An informal look at the Sanyo NC-MQN06U – charging in pairs

[SilverFox]

The Sanyo NC-MQN06U charger charges pairs of AA cells at 300 mA and pairs of AAA cells at 150 mA. This series of tests involved using Eneloop AA cells. The charger was part of the Costco package for Eneloop cells.

When charging in pairs you need to make sure that each cell is roughly at the same discharge level, or you can run into balance problems. The purpose of this test is to demonstrate how the cells can get out of balance. It is possible to balance the pairs if the trickle charge at the end of the bulk charge is high enough and goes on long enough to bring the undercharged cell up to a full charge. The problem is that during the bulk charge, it is possible to damage the cell that is fuller through overcharging. The charger is looking for an end of charge signal based on 2 cells arriving at a full charge at the same time.

Let's get on with the testing…

I am using Eneloop cells for this test that have been recently purchased, have gone through a forming charge of 0.1C for 16 hours, and have about 10 charge/discharge cycles on them.

I charged the cells on a Maha C9000, then discharge cells 1 and 3 at 1000 mA and cells 2 and 4 at 500 mA for a little over 60 minutes. I then put the cells in the Sanyo charger and charge. Cell pairs are 1-2 and 3-4. When full, I discharged them on the C9000 at 500 mA. Here are the discharge results.

1614, 1914, 1631, 1917 mAh.

As you can see, cells 1 and 3 ended up with less of a charge than cells 2 and 4. The imbalance begins…

I then charged the cells in Sanyo charger, and discharged them in C9000 at 500 mA. At this point the cells should be equally discharged and we should end up with results that are much closer. Here are the discharge results.

1897, 1909, 1911, 1913 mAh.

About this time, someone started a thread about the C9000 having problems in that slot 1 seemed to be giving lower discharge capacities than the other slots. I decided to check out my C9000. In this run I charged in the C9000 at 1000 mA, then discharge at 500 mA. Here are the results I ended up with.

1897, 1908, 1906, 1909 mAh.

It looks like my slot 1 also gives a little lower reading…

Back to the Sanyo charger testing…

I charged the cells in the C9000 at 1000 mA, then discharge cells 1 and 3 at 1000 mA and cells 2 and 4 at 500 mA for about an hour. Then charge on Sanyo charger. After charge completed, let trickle charge for 4 hours, then discharge on C9000 at 500 mA. Here are those results.

1912, 1948, 1929, 1953 mAh.

These are closer, but cells 1 and 3 are still lagging behind cells 2 and 4, even after several hours trickle charging.

To compound the problem I decided to combine several runs. Up to this point I have been looking at what happens after one charge/discharge cycle. Now let's look at several, back-to-back, charge/discharge cycles.

I start by charging the cells on the C9000 at 1000 mA. I then discharge 1 and 3 at 1000 mA and 2 and 4 at 500 mA for about 1 hour. Then I charge the cells on the Sanyo charger. This ends cycle one.

I then start cycle 2 by discharging on the C9000 at 1000 mA for cells 1 and 3 and 500 mA for cells 2 and 4 for about an hour. Then I charge the cells on the Sanyo charger. This ends cycle 2.

Repeat the same discharge for cycle 3, and once again charge on Sanyo charger. After 3 cycles of this imbalanced cell charging in series on the Sanyo charger, discharge on the C9000 at 500 mA. Here are the test results after 3 cycles of imbalanced discharging and charging imbalanced cells in pairs.

1891, 1946, 803, 1891 mAh.

It appears that there is some imbalance here. Cells 3 and 4 completed charging quite a bit earlier than cells 1 and 2. I simply let them trickle charge until cells 1 and 2 were done. Then I let the whole batch trickle charge for 2 hours more.

A person that does not understand that imbalance can occur while charging in pairs may come to the conclusion that cell 3 was bad and that rechargeable batteries are a royal pain.

Chargers that charge cells in pairs are a compromise. As you can see from this very limited test, sometimes you can get away with it, but other times the cells come out unbalanced. Please note that this test purposely caused an imbalance that is probably a little bigger than what would be encountered in normal use, but I was trying to illustrate a point.

