Kawasaki Versys Forum banner

Stator Temperatures & Cooling

2K views 11 replies 5 participants last post by  onewizard 
#1 ·
Yes , this is the latest winding configuration, all the delta jumpers on the back, epoxy encapsulated , just curious what $ with shipping? This stator is one of only a few I recommend , RMstator is only recommended if you have a Polaris regulator installed, as it is a cheap China stator with Y connected stator, the lamination's are less so spacers are required on the mounting bolts and they cheat and use 18 gauge wire instead of 15 gauge, so less copper and less lamination's , yes it will output 330 watts for a short time until it burns up.

My logic on accepting RMstator combined with a Polaris regulator is normal warm weather load will be about 60 to 70% of full output.So if you follow the post below, the single 18 gauge wire only loaded to 60% output , it should last equal to OEM, when it gets cold out and you use heated gear, your oil will be cooler and able to cool the extra load on the stator. The ideal stator in my mind would be a 15 gauge wound, star / Y connected, less chance of turn to turn short or phase to turn short as the voltage per turn is less than delta.


https://www.kawasakiversys.com/foru...2017-a.html?highlight=delta+winding+rewinding
Below is post #5, you use root 3 to convert turns, the wire gauge is determined by circular mils per amp, which calculates out at one 15 gauge per phase.

Copied one of my former posts, easier than the links
the winding dope for both a Y and a Delta

Delta which is what OEM is;
44 turns per pole of 18 gauge magnet wire per pole, I would try and get class H---( OEM looks like a class C or F)

length of 18 gauge is 96" per pole, 576" per phase plus 20"= 596" and 1788" total to rewind delta = 150 feet approx.of copper in OEm

-----------------------------------------------------------------------------
Y connected
15 gauge magnet wire, 25 turns per pole= 56" X 6 poles =332" plus 20= 352" or 29.5 feet per phase or approx. 90 feet of 15 gauge connected Y to do a complete rewind

--------------------------------------------------------------------------
as you can see, 60 feet less to wind Y and 450 total turns on Y as opposed to 792 turns on delta-------advantages and disadvantages for each, one advantage of Y is less insulation----as a example if I subtract 25 from 44 = 19 X .004 = .076 inches, I gain roughly the thickness of two 18 gauge wires with less turns, the .004 is the insulation ( .002 on each side);

I am posting this more for interest than anything, this is one very difficult thing to do, especially if you have large fingers, not that it matters as you will need to come up with some sort of winding tool similar to a plastic straw.
 
See less See more
#3 ·
Stator death is a common issue with the stock shunt regulator, and even simply due to bad batteries. As shere are many threads about troubleshooting electrical issues to determine if the stator has failed, if the battery is bad, and also replacing the regulator to help prevent stator failure, that's not the focus here.

I'm curious about stator temperatures and cooling.

I have a few questions:

1. At what temperature do stock and/or aftermarket stators tend to fail?
1a. What kind of insulation do the stator windings use?

2. What is the best way to get an idea of stator temperature considering it cannot be measured directly?
Stator Cover?
Coolant Temp?
Oil Temp?
2a. What are some ways (and challenges) to setup these monitoring systems?

3. What is *your* (and ultimately a normal) estimated stator temperature (and/or proxy temperature that indicates approximate stator temperature) on average?
How has this been for your stator? (has it burned out at this range before?)
3a. At what temperature has your stator burned out?

4. What are ways to ensure good stator cooling? (other than using a series type regulator, and limiting electrical load - see other threads for details)
 
#4 ·
Not sure if this helps, and this is a static measurement, but we used to roughly determine winding temperature from the change in resistance of the coil. Need to know the cold winding resistance, cold ambient temperature and temp coefficient of copper. Get winding up to temperature, them measure resistance and solve for temperature. Please note a that a four wire resistance measurement technique (or dedicated low ohm meter) will be needed to measure the low ohm coil resistance. Not very practical for real time measurements!

