[ 40,000 Lumen Manker MK38? - A Review and Teardown ]

Thank you for digging into this more. I appreciated the battery pack and charging implementation information most specifically.

I’m not surprised there’s a premium on the in built pack, based on the tear down.

I passed during the sale because I wanted to fully understand the light. I regret missing the deal now, but not waiting to be informed before purchase.

Now that I have details, it’s on my want list.

4000K CRI90 sounds nniiccee. I’d definitely get one more if they ever launch it.

However given its Manker, not Emisar or Convoy, I doubt they’d ever do that…

Thanks for the review. I was wondering how the battery pack was constructed to see if it would be possible to build a replacement pack without sending it back to manker, and it looks like it can be.

Thank you!

Was there a o-ring/seal between the end cap and the main body? I could not spot it in your pics, but I am sure it was there.

I’m happy to see the charging board seems to look reasonably well done. I also like that they used a water-protected USB-C slot and a additional flap for dust/debris proofing - I honestly expected they just rely on the flap, happy to see I was wrong.
My MK38 died during it’s 4th charge (something must have shorted on the charger circuit) , the light got very hot (from the bottom up) and the smell of magic smoke was present.

As I sent the light back to be replaced under warranty I did not open the back to have a look inside, although I would very much have liked to find out what exactly failed, and especially why it failed. I really hope it just died because one component failed instead of indefinitely charging the cells to oblivion, because that would not have gone well had I not pulled the cable.

I’m also happy to see that they did not just completely pot everything. While this offers the benefit of mechanical protection it makes a future replacement of the cells impossible. I agree that they could have done some mechanical protection for the bigger elements on the PCB though.

We shall see whether they used a BMS for balancing on the cells. I really hope they did. Even if the cells were well matched to begin with, the chance of drift becomes real in high-drain applications. I really would not like them relying on matched cells in the long run.

Overall I’m still happy with what I am seeing so far. I really appreciate your teardown and the pictures a lot!

Manker is afterall trying to sell flashlights. If there are enough people interested, I'm sure Manker would create such a version. I'd have preferred that over my 5000K version too.

It looks like Manker is trying to do the right things here - yes there is a nice oring seal on the back. One improvement they could make would be adding some o-ring grease. So far all other entry points, ports, screw holes etc all have o-ring seals.

I'm sorry to hear that and I'd be curious to find out what was the cause of the problem. Did you use the included USB-C charger? As I suggested in my post, I'd like to have seen an in-line fuse to the charger section, which I plan to implement on my flashlight.

So far from what I see, it looks fairly reparable. Manker offers a 5 year warranty if the product fails within normal operation, so it would be useful to Maker to design the product to be repairable. Again, I'd like to repeat that Manker states: "DO NOT disassemble or modify this product which can cause damage and will void the warranty.", so I wouldn't recommend anyone disassemble their MK38. In addition, due to the cells inside, any mistake during disassembly could also lead to a short and cause a fire and safety hazard. Proceed at your own understanding of the consequences.

Agreed with that line of thinking; I hope there is some sort of BMS too but I'm not holding out for it. Thanks for your comments Pobel. I'm glad they are at least slightly useful.

they are very useful indeed.
Normally I do many repairs myself and I am also absolutely no stranger to making or repairing Lipo/Liion Packs.

But in this case I have warranty, the light is expensive and only days old, so I naturally do use that warranty, although my curiosity couldn’t be satisfied. The urge to bave a peek on what went was huge.

I did not use the included charger, but with USB-C PD any charger should suffice. The charger I used charged this light before without issues and still charges all my other devices nicely.

Excellent theme for a review, and I’m looking forward to your updates.

By the way, if you encounter elements of the disassembly that are not immediately intuitive, it would be beneficial to anyone who may in the future modify this light to document how it comes apart.

Hypothetically, it could be XHP70.3 HI, 90 CRI.

