[Technical Analysis] Unlocking the Wurkkos TD07: Performance Gains with Molicel P30B vs. Alarming QC Issues

Scruff · March 11, 2026, 6:02pm

1. The Project: High Performance on a Budget

My goal was simple: find a compact, high-intensity flashlight for my girlfriend to carry in her purse for self-defense (due to high occurance of motorcycle snatch theft after workshift in her area).

Criteria: High candela, compact form factor, and budget-friendly.
The Choice: The Wurkkos TD07 seemed perfect, but research revealed a safety bottleneck with the stock 18350 battery.

2. The Engineering Fix: Molicel P30B + Extension Tube

The stock 18350 (1100mAh) is rated for a 1.1A max discharge (according to the datasheet), while the TD07 pulls 4.5A~6A on Turbo. Furthermore, the 2A internal charging is way too fast for this small cell.

The Solution: I sourced Molicel P30B cells (safe up to 3A charging and 30A discharge) and extension tubes.

“Data-Driven Safety Comparison:” I have attached the official Molicel P30B Datasheet highlights below. Comparing these specs to the TD07’s draw confirms why this setup is technically superior for long-term safety and performance. The stock cell is pushed to its absolute limits, while the P30B operates well within its safety margins.

(https://www.molicel.com/wp-content/uploads/Product-Data-Sheet-of-INR-18650-P30B-80111-2.pdf)

The Struggle: Olive tubes were OOS in Asia. I had to buy them at a premium from Amazon Europe and ship them back to Asia. Total cost: €40/unit.

3. Performance Results (Tested with Sekonic L-308X)

I calibrated my tests using a Convoy S2+ (SST40 4000K, 8*7135) as a baseline. I did three sessions of test at differents distances (3, 5 and 9 meters). I saw disrepancy in the candela (90kCd, 72kCd, 58kCd) value but consistency in the ratio:

The “P30B Boost”: Consistent +22% gain in Candela on Turbo and High (and 8% for Medium) compared to the stock cell.
Turbo Stability: While the stock battery sags after 30s, the P30B maintains its output. After thermal step-down, a restart shows the P30B recovers almost all its initial punch.
UI Mapping (P30B curve):
- Steady Green: 100% to ~80%
- Flashing Green: ~80% to ~50%
- Steady Red: ~50% to ~25% (Performance now matches the stock 18350)
- Flashing Red: <25% (Charge immediately)

4. The “Dealbreaker”: Manufacturing Defects

I bought 3 units. When I unboxed the factory-sealed backup (to test it as a reference for the 18350 battery), I finally noticed the foggy reflector and was shocked:

Reflector Outgassing: 2 out of 3 units have foggy/cloudy residues. My 4-year-old Convoys are still crystal clear.

Structural Defect: One unit has a physical “bump” on the reflector edge.

https://imgur.com/a/6lcThFf

5. Customer Service & Conclusion

Wurkkos Shopee support was dismissive: “If it can be lit, it will not affect the follow-up use”. It is heartbreaking to spend weeks on technical optimization only to be let down by basic Quality Control and poor support.

This is technically the best price-to-performance ratio under $50, but the current batch quality makes it very hard to recommend.

Wingman · March 11, 2026, 8:57pm

I’m not doubting you, but I’m just curious where you found the datasheet? Via Wurkkos or did you remove the wrapper/sleeve and figure out which cell was actually underneath?

1.1A max discharge is would be a really(!) low drain battery by the way. Currently I’ve got 2 Vapcell M11 (1100mAh) 18350’s, with respectively 9A CDR and 10A CDR (V2.0), but maybe take that with a grain of salt. Nonetheless, they’re higher drain than most other 18350’s, which is why I bought them.

By the way: what style of wrapper/sleeve does your Wurkkos 18350 have? I also bought a TD07 last year and the 18350 that came with it, was in the full traditional ‘Wurkkos style’ like the rest of their batteries. But when I look at their own website right now, the photo’s of the separate 18350’s they sell now show “generic looking” 18350’s with only a little bit printed on them. No clue if this might be a really old photo or that they switched cells?!?

I completely agree with you on that! Unfortunately I’ve also seen this on other Wurkkos and/or Sofirn lights with small batteries inside (for example 14500 or 16340 batteries) which charge them with way higher current than is good/healthy for them.

Damn, that really sucks!

I haven’t used my TD07 for quite a while now, and didn’t remember seeing foggy/cloudy residue on the reflector, but I just checked it again, just to be sure, an my reflector is fine. (Although that is only a sample size of 1 ofcourse.)

I just looked it up and I bought my TD07 on August 7th, 2025 from the Wurkkos Official Store on Aliexpress. It was basicly an impuls buy, because after applying a store coupon, coins and a promo code and it came down to € 11,56 (from € 34,79), so I wasn’t able to resist it for that price.