Temperature Coefficient of Resistance
 
#5 ·
...I have a few questions:

2. What is the best way to get an idea of stator temperature considering it cannot be measured directly?
Stator Cover?...
It seems that a GOOD way is to measure the temp at the stator cover w/ one of those remote-sensing thermometers, that you point, then squeeze a trigger.

...I have a few questions:4. What are ways to ensure good stator cooling? (other than using a series type regulator, and limiting electrical load - see other threads for details)....
ALWAYS keep your oil-level FULL, as it's the oil that 'cools' the stator.
 
#6 ·
Not sure if this helps, and this is a static measurement, but we used to roughly determine winding temperature from the change in resistance of the coil. Need to know the cold winding resistance, cold ambient temperature and temp coefficient of copper. Get winding up to temperature, them measure resistance and solve for temperature. Please note a that a four wire resistance measurement technique (or dedicated low ohm meter) will be needed to measure the low ohm coil resistance. Not very practical for real time measurements!

Temperature Coefficient of Resistance
Finally someone besides myself that understands, the four wire or Kelvin leads, would be extremely difficult unless you made a male socket to match the 3 phase plug. I am going to copy some of my info. Doing that I am going to start a thread identical to this one, in members only.
 
#7 · (Edited)
If you look at my first post which I copied from my other post , you will see that each delta phase is approximately 576 to 596 inches long, I mentioned each pole is 96 inches long, six poles per phase.What you are suggesting is going to extremes, if you truly wanted this , several options, you could install a RTD or therm o-couple on the winding and bring that out. To measure this there is a standard value of resistance at 20'C , to be honest , even though I own test equipment that could measure this, I would never bother. Since I have discussed this many times, but never included the table / chart, I will repeat it again. On a Delta wound stator you would need to take B to C at 596 inches plus C to A @ 596 , which is 1192 inches in parallel with A to B 596 inches:nerd:;):funnypost: yes absolutely a waste of time >:)
Delta wound has 1 18 gauge magnet wire per phase , the phase voltage and line voltage are one and the same, someone measuring VAC at the stator plug at 2000 RPM will measure 24 to 26 VAC, that voltage is across one winding, say A to B phase. If that was a China stator would Y or Star connected that same A to B voltage would be across two windings, if you were able to measure A phase to the star point , you would measure 13.8 VAC , to arrive at the line voltage it is phase voltage times root of three or 13.8 X 1.73= 24 VAC.
In my discussion about preferring star over delta due to volts per turn, that would be the preferred method for motorcycles due to vibration and expansion contraction, the problem is those winding Star convert the turns ratio but fail on the wire gauge, which should be 15 gauge wire.

Our stator outputs 330 watts , using 14.2 VDC equals 23.23 amp DC, the line current of that is 13.4 amp, the Delta phase current is roughly 7.76 amp AC on 18 gauge magnet wire which happens to be 1624 circular mils, to calculate , divide circular mils by current = circular mills per amp 1624/ 7.76 = 209 cm per amp, so 209 times 13.4 amp = 2804 cm , the closest is 15 gauge at 3260 cm, so using the proper 15 gauge wire, with less turns we gain circular mills per amp of 243 cm per amp or a 14% increase in wire size. That is if they did a true conversion to Star.
So the China stator magnet wire is 1624 / 13.4 amp = 121 cm per amp, and you wonder why the China stators burn out using a shunt regulator.

https://mwswire.com/wp-content/uploads/2016/10/Copper-Magnet-Wire-Data.pdf

As to magnet wire insulation, I am used to Nema standard, I would guess the China and possibly OEM are class A or E , my preference is class F or H , be aware the higher insulation classes have a heavier coating.