Since this isn’t a thrower, I mention HI more for the typically lower tint shift across the beam than for the smaller effective light emitting surface. On the other hand, the HD has a little bit higher luminous output.

To each their own. I agree, a max lumen low cri option makes a lot of sense on paper and for some S&R, but as an enthusiast without any real need, 20k lumens of 95cri is >> than 40k of cree 67cri

I think I’d rather have the low cri with higher output and better efficiency. Yah, warm fuzzy tint is nice, but in a light like this it’s a bit superfluous. Besides, xhp70.2 even if high cri doesn’t give a good beam anyway.

The 4600 (measured) CCT xhp70.2 Thrunite TN50 looks real nice and still has over 17,000 (measured) lumens and good battery life.

Thanks for the patience everyone; I've added more details and photos of the driver. Enjoy, and thanks for reading.

Fantastic review.

Question: Running on turbo at a distance of 30cm, how long does it take to melt a quarter-pound stick of room temperature unsalted butter on a standard ceramic butter dish?

Thank you. I'm not sure about that, but it begins smoking any remotely dark-coloured paper held up near the lens on turbo almost immediately. You definitely want to make sure it's on lock-out mode (4 clicks) when transporting. There is no physical lock-out switch though, so I'd set the single click on to the lowest mid-mode (~900 lumens).

Why are lights with multiple individual cells connected in parallel more dangerous than lights with multiple individual cells connected in series?

I am new to this hobby and have several multi-cell lights with 3P and 4P battery configurations and some of those contain built-in charging ports.

I have avoided purchasing multi-cell lights with batteries configured in series since I was under the impression (apparently a false-impression) that series battery configurations were more dangerous since they do not keep all of the batteries at the same voltage.

With my limited understanding, I assumed that an individual cell (even in a matched set) would be more likely to fail over time in a series connected battery configuration (than in a parallel connected battery configuration) because in a series connected battery configuration, an individual cell could be discharged more (potentially even over-discharged) than the other cells while using the light and an individual cell (the same or a different cell) could also be over-charged if charged in series with the other cells if a "Balanced Charging” mechanism was not implemented.

I also wonder, if an individual Li-ion cell failed catastrophically while inside of a multi-cell light, would this cause all of the other adjacent cells to also fail catastrophically (from the intense heat created by the initial failure) whether the cells were connected in a series or parallel configuration.

Thanks again for this interesting thread. I have also enjoyed reading your posts from other threads.

To begin, I'm not a battery expert and I'm sure that there are lots more people out there who are a lot more knowledgeable about the topic so please correct me if I am wrong and I'd love to learn more about this topic myself.

There is nothing fundamentally wrong with putting battery cells in series or parallel; these is often done in the industry. Electric cars are a great example. However, typically cells which are assembled into packs are often from the same batch and are specifically binned to be matched. Like you mentioned, ideally S packs will have a BMS to ensure charge balancing and pack monitoring as well.

For parallel individual cells, the main problem I see is that it is very easy for batteries to be mismatched by the user - these could be batteries from different batches, completely different cells, or cells with differing states of charge. Typically the cells used are individual cells which have NOT be connected together prior to being installed in a flashlight. It's not too difficult to see a situation where a user could mistakenly put 2 full and 1 empty cell into a 3P flashlight, and cause a fire hazard as the full cells try to charge the empty cell. This does not apply to a P battery pack that cannot be taken apart - such battery packs do have benefits since they are inherently balanced (which is not the case for series cells).

That said, even in situations where the cells are the same, the user has no easy way to determine the specific characteristics of each cell, for example, internal resistance. Even a small mismatch in internal resistance, especially at the huge power draw rates in modern flashlights, can cause significant acceleration in the degradation of the cell.

There are still concerns to be said for individual series cells, but these can be mitigated with a BMS (which ideally should be implemented). For a flashlight of this power, a properly designed battery pack is the safer and more efficient option.

Thanks for the clarification.

