Possible NiMh DIY Configuration

[AliasMrJones]

I've been wanting to replace the 3 12v12ah SLA batteries I've been using with a WE BL-36 hubmotor with nimh and hoping to up the v in the process. Based on $/wh, I'm looking at these D nimh cells:

http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=325

Looks like soldering together with the tabs should be pretty easy. 40 of these in series would give me 48v10ah, less weight and a bit more power than I currently have. I've read the WE controller should handle the 48v OK. Currently the 30 lbs of SLA batteries are on a rack over the back wheel (hub motor is in front wheel). Should I try to find a way to mount the batteries more forward? The bike is a standard triangle mountain bike with front shocks. Maybe triangle configuration in a triangle bag? I haven't noticed any real handling "problems", though it obv doesn't handle like a regular bike any more. With the motor weight over the front and batteries over back it should be more or less balanced.

Also, what would be the best configuration for charging? Should I charge it as 48v? Break it up for charging into 2 24v or 4 12v with separate charges for each individual set? Cost is a factor so I'm not wild about buying a whole bunch of chargers, but I also want to take care of the batteries so if it makes sense in the long run...

Any advise about any facet of this project are really appreciated!

Thanks,

Eric

 

[xyster]

Unless all the cells are precisely matched, which 40 of those probably wouldn't be, then a string of 40 will go out of balance very quickly. Other folks have found that 4, 12v chargers works well enough minimizing imbalance.
As far as physical location and configuration, I like the triangle idea. Might take a look at the configuration on Ypedal's Chaos too.

 

[tomv]

I'm doing exactly the same upgrade. I intend to keep the batteries in the back. Even with my 35 lbs SLA batteries I feel that the center of gravity is more forward than with a regular bike. I keep them in a topeak trunk bag and it's surprisingly convenient to have some extra storage spece for carying tools, extra clothing, snacks, etc.

I've just recieved my cells (same spec as batterspace, a bit lower price):
http://cgi.ebay.com/8-D-Size-10000mAh-NiMH-Rechargeable-Batteries-with-Tabs_W0QQitemZ330120047116QQihZ014QQcategoryZ40975QQrdZ1QQssPageNameZWD1VQQcmdZViewItem

I'm also undeiced on chargers. You'll definetly want a thermistor in your pack and it has to be of the type the charger expects.

The chargers I've looke at sofar are:
48V 2Amps, needs 10k NTC. XLR connector. $56
http://www.powerstream.com/NiMH.htm

http://www.ebikes.ca/store/
48V 2Amps, looks like XLR connector, not clear about NTC. $85

http://www.all-battery.com/index.asp?PageAction=VIEWPROD&ProdID=1597
48V 1Amp (slooooooww), $46

Here's a useful article about NimH charging:
http://www.powerstream.com/NiMH.htm

The interesting piece there is that if you're willing to charge at c/10 (or have a smart charger that tops off at c/10) then balancing problem is sort of solved. Nimh can tolerate overcharging if it goes at a slow rate. Battery life is shortened a bit but they won't explode like lithium-ion

I have no first hand experience with nimh pack charging/balancing, it's just my guess. We'll see once my pack is up and running.

 

[NickF23]

unless all the cells are precisely matched, which 40 of those probably wouldn't be, then a string of 40 will go out of balance very quickly. Other folks have found that 4, 12v chargers works well enough minimizing imbalance.

I've had no problem charging 36 volt with one charger. I can't see any reason why cells going out of balance should be a problem with charger that finishes cutout on temperature rise or absolute temp. any imbalance should be reduced on each charge by the equalisation stage.

Poweridestore also do a cheapish nimh pack

http://www.poweridestore.com/36V-13Ah-NiMH-Battery-Pack

 

[Jay]

After a couple months using SLAs for my WE BL36, I switched to those Batteryspace D cells. I bought the pre-made packs, 1 36V and 1 12V (they didn't have a 48V pack then). They claim the batteries in their packs are matched, though I didn't check. I charge using 4 cheap (no temperature sensing) 12V chargers that came with the pack.

I imagine you could request matched cells, though it might cost a little extra. No telling whether they'd understand your request, though; they seem to have great difficulty with English.

I've done about 160 charge cycles so far with mine & they seem to work okay.

 

[Ypedal]

I've also been charging 30 cells with a single charger ( those Hi-Power brand chargers from Ebikes.ca )

2nd year on the packs, not sure on exact charge cycle count, but extimate aprox 150 cycles.. still going strong.