Final coating also is a factor, in the photo that is epoxy coated, some venders to Kawasaki encapsulated the winding, this allowed it to withstand extremes in the environment it was exposed to , one huge drawback to encapsulated winding's is poor heat transfer,, today they have epoxy varnish , if it is machine wound with proper layering, no crossovers and a tight winding using a class F or H and a epoxy dip, it will last forever.

https://www.superioressex.com/uploa...s/emcwa-nema_magnet-thermal-class-ratings.pdf
Bottom line, switching to a series regulator and keeping your oil level at the upper line,making sure all connections have dielectric grease in moderation, all grounds are clean with dielectric grease, your battery is in good shape, you should never experience stator failure ( I did a post on battery and stator failure , they are linked, that is the regulator can cause the battery to fail due to overcharging, in the case of stator failure and bad battery that is linked to a series regulator and China stator)
 
#8 ·
Heat Cycle / Never Discussed Before

So something I have understood for over 40 years. Electric motors running at full load 24/7 will last longer in hours run by in some cases 2 times that of cycling loads and up to 4 times longer on those motors that start and run for short periods say 10 to 30 minutes at a time. A continuous rating is something that runs for 2 hours or more at full load in the Ontario Electrical code, depending on what information you use, Nema standard is 3 hours at full load, there are other factors such a service factor, most of what I am saying here will mean absolutely nothing to 98% of the membership. I copied the link below as it has ratings referred to as insulation classes and hours of operation, note page #8, as a example I mentioned class F at 155'C , this has a direct bearing on life of the winding. This link would be extremely useful to anyone in the electrical service or installation of electric motors.
Page #8
RATED TEMPERATURE RISE OR INSULATION CLASS

https://www.cedengineering.com/userfiles/Motor Nameplate Information.pdf

So heat Cycle relating to Series and Shunt regulators/ Varying load / Fixed load

Shunt Regulator:
Tow things to be aware of , varying load and heat cycle. Varying load, many don't realize the shunt regulator starts working at or above roughly 14.5 to 15 Volts DC. So at idle the shunt regulator has 6 diodes that transform AC to DC, the stator provides the wattage required to run the motorcycle , I refer to this as base load, as the RPM increases, the stator output in the form of Volts AC increases, this in turn causes a increase directly proportional in Volts DC. I have mentioned cooking the AGM battery, which battery will last longer, one charged at 14.2 VDC or one that sees 14.5 to 15 Volts DC, yes I know the answer. Cycling load, once we activate the shunt part of the regulator 14.5 VDC - 15 VDC , we start wasting Watts to ground. So to interrupt my explanation, those with incandescent lights , incandescent signal lights , heated gear, accessory outlets, switching to LED or trying to reduce load, concern about overloading the stator is unfounded, because anything in excess of 14.5 VDC - 15 VDC, will be shunted to ground in the form of heat.
Now to the meat of this discussion, at 1350 to 1500 RPM idle ( I run 1500 because that keeps output at 13.9 VDC with my fan of something like 70 Watts running), heat cycle with a Shunt regulator, at idle we have what I refer to base load, somewhere around 160 to 180 watts ( keep highlighting watts as that is energy/ work done , work is volts times amps = watts, you can have volts and no work done). So once you reach 3500 to 4000 RPM the stator is capable of outputting full 330 watts , going above 3500 - 4000 RPM accomplishes very little, so your shunt regulated motorcycle stator is wasting if we use the base load of 180 watts , it is wasting 150 Watts.
And finally why this is a problem, copper expands and contracts much more than steel ( laminations https://www.google.ca/url?sa=t&rct=...trical_steel&usg=AOvVaw1ijfMXT5n9jk-NA_w0FU64 expansion and contraction ) The magnet wire https://en.wikipedia.org/wiki/Magnet_wire coating is typically smaller than a human hair, again most of this information will mean nothing to 98% of the membership.
This expansion and contraction will occur every-time you go from 3500-4000 RPM and higher back to idle, as the oil is trying to cool the stator.

When winding is done, it is usually either dipped in a now epoxy coating, encapsulated or years ago a varnish dip and bake, this purpose was to prevent movement of the magnet wire. If anyone has ever watched someone welding using high current with both the ground and electrode cable traveling 15 feet or more, will notice the cables jump every-time a arc is struck , that is a direct result of the magnetic field change.

