SkyRC — IFA 2014 — MC3000 charger-analyzer

HKJ

Flashaholic
I saw something that made me wonder if the MC3000 could be run using DC power instead of AC. Is that possible with the MC3000?

The charger is powered from 12V, i.e. you can easily skip the mains supply and directly feed it 12V.

So based on your comments am I correct in assuming that the power loss could/might be something like 60%? If that were true would that mean that if I wanted the batteries charged to 100% that my power source would have to provide something like 167% of the power?

Why my numbers the total efficiency would be 0.7*0.5 -> 0.35 or 35% of the supplied energy ending in the batteries, that means you would have to supply about 285% energy.
Note: Power is instant, energy is over time, units are watt (W) and watt*hours (Wh or kWh for 1000Wh).

HKJ

Flashaholic
I just checked the numbers for a fairly simple charger (I do have record of power consumption for a few of my tested chargers).
Input power 21.2Wh at 230VAC
Charged into battery: 12.8Wh
Efficiency about 60%, this is rather good, I wonder if the power supply has extra high efficiency (i.e. 80% instead of 70%).

It was a 4 cell charger with one cell in it and charger current around 1.1A, the cell was one of my usual "SA18650-33" test cells.

fmc1

Enlightened
I have a basic MC3000 newbie question that I hope someone can answer.

When running cycle C>D>C N=1 or refresh, is there a way to see the discharged capacity of a cell once the second charge cycle has started?

It seems like if you don’t catch the charger in that D. resting phase you miss the discharge capacity and need to wait until the final charge phase is finished to see the capacity.

I have quite a few 18650’s from old laptops that I need to test and it would save me a lot of time if I could skip that last charge cycle if I knew the battery was going to end up in the recycle bin due to the capacity being low.

For now I am using cycle with mode C>D and if the battery is good I charge it. It works but it’s a two step process.

kreisl

Flashlight Enthusiast
When running cycle C>D>C N=1 or refresh, is there a way to see the discharged capacity of a cell once the second charge cycle has started?
let's assign numbers for easier communication, you wrote "When running cycle C0>D1>C1 N=1 or refresh … For now I am using cycle with mode C0>D1 and if the battery … " (0 = starting routine, not a full charge or full discharge). All right then.

In Refresh operation mode, when battery is in C1, you would like to see the result of D1, without having to wait until C1 has finished oic:
1) as you know, the charger beeps when D1.RESTING starts. if you're not in the room, you can't hear it oic.
2) using Cycle operation mode C0>D1 N=1 instead is a good option imho.
3) if someone is serious or ambitious with the capacity testing or cycling of an entire batch of batteries, i would highly recommend using DEX for assistance (logging, saving, restoring, etc). If it is inconvenient to let a full 350W PC (WinPC, MacPC, or LinuxPC) be running all the time, i would highly recommend a headless raspi instead, only 2W. With the raspi logging, you can view all the graphs and data in real-time on your smartphone and operate DEX from your phone, even when you're on the road: a low-bandwith mobile inet connection is sufficient to stream the raspi screen onto your phone screen. the phone app is called VNC Viewer. i know nothing about raspi, linux and what have you, but even me, i managed to setup a raspi, to install DEX on it, and operate it remotely. Yeah a raspi costs 35\$, a 16GB microSD card 5\$, a raspi case isn't free either, but at least i can offer you a solution and it works flawlessly and is really satisfactory/fun/cool .

There are technical reasons (which i don't understand hence can't explain) why SOV shows the result of D1 (D2, D3, etc) always after 1 full cycle has finished. Do you know a 4bay charger which shows D1 during C1, maybe the MH-C9000 does so? I can't remember.

I hope that you can consider the raspi option. Over 10 million copies have been sold so far, wow!

Last edited:

fmc1

Enlightened
Do you know a 4bay charger which shows D1 during C1, maybe the MH-C9000 does so? I can't remember.

yes the opus bt-c3400 does.

I don't like using the Opus for testing for lots of reasons. The main one is the Charge rate and discharge rate must be the same in test mode. So if you want to charge at .5c and discharge at.2c it's a three step process.

