Thanks for all the input.
I haven’t duplicated a connection-break, but now it simply won’t charge all 4 cells at 3a simultaneously. It instead finds one of the cells to be almost full and reduces current accordingly.
It doesn’t matter which cell, and it’s done it in more than one slot. I just picks a cell and decides it’s almost full.
Do you see the problem if you reduce the charge current? Maybe to 2.5 amps?
Not making excuses for the MC3000, just wondering if it actually is an input current limitation. If so maybe it is the power supply.
What are the ratings for output (volts/amps) on the one that you have?
Of course that reduces the current for the cell and the overall load. If you kill the 4 cell charge and check that single cell, what does it show as voltage and mA (charge rate)?
Just trying to help you figure out what is going on.
I gather that the cells you’re currently trying to charge are 4 brand new cells. Two questions just for the heck of it are:
What are the internal resistance measurements being reported by the MC3000 for these 4 new cells you’re currently using and experiencing strange results with?
Do you experience similar strange results when charging other known-good 18650 cells (using the same ‘program’ / parameters) that you’ve previously been using up to this point?
The ir of all cells are reported to be similar (8-10). Voltage however, does not match on the charger, though a multimeter reported them to be the same.
I haven’t tested other cells yet, I haven’t drained any other Molicels low enough yet. That is my next thing to check though.
It behaves the same when set to 2 amps as well.
The power supply came with it, it is new. I am away and can’t actually check the numbers currently.
MC3000 Internal Resistance Measurement Consistency Check
After recently acquiring an MC3000, I did a very simple, small-scale test to get a sense of its IR test measurement consistency / repeatability under real-world conditions, when used as intended / directed by a typical user, while observing best practices and introducing no other variables. I’m posting the numbers in case any readers might for various reasons find them interesting or useful.
My unit is box-stock, and has not had a screw turned. It self-identifies as being FW Version 1.15 and HW Version 2.2. It was purchased directly from an extablished U.S. distributor.
For my test I used a single brand new 16340 (‘protected’) cell which had only 2 discharge cycles on it to verify its health, was then charged to 4.2V and had then sat for several weeks, which allowed it to settle and reach a stable resting voltage. I chose the 16340 as the smallest cells I use, and to minimize cell terminal contact spring pressure, which if anything would be ‘worst case’ for the test / contact resistance.
I inserted the cell into Slot 1, waited for a stable display reflecting that, pressed the ‘UP+DN’ keys to initiate an IR test, and recorded the result. I then removed the cell, and waited for the display to reflect the empty slot status.
I then re-inserted the cell into Slot 1 and repeated the above process 3 more times (for a total of 4 measurements), then repeated that entire process for each of Slots 2-4 (for a total of 16 measurements). Those numbers are listed as Tests 1-4 in the chart. When I shut the unit down and sat down to actually look at the numbers, I decided I needed more data points. I then got the unit back out and repeated the entire procedure above, thus providing me with 32 measurements distributed across all 4 slots. That second set of numbers is listed as Test Numbers 5-8 in the chart.
Looking at the numbers for a couple of minutes, I realized there was no need to grab my trusty old HP11; the numbers completely spoke for themselves and answered the question which prompted me to do the little test to my satisfaction.
I note little pattern, aside from perhaps a seeing a slight trend as I look diagonally down the chart towards the area of lower right corner, possibly seeing 1 or 2 of the ‘higher’ numbers (by 1 or 2 counts) perhaps appearing more in that area. Those are the last measurements taken on the second pass. If so, that possible slight trend would likely be accounted for by the fact that the MC3000 uses a couple of discharge pulses to do the DCIR test, and while not a big thing, the little cell did see rather a lot of them during the testing, and they do add up. I believe that as SOC decreases, DCIR increases, which could have shifted the average measurements higher by a fraction of a ‘count’ / milliohm or so. I’m talking about mouse attachments here though. In practical terms, these numbers don’t require much analysis on their surface. Note that possible factors behind the numbers is another matter entirely though.
I may have some additional comments regarding these numbers and what’s behind them later, but for now will just let them ‘speak for themselves’. Any comments or questions about what I did are welcome, and if anyone has representative data from any other unit / device which would correlate / be comparable with this data (gathered using the same real-world methodology apples-to-apples), that would also be welcomed. “What’s in your wallet?” 
Thanks for this information. I don’t have a MC3000 charger. My many years of battery use starting with RC cars in 1990 gives a slight perspective of performance.
