Mistakes that are easy to make, and how to avoid

[parajared]

Melting your phase wires: $20
Melting the varnish off your coils: $250
Melting your speed controller: $150
Overdischarging your battery: $40 a battery
Voltage sag killing your battery: $40 a battery



Melting your phase wires:
Steep rugged hills can demand 2000 watts and 3000 isn't unheard of. I have observed that wires start go get warmer when ran above 20 amps. Heat them up too much and the insulation will literally melt right off. I have observed cooler phase wire temps by overvolting and keeping amp draw low. Another option is modding your phase wires.
The phase wire mod where you run all new wire is tough. I haven't had any luck with wadding wires down the hub shaft. I did the mod where you just cut phase wires close to the hub and run heavy gauge wire the rest of the way to the controller. The wires get warm when you do this as opposed to unmodified hot and melty. Turnigy 8 and 10ga wires have worked pretty good for me.

Melting the varnish off your coils:
I killed a 9c hub by doing this. Using your hand to determine how hot the motor is just does not work. You will need to install a temp sensor if you use your bike for anything demanding. Fortunately this is pretty cheap and easy. I have seen several builds where people pull apart a $5 bbq sensor, but I found the more expensive TG3 on board sensor sold at amazon.com is easier to work with. I tried hobby king's vt sensor and do not reccomend because you cannot read the display in the sunlight. Temp sensors are easy to install. You just pop the cover off your hub, wad the sensor under a coil, and feed the sensor wires out to the display.

Melting your speed controller:
I have done this two different ways: running 20amps continuously through a 20amp controller, and running 4000 watts down a controller with IFRB4110 mosfets. 4110s are only good for about 2800 watts, they die after much more. Most controllers have 4110s in them, so if you are running a 12s setup don't run more than 55amps.
Full throttling a 20amp controller up a long steep hill heats them up and kills them. It would seem that if you want to run at 20 amps you would need a 30amp controller and limit to 20.

Overdischarging your battery:
I heard that the golden rule of lipo is never below 2.7volts per cell and never above 4.3. When I first got into e-biking I thought that you could discharge to 2.7volts per cell. Ya, don't do that! Things start to quite wrong around 3.4 volts per cell on turnigy lipos. I reccommend that you don't discharge below 3.65volts per cell. So if you are running a 12s setup that would be 43.8 volts (12x 3.65).

Voltage sag killing your battery:
I have been playing with the 24s setup and the problem I have is that the voltage sags a lot. I have seen the voltage dip as much as 3 volts when throttle is applied. I have killed batteries thinking if I stay above the throttle off cutoff of 87.6 it would be ok. It turns out that lipo batteries go out of balance around 3.4volts per cell even if it is just voltage sag.

Hope this helps,
Jared

 

[cal3thousand]

Looks like the start to a good sticky

 

[akatherder]

I was really excited about the title but this all flew way over my head. I don't think my bike is powerful enough to scoff at the first three and it's pretty flat around here. I do wonder about overdischarging and voltage sag on my battery. Then again I'm in the minority with 36v SLA batteries (first ebike and I figured I'd use them until they croak on me).

 

[dogman dan]

Good one. And yeah, I've pretty much done all of it, except somehow I never blew a fet on a controller.

I learned the same basic lessons.

Actually monitor temperatures.

I live in the desert, so air cooling sometimes doesn't work so great. Lots of folks drill holes to cool the motor. Personally, I find that effective only for letting the smoke out when it comes. It WILL help a motor cool quicker once you stop. That can keep a heat spike from taking the motor over the line, so holes to cool do some good. But you'll still heat up a motor fast if you feed it enough watts.

Don't overdischarge.

I thought the same thing. 2.7v is the limit, so as long as I keep it above 3v right? Those Hobby King packs lasted 50 cycles. Now I try to stop at about 3.7v per cell. They stay balanced better, and have lasted two years.

The thing that gets me up hills is slower RPM motors.

Slow windings run on higher voltage scamper up hills much better than a normal winding on a few volts less. Same wattage, but the slow motor tolerates a slower speed better when the hills get REALLY steep.

If you want real high wattage, get a bigger motor.

Though you can put 4000w into a 500w motor and zip along, it's not liking it. We have big motors avaliable now, so use em if you want much more than 2000w. with a truly 5000w capable motor, you can zip up the hills on any winding, since it won't be slowing down.

 

[Punx0r]

Following the advice on this forum I charge to 4.10VPC and discharge to 3.67VPC. I've found this gives almost exactly 80% of rated capacity. Since the other good advice is only to use 80% of capacity before recharging, this works out nicely 8)

 

[mvly]

I would clarify if you have enough FETs, you can go over 2800W. I run my 18FET Lyen at 74V Lipo and almost always go over 3.5KW during startup. Though I do keep it around 2.5KW continuous during normal riding most of the time.

 

[parajared]

Ah, I put the wrong data in. I maxed mine out at 2800 but I could go to 3500 on my 12 fet controller. Here is a e-mail I recieved from Lyen

Hi Jared,


The recommended maximum output for the 12 FET 4110 controller is around 3500 watts and 18 FET controller is around 5250 watts. Power=V/I. So let say you are going to run 22S lipo at 81.4v nominal voltage. You then divide 3500 watts by 81.4v and you get around 43A for the 12 FET, 5250/81.4=64.5A for the 18 FET. Or for 24S is 3500w/88.8v=39.4A for the 12FET and 5250/88.8v=59A for the 18 FET.


Regards,
Edward Lyen

 

[spinningmagnets]

Concerning the charge of a LiPo over 4.10V per cell, I have found that my battery voltage drops off rapidly when first used, down to a fairly stable voltage near the nominal voltage for most of the ride. This has led me to believe that any battery charge over 4.10 will add only a tiny sliver of distance to your ride, in exchange for added risk and a possible shortening of battery life.

A higher voltage will give you a higher top-speed. So, try this on your own bike, keep an eye on your top speed when you put a freshly charged pack on it, and see how long before the top speed stops decaying and stabilizes.

 

[Ykick]

IMO - build to slightly above performance of a top athlete on a modern road bike (instead of small ICE motorcycle/scooter) you'll rarely experience these "mistakes". Plus, you'll enjoy a reasonably lightweight ride with more than adequate brakes/suspension. While accommodated by most bike racks/carriers and more easily managed in/around apartment buildings.

That said, I feel the need for speed too and I'm all for racetrack/course efforts but when it comes to everyday reality and transportation in my life, a little restraint and remembering "it's only a bicycle" has kept me from experiencing any of these mistakes.

9C 2807 under 2kW is very hard to kill...

 

[dogman dan]

Damn right, Look how hard I tried to kill one at 1200w and failed. Only when stalling them straight up off road hills, or racing do they melt.

And at that power level, 48v 20 amps, they only go 27 mph! Plenty fast for riding anyplace it's not stupid to ride a bike.

Fun as hell to go melt a few things though, playing with off road and race track bikes.

 

[cal3thousand]

IMO - build to slightly above performance of a top athlete on a modern road bike (instead of small ICE motorcycle/scooter) you'll rarely experience these "mistakes". Plus, you'll enjoy a reasonably lightweight ride with more than adequate brakes/suspension. While accommodated by most bike racks/carriers and more easily managed in/around apartment buildings.

That said, I feel the need for speed too and I'm all for racetrack/course efforts but when it comes to everyday reality and transportation in my life, a little restraint and remembering "it's only a bicycle" has kept me from experiencing any of these mistakes.

9C 2807 under 2kW is very hard to kill...

Very good points.

 

[parajared]

Just "peddle assisting" yourself every now and then puts little stress on your components. Congrats to those who do that, I bet your bikes are really light. I think this is misleading for those doing a first time build however. If you are using your bike just for exercise this is fine, but E-biking to work peddle-assist style is tough! Think of yourself on those really groggy roll out of bed mornings. Last thing you want to do is start hammering on the peddles.

Noobs fall into the trap of going too small at first and having to upgrade later. For me it went from just enough to peddle assist, then just enough to coast at 25mph, and now enough to singletrack rugged mountain trails. I would have been much better just to build a hardy machine in the first place.

 

[doc007]

9C 2807 under 2kW is very hard to kill...

What about a 9c 2807 at 3600W?

 

[amberwolf]

If it's unventilated or otherwise unmodified cooling-wise, you probably won't be able to run it at that very long before you melt the phase wires, and short stuff out at the axle. If it manages to not melt the wires, it'll probably melt the solder on the phase wires and disconnect one or more, and save the motor then. (like happened to Dogman once). If that fails to happen, it'll probably cook the windings after a while.