Thank you for the suggestion about using DEX. I'll give it a shot, first on the PC since I already have one.

Thanks again

pgogborn

Newly Enlightened
Darn, I purchased my MC3000 in the UK a couple of weeks ago, I thought I would be safe and then I heard a loud pop.

I have ordered a new case from GearBest, it hasn't arrived yet, but has anybody got any tips for installing the new case?

kreisl

Flashlight Enthusiast
Darn, I purchased my MC3000 in the UK a couple of weeks ago, I thought I would be safe and then I heard a loud pop.
Sorry to hear!
From which UK seller did you buy? Amazon UK inventory could be quite old stock that's true. Then it wouldn't make sense to ask for free replacement from them, i also agree.

Disassembly of the charger is easy, there are six(6) screws on the bottom. Reassembly is clear too, but requires some patience is all. Some nut posted the following video, reassembly of the old case, on youtube, you might find it helpful good luck:

Do be careful when opening the case after the 6 screws: the ventilation fan is wired to the PCB and the little white connector might have some glue residues on it. The glue is easy to scratch away, just be careful. Then you can unplug the little white connector.
:tinfoil:

pgogborn

Newly Enlightened
Thank you for the tip and the video. It wasn't Amazon but it was a seller with a 'large' Internet store.

I will name names and evaluate how good their response was when/if I succeed in switching boxes.

My pin broke after several long Break_in cycles with the sliders fully extended by 18650 batteries - I suggest anybody worried that they have purchased old stock run a stress test with long batteries inserted for several days not necessarily with the box powered up. In my case I do not think heat was a major contributor to the failure.

In the meantime I have ordered a *compression* spring from eBay to push against the slider on top of the box, I will see how that works out.

kreisl

Flashlight Enthusiast
a *compression* spring from eBay to push against the slider
I think I saw maybe in this thread someone having fixed the broken pin situation with a very small commercial zip tie. You loop the zip tie around the hole of the PCB and through the end of the pull-spring. This fixes the spring end very securely to the PCB.

Last edited:

canonite

Newly Enlightened
i like my MC3000 ... but it seems to have a loud whirring sound sometimes which last 1-2 secs, when i switch it on .. during charging , if the fan comes on... it does not have that loud whirring sound at all..

wonder if anyone else has the same ?

DeJaVu

Newly Enlightened
i like my MC3000 ... but it seems to have a loud whirring sound sometimes which last 1-2 secs, when i switch it on .. during charging , if the fan comes on... it does not have that loud whirring sound at all..

wonder if anyone else has the same ?
Yep, i have same here.
Its a common thing on old pcs, fans are loud until system/fan bearings warms up.

Fan needs to be oiled/lubricated to stop doing that.
I would guess this is a common issue with the charger, and most other chinese charger with fans like the opus, litokala etc.
They just cant find it in their hearts to use good quality fans , even in a charger that costs almost 100\$...

Last edited:

canonite

Newly Enlightened
Yep, i have same here.
Its a common thing on old pcs, fans are loud until system/fan bearings warms up.

Fan needs to be oiled/lubricated to stop doing that.
I would guess this is a common issue with the charger, and most other chinese charger with fans like the opus, litokala etc.
They just cant find it in their hearts to use good quality fans , even in a charger that costs almost 100\$...

hope its nothing more then that.. i think i should be able to find a replacement fan if needed... was thinking of attaching those usb fans to the USB output for 'additional' cooling ..

DeJaVu

Newly Enlightened
hope its nothing more then that.. i think i should be able to find a replacement fan if needed... was thinking of attaching those usb fans to the USB output for 'additional' cooling ..

The fan inside the charger is not 12v like most fans ive come about, but its rated at 15v. So have that in mind when looking for replacement.

kreisl

Flashlight Enthusiast
Today i followed Alt2 to install 3.2.9 and i did some mc3k logging (click to enlarge!):

Then i grabbed 3.3.0(beta) and followed Alt1 to install it.
Sharing time. Maybe nothing new to the people who followed the Raspberry Pi adventures from the very beginning, but afaik nobody posted so far this very setup in picture form, we only talked about it ("headless raspi", "DEX on raspi", "raspi powered by mc3k") and i would believe that not everybody could picture the setup properly in the mind. I didn't share a pic earlier because i wanted to wait for my aluminum case to arrive. Finally we're there!

