Qunfense Hub Motor? My self introduction

[Electrobent]

Hi:

I bought a used under-seat steering recumbent a few years ago and loved it.  But it was not safe once I got tired because I could not get going at lights.  I figured that it would be perfect if it only had an electric front wheel.  I looked around on the 'net and found expensive Canadian and cheap Chinese options.  I went with the Chinese, found the Golden Motor website and emailed sales and got no reply so I emailed Raul at Perfect Team in Oakland and he eventually replied. I ordered the 36V Brushless motor kit and SLA battery pack and some very cheesy accessories that I could not bring myself to put on my beloved Turner T-Lite. I asked for a regen controller but was told that there were none available and that it would not net me much juice anyway. When it came, the motor was missing one of the pieces that go between the bearing and the fork. The wheel was horrifically out of true.   And my new fork (to go from 16" to 20") needed splaying as well.  With some invaluable help from the machinist at work I finally got it all put together. Oh, and I should say that Raul sent me the missing piece post-haste but it was nowhere near as nice as the stainless ones Mike the machinist made for me.
 
I soon discovered that it did not take a whole lot of throttle get going and get going fast.  I went down a tooth in the rear from 12 to 11 if I remember right.  That did not make it much faster.   I rode it fast and hard and uphill and filled my giant baskets that I bought to have the giant rack to hold the silver plastic SLA case.  My SLA batteries did not last long and cracked and oozed but were taken off my hands for their lead content. 

I looked around at replacement SLA's and was not really satisfied with the weight and cost of the options I needed to give me the 20 Amp discharge rate necessary to go up the hill at 25 mph. 

I was told that lithium would blow up and to stick with NiMH.  I bought a 39.6 V 10ah D pack from Battery space.  It never did seem quite right so I bought another one and put them in parallel.  The Batteryspace packs are a pain because they divide them into 12V (or 13.2V in my case) segments for charging.  I learned today that they do this because it works with the lower voltage charger that they are selling. So I strap their two little camera-bag cases to the front of my baskets to get the weight down below my own weight so I don't fall over again.  That made it faster--a lot faster--until I melted the molex connecter on the fat power wires from the controller to the motor.  I replaced the connector with a bakelite screw panel and my top speed for motor only cruising on the flats went up 2 mph. 

This is great, I get paid for going to work instead of paying to go to work.  With diesel at $5 a gallon this is worth about $4.50 per day for me and that will almost buy lunch . . .

All is great until my high school buddy calls and wants me to find him a motor he can use to convert a VW bug to electric.  I figured I'd go back to Golden Motor and see if they had anything bigger.  I did not find what I was looking for but I did see that they now have regen controllers.  I transfered some money to pay-pal the slow, cheap, way and by the time I went back to make my purchase they were allowing the selection of a 48V regen controller with the 36V brushless motor kit.  I emailed sales and asked if it would work with my motor.  No response.  Raul again.  He wants $20 more and does not have the 48V model.  I ordered the 48V model because I figured they would not allow it as an option if it would not work well.

Then I start wondering if it is going to blow up my 39.6V packs by trying to shove too much juice back in so I start reading here and at endless sphere and realize that my problem will also be falling below the 42V cutoff on the 48V model. So I need more batteries . . . and maybe I'll melt the wires to my motor.  I am hoping it is one of the old ones.  It is painted silver where it is painted and says "QUNFENSE" on it along with SWX173180A and what looks like a serial number: 52779.  Does any of that make any sense to anyone? 

I googled "Qunfense" and came up with this link:

http://www.made-in-china.com/showroom/successmotor/product-listjeMEwzWufncs/Brushless-Hub-Motor-catalog-1.html

The wattage there is not clear as most claim to be 500-1000W.

Anyhow, how do I tell if I need to replace my wires before they tell me?

And any body build any high performance NiMH packs to be able to take a boat load of current coming back in fast?

I happened across a Honda insight junkyard and got to see their battery packs:  120 NiMH D cells assembled into rods with 3-4 inches of airspace between them and a fan blowing over the whole assembly.

I was shocked to learn that there were less than twice as many D cells in an Insight than I had strapped to my bike and figure that this air gap must be what allows them to get away with so few cells.  My own battery packs are bricks and sometimes over temp in charging.

So my plan is to take apart the two packs that I have and get enough additional cells to get to 48V (well maybe 49.2).  The Honda packs are held together with shrink wrap but I think I want something sturdier--and something that can be attached to the bike in a theft resistant fashion. 

So I am thinking long tubes of batteries. I walked around Lowe's trying to shove one of their D cells into various pipes and conduits and there were no snug fitting metal tubes but 1 1/4 inch Schedule 40 PVC pipe is pretty close plus its cheap and easy to work with, and if you paint it, it will last in the sun.  It will retain some heat but it could be cross-drilled for ventilation. So now I am thinking of building geometric shapes with 1 1/4 sprinkler pipe and going at them with a drill press.

