Parallel charging

[Tiberius]

flecc, in your AA pack experiments, you have Schottky diodes to connect the battery stacks when using them, and you switch these out for charging.

You also talk about charging two stacks in parallel. What did you do to stop transfer currents in this case? Or are you assuming that if both stacks are discharged any cross currents will be low and will be short lived?

What about using low value resistors to balance them for charging? Or can we assume that the voltage drop of diodes is not enough to confuse a 36 V charger?

Nick

**WARNING** 2/2/08 I'm adding this warning for anyone coming to this thread later. One of our members had a battery fire while parallel charging.Battery fire. The exact cause is yet to be determined but useful precautions to take would be:
Ensure that the trickle charge current for NiMH cannot exceed the safe level for the cells in any circumstances.
Use something that detects a temperature rise and terminates the charge
Do not leave charging batteries unattended.
****************

 

[flecc]

When I reduced the battery to two packs in parallel, I did away with the Schottkys Nick, just charging in parallel and relying on self balancing. That's worked perfectly ok to date.

The main reason for the Schottkys originally was just to allow me to keep an eye on the packs individually while experimenting, ditto the switch, though I could wire that to charge one at a time.

I also played with using the Li-ion charger, since that peaks at 41.4 volts so couldn't overcook the cells, but now use the normal eZee 36 volt charger. I've yet to try the Powerstream charger on that battery, since I've been avoiding biking in the recent severe cold and using my tin overcoat (car) instead for a few days.
.

 

[Tiberius]

Thanks flecc,

I want to try parallel charging, if only because the charger current is a bit fierce for one battery stack - I assued that if the batteries were kept at similar levels of charge it would be ok.

The next question is how to place and wire the thermistors on parallel packs.

The cold and ice has kept me off the bike too.

Nick

 

[flecc]

On the centre line and in between the cells about one third of the pack from the top of the upper pack will be fine Nick. That will gather the heat flow from both packs and avoid any cooling airflows at the top.

Normally hard against a cell wall, but with AA cells it may well be squeezed in anyway.

I found the same on the charger fierceness, too much for one pack, and that's when I first tried the more gentle Li-ion charger.
.

 

[Tiberius]

I was thinking of how to do the thermistors on two independent packs. Even though I may always charge and discharge them together it is easier to mount them separately. I suppose I could use 5K thermistors or 2 x 10K in each pack but then I lose the ability to run them independently. Ok, I know if I use them independently then I shouldn't charge them together. I just want cake and eat it.

 

[flecc]

You could use double the resistance value with one in each pack and run the pair in parallel. Then when one pack starts cooking, it would cut the charging for both packs.

Giant must have an answer to this on their new Twist 2 model, since that has twin 9Ah NiMh packs, pannier fashion. Giant usually use off-bike charging, but they wouldn't separately charge, so must have a dual pack charger.
.

 

[Tiberius]

Right, I have set up my two pack charger. There are 2 thermistors in each pack and there is a fancy harness to connect the packs to the charger so that it sees the right thermistor value. Exactly as you said, flecc. I expect it would spot one pack cooking; it would only see half the rate of temp rise, but with small batteries, big charger, it ought to be enough.

I decided to put a pair of Schottkies in to stop cross currents if the batteries were different when first connected. Unfortunately that also stopped the charger - I guess it looks for some battery volts before turning on the current and got really stroppy when it could see the thermistor and not the batteyr. So I shunted the Schottkies with 6 V bulbs and that makes the charger happy again. I don't expect them to ever light up, but it will be illuminating if they do.

Its running now. Time will tell.

Nick

 

[flecc]

Fingers crossed Nick, hope all's well.
.

 

[Tiberius]

Well, my double charger seems to work fine. It charged them up and turned off when I expected; I ran the batteries about half way down and back up again* and it behaves. Next move is to try it with batteries in different states of charge.

BTW, when I say the charger is too fierce for one battery pack, its not the actual charge current that's the problem. That's in spec for one battery pack. I'm more concerned with the trickle current that it switches to at the end. The charger documentation doesn't state that explicitly but only what Ah cells the charger is to used with. My view is that leaving it on trickle for a while is a necessary part of the cell balancing, so I can't just turn it off.

Nick

*This was a road test, with hills - report in due course.

 

[Jeremy]

Interesting stuff, Nick. I'm currently pondering about batteries. I don't need much capacity, 6Ah of NiMH will do fine, based on my test runs with the small capacity SLAs I have.

At the moment the choice is either to buy individual cells and wire some packs up the hard way, as you've done, or just buy some RC model packs. 12V 3.7Ah sub-C packs are around £20 or so each, so I'm thinking of getting six and running them as a series/parallel pack.

I was wondering if there is any difference in the various ways of connecting packs like this. I could opt to connect pairs of 12V packs in parallel, then connect the three combined packs in series (which would be more lossy with diodes) or connect three packs in series and then parallel the two packs.

The first method may keep the cells in balance better, with a lower risk of cell reversal perhaps.

I await the outcome of your experiments with interest as to whether or not I need to add isolating diodes.

Jeremy

 

[Tiberius]

Hi Jeremy,

I didn't quite start with individual cells. Batteryspace had a BOGOF offer on 12 V packs of AA cells (ie 10 cells already made up), so I got 6 of those. I built them into two 36V assemblies which I am using in parallel. Each assembly has 2 thermistors in, so I can charge them together or separately. I haven't bothered with any pack matching; we'll see if that's a problem.

