drop much at all. This is all indicative of the springs combined with a direct FET driver. I've seen this type of variation a few times before. It's also clearly in the ballpark for 6.6 amps, and the light got quite hot to hold in 30 seconds. So the amps readings, heat, and lumens reading all correspond.
Dunno, maybe I got a new version, or a buggy one, if everyone else got much lower readings.
My mistake - the bug caused turbo to come on on for both LED's, not just the spot!!
Okay, I’ll have to give it another test. Are you certain it was only the spot led that activated?
Do you have a tube or sphere?
As I said above, I have a PVC light box - guess you call it a tube? One of the original ones, before TA was around, before Dale got one. Because a rt. angle light doesn't fit, I took a somewhat similar light with about the same optic width, the SC31 Pro, and used it to calibrate the position and configuration. I held the light centered over the glass, but up a couple inches above the glass - normally the flashlight sits flush on the glass with dark gray foam blocking the rest of the opening. The conversion factor I came up with was adding 5.8%. It's approximate for sure, but if you look at XPL HD LED tests, at 6.6 amps, by TA here, and djozz here, 1500 lumens is well under what they tested.
It could be my driver has an issue in that something that should be throttling back to the FET on turbo is not working, like a resistor that is short circuited.
I haven't seen any amp readings posted on the HD20, so that's the ultimate data to compare. Btw, it's not easy to do an amp reading unless you got a third hand helping. I was able to do it myself in my lap, and able to apply enough pressure to get the spring about fully compressed, maybe as much as when the tailcap is tightened down.
After a bypass I get 6.8 amps and 1680 from both. I can’t wait for my clamp meter to arrive so I can compare. All I can say is I get the same reading on both units and it certainly doesn’t look like 1500 lumens from the xpl spot led. Who knows until more share their findings but ’m certainly getting a lot less amps then you.
I’d like to think yours is working as it should.
I'll sanity check everything again later today. I'm pretty sure it was only the XPL, but I'll check.
There’s a bug that activates both leds on turbo after switching off for 5+ seconds. Only happens with the spot led enabled
Correct. Among other things this glitch has been reported to Wurkkos already and is reflected in my updated UI chart that I posted in Wurkkos's original thread. For reference, I post this updated UI chart here as well. The revised parts are highlighted with red text on yellow background. The glitch you noticed is mentioned at the end of the chart.
Ahh, ok - that was it. It dbl clicked to both on and I didn't even notice that it had. So, with both on, the #'s I'm getting match up well to about 1500 lumens, ~3.4 amps.