But then I got to thinking that its probably best to charge each cell fully (independently if I can get down that low) and then measure them and match them together in pairs of similar voltages and then put these pairs in series. That would be less elegant in the pvc pipe unless I built ladders with wires going across and that would be dangerous.

I searched on "cooled battery pack" and found some RC ones where the stand single cells up apart from each other and then sandwich them between two sheets of something with a fan on one end. So now I am thinking a couple of Harbor Freight plastic cutting boards or maybe the corrugated plastic that they use in Post Office tote tubs?

Any body done anything like this?

[philf]

Wow!

A stunningly long post, and lots of effort.  I, too, am going through exactly the same exercise with NiMH D cells (the 10Ah type) that you are at the moment.  I haven't found a perfect solution, either - my first arrangement was a compact "brick" which I made up of 6 sticks of 5 batteries each, sandwiched together.  I was impressed with the power delivery, but did notice that the batteries became warm with heavy use.  I was actually given to wondering if the chemical action within the batteries was causing the heat, or whether the nickel tabs (and those tiny spot welds on each cell) were adequate for continuous heavy discharge.  This head-scratcher was supported by the fact that one of the connections, which I didn't get enough solder into, actually burned the inside of my carry-bag on the first ride.

I'm using the "Tenergy" 10000mAh D-cells from all-battery.com.  The first time I charged them, I didn't use an NiMH charger - knowing that the things will start to warm when they get to charge, I used one of the chargers from the the GM kit (all of which I've had to fix the polarity on before use), and simply kept an eye on the temperature and voltage manually while I waited for a proper charger to arrive in the mail.  This worked suprisingly well.

The second time I did this, it didn't go as swimmingly - I got distracted and didn't visit the batteries for a couple of hours, and during that time they peaked.  By the time I got to them, they were too hot to handle and one of the cells in the middle of the pack was starting to vent.  Doh!

The charger that I got for continued use is one of these http://www.powerstream.com/nimh-3pn9-36V.htm.  It was comparatively inexpensive, would charge the entire 36V pack automatically, and provides a discharge feature which will take the entire pack down to 1V per cell and bring it back up to charge again automatically.  The only thing you have to be sure to do is to stick an NTC 10K thermistor securely into your pack (Digikey or Mouser, and they're cheap - make sure it's 10K and B=3950), and wire it so the charger can see it.  The charger looks at both voltage and temperature to detect end-of-charge.  (They make a 48V version, BTW).

I'm continuing to play with all of this, but am as interested as yourself to see what others have done to package their NiMHs.

Cheers!

[philf]

An addendum to my previous post (something I meant to include)...

When I got that charger from Powerstream, I had to giggle.  It's made by Shenzhen in China.  Same people who make the controllers...

[Electrobent]

Thanks for the tips--especially the specifics on the thermister.  But where to put it and can you only have one?  Ideally there would be one on each cell.  I wonder if the 36 and 48 volt chargers are the same.  They both say 56 volts peak charging.

[philf]

I suppose it's theoretically possible to put more than one thermistor in the pack, but I can't imagine that it would be necessary, as long as the cells you are using aren't being swapped around a lot.  In other words, assuming you're not combining a bunch of "sub-packs" to make up an operating "full pack" at any time in such a way as to make it possible that you're mixing charged cells with fully/partially depleted ones.

What the charger is looking for is a change in temperature over a defined (short) time period.  This only figures into the equation if it misses the slight 'dip' in voltage that occurs when the pack is charged up (on NiMH batteries this is much smaller and harder to detect than with NiCads).  The third safeguard is just time itself...  If the pack never satisfies any of the other two criteria, most chargers will just quit tryng.

If you had two or three separate bricks of batteries making up your working pack, putting a thermistor in only ONE of them would be fine, as long as the pack was always charged/discharged as a unit.  I think you said you were building your packs with the batteries "nose to tail" in tubes.  A thermistor is only going to see the state of the battery you stick it to.  I've built my packs like bundles of dynamite (heh heh heh - I think about airport security every time I look at them).  With the thermistor located in one of THESE packs, it can react to the temperature of the cell it's stuck to, as well as two of its neighbours.  A pretty good indication of what the rest of the pack is doing is had this way.  Make sense?  I should post a picture.

Observing what the charger "sees" through the thermistor...  It looks like the thermistor shows a nominal 10K resistance at room temperature.  When I've measured its resistance at the point where the charger is happy (and the pack is warming up) that resistance has dropped to just less than 3.5K.

Combining multiple thermistors in series/parallel will most likely reduce the resolution of the whole affair, and I can't recommend it without additional and un-needed electronics.  Someone else may have a different approach, but I think simple is better in this case.  However you build your packs, there will always be some inequality between the cells.  Sticking a thermistor on a single isolated cell which peaks too early will result in a consistently undercharged pack.  The converse will hold true, also.  An argument for matching the cells as best you can at the outset, and not mixing them up after they've been put into service.

'twould be nice if there was a more plug-n-play way of doing things, but at the end of the day we're dealing with chemistry :-)