Need advice on mixing chemistries

Hi,

I have a front hub Ezee kit with a 48V NiCad pack. I need to extend the range of my eBike 3x from what I have now. I’m a neophyte with battery pack but I’m willing to learn. I need 1) advices on which battery pack I can connect with my NiCad using a schottky diode parallel "Y" battery joiner (for mix matching chemistries). Then I need to know 2) where I can buy this pack with a proper charger (cherger model please).

From what I’ve read on the forum, you have a good experience with batteries.. :wink:

Thanks for your time,

LogRaam

Typically SLAs are safe to parallel as their exponential increasing internal resistance in regards to current output ensures they don't put out exceedingly high currents, but what amount to get depends on a quantification of your desired range and it might also depend on the existing capacity of your niCad pack if it's "not normal". The problem with this strategy, though, is that SLA is pretty heavy and you might not like that.

So far, it doesn't seem like there's much experience mixing other chemistries but it seems to be the case that you want to "match" the internal resistances of two different chemistries to get nearly equivalent output currents or a different ratio to get the desired current contribution ratio. So, this calculation depends heavily on your NiCad's internal resistance and capacity and if the internal resistance isn't known, then you might have to engage in some expensive experimentation and/or learn how to measure it. Also, different chemistries have different "internal resistance" curves in regards to how to full it is and the temperature, so it isn't anywhere near an exact science with our limited knowledge, but they make good approximations for prediction.

Also, one warning many n00bs get is "get more than you need" because variances in regards to energy usage is only inevitable and you may be cursing the day when the winds are gusting against your favor.

By the way, what problems in Computer Vision are you tackling? How's the "State of Art" in regards to object recognition of general objects?

I have been running a 8 ah nicad pack from ebikes ca, in paralell with a set of 12 ah sla's. Not many cycles though, I only ride the two sets together for longer rides, and it's a weekend bike. But it seems to be working well, with both batteries supplying similar ah. I haven't used any diodes, though. Because of the weight issue, I have another nicad pack coming as I speak. It seemed to work great at least for a short term solution. There seems to be a general agreement that mixing sla's with anything is ok. Using the diodes, I don't see why you couldn't mix any chemistries as long as voltage was fairly similar. That is to say, 48v with 48v, but not 48v with 36v. And I refer to paralell connecting.

Thanks folks,
I purchased 2 new battery packs, 48V/20Ah LiFePO4. Probably I won't use NiCAD anymore, depending on the performance of LiFePO4.
Any experience with LiFePO4?

To swbluto: I work with object recognition, data compression, and several image descriptors for scientific purposes. I own 2 US patents, 1 in data compression and the other is with psychology/image matching. I'm active in the medical, geospatial, image broadcasting and the field of creativity. The start of the art today about object recognition is CBIR (content-based image recognition). If you are interested in the subject, I can refer you to the work of Dr. Michael Irani at http://www.wisdom.weizmann.ac.il/~irani/. Check the publications.

Log

Well hang onto the nicads. Their shelf life is forever, and someday a second battery that you know is functioning will be priceless when you try to figure out if the controller or the lifepo4 bms is causing a problem. I like redundancy and emergency back up for stuff. Part of why I have three ebikes now.

I purchased 2 new battery packs, 48V/20Ah LiFePO4. Probably I won't use NiCAD anymore, depending on the performance of LiFePO4.
Any experience with LiFePO4?

Click to expand...

Just one of those 48V20Ah packs is some serious range. Two packs and you'll be able to go a long, long way.

Yeah, 40 ah is enough to melt down some motors for sure if you don't watch out for heat buildup. That's why Justin went to the big clyte before he even got out of town. The designers of most of the motors are sitting there thinking, Ok the motor will run 10 miles on sla's and then get to rest for 6 hours.

Burning the hub huh? That sound bad. But if I use only 1 20Ah pack, then switch to the other pack if needed, do you think Ezee motor should work witht he 20Ah?

I have found that on my motor, An aotema 600 watt brushless often sold as Wilderness Energy, I can ride continuously at lower throttle settings on 36v for just about forever. At 3/4 throttle a slow ramp of increasing temperature begins. At full throttle, or higher voltages, the temp simply increases faster. In hot weather, the motor cools slower so the over heat line gets to be less and less miles. I've been gathering info on this with a thermometer sensor on the winding inside my motor.

The main thing I learned is, a relatively short time stopped really helps. A 5 or 10 minuite break every 30 may be all you need. Or riding slower, so that the stops that naturaly occur for stop lights or whatever, are enough to keep the motor cool enough. A really long hill, several miles long, might require a stop midway. It just depends on how you are riding and the terrain and the weather. I found out the limits on brushed aotema motors last summer was about 10 miles of full throttle at 36v when it is 100F out there. Sla's in 12 ah size would run down with about a 40% margin of safety, but a 36v 20 ah lifepo4 battery would have close to twice the safe range for the motor in that situation. So I easily fried a motor last summer finding this out.

Justin did a similar thing at the start of his ride across canada. On the first day or so of the ride, the smaller motor he started on cooked on a long long hill. The rest of the ride he used a crystalyte 5000 series motor for it's abilty to take the heat. You can keep an eye on how fast heat is building up in a motor by using an infared thermometer to sample the outer case temperatures, and look for a riding style that keeps it rising slower, or that rises and then stops getting hotter.

Thanks dogman, this is very interesting. From what I see, my Ezee geared hub doesn’t heat at all after a 1 hour full speed ride using 48v/8Ah NiCad, BUT the controller and the battery pack did: becomes very hot. This winter, I had a controller shutdown after a climb in snow due to controller overheats; the current stop flowing and the controller light start to blink. But after 2 minutes and a bit of snow the controller cooled down and the current start flowing again. Probably the 48v/20Ah LiFePO4 should bring its heat problems, and more during summer time. Have you guys tested some heat sink solution or cooling device for battery packs and controllers? 8)

My 24v 8 ah nicads get hot too, it's the nature of the beast, but if I take both of em, and paralell for 16 ah, they stay cool. Heating is all about how many watts for how long, so the higher wattages will heat more. Lower amp controllers help, with the lower wattage they deliver, so that may be why your ezee stays cool. Mabye it draws less watts than my motors. My aotema runs at about 800 watts when climbing a hill, so it will heat up after about a 45 minuite ride. My EV global is running on too high amp a controller, so it heats up fast! I get about 30 minuites out of it.

A better heatsink on the controller is always a good idea. I like to mount mine touching lots of metal, and in the wind.