Since our Stator has a rotating permanent magnetic field, every-time it passes across the magnet wire, the collapsing field of the magnet wire is equal to and opposite the applied force, between the expansion and contraction of the winding on RPM change, eventually some looser turn start moving and that less than a human hair insulation wears off and a turn to turn short occurs. So consider a day ride where you start and stop say 3 times, all lasting 10 minutes or more, in addition to the numerous heat cycles during RPM change you also have cooling below normal operating temperature. Will stop here, next Series Regulator and why your stator will last indefinitely.

Series Regulator
I am not going into detail how it works , but the why and differences. First I am going to point out for those that have no clue what I said above :eek::funnypost::forgetit::thanx: In my stator testing and my Polaris install I mention to test your stator before you convert from Shunt to Series at 2000 RPM using your idle screw, if you have more than 1 VAC difference between phases your stator is toast or to explain what will happen next if you switch to a series regulator, imminent failure will occur. The reason is because you already have a turn to turn short, in fact for those reading this and they are switching, running at 3000 RPM with the idle screw would be even better, strictly to find if you have a turn to turn short, and this test would be better run once the motor is up to temperature, at that point the copper winding has expanded . The reason for this is because instead of loading / shorting the stator at or above 14.5 to 15 VDC, the series regulator starts switching around 14.2 VDC, what happens is the total watts output never changes , that is if you need 180 watts , that is all that is produced, yes there is a catch , much like a light dimmer in your home, the Series regulator starts switching, the higher the RPM the higher the switching, the end result is the VAC will rise, only the portion that is required to keep at 14.2 VDC is used, so it is possible to have 36 to 48 VAC across phases with a series regulator, so with a already damaged magnet wire stator, that higher voltage will speed up the failure .
Two very important points to remember, my stator with what is now a 4016868 Polaris regulator , only sees roughly 180 watts all the time, which is roughly 55% load, the heat produced is much less as the circular mills per amp is greater because of the reduced current. When I am riding with my heated gear and heated grips, it is colder out, again that load is fixed, some cycling does occur because at idle my stator can't keep up with my Gerbing jacket load, howver my stator never sees full load, unless I am riding at night, high beam on with heated gear, and the fan running. So with a series regulator , a fraction of the heat cool cycles of the shunt.
I am sure I missed something, this forum is open to questions, feel free to ask.
 
#9 ·
Post went astray

Somehow this morning I timed out, checked with a second browser, logged in but even trying to refresh the page failed to post. I copied and pasted in a second post, tonight I discover it was a new thread. Also I copied post #2 from another thread, to give a example of a series regulator VDC output. All good!
 
#10 ·
Thanks for all the incredibly detailed information on the stator/regulator topology. Just some thoughts/questions rattling around in my head ...

Reducing system load in a shunt regulator set up just means the shunt kicks in earlier (lower RPM) but either way isn't the (approximately) same amount of power being dissipated in the stator?

Do you have any idea what the max voltage rating of the insulation is? As you said there will be much higher voltage across the stator windings with a series regulator.

Going from theory, and information you provided, there will be large power dissipation in the stator at higher RPMs. Just wondering if stator failures are more likely when the bike is ridden for extended periods at high RPM versys (pardon the pun) a bike used for commuting where the RPMs are lower and high RPMs are very limited in duration.
 
#11 · (Edited)
Thanks for all the incredibly detailed information on the stator/regulator topology. Just some thoughts/questions rattling around in my head ...

Reducing system load in a shunt regulator set up just means the shunt kicks in earlier (lower RPM) but either way isn't the (approximately) same amount of power being dissipated in the stator?
Exactly, which is why I mention adding LED headlights to reduce load, it doesn't , the reduction in amps using LED lighting only increases the amps dissipated by the regulator to maintain voltage below 15 VDC. Only advantage of adding LED lighting or other amp reductions such as using LED city lights, would be if the extra amps will be used for a second set of heated gear for a pillion passenger. You may then create another problem in that the regulator may be dissipating a larger Wattage than designed for. Again a valid reason to go Series regulator.