If someone wants to log mc3k without a laptop, full-size PC station, or a PC monitor, OR if someone wants to log mc3k for 24/7 (say 99 cycles), this is the recommended way to go. As you can see the raspi is at ~40% GPU-CPU, and the energy consumption is ~2W, which is a fraction of the mc3k's energy consumption. The mc3k's power brick adapter is rated at 60W iirc. Personally i haven't actually tested 99 cycles with this setup for the sake of testing, but i will. Summer is long. So.

In case you're wondering.. while a (beta) version of DEX for Android does exist :twothumbs, the above setup has nothing to do with it. The above demonstrates how i am using my smartphone to view (and actively control) what's going on on the raspi. There's a smartphone app called VNC Viewer. The app mirrors the raspi desktop on my phone in real-time, and there's no major time lag unless you run a high-res HD video on raspi's VLC Media Player. So, DEX is installed on raspi, and i use my phone's viewer app to remotely access raspi and its DEX. I can do so, even when i am on the train, on the road, etc., the VNC Viewer app is crazy awesome, stable and modern. And easy to use.

If you're on the road in your car and really wtf need to transfer files between smartphone@car and raspi@home, then upload the file to the cloud or as email attachment and download it on the system where you need it on (Raspbian or Android, either accessed through your smartphone@car); there are also dedicated tools for any kind of file transfer between computers, for example Send Anywhere.

Everything works beautifully like that.

Once in a while i would update my raspi thread, whenever there is something new or neat to share. Hope you find my posts worthwhile reading or even inspiring (any mc3k owner interested in buying a raspi now too? i know at least one other person :huh. The content is very basic, i am a raspi beginner, but scratching at the raspi world is absolutely enough to keep myself hooked. And as a logging machine for mc3k (and UT61E) it found a serious purpose in my hands.

Last edited:

DeJaVu

Newly Enlightened
YEEEESSSSSS, WOOOOHOOOO
I was starting to feel left out, but no more. Full fledged member of the brotherhood, baptized by snap.

kreisl

Flashlight Enthusiast
I am not sure what you're saying but you do sound delighted about something with regard to this thread. Must be something good positive then

Btw to whom it may concern:
Since this great flexible method of "quasi-offline" logging (technically "offline" is incorrect word choice because raspi is a standalone computer and has wifi and www capability) of mc3k exists, works and pleases, I won't request true offline logging, i.e. on a SDcard directly inserted in the charger, if sky ever thought about a "MC5000"in a dream future and asked me about ideas for it. With DEX, you can control which data columns and in which format (how many decimals, which time format, etc) and which sessions and slots are to be included in the exported *.CSV-file. One wouldn't have any such controls if a "mc5k" logged directly to card (unless a raspi were integrated in the mc5k casing - crazy idea).
Logging directly to card or to USB thumb drive made sense to me like 5-10yrs ago. The game has changed since. The new relevant co-players are wifi, internet, cloud, smartphones and raspi, not even Bluetooth imho. And of course also thanks to the dedicated geman team behind DEX, the software is key in the game.

When people, from this very day on, approach sky suggesting offline logging capability to built-in SDcard, the rep will probably respond with "aha yes sir do you already know about the raspi dex solution to your request?" and then ignore and move on, how rude lol

Last edited:

DeJaVu

Newly Enlightened
I am not sure what you're saying but you do sound delighted about something with regard to this thread. Must be something good positive then

Yes, im ecstatic. For joining the brotherhood of snapped spring posts.
Not a half hour had passed after me badmouthing skyrc for using cheap fans, and the skyrc gods punished me with a snap sound .
1 post snapped while charging single 18650(not that oftenly used slot), 2 more were about to give out and only one was intact.

