3 phase motor 6 wires?

[Eastwood]

So I’m in the process of varnishing a hub motor and I just realized that the 3 phase wires are actually 6 wires going into the hub through the axle, why is this?
So why did QS motors put 2 wires for each phase connection in the hub?

Never seen this before, would it not be better to have one thick wire for each phase?

 

[calab]

Maybe they ran out of bigger phase wire, or maybe its a feature they promote like double sensor wire bs.

 

[DogDipstick]

Flow analogy? WHat flows more?

Two ( two inch ) small pipes can be sized to flow more than one ( three inch?) larger pipe . Exhaust systems in cars use this often, this knowledge.... many cars have double pipes where one big one would fit, and snake along.

Back to electronics.

Two 10 gauge wires can carry more than (one) 8 gauge... but not as much as a six gauge.

Two 8 gauge can flow more than (one) six gauge.

Two 6ga can flow more than a single four gauge.

Two 4 gauge can flow more current than a single two gauge. Go check it, google it. Its the way the AWG works ( logarithmic).

 

[calab]

two 10 gauge equals one 8 gauge but not in size.
To stuff needs to take up space, two of something is more then one.
Maybe it works for them and thats the biggest wire can pull through in quickness.

 

[Eastwood]

Flow analogy? WHat flows more?

Two ( two inch ) small pipes can be sized to flow more than one ( three inch?) larger pipe . Exhaust systems in cars use this often, this knowledge.... many cars have double pipes where one big one would fit, and snake along.

Back to electronics.

Two 10 gauge wires can carry more than (one) 8 gauge... but not as much as a six gauge.

Two 8 gauge can flow more than (one) six gauge.

Two 6ga can flow more than a single four gauge.

Two 4 gauge can flow more current than a single two gauge. Go check it, google it. Its the way the AWG works ( logarithmic).

Ok thanks, makes sense. Yeah the wiring on this new QS205 V3TI is different than the old V3. So two wires for each phase connection through the axle, then the phase wires coming out of the axle to the controller are flat. So was reading that flat cables have greater surface-to-volume ratios. So it seems like they’re trying to increase the efficiency for the QS205 V3TI compared to the original V3.

How thick of wire can be squeeze into the QS205? Considered upgrading the wires since I have the motor open for varnishing.

Also maybe you seen my other post recently but I’m upgrading the wires from the controller to the axle and creating a quick connect/disconnect so I can switch between motors frequently. One QSmotor is for Enduro and then the other motor set up is for the Street. Hopefully it doesn’t create too much resistance with having a connection in the phase wires but I think it’s practical since I like to ride both Enduro and on the street.
Sidenote with the older QS V3 motor, I’m putting a street tire, I would love to put a 14 inch rim for the most possible off-line torque but I don’t think I can get 10 gauge spokes that short. So might lace the older QS in a 16 inch rim just like my off-road version.

 

[AHicks]

Packing smaller diameter wires will leave less/fewer voids than packing larger wires....

 

[Eastwood]

Packing smaller diameter wires will leave less/fewer voids than packing larger wires....

So I guess you end up with more copper with two thinner wires versus one thick wire when it comes to packing wires through the axle, am I getting this right?

If so it kind of makes sense there would be larger gaps in the axle with bigger wires versus smaller wires packed in the axle.

On the other hand if you’re not squeezing wires through a small diameter, guess it would be better to have one really thick wire, but with hub motors I seems different because of the axle bore size.

 

[Eastwood]

I really like the design of this axle.
Amberwolf recently posted those pictures on a different thread.

 

[E-HP]

We need the CAD guy, but attempting to show dual 14 gauge vs single 10 gauge and relative minimum diameters when packed together. 14 guage wins for overall diameter (blue vs gray), but still appears to have room for the halls.


I used the specs from this table.

 

[Eastwood]

We need the CAD guy, but attempting to show dual 14 gauge vs single 10 gauge and relative minimum diameters when packed together. 14 guage wins for overall diameter (blue vs gray), but still appears to have room for the halls.
14 10.jpg

Ok thanks!
The picture/diagram really helps to see how much wasted space with the large phase wires. Really interesting to see the differences, 2 smaller wires vs 1 large wire.

