HOT phase wires: does lazy man's phase wire mod work?

[parajared]

I just got my Magic Pie II today. One of the reasons I moved away from my 9c was because of scary hot phase wires. Commuting was fine, but when I would go dirt biking in the mountains the insulation would get all melty and super hot (no failures yet though). The hub shaft is too small, there is just no way to get 10ga 12ga wire down into hub. My MP2 has the same problem: 16ga wires and no way to cram anything much bigger down the shaft.

Sacman mentioned to me that running 10 ga wire from the controller to the hub helps this. He says that the larger gauge wire acts like a heat sink. My previous setup was 16ga all the way to the hub.

I have tried to incorporate Ohm's law with AWG chart into figuring out mathematically if it would be okay to have 6 inches of 16g and 6ft of 10ga, but my calculations keep coming out funny.

Is this an acceptable way to do a phase wire mod, or are you just going to end up melting the 6 inches of 16ga on a 50amp uphill run?

I have a 48v 50a speed controller btw.

[wesnewell]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage.

[TylerDurden]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage.

What will the resulting phase amps be?

[neptronix]

I ran 12G from about 1/2in. out of the axle to mine and pumped 8kW peak into it, 3k-3.5kW continuous. The lazy man's mod works well. There is not much room to cram thicker wires down the axle anyway, even if you bore it out.

[parajared]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage

I don't really want to. I have eight 6s batteries that fit perfectly in my triangle so that only means 6s 40ah; 12s 20ah; or 24s 10ah. I could do a 24s setup, but that would mean I would need a new controller.

[TylerDurden]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage

I don't really want to. I have eight 6s batteries that fit perfectly in my triangle so that only means 6s 40ah; 12s 20ah; or 24s 10ah. I could do a 24s setup, but that would mean I would need a new controller.

Plus new motor or smaller wheel.

[dogman]

Wire mod better than nothing. But if you are melting the motor riding too slow up hills in the mountians, your wires will melt as easy as the motor.

You need to build for the trails. Slower motor, or smaller wheel, or something. The pie should help some, I hear it's a bit slower winding than a 2807 9c.

Either way you arrange your watts, 2500 is plenty to get a motor hot when you run it slower up steep hills in trail riding. Even flat trails can heat a motor, if you keep slowing down for corners and speeding up again repeatedly.

[parajared]

Ok, I did it.
It was worth it because it lead to hilarious conversation with my hub. Yeah, my hub talks to me, it's wierd.

Me: Hi I'm your new owner
Hub: I see your installing a temp sensor in me; that's not necessary if you stay within the manufacturer's param... wait what are you going to do with that gigantic wire!
Hub: AAAGH oh no! Your'e a monster!

[neptronix]

Have fun!

[Lessss]

Only other mod worth trying is to go sensorless rip out the 5 hall wires and see if that provides enough room to run better guage phases.

[neptronix]

Sensorless brings it's own probs though. Jittering and such has been reported by at least one person on this motor who took that route.

[John in CR]

If you got the wires hot enough to get soft and melty, then you need to do a full harness replacement including the hall wires. After melting it becomes brittle and the risk of failurefrom cracked insulation is high, and hanging a lead tin and copper bundle off the end of it like a fishing weight will help speed the process. Before getting so hot, yes, the lazy mod helps.

If you have a programmable controller, experiment with lower phase amp limits leaving the battery limit the same. Maybe you'll find a setting that works better for you and isn't stressing your system with spiking phase currents.

[parajared]

If you got the wires hot enough to get soft and melty, then you need to do a full harness replacement including the hall wires. After melting it becomes brittle and the risk of failurefrom cracked insulation is high, and hanging a lead tin and copper bundle off the end of it like a fishing weight will help speed the process. Before getting so hot, yes, the lazy mod helps

Yes, I had a hall sensor wire break just from handling it.

[neptronix]

Wow! Congrats if you can melt the insulation off the wires but not fry the hub motor itself.

That's an achievement

[mvly]

The only downside I found out about doing these type of mods is when you need to fix the motor itself, you will have to desolder it to open the motor and solder it again once you finish the fix.

For geared motor which requires greasing, I would stay away. For direct drive which can last a lot longer, I think it would be OK.