At least i had a good excuse to open up the charger and tidy it up a bit.
1.Removed all posts, replaced with zip tie
2.bent the lower portion of positive contacts so now all flattops can be charged without lifting
3.Added long overdue "twin turbine" 40mm fans at the back connected to the chargers usb port , got about 7 degrees lower system temp.Considering that the back fans are running at 5V and are inaudible its a good gain.
Bit questionable mod though, even though you get lowered system temp, when operating it at full power like discharging 4 batteries battery temp rises because the fans are cooling the heatsink and blowing the hot air toward the batteries. Still havent tested how it would behave on charging with high current, but i think it wont bother the batteries much, and those rare times the fan starts during charging should be kept at minimum or posibly eliminated alltogether. Making a side grill so the air escapes from the sides along the airflow(fins are horizontal so air doesnt escape from the top vent but goes along the sides) of the heatsink could help resolve the battery heating issue when discharging.
Not sure if i will try that, the little bugger got me tired, removing that chewing gum from the connectors was a real (challenge). Took me more than half an hour, and was actually the hardest part of the whole process if you ask me.
4.And at the end removed that cheap paste and replaced with good one, battery temp sensors now seem much more precise to me. Gained a few degrees of system temp upon doing this grrrrrr. I know, i know, the degrees were there but i just couldnt see them, and now i have the gift .

All in all, i found the good in the bad.

Last edited by a moderator:

Ravel

Newly Enlightened
How do you transfer files from the raspi to your (new/old) PC workstation or to your smartphone in the most elegant and efficent way?.

Congrats on the whole setup
Happy to see a convert :twothumbs to Pi

About transfer, samba is supported also,
and sshfs allows to mount the raspi memory card as a remote drive/folder from another linux machine.

kreisl

Flashlight Enthusiast
Congrats on the whole setup
Happy to see a convert :twothumbs to Pi
@Ravel , thanks for your cheers!!
Without you guys' help and the recipe it would have taken me much longer to install DEX on Pi. Using the install script made it so much easier! A RS232 serial board in form of a "Pi Hat" is on my way, that should complete my Pi-related purchases. Then i'll be able to log my multimeter on Pi too, not only the mc3k.
So you like samba and sshfs for remote access? If i understood correctly, SSHFS is a thing between two Linux machines; i couldn't use it on my Win-machine or on my Android phone. Samba seems popular, versatile, powerful, and has functionality like the control of printers which SFTP doesn't have. But for file transfer/file management, it seems that i would use my Win file manager here too, the raspi drive/folder would become available as a new shared network computer entry in my Win file manager. Samba must have its strong advantages over SFTP but the latter still seems to me the most direct primitive easiest way to connect and transfer, since modern file managers already have an integrated FTP-module and one does not need to create/modify/manipulate linux files on the raspi.

kreisl

Flashlight Enthusiast
Hello. Which charger currently has most reliable battery internal resistance measurement, for both Lithium and Nickel cells? (Except hobby chargers and MC3000 please).
Since the author doesn't want mc3k discussion in his thread.. :mecry:

I saw a similar question on the geman forum. The OP couldn't produce consistent Batt IR measurements with the mc3k and questioned the acquisition. I commented that, unfortunately as so much in serious life, if one wants to produce serious or ambitious measurements (like HKJ) which are suitable for publication or future reference, the mc3k device takes a minimum of personal efforts to produce such measurement results. The key reason for this being the various contributions of contact resistances per slot (dtsch. Kontaktwiderstände, Übergangswiderstände, Kontaktübergangswiderstände). Let's count, (+)TERMINAL-(+)BATTERY, (-)TERMINAL-(-)BATTERY, SLIDER-RAIL; that's 3 contributions. If you modded (it is possible but don't ask me how lol) the mc3k to having FourTerminalSensing (actually, it is ThreeTerminalSensing in practice) slots, you would eliminate the SLIDER-RAIL contribution only, and the total contact resistance would still depend on the contact pressure. Contact pressure is highest and optimal in a fixed bay charger like the MH-C9000; the AAA and AA batteries are really firmly seated between the contacts.