Let's look at these cells again, this time charging them individually on the C9000 at 1000 mA, then discharging at 500 mA. As you can see from these results, cell number 3 is just fine.

1907, 1909, 1910, 1914 mAh.

Tom

 

[Hitthespot]

Tom,

These may be stupid questions but......

If chargers that don't have individual bays and charge pairs or 4 batteries at a time can cause such an imbalance, why are they made? Is this not dangerous to the general public, who doesn't really know that much about batteries, other than my camera doens't work so plop them in the charger?

I don't think I have any such chargers but my daugher may?

Bill

 

[Mr Happy]

Forgive me, but unless I am missing something, the situation is not as bad as it might seem.

Your tests have deliberately created a mismatched starting condition of discharge on the C9000 before charging on the Sanyo charger. And yet, to the best of my understanding, the mismatch reduces after you use the Sanyo charger to recharge them.

I think a fairer test would be to discharge batteries equally in pairs and then charge them in pairs, and repeat that for a few cycles. At the end, discharge in the C9000 to find out what imbalance exists, if any.

Very few consumer devices use single cells (I don't count flashaholic 1 AA lights). Most items like cameras use two or four cells, with some perhaps using three. Three of course is as bad as one, but I do think those instances are rare.

 

[SilverFox]

Hello Bill,

With NiMh and NiCd chemistries you simply end up killing the cell and that is not all that dangerous. The charge rate is low enough that even when the cell vents there is little danger involved.

The main reason these chargers exist is because many people want a "free" charger, or place "value" on the lowest priced charger they can find. Of course they also complain when they run into a drop in performance...

If no one purchased them, their availability would most likely drop off.

I think it is better to shop for "value." Value, to me, means very strong performance, and a decent price. A charger that offers good value may not be the lowest price one, but I have found that it will save me money in the long run by taking very good care of my batteries.

Now for the hard part...

You decide to give your daughter some Eneloop cells for her to use. The package from Costo comes with this charger that charges in pairs included in the package. Do you simply give her the whole package and caution her to make sure she charges cells that are similarly discharged, or do you spend extra for an independent channel charger and instruct her to throw the charger that came with the Eneloop package away... ?

Tom

 

[SilverFox]

Hello Mr Happy,

You are correct in that if you keep track of the cells and charge cells that have been discharged about the same amount you should end up with cells that are about the same state of charge when fully charged.

However, what happens if you mix the cells up?

Let's say you have a camera that uses 2 AA cells. You have a pair of cells on the table that you charged some time ago. An occasion comes up and you take the cells out of your camera and insert the cells that were recently charged.

Afterward, you dump the cells out of the camera and decide to charge all 4 cells back up. In the process, you loose track of which cells were which and end up charging similar to what I did.

Tom

 

[viorel00]

...
About this time, someone started a thread about the C9000 having problems in that slot 1 seemed to be giving lower discharge capacities than the other slots. I decided to check out my C9000. In this run I charged in the C9000 at 1000 mA, then discharge at 500 mA. Here are the results I ended up with.

1897, 1908, 1906, 1909 mAh.

It looks like my slot 1 also gives a little lower reading…

I believe that was my thread about Slot 1, and I am about to start a comprehensive test to study this problem. I plan to use 4 good eneloop cells, let's call these A, B, C and D and move them in the charger as shown below, and do a charge/discharge each time.

A B C D
B C D A
C D A B
D A B C

Then I will make averages, each column is an average for that slot (with 4 different cells), to account for the fact that the cells might not in fact be 100% identical.

I will post the finding when the test is done.

 

[Bones]

Thanks for another eye-opening series of observations SilverFox.

Until now, I've been under the impression that chargers that charged cells in pairs would over-charge the first cell to reach a fully charged state in order to finish charging the second.

This doesn't appear to be the case here, at least with this charger. I would be interested in your thoughts on what the charger is actually doing, and why it is doing it?

 

[SilverFox]

Hello Viorel00,

That looks like a good test matrix. Interestingly enough, my slot 1 seems to have come back to normal. I am not sure if it is a quirk with the cells, or if it is running a little lower than the other slots. I can't say I have noticed this before, but I usually am discharging on the CBA to get better accuracy on cell capacities.

Let us know what you find out.

Tom

 

[SilverFox]

Hello Bones,

If you look at the data, I believe that is what happened. Cells 2 and 4 ended up with higher capacities. I think this was caused by those cells being over charged.

When you charge in series, you are looking for the combined voltage of both cells. If one cell starts off with more charge, it will give a peak voltage first, then its voltage will begin to drop off. Now we have the condition where one cells voltage is dropping and the others is increasing. When the combined voltage triggers the charge termination, we end up with an over charged cell and an under charged cell.

Judging from the results several of my friends have given me, in normal use this degree of imbalance takes about a year to form in somewhat normal use. As you can see, the test that I ran accelerated this process. At any rate, I have spent a lot of time helping my friends figure out what was wrong with their batteries. The problem was not with the batteries, but with the charger they were using.

Tom

 

[Bones]

...

When you charge in series, you are looking for the combined voltage of both cells. If one cell starts off with more charge, it will give a peak voltage first, then its voltage will begin to drop off. Now we have the condition where one cells voltage is dropping and the others is increasing. When the combined voltage triggers the charge termination, we end up with an over charged cell and an under charged cell.

...

Thanks SilverFox.

I'm now wondering if there is some type of charging practice that could be implemented to minimize the imbalance with the type of charger. Especially when you can end up with one cell receiving less than half of full charge. Perhaps by always following the primary charge with a minium period of time on trickle charge?

 

[coppertrail]

I completely agree with the point about keeping track of which cells are used in pairs or in a particular device. With all my cells, i occasionally get them mixed up.

Although I own 2 paired channel chargers, I've not used them in over 2 years. The C9000 and C800S are my primary chargers.

 

[Black Rose]

I completely agree with the point about keeping track of which cells are used in pairs or in a particular device.

I keep all my cells numbered, with each set of numbers resetting for brand or battery type.

For example I have batteries 1-44 in Eneloop AA, 1-12 in Eneloop AAA, 1-12 in Rayovac Hybrid AA, etc.

I keep all this information, along with charge dates, capacity information, and which device the cells are installed in an Excel spreadsheet. That way I know which cells are getting what type of treatment, when they start to fall of, etc.

 

[Hitthespot]

About this time, someone started a thread about the C9000 having problems in that slot 1 seemed to be giving lower discharge capacities than the other slots. I decided to check out my C9000. In this run I charged in the C9000 at 1000 mA, then discharge at 500 mA. Here are the results I ended up with.

1897, 1908, 1906, 1909 mAh.

It looks like my slot 1 also gives a little lower reading…

Tom

Tom

I was curious about this so I made note of my first experiences with my C9000 (#OHOFA) noted below. While it isn't by much, I am also getting lower discharge capacity readings in bay 1. Both reading were after the break in cycle on new batteries. This may need further investigation or at least explantion?

I have some older Eneloop batteries I will be running the Refresh Analyze on later this week. I will note the discharge capacities on these also.

PowerEx 2700 AA 1/7/2009
1 2544
2 2549
3 2564
4 2547



PowerEx 1000 AAA 1/9/2009
1 927
2 945
3 928
4 936

 

[lyyyghtmaster]

WRT two accidentally mixed-up pairs with different discharge levels, I've found with Eneloops previously charged and used properly I can simply measure the voltages to determine discharge depth. The Eneloops are so good, and consistent, that this method has yet to fail me, and I use it A LOT. It's very sensitive, probably to within 10% discharge difference or less. (Of course all my Eneloops, though from '06, are still in excellent shape) I'm charging mostly in Maha C-204F pair chargers, and occasionally in the LaCrosse to check how things are going. I tend to leave the pairs on trickle charge for a couple hours to a day, at random, before taking them off when convenient.

I cannot say this method works for non-Eneloops, including, unfortunately, some of my ROV Hybrids and now Titanium Enduros, a few of which have developed alarmingly high self-discharge rates. Indeed many of my 2300 mAh and larger standard Nimh cells have developed so much internal resistance and/or fast self-discharge that I've given up trying to charge and use them equally. When they die they die. I'll just buy more Eneloops! :twothumbs

 

[PeAK]

Hello Bones,

If you look at the data, I believe that is what happened. Cells 2 and 4 ended up with higher capacities. I think this was caused by those cells being over charged.