NoEconomist · March 11, 2026, 9:54pm

Take a look at this post I made a few months ago.
https://www.reddit.com/r/flashlight/comments/1ot268k/td07_world_record_19_km_6233_ft/

smurcod · March 12, 2026, 12:01am

Take the bezels off and very carefully clean the reflectors? It looks like it can be disassembled.

Scruff · March 12, 2026, 12:07pm

Hi, I found the datasheet right here on BLF: Sofirn/Wurkkos 18350 1100mAh Datasheet. I didn’t remove the wrapper; my units are the standard orange Wurkkos branded ones. While I can’t be 100% sure of the exact OEM source, this datasheet seems to be the community consensus for this cell.

I ordered the TD07 first, and then saw a YouTube video about original battery safety concerns. At first I wanted the Vapcell M11 V2.0, but I ended up choosing the Molicel P30B for a few key reasons:

Charging safety
Several community reviews (like ZeroAir) mention a ~2A internal charge rate. The M11 is rated for 1.5A “Quick Charge”, whereas the P30B handles 3A “Standard Charge” (up to 9A). Since this light is for my girlfriend who only uses internal charging, the P30B offers a much safer overhead.

Performance & Ergonomics

• Internal resistance: The stock cell’s resistance is roughly 3.8x that of the P30B and 2.6x the one of the M11 V2.0. This explains why the Molicel stays “punchy” even at lower state of charge. For comparison, Flashaholic also measured a 22% boost in High mode and a 15% boost in Turbo using an M11, which perfectly aligns with the performance gains I observed.

• Thermal & Grip: The extension tube gives more surface to cool and arguably better handling in a high-stress “snatch theft” scenario.

• Discharge curves: I made a quick overlay (50% transparency) to visualize the difference. Even ignoring the capacity gap, the voltage stability of the P30B is on another level. It ensures she has maximum “dazzling power” available even if she forgets to top up the battery frequently.

Pricing
The P30B setup was actually supposed to be cheaper in Malaysia:
Vapcell M11 V2.0: RM 45 (~€9.50) VS Molicel P30B + Extension Tube: RM 24 + RM 8 (~€6.80 total), though scarcity made me pay ~€15.3 in the end.

I bought mine in February 2026 via the Shopee official Wurkkos Store; it’s likely a different batch with poorer Quality Control. I’m quite envious of the pricepoint you got

That 2 km reach with the Osram NM1.F1 is insane! Since you already swapped the tail spring for a 10A Convoy one, I’m curious: have you considered trying a Molicel P30B with an extension tube? With its much lower internal resistance compared to a 10A 18350, you might squeeze even more performance out of that FET driver, if you don’t mind the extra length!

While the shared mod confirms the head is accessible, I have decided not to open or touch it for now. As you can see on the photography I uploaded on Imgur (https://imgur.com/a/6lcThFf), the reflectors arrived in a deplorable state.

Between the heavy “outgassing” haze and the physical pitting on the edges, the optical path is already compromised. Attempting to clean such a delicate surface would likely add micro-scratches, making a bad situation even worse.

I don’t want to provide Wurkkos any excuse to void my warranty by claiming I damaged them myself. Since 2 out of 3 units are defective, I am expecting a professional resolution—ideally replacement heads that actually passed Quality Control.

PiercingTheDarkness · March 12, 2026, 12:31pm

That haze is actually on quite a few, also on many Convoy models, it has little to no visual range impact, I’ve measured and it was next to no difference. The pitting however, could be something if it bothers you, but only if you can visually see it in the beam, otherwise they won’t replace it. When buying from these cheaper brands you need to be prepared to accept small imperfections. Take it from me who sells Convoy, Sofirn, and Wurkkos, I QC check thousands of units. I’ve personally never claimed warranty for this type of thing, but that’s me.

You’re well within your rights to complain though, but not sure whether they’ll offer replacement for that.

Scruff · March 12, 2026, 1:34pm

Thanks for the updated feedback! It’s actually very helpful to have your perspective. Since you check thousands of units, I’m curious about your experience:

Proportion: Is this a “2 out of 3” situation for most current batches, or did I just get unlucky with this specific lot?
Aging: Have you noticed if this haze/pitting remains static or if it degrades further over time (oxidation, etc.)?

Regarding the ‘budget’ expectations: I bought 4 Convoy S2+ units three years ago, plus a C8+ and a C8 UV. I took them to tropical forests, flood waters, beaches… and every single one of those 6 reflectors is still crystal clear today. This is why I found the state of these TD07s so disappointing. It feels like a significant drop in QC standards compared to what budget brands used to provide.