That sounds like a good price on sub C packs so you could do exactly the same as I did but with bigger cells.

If I knew how to post pics, I could show you the finished articles.

Nick

 

[flecc]

Hi Nick

If you have somewhere to post pictures online like flickr.com or similar, you can display those here in this way. Just go to your web image, right click it and select Copy Image Location from the context menu, or if that doesn't show, click Properties on the context menu that appears and highlight and copy the URL there on the Properties page. Then return to the Reply box here, position the cursor below the text where you want the photo to appear and click the yellow mountain scene icon above. You'll get a box to paste in the link and ok it.

Then submit the reply and your photo will appear.
.

 

[Jeremy]

I've been researching sub-C packs for a while now and it looks like the Vapextech ones probably offer the best price/performance balance, with some quite positive reviews about them.

The best deals I can currently find are from here: Home who also sell the connectors needed to make a pack up easily. Interestingly, they sell individual tagged cells and all the bits of hardware, like large diameter heatshrink and plastic end caps, to make up any size pack you might need. The additional cost of buying a ready made pack is so small that I decided it wasn't worth the hassle of soldering up tens of cells.

I'm about to take the plunge and order six of the "short/fat" 12V 3.7Ah packs, as these will assemble into a box about 120mm wide x 50mm deep x 290mm long, which will neatly fit on the underside of my rack. There will be room to fit the tiny TongXin controller under the rack in front of the box, which should make for a quite stealthy installation. Total cost for the six packs, which will give me a nominal 36V, 7.4Ah pack, is £121.60 delivered to my door.

Jeremy

 

[Tiberius]

Hi Jeremy,

As you say doing 3 x (2//12) instead of 2//(3 x 12) - if that makes sense - requires more diodes in series with the output current and I'm not sure how much it would gain you on the balancing. It would certainly complicate the charging, and you could even introduce imbalance there. You'd also end up with a lot of 12V switching. So far I'm in favour of paralleling them up at the end voltage, then each pack operates in its own right and and could be removed or replaced arbitrarily.

Nick

 

[Tiberius]

Hi flecc,

I'll see what I can do about pictures, thanks.

Nick

 

[Jeremy]

Nick,

I think I'm just going to stick with two packs of three in series, as it's simpler.

I had a good find in Homebase this morning. I've been looking at a way to make a slimline, unobtrusive, battery pack that will fit in the space underneath the rear rack. Plastic housing from the usual suppliers are all the wrong shape, but luckily "100mm" flat plastic ventilation duct looks to be ideal. It has external dimensions of 110mm x 54mm, so will comfortably take the "short & fat" NiMH sub-C packs with a thin layer of foam wrapped around them, and with a couple of end plugs should make an ideal case. If I spray it black to match the bike and rack I doubt it will look too out of place.

Jeremy

 

[Tiberius]

An update on this:

The parallel charging scheme with 2 Schottkies and 2 bulbs is working fine. With batteries at different states of charge, one of the bulbs lights up, showing that there is a voltage difference, and a very small balancing charge. As soon as the charger is turned on, though, the bulb goes out, showing that the charger has raised the voltage enough for both batteries to be charging. At this stage the lower battery will be taking most of the current.

At the end of charge the voltages on each battery pack look good, and each will independently run the bike.

Nick

 

[Tiberius]

**WARNING** 2/2/08 I'm adding this warning for anyone coming to this thread later. One of our members had a battery fire while parallel charging.Battery fire. The exact cause is yet to be determined but useful precautions to take would be:
Ensure that the trickle charge current for NiMH cannot exceed the safe level for the cells in any circumstances.
Use something that detects a temperature rise and terminates the charge
Do not leave charging batteries unattended.
****************

 

[Merv]

NiMh charging

Hi, being new to electric bikes and keen to pick up tips I've been looking around the various contributions and topics, having just fitted my new SLA battery packs and put on charge. I was interested to look at batteries as a topic and noticed many experimenting with NiMh types. Fine for those who know and understand their characteristics but dangerous if not charged and monitored correctly. A sensible warning I noticed previously posted as a result of a fire. These batteries should be charge controlled as an individual series combination_ the characteristics being that as they reach their fully charged point (constant current of course) the terminal voltage begins to fall and this is when the charge should be terminated, and reduced to a low trickle charge state. If this situation is overlooked the cells will go into thermal runaway and eventually explode! I'm not familiar with bikes that use these NiMh cell packs, but i would not trust charging methods that don't detect a change in the charging characteristics_ most low voltage chargers either use dedicated micro controller devices and sometimes include thermistors as an added precaution should the charge not terminate. I've designed my own low voltage chargers using a combination of charge level termination, thermal sensing and time out. A bit belt and brases but better safe than sorry. Nothing wrong with experimenting and learning from the results but please be careful. Regards Merv

 

[flecc]

In fact Merv, NiMh has come and virtually gone on our bikes now, like SLA, now just hanging on in bikes. All the supplied chargers with bikes use three charge termination methods in layers, from thermistor measured temperature through to peak permitted voltage and are very safe in practice. We've often advised on replacement chargers and have some reliable recommendations.

Some have gone back to NiMh due to Li-ion reliability and life issues, but as they improve I think lithium batteries are all we'll see on bikes before long. Of course with those, the charge management is entirely or largely self contained in the battery.
.