I'm definitely not a battery expert either, but have become a frequent user.

Hopefully I will not have any adverse battery issues with my multi-cell lights since I have worried enough about the potential dangers of Li-ion batteries to become somewhat knowledgable, but still feel safest with single-cell Li-ion lights.

Also, your website listed in your signature is fascinating. Thanks for sharing those projects.

loneoceans nailed down most of the issues. In this hobby most series set ups don’t have a BMS. In parallel BMS isn’t needed as the cell are forced to balance. Biggest concern is picking the right cells, in the same batch, with the use, at the right voltage. Most recommend buying the cells together, marking them and never using them for any thing else. If my flashlight maxes at 10 amps per cell then I get at least 15 amps cells. Buy only the same cells. Mark them with the light and 1,2,3,etc as needed. Place them in the light fully charged as this keeps them all at the same voltage. We are safe. Mixing different cells, with different cycle counts, with different capacities. Danger is there, the more variables the greater the risk.

Loneoceans thanks for the review! Your review leaves me with the feeling that Manker is running this light past max. That this light really needs twice the electronics to have a longer life. That 4 batteries instead of 3 would be safer and more durable.

Thanks for the write up! If this came with XHP70.3 HI 4000k 90CRI, I would buy instantly

Having cells in parallel is inherently safer. But it has its drawbacks.

The current (from battery to driver) is much higher, thus you will need to design all components (springs, cables, switches, traces) with this current in mind. It can also be less efficient, since more current means more heat related losses, because in a flashlight you will always have to compromise because of weight, cost, space.

Its always a very good practice (and recommended) to handle devices with several individual cells like Texas Shooter has described above.

- get matched cells (same type, same batch, same age)

- always use and charge them together

  • only use them in this light

Parallel configuration:
Even if you have cells from different manufacturers or even (slightly) different states of charge, it will most likely work. You will lose performance and you will tax individual cells beyond what is recommended.
The load distributes in relation to the internal resistance of the cells, so the “fitter” cells (low IR) will have to do the work harder while the “not só fit” cells (higher IR) will carry less load. If there are differences in SOC to begin with, the cells will balance. If you mix empty and full cells, the balancing currents will be extremely high in the first seconds.
So always use matched cells! Do not mix cells!

Serial configuration
is a different story. It offers potentially more efficiency, as you have lower currents for the same power request. All cells will have to carry the exact same current. If one cell has worse IR than the others, it will sag a lot more and this can potentially end catastrophic if the load is not reduced.
If the cells are of high quality and matched to begin with, cell drift can be minimal. You might get away without balance charging (for a while). Cell drift becomes evident with higher loads. So if you often run cells at the max of their rating or beyond, then you absolutely will need to balance them.
Balancing just means, that during charging the voltages of all the cells are equalized, so that they all have the same voltage. This can either be done via BMS (battery management system) which is permanently fixed to the battery or it can be done by the charger (either by charging cells individually or, in the case of a pack, using the balance leads of the pack)
You MUST NEVER mix cells in a serial configuration. In contradiction to the parallel setup this will get you into trouble very quickly. The “fitter” cells cannot work harder as we experience in a parallel configuration, since all cells carry the exact same load.

It was foggy yesterday so I was able to do a quick beam-shot.

The MK38 with XHP70.2 LEDs is quite floody (as expected) with a bright central spot - this is my preferred use case. When using it to walk around at night, the XHP70.2 version excels for wide-illumination of open spaces (e.g. good for group events, work light, etc), or for enclosed spaces such as hiking in a dense forest. However, it's wide-nature is not ideal for throw (as expected), and I wouldn't consider the XHP70.2 an effective search light - the SFT40 should be far more ideal, and may even appear more impressive, as throwers often do.

Thanks for the nice review with teardown. :sunglasses: :+1:

I wanted to know what this light looks like inside. I already asked Manker on AE if they would sell the MCPCB and the reflector but they said no.