I tested the packs after the first year of use, ( Opened the shrink ) and found one bad cell on one of the 2 packs, replaced it, since i had it apart i changed the pack's shape to fit along the downtube on my full suspension bike.

As long as the cells are used regularly, they are fine.. if left un-used for a month, plan to take it easy for the first few cycles, keep below 1C and do shallow discharges 50 % or less of capacity.. fully charge and leave on trickle for a few hours and after the 4th cycle the cells are usually back up to snuff.

It's like having 30 glasses of water on a table for a month, if the window lets the sun hit a few of them but not all of them , they evaporate faster than the others.. Ie: Self discharge. If you use an unballanced pack that has a weak cell , you force current into it below 0v and it goes reversed ( + becomes - ) . You can permanently damage a cell this way.

When a pack is ballanced, the cells all hit 1 v within minutes of eachother, so the bike stops and everything is just fine. Monitor your voltage and don't allow the pack to sag below 1v per cell while under load.

Nimh/Nicad can tolerate over-discharge and over-charge at a slow rate, without much harm, but do this at high rate and you start to have problems.

Never plug a charged pack back into the charger, always make sure you use at least 10 % of capacity before you re-charge.

If leaving a pack unused for a week, leave it partially discharged and plug it in the night before you plan to ride, i leave my on overnight every night.


Oh.. if you plan to buy raw cells to make your own pack, it's worth testing the cells yourself individually before you make your pack, Bob Mcree did this with batteryspace cells and had good success.. look in the Power-Assist archives on Yahoo groups and search for Bob's posts.. I think he may even have posted this info here as well..

 

[NickF23]

Good post Ypedal,

Quite a lot to learn with nimh. I think perhaps that's why some even with the top quality cells that justin cells people seem to report problems.

from ebikes.ca

"s well, we found that despite our best efforts at screening and testing every battery before sale, there were still customers who were experiencing problems down the road. The most common issues were individual cells going bad and tab welds coming loose. This was the case with pretty much every manufacturer we dealt with"

and also perhaps why Rabbit Tool insist on people buying a nimh book with every pack they sell. Has anyone read that btw?

 

[xyster]

NiMH sounds like much more of a general PITA compared to lithium. Lithium cobalt either blows up, or keeps working as advertised. NiMH packs don't explode, but suffer all sorts of niggling problems requiring special attention.

 

[Ypedal]

My biggest beef with NIMH is the self-discharge factor.

A fully charged pack that has been sitting for one week has lost a considerable amount of power, and canot be plugged back into the charger without over-charging certain cells in the pack...

I need to make/find a trickle charger for this purpose that does 30 cells at C/10 or something funky like that.

They are not that bad, certainly better than SLA imo, but then comes Lithium !!!!! yeah baby ! yeah !! Low-self-discharge and you can top them off anytime you want.

 

[oatnet]

I need to make/find a trickle charger for this purpose that does 30 cells at C/10 or something funky like that.

For a 10ah pack a 36v 1ah power supply or charger would be c/10, right?

Thanks for the tip on not re-charging nimh without 10% discharge, I had no idea. Can you illuminate the issue behind this so I can better understand?

-JD

 

[Ypedal]

http://www.greenbatteries.com/bachfa.html

This relates to AA cells mostly, but the same principle applies to larger cells.

In a series pack, and using a Hi-Power charger, the charger provices the full current to the entire pack when switched on.. once it detects a voltage drop, max temp , etc.. it shuts down.

If a pack has been used, even slightly, this will allow the pack to absorb charge for X time and gradually warm up and shut down the charger safely.

But, if you plug in a pack that has been sitting for a while, but not long enough to drain it significantly, some of the cells in the pack will still be almost fully charged, while some others have lost 10 % or more just sitting.

The cells that are almost full will heat up quickly, while the drained cells willnot, depending on wich cell the temp sensor is touching it can create problems in various ways.

A pack that has been left partially discharged can be safely plugged in and charged before use at any time.

 

[Ypedal]

[/quote]

For a 10ah pack a 36v 1ah power supply or charger would be c/10, right?

-JD[/quote]

My C/10 trickle specs may not be exact, would depend on the cells, but C/10 or less should be a fairly safe bet for overnight top-off.. i would not leave it on for days on end this way.. but to ballance out a pack overnight should be fine. Some of the EE's out there may be able to give more precise details on this.

 

[oatnet]

http://www.greenbatteries.com/bachfa.html

In a series pack, and using a Hi-Power charger, the charger provices the full current to the entire pack when switched on.. once it detects a voltage drop, max temp , etc.. it shuts down.