This is based on various reports, troubleshooting, etc., from assorted forum threads I've read or participated in the last couple years, plus my own recent experiences as I now have 4KW on tap to briefly goose my older style 9C 2807 with on CrazyBike2 (or at least I did until I broke the frame in a skid/crash the other day).

With mine, which is ventilated, I can SMELL the (stock) wires overheating, with just bursts of 80-110A battery current, I think it's at 15s right now. I try to keep those under a second or two, but it's kinda fun to be able to accelerate 0-20MPH in under 3 seconds on that big and heavy a bike, and know that if it weren't for the legal limit on ebikes here, I could keep going from there pretty jiffily. So I am not as easy on it as I should be, and it is often too hot to touch the windings thru the vent holes when I stop. :lol: I can only imagine what it would be like if I had it still sealed up, unventilated! And that is with a very short <2.5mile work commute, which is mostly stop-and-go traffic, with some stretches of 20MPH cruising.

 

[Ykick]

9C 2807 under 2kW is very hard to kill...

What about a 9c 2807 at 3600W?

NFL = not for long.

Don't understand the OP assisted comment? I'm not talking a Pedlec. I'm saying if built to the performance of a top roadie using these components such "mistakes" will be rare.

 

[dogman dan]

In my experience, above 3000w continous, or continous enough, about 30-40 min will cook off a 9c motor every time.

 

[parajared]

ykick wrote
Don't understand the OP assisted comment? I'm not talking a Pedlec. I'm saying if built to the performance of a top roadie using these components such "mistakes" will be rare.

If you do most of the peddling, and just give yourself a squirt of power every now and then (ie peddle assisting yourself), your 18ga phase wires and 20a speed controller will hold up just fine, but it seems that after a while you find yourself wanting to move away from this and going on fun e-bike no peddle rides. Just build your bike up beefy in the first place. 8)

 

[Ykick]

ykick wrote
Don't understand the OP assisted comment? I'm not talking a Pedlec. I'm saying if built to the performance of a top roadie using these components such "mistakes" will be rare.

If you do most of the peddling, and just give yourself a squirt of power every now and then (ie peddle assisting yourself), your 18ga phase wires and 20a speed controller will hold up just fine, but it seems that after a while you find yourself wanting to move away from this and going on fun e-bike no peddle rides. Just build your bike up beefy in the first place. 8)

I don't really pedal but I can more than keep up with top road riders and using +30A battery current I've never suffered these "mistakes", as you call em. My ride is plenty beefy yet still fits most bicycle racks, carry up/down stairs, bus, ferry, train, etc.

Of course, if you're chasing motorcycle/scooter performance then it must be built to a different standard or you will likely suffer the failures detailed in your OP.

 

[John in CR]

Of course, if you're chasing motorcycle/scooter performance then it must be built to a different standard or you will likely suffer the failures detailed in your OP.

None of these failures here since I stopped trying to squeeze too much out of lesser controllers. With the proper motor and proper size wheel on a bike with enough wheelbase and lowered CG, then moped performance is easy. Get enough motor and controller along with proper ventilation, and motorcycle performance can't be too hard, since I did it.

 

[dogman dan]

Amen to that. Despite how long you've been showing the way, sill a lot of guys wanting to ride 3000w on little bike hubmotors. Might work ok for a few laps, or a quick blast near the house, but not for a longer ride.

 

[Ykick]

Of course, if you're chasing motorcycle/scooter performance then it must be built to a different standard or you will likely suffer the failures detailed in your OP.

None of these failures here since I stopped trying to squeeze too much out of lesser controllers. With the proper motor and proper size wheel on a bike with enough wheelbase and lowered CG, then moped performance is easy. Get enough motor and controller along with proper ventilation, and motorcycle performance can't be too hard, since I did it.

John - you're an excellent example for building motorbikes to high standards required by your desired level of performance. If people wanna go there and build motorbikes I'm all for it just best not be using a 9C to do it.

Perhaps people don't understand when I compare to a top athlete road bike? I often pace these guys and really, really good ones can haul 20MPH uphill, maintain upper 20's or more on flat ground and reach +35MPH downhill. A stock 9C can do that without suffering any of the example "mistakes" provided in the OP. I've done it for over 4k miles while enjoying a very close relative to a bicycle.