Test/Review of MXJO IMR18650 3000mAh (Yellow)

MXJO IMR18650 3000mAh (Yellow)

Official specifications:

  • Capacity: 3000mAh
  • Nominal Voltage: 3.7V
  • Max. Pulse Discharge Current: 20A
  • Dimensions: 65*18.24MM (+/-0.05)mm
  • Operating Discharge Temperature: -10 to 60 Degrees Celsius
  • Standard Charge: CC/CV
  • Cycle Life: > 500 cycles
  • Weight : 65g

This cell is rated at 3000mAh with a rather high current.

The discharge curves looks very fine up to 10A, but 15A and 20A looks like too much.


This cell has very good performance up to 10A, but cannot really deliver 20A.

Notes and links

How is the test done and how to read the charts
How is a protected LiIon battery constructed
More about button top and flat top batteries
Compare to 18650 and other batteries

Thanks for the review. So im better off with NCR PFs for less money?

HKJ thanks for the review. Checking your 18650 comparison charts this battery performs quite well against others for <10A. Depending on the price it may be a good option


It may be the 30Q, but if poor quality cell, you probably will be bad for a company that so far gave good batteries on the market and within a short time, did good name .
It is wrong to label such a battery with false information.

We have to wait and see until hkj reviews the 30Q and the other new medium drain 3000mha cells, before we can draw any further conclusion if MXJO are simply rewrapers or not.

proper description

The discharge curve looks neither like SDI 30Q nor like LG HG2. At least not like the ones I tested.

Interesting so it might not be a 30Q or a HG2, so of these 3 20A 3000mha cells you have tested, what are the strongest for DD FET driven lights performance = 3-10A discharge?

I have tested 2 3000mAh cells: HG2 and 30Q up to 15A. Not sure which is better:

LG HG2 vs Samsung 30Q ……
I thought not worth talking about differences in these two batteries .

The differences that have in their yield , unworthy reference .

When two batteries of the same class are differences in performance 0.02Volts, I do not think anyone deserves to become stressed about what is best …….

H MXJO, is not confident it has the 30Q under the label of .
So we should also consider the fact that residents from its own suit, be some battery with 10 A performance, but the most common way to call it 20A .
From the picture of the top , I see similarity with the Samsung, but the company has other batteries with lower yields .
It can be one of these and not necessarily the 30Q ………

Agree, the HG2 & 30Q are very similar.

I would really like to know how they ages after a bunch of DD cycles, at for example a 5A discharge. To bad nobody test those aspects of the best (or maybe i should say strongest at the best price like these two candidates here) but similar cells.

Sorry, I don’t have such data for the 30Q, but for the HG2:

Though this isn’t the most precise test (ambient temperature fluctuations from 22°C in the morning to 30°C in the evening being the main contributor to fluctuations), there is a clear tendency: downhill. From 2.921 Ah in cycle #3 (first 5A discharge) to 2.624 Ah in #70 (30 days later).
But this early capacity loss seems to affect only the end of a dicharge cycle. Internal resistance and voltage don’t change. Average voltage even increases, but only because the low-voltage-part at the end becones shorter.
Between cycle #72 and #73 the battery rested 60 days at full charge, July-September, at room temperatures of 25-30°C.
Unfortunately, there was a partial data loss in cycle #73 (discharge after storage).
Comparing cycle #70 (last 5A discarge before storage) and #74 (first full cycle after storage) one can see the irreversible capacity loss during storage.

Hi light-wolff,

fantastic data. Very intriguing
I had read somewhere that you discovered a significant capacity loss with the HG2 after what sounded like a relatively small amount of cycles. This chart makes things very clear.

I’m glad to see that the loss of capacity (which all Li-Ion chemistries suffer from, more or less) does not come along with increased voltage sag (a higher internal resistance). Looking at the 5A graphs - practically the main current we need these cells for - green, orange and light blue are very close and almost overlap in the first 70% of time. After that you want to recharge it anyway for a DD/FET/LD2 light.

Honestly I can live with the loss of capacity. If I need runtime, I use the NCR18650GA or any other LiNiCoAl cell. These cells are for showing off. 8)

You might add the chart to the 30Q thread, as long as no dedicated HG2 thread exists. This is really really valuable and much more precise than most of us could ever achieve.

Amazing light-wolff :slight_smile: i am so glad i asked :bigsmile:

This is great data and so much we can draw from it, i agree with everything HarleyQuin said.

Good to know it is only losing capacity in the low end that aren’t very usable in DD driven lights anyway, i almost always recharge before i get below 3.5v, because of the high Vf of today’s top flux bin cree led’s.

I wonder if all INR’s have similar characteristics as this HG2 have shown. And i wonder even more about how IMR’s cycles now & if they also are losing capacity from the low end first.

@HKJ, you have put 65g weight under the “Official specifications:” but i assume you mean 45.6g? or is the official specifications from the manufacturer really off by 20g from what you reported lower down in the measured specs?

The 65 gram is directly copied from the MJXO website

Ok, strange how they could be off by that much.

They really need to invest in some new scales because that doesn’t really inspire confidence in there official specs.