What kind of personal efforts? Well, amongst others, a fully serviced, maintained, re-calibrated unit, plus the manual minimization of the total contact resistance. Clearly, the ambitious tester tries to find the minimum of the total contact resistance (in a particular slot), before he proceeds with the test (Batt IR test, capacity test, whateva test) and publishes the test results.

Getting consistent and reliable Batt IR measurements with the mc3k is, due to the personal efforts, a bit challenging at first. Same with sharpening knives, tuning an old violin, or frying perfect eggs sunny side up! We are talking here +1 milliohms(!!) resolution, so any disturbing factors must be taken into account, eliminated or minimized!! Unless the battery is soldered to the PCB, you cannot eliminate contact resistance. So you can only minimize it by maximizing the contact pressure, e.g. as seen in the MH-C9000 or NC2500 fixed battery slots.

Nuff with the blablah. My unit was serviced and re-calibrated over a year ago. So, for my test videos, i only had to look out for the minimal total contact resistance and then try to reproduce the minimum before turning on the camera! The "trying to reproduce" took me an average of ~5min. So the time difference between each of the 4 video clips is ~5min. I didn't feel like video recording the 20mins of efforts (= 4*5min), i am showing the end result only. As any reviewer/tester does. As any pianist does. No pianist video records how he progresses studying a new piano piece until he finally gets to public-worthy performance quality!

The point of the footage is NOT to teach you how (what i did in the 5mins) to get to each of the four video results. The point of the footage is to prove that consistent Batt IR measurement is possible on the mc3k. Note: The test results are also reliable, as long as you repeat them under the same battery conditions: the battery should be well-conditioned (i.e. not a sleeping battery), fully charged, cooled off, and clean. lol.

Through rotation of the 4-bay configuration each battery was tested in every slot.

WHITE: 94, 94, 94
BLUE: 86, 86, 86
VIOLET: 86, 86, 86
YELLOW: 81, 81, 81

YELLOW: 81, 81, 81
WHITE: 95, 95, 95
BLUE: 86, 86, 86
VIOLET: 88, 88, 88

VIOLET: 86, 86, 87
YELLOW: 81, 81, 81
WHITE: 93, 93, 94
BLUE: 84, 84, 85

BLUE: 87, 87, 87
VIOLET: 87, 87, 87
YELLOW: 81, 81, 82
WHITE: 95, 95, 95

The batteries are from my set of Eneloop Panasonic Tropical (manuf. 2014), have been re-conditioned just now (BREAK_IN, followed by 6 cycles at -0.5/1.0A with MH-C9000, avg. mh-c9000 capacity reading ~1890mAh) and have been discharged to 0.9V without D.REDUCE prior to the above Batt IR measurement tests. Afaik Batt IR measurements should not be taken with discharged batteries, because this would produce unreliable or unstable results, but .. the heck i thought and did it anyway haha, with no detrimental effect in this clip series, as we could see.

To cut the whole post short:

If i gave my mc3k, this very unit!, and my Eneloops, this very set of 4 well-conditioned samples!, to my RC hobby savvy neighbor and asked him to shoot a series of videos showing 4-bay configuration rotation and measurement of Batt IR (3 readings per battery per slot) — basically doing exactly the same what i did in ~22mins of efforts! — he would have shot everything in 1 single video of 1min length and gotten completely different measurements showing in his video footage: whose fault for the discrepancy between mine and his footage would it be then? Is it the device's fault? Or is the mc3k user to blame? I will let the buyer/shopper decide upon this question.

The gist is, it really does take some efforts (time, energy, efforts, experience, practice, patience, frustration, knowledge, understanding, etc) to get things as beautifully right as seen in HKJ's review or the above set of 4 video clips ... regarding the ambitious measurement of battery internal resistance with the mc3k charger. ( Btw before i start very serious capacity tests, i use the Batt IR function too to learn the total contact resistance minimum per slot. Extra contact resistance, for example caused by my flaking dandruff, would ruin the beauty of the capacity test results. )

Last edited: