Kawasaki Versys Forum banner

Stator Output/ Stator Testing/ Device load Ratings 2015

19K views 88 replies 12 participants last post by  onewizard  
#2 · (Edited)
Use Post 14 and video post 16 for testing, and post 34 in reference to China Stator RM Stator


At some point I will extract all the testing and place it in an appropriate thread. For now, this is going to do it for me.
More of a simple, very quick and extremely accurate test, to prove if any stator damage has occurred.

Basically, use your idle adjustment screw and get the RPM around 2000 RPM, this is warmed up RPM, do not try holding the throttle and measuring this.

So depending how fast you are, you may need to hook up your battery tender.

So what you need is some fine jewelers screwdrivers, straight pins or something that can be inserted in your socket of the stator plug, also a meter that reads volts AC, preferably with alligator clips on the probes . This is the 3 wires coming from the stator, to a plug close to your throttle position sensor. This connector has a latch locking it together, to release you need to squeeze down on the latch and wiggle / pull at the same time. When apart you are measuring the output from the stator under no load conditions, for your purpose, make a drawing and identify the 3 female crimps as #1,#2,#3, as long as you know what you are calling when referencing your measurements. So at 2000 RPM measure 1 to 2; 2 to 3; 3 to 1******that is your 3 readings, they should be around 24 to 28 VAC at 2000 RPM, the readings should be 0.5 VAC within each other, that is 1==28.0; 2==27.5; 3 ==28.3----

-if any readings are like the following ****1==24; 2==16; 3==22, you have shorted turns.

There is a third test that can also be done, measure 1,2,3 to ground, record these three readings, should be around 17 volts AC


Note:
One thing I have never mentioned, between my test and Kawasaki. Kawasaki requests 4000 or 5000 RPM, at that speed the rotor is producing maximum flux density, at 2000 RPM it is about 35% of maximum.



Note Added This in February 2017 To Test for Phase Loss on a Series regulator under Actual Load Conditions


Added additional info March 2019


in reference to testing Polaris regulator 4012941 and Polaris 4016868 , both have the same pin configuration except the 4016868 has less losses and a lower forward voltage drop / loss .

For those trying to test to see if the Polaris regulator is OK, a initial crude test would be to use a ohmeter on the 3 phase input, A-B,B-C,C-A all should be in excess of 1 million ohms. You can then hook up the 3 phase input to your stator and follow the 2000 RPM test, only difference is do not connect the battery to the Polaris output. With the bike running you should get no output, also if your battery is at or below 8 VDC I recommend charging the battery first as the Polaris has reverse polarity and low voltage protection built in.

It will not output any VDC no matter even if you are 48 VAC input and 4000 RPM, without having the correct VDC polarity at 8 VDC or greater connected to the output ( this is used to fire the SCR gate pulses) . Lastly connect the proper polarity with a semi or fully charged battery, at 2000 RPM you should get 14.2 VDC with the bike running and base load ( including low beam headlight). At this time you can follow my 3 phase input testing below, be aware not all meters can recognize the 3 phase as it is being rapidly switched, if your VAC readings are within a volt on all 3 phases and you are getting 14.2 VDC out, all is good!!

Original before March 2019;
A better way would be using a small AC clamp on current probe rated for 400 HZ @ 25 Amp AC, pretty sure no one on this forum has one.So I came up with this:
This test is to prove all 3 phases of a Series regulator are functioning, such as Polaris or Compu Fire. What you will need is some straight pins, needles or jewelers screwdrivers, to insert from the stator wire side of the connector. You are measuring VAC, you need to set idle around 1800 to 2000 RPM, have the headlight on, measure A to B, B to C, C to A , record the readings, they should all be around 16 to 18 Volts AC , battery voltage should be 14.2 VDC. If you get one of the phases reading 22 to 24 VAC or more, that phase isn't conducting/ being fired, which means you have a single phase condition on the stator. The stator is rated at something like 24 amp output, phase current is rated at 14 amp maximum, if you have a phase loss, you will get a single phase condition, with a maximum 24 amp available, hence the burnt stator.

If you have any questions ask me. Very little about induction I don't know, that has been my specialty for over 40 years.
The difference is that, at 25% flux density, a small turn to turn short or turn to line short will have a large impact on the AC output. At 100% output, the imbalance between phases will be less noticeable.
 

Attachments

  • Like
Reactions: mikesundrop
#83 ·
One thing and it is too late tonight to post my scope waveforms of the Polaris 4016868 50 amp series regulator. Everything produces harmonic distortion. Ripple is non existent on my 2015, that is, there is no AC ripple found on my DC of the battery. I havn't tried capturing a narrow microsecond wave form, but AC coupled shows no ripple. I see you are 5 millisecond and 10 millisecond .
Now about the series regulator, my waveforms look similar , but different. can't find ones I took in late 2023. What I found on the Polaris 4016868, rated at 50 amp. Our stator outputs something like 22 to 23 amp DC maximum. The Polaris regulator switches the 3 phase SCR's until a voltage exceeding about 14.2 to 14.4 VDC , at that point is switches A&B phase for 20 seconds, then B to C phase for 20 seconds, then C to A for 20 seconds. The instant you add load, say turn my 36 watt high beam LED on, it brings all 3 phases back online. At first I got worried, single phasing is a good way to get a burnt stator. I have reduced so much load on my 2015, it reaches a point where it can't reduce the output DC as there is a time required for the SCR to turn off, plus the gate pulse turn on time. I would be interested to know where you were connected to. One thing I found, proximity in respect to the OEM regulator wiring and wire size, had a huge effect . I have precision 1 ohm 50 watt transducer resistors - so in parallel it is 100 watts at 0.5 ohms . What I was looking at was adding some filter capacitiors. Well the Polaris regulator sure didn't like that. I was also looking at monitoring current output. In another thread, I discovered the regulator was doing it's job, I had created a bad frame ground which affected my VDC monitor ( neat little voltmeter), telling me my stator was failing, hit a bump and it was fine again.
One of those-just take it apart-why bother-only I would think of something stupid, like adding filter capacitors.
 
#84 ·
OK you got my attention, explain "I carry my trusty OEM one with the tools. It's small enough. "
I only meant, the regulator is small and can be carried along with the usual paranoiac loner toolkit that I have... !
I'm half joking about the "trusty" part, there is a reason I replaced it, but it's better than nothing.


I would be interested to know where you were connected to
Directly on the battery.


Judging by the ~70 ms for all 6 humps, "something" is at 14.3 Hz = 857 rpm. I'm gonna trust that it was idling (since the left hand was holding the scope and the right hand taking the picture...), so I guess the engine was at 1704 rpms (assuming an integer 2x faster than a phase cycle). It could likely be explained by gearing of the rotor, number of poles on the rotor, and the 3 phases 12 coils stator arrangement, but I'm no longer that curious to validate it...or so I say now...(sigh)).

What I getting curious about now is to capture the kimpex, and see what it does with high beam on, heated gloves, turn signals blinking, horn...
Ah, there's always something...
 
#85 ·
I only meant, the regulator is small and can be carried along with the usual paranoiac loner toolkit that I have... !
I'm half joking about the "trusty" part, there is a reason I replaced it, but it's better than nothing.



Directly on the battery.


Judging by the ~70 ms for all 6 humps, "something" is at 14.3 Hz = 857 rpm. I'm gonna trust that it was idling (since the left hand was holding the scope and the right hand taking the picture...), so I guess the engine was at 1704 rpms (assuming an integer 2x faster than a phase cycle). It could likely be explained by gearing of the rotor, number of poles on the rotor, and the 3 phases 12 coils stator arrangement, but I'm no longer that curious to validate it...or so I say now...(sigh)).

What I getting curious about now is to capture the kimpex, and see what it does with high beam on, heated gloves, turn signals blinking, horn...
Ah, there's always something...
Stock stator since 2007, is 3 phase , 6 pole , 18 coils. 18 gauge magnet wire connected Delta. I have a stock stator compliments of @Smiley . One consideration was to wind it Star, less turns and volts per turn is less. The biggest drawback is I would need to use 15 gauge wire, I also considered going to 14.5 gauge, and dropping maybe 2 turns, every other pole. This would give a higher current output. The fact my 2015 stator looks new, I have the 50 amp polaris, and sorted out the problem I created. Just no ambition to try hand winding with 14.5 magnet wire, plus I would be looking at epoxy coating the raw stator, before winding. I have the total length of magnet wire needed for either delta or Y.
Yes as to RPM, next time I intend to use my Tektronix current probes, P6021, right in the ball park rated;
120 Hz to 60 MHz -10.6 A RMS, 250 A peak, 10 mA sensitivity
 
#87 ·
WOW I need to ask, what were you using for probes? I see it is AC coupled, 5 mS . The frequency 231 HZ . 0.2V AC . I would be interested in using DC -direct coupling to see the AC component on top of the DC. 4.98 V P-P

I assume this was at the battery.
 
  • Like
Reactions: JT in GA
#88 · (Edited)
WOW I need to ask, what were you using for probes? I see it is AC coupled, 5 mS . The frequency 231 HZ . 0.2V AC . I would be interested in using DC -direct coupling to see the AC component on top of the DC. 4.98 V P-P
I assume this was at the battery.
You should ignore the frequency, this little DSO 150 scope is probably fooled by the noise.

The 3rd one is at the accessory relay output (no heavy accessory plugged but the small voltmeter and an empty usb port (maybe 20mA draw))... I'm not expecting much difference from direct battery measurement, but you have me doubting...

The 4.98 V peak-to-peak (Vpp) makes no sense to me (likely another glitch in the DSO 150), the graph shows about 3 divisions at 0.2v so maybe 0.6 Vpp. The voltmeter show 14.6V dc. If I presume it's an average, the peak voltage is probably 14.9V, getting seriously close to problematic.

I'm getting really tired of searching and only finding Chinese crap with no specs or finding items that can't even match the plug. I know I could make my own, but I have no confidence that I can make a terminal that can carry dunno-how-many-amps without melting.

I really need to double the effort on finding a good, clearly labelled "series" regulator...