So makes me wonder since I have flat phase wires coming out the axle, if they just split the strands of wire and made it two single wires going through the axle for each phase. Or if they actually used different wires and connected them. I don’t see any bulge in the wiring/heat shrink so maybe they simply split the strands of wire.

 

[DogDipstick]

two 10 gauge equals one 8 gauge but not in size.

NO. No, it desnt, not at all.

nooope.


Not in size, nor in its ampacity.

two 10 gauge > one 8 gauge (in size, and in ampacity).

Rules of thumb

The sixth power of 39√92 is very close to 2,[4] which leads to the following rules of thumb:

When the cross-sectional area of a wire is doubled, the AWG will decrease by 3 . (E.g. two AWG Nr. 14 wires have about the same cross-sectional area as a single AWG nr. 11 wire.) This doubles the conductance.
When the diameter of a wire is doubled, the AWG will decrease by 6 . (E.g. AWG nr. 2 is about twice the diameter of AWG nr. 8 .) This quadruples the cross-sectional area and the conductance.
A decrease of ten gauge numbers, for example from nr. 12 to nr. 2, multiplies the area and weight by approximately 10, and reduces the electrical resistance (and increases the conductance) by a factor of approximately 10.
For the same cross section, aluminum wire has a conductivity of approximately 61% of copper, so an aluminum wire has nearly the same resistance as a copper wire smaller by 2 AWG sizes, which has 62.9% of the area.
A solid round 18 AWG wire is about 1 mm in diameter.
An approximation for the resistance of copper wire may be expressed as follows:

 

[Eastwood]

The sixth power of 39√92 is very close to 2,[4] which leads to the following rules of thumb:

When the cross-sectional area of a wire is doubled, the AWG will decrease by 3 . (E.g. two AWG Nr. 14 wires have about the same cross-sectional area as a single AWG nr. 11 wire.) This doubles the conductance.
When the diameter of a wire is doubled, the AWG will decrease by 6 . (E.g. AWG nr. 2 is about twice the diameter of AWG nr. 8 .) This quadruples the cross-sectional area and the conductance.
A decrease of ten gauge numbers, for example from nr. 12 to nr. 2, multiplies the area and weight by approximately 10, and reduces the electrical resistance (and increases the conductance) by a factor of approximately 10.
For the same cross section, aluminum wire has a conductivity of approximately 61% of copper, so an aluminum wire has nearly the same resistance as a copper wire smaller by 2 AWG sizes, which has 62.9% of the area.
A solid round 18 AWG wire is about 1 mm in diameter.
An approximation for the resistance of copper wire may be expressed as follows:[/size]

So wanted to follow up on this topic. What’s your opinion on the phase wires that came with my motor?
Would it be better for me to replace the dual wires with one thicker single wire for each phase?
Still varnishing this motor and I still have it open.
I understand that the larger wires waste extra space in the axle but I’m not sure which scenario would be better? would the one thicker wire handle higher current more efficiently?

 

[liveforphysics]

Two similar length and diameter wires that both have similar low resistance termination at each end will share current decently well, generally tighter than 5% current flow imbalance between the wires. Sometimes bring my phase leads out in pairs if it packages better.

 

[Eastwood]

phase wires look thin
QS motors lied. They claimed that the hall sensor wires are now integrated and not loose. There was only heat shrink at the very end, as you can see in the picture there still loose wires
Not that big of a deal as I put heat shrink the hall sensor wires but they shouldn’t make claims that they can’t fulfill. Especially as simple as heat shrink

 

[amberwolf]

What would be more useful than just heatshrink or any kind of plain cable housing on the halls is a braided shield around them and their power wires, all the way from the controller to the actual halls inside the motor, grounded only at the controller. That would reduce induced interference from the currents in the phase wires and the motor coils, which would make for cleaner hall signals, which should make for an easier time of the controller reading them and properly driving the motor especially under adverse high-load high-current conditions. (which have been problematic for some systems)



It'd be even more helpful if they'd adopt (at least as an option) that large-diameter hollow tube axle design, for the high power motors. They could even do really big phase wires, with two out one end and the other out the ohter end with hall/etc wires in a separate shielded bundle.

 

[liveforphysics]

It'd be even more helpful if they'd adopt (at least as an option) that large-diameter hollow tube axle design, for the high power motors. They could even do really big phase wires, with two out one end and the other out the ohter end with hall/etc wires in a separate shielded bundle.