Do you have any idea what the max voltage rating of the insulation is? As you said there will be much higher voltage across the stator windings with a series regulator.
Good question, I looked up class F which is 115'C,all kind of ratings, if you followed my earlier posts, 44 turns and six poles so let us say maximum voltage is 60 volts, that would be open circuit, we will use that, so 60 volts divided by 264 turns = 0.227 volts per turn, I would say that class F is good for in excess of 50 VDC per turn so over 200 times what is needed. The volts phase to turn could be 45 VAC however I picked 50 VDC , it is more likely over 200 VDC.

Going from theory, and information you provided, there will be large power dissipation in the stator at higher RPMs. Just wondering if stator failures are more likely when the bike is ridden for extended periods at high RPM versys (pardon the pun) a bike used for commuting where the RPMs are lower and high RPMs are very limited in duration.
Rather than continue about insulation voltage, I thought this last question will answer many others, surprisingly the guy doing long distance riding at high sustained RPM is likely to last 10 times longer with the same shunt regulator :surprise:.

The root cause of failure is turn to turn short or crossover to phase short or output connection to phase short, lastly single phase condition, all what I described the root cause is vibration of the magnet wire and insulation breakdown,( the movement causes the insulation to wear / rub, eventually bare copper, remember it is around .0015 of a inch coating ) off primarily expansion and contraction / heating / cooling / numerous cycles while riding, last is low oil level. Again all this can be virtually eliminated with a series regulator. Thanks for asking, as you can see it is complex, I can explain many ways, one really stupid way is saying two guys with identical cars, identical motors, both wish to travel at 60 miles per hour, Dude A is running a series equivalent car Dude B a shunt car, ** so Dude A accelerates to 60 % power = 60 MPH ** Dude B has no control he accelerates to 100 % power and uses his brakes to slow down to 60 MPH. Which engine is going to fail first? That is as simple as it gets and 100% accurate, the" brakes" 40% wasted at or above 3500 to 4000 RPM is a typical shunt, my opinion in the year 2019 Stone Age technology.
 
  • Like
Reactions: Fastoman
#12 ·
Insulation Breakdown / Temperature Insulation Class/Carbon

Something I talked about, and you hear this, insulation breakdown, what is it. https://www.engineeringtoolbox.com/nema-insulation-classes-d_734.html
Many factors that go beyond what even the most advanced member on this forum, service factor and temperature rise.

oC oF
A 105 221
B 130 266
F 155 311
H 180 356

So insulation breakdown, and electric motor winding, much different than our stators, our stators are continuous wound, the only connections at the power leads out. Motor winding individual poles were wound , talking 3 phase there is winding holding the pitch up and winding with the pitch down and then basket winding, typically the individual winding poles had to be connected, what we did on gauges 20 gauge and larger is use a scraping tool which best described is like a huge set of tweezers with a straight blade on each arm. On gauges larger than 20 gauge we would use a match or brazing torch to burn off the insulation, usually took less than 2 seconds.

The point is, you overheat the winding and that insulation starts burning and becomes charcoal , charcoal is a good conductor, add some vibration and it doesn't take much imagination to see 2 burnt turns first starting to conduct between each other, then short out, eventually causing a increase in current within the pole or phase.

What I would hope is our stator is wound with a minimum class B at 130'C. So my calculated voltage between turns of 0.227 VAC isn't much, what can happen is the adjacent turn could be 20 turns in, which would be 4.414 VAC.

Bottom line, converting from a Shunt regulator to a series, testing your stator using my test method, ( best to test after completing a ride while everything is heated up and expanded ) a difference of more than 1 VAC between winding's at 2000 RPM is suspect, greater than 2 volts at 3500 RPM , your stator will fail using a series regulator as the damage is already done.

How long? It might last a long time as the load is reduced and the heat cycling is also reduced.I mention this as some have converted to series and then had the stator fail,
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top