When you charge in series, you are looking for the combined voltage of both cells. If one cell starts off with more charge, it will give a peak voltage first, then its voltage will begin to drop off. Now we have the condition where one cells voltage is dropping and the others is increasing. When the combined voltage triggers the charge termination, we end up with an over charged cell and an under charged cell.
.
.
Tom

The inexpensive MQN06 charger is probably similar or less featured that the MQN04 in the summary sheet below:

​

From the spec sheet, it is also possible that the charger termintated on the temperature due to the pumping of charge into the fuller battery.


PeAK

 

[NiOOH]

AFAIK MQN04 is only timer controlled. It is sold over here bundled with Eneloops, but also with 2700 mAh Sanyo cells.

http://www.eneloop.info/products/chargers.html

Quite bad combination actually. IMO MQN04 was designed as an overnight charger for 2500-2700 cells. It will overcharge Eneloops.

 

[PeAK]

AFAIK MQN04 is only timer controlled ... IMO MQN04 was designed as an overnight charger for 2500-2700 cells. It will overcharge Eneloops.

If you look at the charger summary table (earlier in thread), the variable charge duration with different size batteries indicates that some sort of charge detection is at work. The low charge rates would suggest that "zero delta-V" might be at work which they call "peak cut-off". Measurements by SilverFox and the ensuing discussion in that thread makes me wonder why more manufacturers do not use this termination condition and err on the side of undercharge. It seems to work for all formulations at all charge levels!!!

I have a recent charger which I suspect uses "zero delta-V", the window over it determines the slope seems to be about 20 seconds versus the 200 seconds found for his units. Progress ?



PeAK

 

[NiOOH]

If you look at the charger summary table (earlier in thread), the variable charge duration with different size batteries indicates that some sort of charge detection is at work. The low charge rates would suggest that "zero delta-V" might be at work which they call "peak cut-off". Measurements by SilverFox and the ensuing discussion in that thread makes me wonder why more manufacturers do not use this termination condition and err on the side of undercharge. It seems to work for all formulations at all charge levels!!!

I have a recent charger which I suspect uses "zero delta-V", the window over it determines the slope seems to be about 20 seconds versus the 200 seconds found for his units. Progress ?




PeAK

If you follow the link http://www.eneloop.info/products/chargers.html you'll see that Sanyo states it is a timer-based charger. I don't have it so I cannot be 100% certain, but this is what the specs are telling me.
Besides, what is the charger that you think uses 0dV?

 

[PeAK]

If you follow the link http://www.eneloop.info/products/chargers.html you'll see that Sanyo states it is a timer-based charger.
.
.
.
Besides, what is the charger that you think uses 0dV?

You're link does not line up with the data posted in the table/image. The data in the link seems to line up more with the specs of the MQN03. The variable time (with capacity) indicates that MQN04 uses a termination dependent on capacity (i.e. non timer). This also seems to be the case with the MQN06. There might be different versions of the MQN04 which has a suffix tacked on (i.e. MQN04U) which might explain the discrepancy in the descriptions/specifications(???).

PeAK

 

[Bones]

You're link does not line up with the data posted in the table/image. The data in the link seems to line up more with the specs of the MQN03. The variable time (with capacity) indicates that MQN04 uses a termination dependent on capacity (i.e. non timer). This also seems to be the case with the MQN06. There might be different versions of the MQN04 which has a suffix tacked on (i.e. MQN04U) which might explain the discrepancy in the descriptions/specifications(???).

PeAK

The Sanyo Hongkong site, which is usually quite accurate, agrees with the Sanyo Europe site that the MQN04 is controlled by a 16 hour timer.

Unfortunately, the link to the manual on the European site is no longer valid, but the TipidPC Forum, which linked to the manual when it was valid, offers this supposed excerpt:

MQN04 pdf manual is available here

<click here for link>

sorry it says

Charging control method : Timer cut​

so its dumb.

Dosent have a trickle charge function, it just has a 16 hour timer.

I speculate that the specification sheet is referring to the time it would take the MQN04 to actually charge the referenced cells, and not to when it would terminate the charge.

Sorry PeAK, but in this debate, my monies on NiOOH.