That would be a game changer for hot rodding hubs, as we know from the few folks who've done it. Maybe if enough people ask QS motor they would offer it as an option. It would cost at least one side-plate re-tooling and a different axle, but the benefits would be huge for folks trying to put big power into hubs. I know I would be more inclined to buy a hub that had a wire exit like you described, and I bet there are a lot more folks who would also want it, even at a +$100 upgrade cost or whatever they need to justify the tooling costs.

 

[Eastwood]

Two similar length and diameter wires that both have similar low resistance termination at each end will share current decently well, generally tighter than 5% current flow imbalance between the wires. Sometimes bring my phase leads out in pairs if it packages better.

OK good to know, thank you!

What would be more useful than just heatshrink or any kind of plain cable housing on the halls is a braided shield around them and their power wires, all the way from the controller to the actual halls inside the motor, grounded only at the controller. That would reduce induced interference from the currents in the phase wires and the motor coils, which would make for cleaner hall signals, which should make for an easier time of the controller reading them and properly driving the motor especially under adverse high-load high-current conditions. (which have been problematic for some systems)

Yeah that’s a really good idea, maybe I will at least do this with the hall sensor Wires. Like you said it’s probably best to do it on all the wires.

That would be a game changer for hot rodding hubs, as we know from the few folks who've done it. Maybe if enough people ask QS motor they would offer it as an option. It would cost at least one side-plate re-tooling and a different axle, but the benefits would be huge for folks trying to put big power into hubs. I know I would be more inclined to buy a hub that had a wire exit like you described, and I bet there are a lot more folks who would also want it, even at a +$100 upgrade cost or whatever they need to justify the tooling costs.

Yes! They know we hot rod the QS 205, I don’t know why they don’t allow for thicker phase wires. They probably simply don’t care, they’re like buy the QS273 lol
Wish I had access to a machine shop, would make the new axle and remake/rework the side plate cover. Somebody should start producing a kit and selling them, new axle with big bore and new side plate cover. Would be awesome!

 

[amberwolf]

What would be more useful than just heatshrink or any kind of plain cable housing on the halls is a braided shield around them and their power wires, all the way from the controller to the actual halls inside the motor, grounded only at the controller. That would reduce induced interference from the currents in the phase wires and the motor coils, which would make for cleaner hall signals, which should make for an easier time of the controller reading them and properly driving the motor especially under adverse high-load high-current conditions. (which have been problematic for some systems)

Yeah that’s a really good idea, maybe I will at least do this with the hall sensor Wires. Like you said it’s probably best to do it on all the wires.

It isn't needed on all the wires, just the hall signal and power/ground wires. If there are other sensors (temperature, etc) it wouldn't hurt to include those, but it isn't needed for the phase wires--they aren't going to be interfered with, and you don't really need to pump all that induced current energy into the ground shield.

Somebody should start producing a kit and selling them, new axle with big bore and new side plate cover. Would be awesome!

Keep in mind that you'd also want to sell a mounting clamp for the pipe axle that could be bolted to the frame of a bike, as a start to getting them mounted for various bikes, since the pipe axle will not fit the typical dropouts the flatted axle would.

 

[Eastwood]

Keep in mind that you'd also want to sell a mounting clamp for the pipe axle that could be bolted to the frame of a bike, as a start to getting them mounted for various bikes, since the pipe axle will not fit the typical dropouts the flatted axle would.

Oh, was referring to the type of axle it has now with the axle flats. the axle could be made larger on the exiting side with a bigger bore to allow for bigger phase wires. So the overall diameter on one side of the axle could be increased to be able to have a bigger bore. Then remake just the one side side plate cover that allows for a bigger bearing to accommodate for the bigger axle.
Also the side where the wires exit the axle is already much larger than the other side so they could just increase it more.

But also yes that other axle as well, “the pipe axle” but with this method, like you mentioned would have to Modify the dropouts.

You probably already know :wink: but you can see in the pictures the wire “exiting side” is much larger so it could just be increased more, with the new side plate with the bigger bearing.

 

[amberwolf]

Also the side where the wires exit the axle is already much larger than the other side so they could just increase it more.

But also yes that other axle as well, “the pipe axle” but with this method, like you mentioned would have to Modify the dropouts.

You probably already know :wink: but you can see in the pictures the wire “exiting side” is much larger so it could just be increased more, with the new side plate with the bigger bearing.