If a pack has been used, even slightly, this will allow the pack to absorb charge for X time and gradually warm up and shut down the charger safely.

But, if you plug in a pack that has been sitting for a while, but not long enough to drain it significantly, some of the cells in the pack will still be almost fully charged, while some others have lost 10 % or more just sitting.

I'm not coming away with the same message, am I looking in the right place? Here is the quote from their website that I am looking at:
=================================================
Another common situation is for the charge cycle to be interrupted part way through the charge. The charger is unplugged to see how warm the batteries feel or to use the electrical outlet for something else. Then the charger is plugged back in. Unfortunately, this will cause a complete charge cycle to start again, even if the previous charge cycle was almost complete.

The easiest way to avoid these scenarios is to use a smart charger, a charger with microprocessor control. A smart charger can determine when a battery is fully charged and then depending on its design, either shut off entirely or switch to trickle charge. Most of our chargers use microprocessor control.
==================================================

So what this is saying to me is that if you are charging NiMh with a Dumb/timed Charger, it will reset the timer and start a new cycle. I can't imagine using a dumb or timered charger on NiMh in any scenario, for me that only leads to baked batteries unless I am sitting there watching them. As the author comments, a smart charger will avoid the problem. I think a more pertinent message is "Don't use a Dumb Charger with NiMh".

Since a smart charger is looking for changes in voltage and temp, there is no 'cycle' to be reset. I think the author is trying to tout his chargers as better because they are smart, trying to make buyers think other seller's nimh chargers are not.

The cells that are almost full will heat up quickly, while the drained cells willnot, depending on wich cell the temp sensor is touching it can create problems in various ways.

I would argue that the recharge process is exactly the same as the initial charge cycle: some cells in the pack get overcharged, heat up, the heat is detected by thermo, and then the cycle drops to a trickle to top off the other cells. Even if you used 10% of the battery, those same cells will charge, then heat, then cutoff when the heat hits the thermostat. Heating during the bulk of the charge is minimal at best, it isn't like it is slowly building up through the charge cycle, it comes on suddenly when the cells approach full.

The period between when the fully charged cells start heating up and when the heat is detected should be identical, whether they heat up at the end of an initial charge or from a recharge when full. If the period is the same, then the degree of damage is the same. Yes, there will be a 2nd 'heating up' of the pack, but if it is OK for the first charge I expect it to be OK for a recharge.

That is my understanding - am I missing something?

-JD

 

[Ypedal]

You are on the right track..

Example : I received my Nimh packs from Justin in a discharged state. so when i got them i plugged the pack in for charge and away i went.

When i received mi Nicad packs, without checking the voltage i assumed the same and plugged them in... within 10 minutes i heard a gurgle sound and was like whoa... wtf... ?

The packs had been shipped fully charged, and only a few weeks had passed.

The pack was still cold, no heat at all.. but some of the cells were over-charging and venting, while some of the cells were hapily taking charge.

The overall voltage of the pack was low enough to allow the charger to do it's thing, but not high enough to trip the charger on Max V..

And these are " Smart Chargers " . :lol:

Single cell chargers ( like in the article from the link ) would detect peak volts and go green.. but in a series pack it's not as easy.. bonus for using multiple chargers i guess.. but if you know the basics, it's ok to charge with a series charger...

 

[tomv]

I've finieshed up my pack over the weekend and took it for a test ride. Building took about 5 hours. One caution: it's very easy to short a cell if a drop of solder from the + tab drops on the metal case (-). Impressive smoke/flames

On a test ride I got a very unexpected and worrying surprise. Voltage sag is huge compared to my SLAs! Under no load batteries show arround 48V, and minimum voltage was 28V! 20 volts sag at 20 amps? Does anybody else see something similar? My SLAs drop by about 2-3 volts at full 20 amps.

 

[xyster]

On a test ride I got a very unexpected and worrying surprise. Voltage sag is huge compared to my SLAs! Under no load batteries show arround 48V, and minimum voltage was 28V! 20 volts sag at 20 amps? Does anybody else see something similar? My SLAs drop by about 2-3 volts at full 20 amps.

Yikes, that's terrible. A possible sign of the problem:
NiMH cells are 1.4 volts at full charge. A fully charged "48 volt" pack of 40 should be 56 volts resting, not 48 volts. I would not run the pack again until I tested the voltage of each cell -- you might have some that are almost fully depleted, or dead now. This would explain the excessive sag.