Actually, following your suggestion to look closer at the beam, I noticed that the “lottery” goes even deeper than just the reflectors. Looking at them side-by-side reveals a total lack of consistency in assembly:

Unit 1: Has the haze and a greenish, irregular beam. (I wonder if the tint shifted after a few heat cycles, as the only ‘white’ one I have is the one I barely used yet).
Unit 2: Has the haze/pitting, but the beam is perfectly circular, white, and centered.
Unit 3: Has a clean reflector, but still has a greenish, ovoid/irregular beam.

Even the extension tubes have machining variances (one has an off-center inner bore). It seems every unit has its own unique ‘personality’ from the assembly line.

I agree that for €25, one shouldn’t expect a custom-shop finish, but I’ll still reach out to Wurkkos. Not because I expect a miracle, but to let them know this specific batch is quite ‘messy’. If nobody speaks up, the standards will just keep sliding.

PiercingTheDarkness · March 12, 2026, 2:06pm

Actually, I need to correct myself, the hazing is more something I’ve seen on the Convoy C8+, L21B, 3X21D models the most, that’s mostly what I sell from Convoy. Sofirn and Wurkkos have been clean, but I have seen quite a few reporting hazing on various models either on Facebook groups or Reddit. My last bulk order was Nov, so it could very well be some recent QC issue. I guess it to be caused by flux residue fumes when turning the light on turbo the first time at the factory. I’ve tested models from Convoy and the range wasn’t really impacted, well not something you’ll visually notice.

If there’s noticeable beam differences that you aren’t happy with, then you may as well try get a replacement. I’ve had a rollercoaster with Convoy when it comes to reflector batch issues over the years, so this haze is very minor in comparison haha. If haze and pitting were all that Convoy had wrong I’d be so extremely happy. Most users on here and reddit will say it just comes with the price… I’ve just come to accept it.

NoEconomist · March 12, 2026, 3:41pm

This LED is designed to run at a maximum (already overclocked) of around 5A, but they are quite resilient to high currents. I believe that with the battery I installed, I am already exceeding the emitter’s limits, so I probably wouldn’t get any more performance with another battery

Wingman · March 13, 2026, 9:23am

I completely missed that topic; thanks!

Damn, those are some really weird/unusual specs in the datasheet of that battery! It’s already ‘pretty rare’ of a cell having only a max discharge current of 1C, but it’s even more unusual that the standard and max charge current are identical to the discharge currents. Certainly when you have a (really) low drain cell, the charging current is (in comparison) equally low(er).

Scruff · March 13, 2026, 2:32pm

If it’s flux residue, it’s not just an aesthetic ‘haze’. These fumes are often acidic or contain halides that corrode the vacuum-deposited aluminum over time. My concern is: will these reflectors be unusable in 2 years?

Since you QC thousands of units, are you following a specific AQL (Acceptable Quality Level) standard like ISO 2859-1 or MIL-STD-105E? For a batch of 1000 units (Level II, AQL 2.5), the standard procedure is to check 80 units (n=80). If more than 5 units show this haze or pitting, a professional QC would typically reject the entire batch (Ac=5,Re=6). Finding 2 defects in my personal sample of 3 suggests a failure rate far beyond any industry standard. It would be very interesting to see a real statistical audit on a recent batch.

True, but won’t 5A at above 4V (for 100mAh) get even better results than 5A dropping fast (after 50mAh) under 3.75V? Even with a good cell like the Keeppower 10A, the voltage drops much faster than an 18650.

Exactly. If the stock battery is really a 1C cell, charging it at 1A via USB-C is pushing it to its absolute thermal limit. Does your Vapcell M11 units heat up abnormally during internal charging? I’m curious if there’s any safety margin left at 1500mA, or if a dedicated external charger is mandatory for these small cells.

PiercingTheDarkness · March 13, 2026, 3:37pm

Nope, I just do my own thing, it’s the most thorough check you’ll get though, but I’m not comparing the beam of each, I just measure range and lumens, if it meets specs or matches older stock within X percentage, there’s nothing to complain about. If there’s something really off, like a bad hotspot pattern or general beam, then I’d complain and request a replacement, I’ve only faced beam issues with Convoy though, to the point of it being a problem.

I just price things down or sell at cost if there’s any issue. Would I decrease the price for haze? No, I wouldn’t, not with my measurements and the price of these brands. If a customer complained, then I’d offer a discount or offer a replacement. I’ve never had a customer complain though.

I check:

Lumens
Range
Visually for scratches etc
Tighten retaining rings, if any
Check parasitic drain
Cell voltage
Charging voltage and current
Clean the LED surface with alcohol
Confirm mcpcb screws are tight and any others
Use alcohol to clean all contacts
5-10 min burn in test on turbo

Wingman · March 17, 2026, 3:36pm

I haven’t actually tested that yet. Regardless of it getting warm, the M11 (V2.0) is still only a 1100mAh cell and personally I always prefer to charge a cell with a current below 1C, just for longevity, so when a light charges it with a current above that, I always prefer an external charger, but that’s just my preference.