These numbers are very consistent and repeatable. I don’t think one could ask for better performance with a charger in this price range.
It would be interesting if the numbers are similar with larger capacity cells.
Quick heads up- My charger only has issues with button-tops. It charges all flat-tops simultaneously at 3-amps, no issue.
So button-tops appear to introduce significant resistance. I charge button-tops at 1 amp or so now.
I am interested in these result, as we have discussed in another thread.
I will, at some point, try a similar test with my over 12 year old MC3000.
I specifically am curious about comparing readings from cells that I use the most. Meaning 18650 and 21700 cells. Without protection.
I want to see repeatable results with spring tension involved. I also have some doubts about how the protection circuit will impact readings I want to remove that variable for one thing. But, also because I very seldom use protected cells. Again, I am more interested in testing that reflects my typical usage. In addition, I want to use cells with lower IR, that better represents the range that I usually see (maybe 20 mOhm to 80 mOhm or so). I start being skeptical about non-protected cells health that are in the 150 mOhm range. I am curious to see how repeatable the results are with lower values where a single digit of change is a higher percentage of the measured value.
In any case, I definitely appreciate your efforts. The fact that you did this may give me motivation to do my own, more comprehensive, testing.
Thank You!
PS, the repeatability range that you got is great. Well within what my requirements are.
I took a bunch of measurements today. I used a Vapcell 26650 K62. Mostly because it was handy and I had fully charged it yesterday. But also because it fits snuggly in the bay and makes good contact. My older MC3000 does not have the updated “bumps” on the contacts, so some cells are fiddly. It showed at 4.16 volts after sitting for about 24 hours. Sorry, I don’t know how to make one of those nice tables. I am leaving out units for ease of typing, but the are all in mOhms as displayed on the MC3000
Slot 1 38, 35, 36, 36
Slot 2 36, 35, 35, 33
Slot 3 36, 37, 34, 35
Slot4 36, 35, 35, 34
SO my results are not as close as what @Desertcat observed. The widest spread was from 33 to 38 across all of the slots. Which is about 14% of the average of all of the readings. But variations within a slot are better. The max being about 8.5%.
This is close enough to do what I need … observing IR over time to help monitor cell health. Would I like it more if I could get closer readings? Certainly, I would love it if it could be kept under 5%.
So this series of measurements is much better than some of the worst cases that I remember where consecutive reading varied by 20% or more… But those may well have been with some of those cells that have to be “suspended” between the contacts to work at all.
Just checking in to see if anyone has perhaps done any further investigation into better understanding the rather serious issue (as documented in post #117) w/ IR measurements using V1.18 firmware(?). Thanx!
I just know that there are numerous posts here and at CPF that document the change in readings from 1.15 to 1.17 and 1.18.
At one point @dmenezes said he was able to flash back and forth between revisions. He said that he intended to do some comparative measurements between them.
But I haven’t seen him posting for quite a while. Though it is summer where he is and he seems to do lots of outdoor stuff when the weather is good. So we may have to wait a bit for results.
Honestly, my two units have been working well on 1.15. So I have no motivation to upgrade the firmware.
I think you said that you bought a new model MC3000 recently. What firmware version is loaded on it? Have you seen a change in IR readings from what you are used t seeing with the older unit?
Someone HELP!! out of the blue error code MCP 3424-2 ERR popped up on my charger right after i took a battery off it won’t let me do anything has anyone ever encountered this before??
I couldn’t find the error code listed anywhere… except in a MCP 3424 micro controller discussion…and then not this specific error.
Have you tried pulling the power supply and holding the enter key for maybe 10 seconds…then plug in supply and see what it does?
Also, you might try flashing the latest firmware if the above does not work.
I did and nothing gives the same error i will try a firmware update but this is brand new and has version 1.17 in it that’s what’s so confusing thank you I’ll keep u posted
The Mc3000 will allow either newer or older firmware to be flashed. I believe the 1.18 is available so that is an option.
It may be easier to explore a warranty replacement as it is new. Good luck with it.
Thank you for the insight and help
So i heard back from skyrc they said its a faulty component i have to return it
Well it sounds like you will get a new one. PIA on the shipping etc.
I think that the MCP 3424-2 is the microprocessor that they use. I guess it died…
Yes - rotten luck. I have been ‘following’ the MC3000 since before it came out and never heard of such an error.
But sounds like you will get a new one - as you should.