The axle itself, if made larger, could accomodate larger phase wires. But if it's not made proportionally larger to that, then this weakens the axle, so the higher currents you can now get can more easily break it. This has happened to people in the past with other motors when they bored the axle wire hole larger. (it has even happened without boring it larger on some smaller motors when run at significantly higher power levels than designed for).

Breakage of the axle at the shoulders where it narrows to the flatted/threaded portion is also a significant issue when you begin being able to pour that much more current into it. (the one I have here with both ends sheared off is an example).

Another issue with larger axles is you run into a limit because of either disc rotor mounting or freewheel/sprockets, for systems that use those.

If the pipe axle is used, it easily fits within those limits and in fact could be used within the existing covers, as long as "thinner" bearings are used (same OD larger ID), but gives tremendously more room for wires than any hole you can put in a typical flatted-style axle. Additionally, it is much stronger than such an axle, and further does not have the torque transmission problems of the flatted axles.

That's why I would like to see those used (or made as alternatives).

 

[DogDipstick]

Interesting about your motor. Could you show us the serial? I just keep thing like this in my head for future references and I don't think you showed the motor serial number yet?

Mine, I bought new... Old wire exit style... relatively recently... Came with sheathing protecting both the hall connectors together and separate from the phases ... Wrapped in that fiberglass sheathing the whole way to the plugs. Single 6 mm diameter phase wires. Serial number I can't figure out. Idk. I am loving it though. My first big hub.

 

[Eastwood]

Interesting about your motor. Could you show us the serial? I just keep thing like this in my head for future references and I don't think you showed the motor serial number yet?

Mine, I bought new... Old wire exit style... relatively recently... Came with sheathing protecting both the hall connectors together and separate from the phases ... Wrapped in that fiberglass sheathing the whole way to the plugs. Single 6 mm diameter phase wires. Serial number I can't figure out. Idk. I am loving it though. My first big hub.

Yeah my older V3 came with the sheathing, but this one just has a orange heat shrink, V3TI

On my side plate cover you can see I installed Vents on opposite ends that can be manually removed. To allow water to drain.

 

[Eastwood]

Does anyone know the largest size phase wires that can be squeezed into this axle without boring, QS205 V3?
would like to use this 6awg but the silicone casing is really thick. I have some marine grade heat shrink that’s much thinner but I’m not sure if it could handle the high current? Definitely don’t want to short anything out :?
Any suggestion Guys?

Still have my motor taken apart varnishing and now would be the time to replace the phase wires if ever. The more I look at these dual thin phase wires I keep thinking I should replace them for something much thicker. Especially for off-road use it would help with heat. I could remove one set of the hall sensor wires to allow for more room if that helps.

 

[DogDipstick]

The more I look at these dual thin phase wires I keep thinking I should replace them for somethin... with heat. I could r...

Geek solution/ backyard scientist:

YOu can always set up a teststand and watch the heat created from a certain amperage graveling through the wires under question. Check it against a control...... The single wire.

Check temps of WIRE(single) against WIRE(double) after the same time of a significant current traveling through it for a certain equal time. Same power source, same load, same current and see which dissipates ( wastes) more current at the end of the day.

Empirically.

Man, my 8ga wires from my main traction battery have been getting.. reallly hot since I did my new install. i certainly thing I am dropping volts since I went from 50A on 8Ga, to 100A on 8Ga. I can feel it, dont even need a thermometer.

 

[Eastwood]

Geek solution/ backyard scientist:

YOu can always set up a teststand and watch the heat created from a certain amperage graveling through the wires under question. Check it against a control...... The single wire.

Check temps of WIRE(single) against WIRE(double) after the same time of a significant current traveling through it for a certain equal time. Same power source, same load, same current and see which dissipates ( wastes) more current at the end of the day.

That’s a great idea but in my situation I don’t have a power supply for such testing. I’m sure the 6AWG would help but as far as knowing how much it would be a guess.

My main concern is not sure how difficult it would be trying to squeeze 6 AWG through that 90° angle in the axle as I’ve never done this before. I’ve read this can be quite challenging and they use some type of fishing tape to pull the wires through. I just don’t wanna get in a position where I pull out the old wires and realize I can’t squeeze these 6 AWG wires through the axle lol