 

[tomv]

It's not fully charged actually. It only has the charge that I've bought the cells with. I guess it's 30% charged or so. I'm still waiting for my charger... Couldn't resist trying it out

I did measuere all the cells after the ride (total discharge of about 50 Wh). Each cell pair measured in the range of 2.37 - 2.32 volts (it's a pain to measure each cell individually, but the wakest one I could find read arround 1.12 volts). All cells were slightly warm but nothing was hot.

Xyster, what sort of voltage drop were you getting when you rode your NiMHs?

I wonder if I demand too much of those cells. They are 10 Ah, rated for 30 amps continous, 40A peak, but I guess that's way too optimistic.

 

[Ypedal]

My 72v 8ah Nimh pack, fully cycled and charged, will drop 14v from 84v to 70v at 20 amps.

At 50 % discharge, rest voltage 72v, sags down to 62v at 20 amps.

Rule of thumb, or my own rule if you like, don't allow the cells to go below 1v per cell at any time..

 

[xyster]

It's not fully charged actually. It only has the charge that I've bought the cells with. I guess it's 30% charged or so. I'm still waiting for my charger... Couldn't resist trying it out

I know that urge! I still need to finish truing my new 20" wheel, but I've been riding on it anyway.

Xyster, what sort of voltage drop were you getting when you rode your NiMHs?

I've never used NiMHs on my EVs. Before jumping straight from lead to lithium, I studied the practical considerations of nickel chemistries in depth, and read lots of reports from other nickel users like Ypedal. I initially planned a 72v NiMH pack for my X5, but lithium 18650s weren't any more expensive, and I was put-off by the pervasive QC issues with the available D & F cell NiMHs.

I wonder if I demand too much of those cells. They are 10 Ah, rated for 30 amps continous, 40A peak, but I guess that's way too optimistic.

These are the cells, correct?
http://www.all-battery.com/index.asp?PageAction=VIEWPROD&ProdID=568

All-battery's packs of these are spec'd for 10 amps continuous, 30 amps burst. I looked into these too, but by all reports these are effectively 1C cells, unsuitable for our use unless paralleled.

http://www.all-battery.com/index.asp?PageAction=VIEWPROD&ProdID=1344

"Technical Specifications
...
- Current: Standard discharging: 10Amps (Warranted) and Highest discharging rate for short-time 30 Amps"

Batteryspace sells similarly-priced 10ah NiMH cells with a 3C spec that *should* work much better:

http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=325

I'd return for a refund and then try the BS cells.

 

[NickF23]

yeah 20 volts sag is not good.


I'd be surprised if don't get much better figures once you fully charge them. They need to be cycled a few times to know the true performance.

 

[tomv]

Yep, I need to keep playing with them a bit more. I can live with voltage sag. My 48V would effectively be more like 36V. I'm much more curious to see if the cells can really hold 10Ah...

 

[AliasMrJones]

- Current: Standard discharging: 10Amps (Warranted) and Highest discharging rate for short-time 30 Amps"

Batteryspace sells similarly-priced 10ah NiMH cells with a 3C spec that *should* work much better:

http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=325

I'd return for a refund and then try the BS cells.[/quote]

Those are the exact cells I was looking at. I'm a bit concerned by this on the batteryspace page for those cells:

# Standard discharging rate: 10Amp ( Recommended and warrantied)
# Highest Discharging rate for continuous running: 30Amp ( high discharging rate will recue battery capacity )
# Max. Discharging capability for short time: 50Amp ( Not recommend as working discharging rate )

Looks like they are are really 1C, but you can run them at 3C for shortened life. Just got my Watts Up hooked up last night so now I can see what I'm really pulling from my SLAs. Didn't want to buy/build a new battery until I knew what my actual usage was. Now I know...From this morning's ride to the train appears I pull max 20 amps, avg. about 10 amps. In fact, it seems to hit a hard threshold at 20 amps that it won't go past. I'm guessing controller limit?

 

[xyster]

Yes, that's the controller's limit.

If you want to try those cells, I'd suggest buying one, then load-testing it at 1C and at 3C with something like power resistors, and measure the voltage sag. Multiply by the number of cells in your pack to estimate what the pack voltage sag will be (which doesn't take into account resistance in the pack wires, solder, tabs, etc...).
IMO, total sag more than 15% with wide-open-throttle (7 volts on a 48 volt pack) is excessive -- all that power loss is dissipated as battery-damaging heat.