[dnmun]

i thought it was possible to get three 12AWG wires through the hole along with the hall sensor leads.

i was able to get that through the golden motor and i think the 9C is the same.

very hard though because i used solid, not stranded. find the stranded with the smallest overall diameter including the insulation. you need a pick to get it to turn inside the hub where it reaches the 45o angled exit hole. you have to pull the ends up outa the hole as you push it through.

you have to wrap the bundle of wires together, colinear tight packed and not twisted, when you push them in through the axle because it flexes and the bundle gets distorted and jams when you push into the hole, and tears the hall wires if they get twisted over the phase wires and rub against the inside of the axle hole.

but 12G will carry twice the current of the 16 gauge stock.

[neptronix]

Which golden motor hub did you stick that through, dnmun?
The magic pie, the 9C sized DD, the geared motor, one of the scooter hubs, etc?

I had very carefully grinded away at my magic pie's axle and was barely able to get 4x 24awg additional through it - that's *after* replacing some stock insulation with heatshrink. Anything over that would have both not fit through the side cover bearing, and grinding the axle would have weakened it too much since it was verrrry narrow / small to begin with.

Maybe i could have jammed 13ga down it if i had not installed a temp sensor and fan wires.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=35060&start=345

^-- some pics here.

[dnmun]

this is so old, this where jeremy educated me about hall sensors, here i am doing the same thing all these years later.

i think the heinzman rebuild was my first, from keith barrett.

anyway, this motor had only gone about 2 feet before i got it. why torque arms are critical and lawyers lips will ruin your life:

viewtopic.php?f=2&t=10667

[neptronix]

this is so old, this where jeremy educated me about hall sensors, here i am doing the same thing all these years later.

Good thing we don't sticky posts with good info in them around here. We'd be out of a job

[dnmun]

this is what got me started on relearning electronics.

that one tip that richard makes about adding the protective resistor on the hall sensor lead is so cool. it was totally and impossibly over my head.

i did not even understand things would blow up for a reason. i was so clueless, never even did a proper analysis of the normal operation of the hall sensors in spite of jeremy warning me.

just replaced all three, why not? they most be dead i assumed. no way i could tell from what i knew then. so stupid.

[liveforphysics]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage.

Phase current is all the motor sees. The phase current will be the same.

[TylerDurden]

Up the voltage and cut back on the amps. VA=W. 48*50=2400W. 72*34=2448W. 84x30=2520W. More torque, more speed, less amperage.

Phase current is all the motor sees. The phase current will be the same.

Lol, for some reason wesnewell disappeared...

[parajared]

I think I know what Wes is talking about. You can dump 100,000 watts down a tiny wire and not overheat it if you keep the amps low (say 100,000v 1amp) but if you have low volts, high amps, all sorts of stuff gets hot (speed controller, phase wires, motor coils ect..).

Upping voltage would also yield the benefit of higher top speed, but watts are watts, the motor will handle wattage the same and can still run into trouble if you put too many watts into it.

In other words, if I want to go up hills at 2500watts the motor will act exactly the same, but I could run 28 amps instead of 52; all I need do is wire my batteries in series instead of parallel so that I am running at 90v instead of 48.

Well... that and buy a whole new speed controller

[Punx0r]

Has anyone used thinwall cable to help with this sort of problem?

I'm not sure if it's available in ~200*C rating, though.

[Alan B]

I think I know what Wes is talking about. You can dump 100,000 watts down a tiny wire and not overheat it if you keep the amps low (say 100,000v 1amp) but if you have low volts, high amps, all sorts of stuff gets hot (speed controller, phase wires, motor coils ect..).

Upping voltage would also yield the benefit of higher top speed, but watts are watts, the motor will handle wattage the same and can still run into trouble if you put too many watts into it.

In other words, if I want to go up hills at 2500watts the motor will act exactly the same, but I could run 28 amps instead of 52; all I need do is wire my batteries in series instead of parallel so that I am running at 90v instead of 48.

Well... that and buy a whole new speed controller

This holds true for the wires from the battery to the controller. But it does not work that way for the controller to motor wires. The conditions there are determined by the power being delivered to the motor and the motor speed (back EMF) and a few other things like wire resistance. The controller converts the higher battery voltage to the lower motor voltage - it is essentially a buck voltage converter. So raising the battery voltage doesn't reduce motor phase current. Raising the battery voltage won't cool the phase wires. It will increase the power that can be delivered to the motor, and the top speed, and so will likely make the phase wires hotter because it will be possible to put more power in, and it will make the controller run hotter and less efficiently and put more stress on the FETs.

One other thing that raising the voltage will do. It will increase your ebike smile.