GS Stators: Revised Phase B Tests

[posplayr]

Most everybody that has been here at GSR for any amount of time has heard of or read the famous "Stator Pages".

This page is for a large part a result of the knowledge of Ritzo Muntinga, top dog of ElectroSport Industries

http://www.thegsresources.com/garage/gs_statorfault.htm

The pages are basically a derivative of other copyrighted material from ElectroSport

http://www.electrosport.com/technical-resources/diagnosis-center/fault-finding-guide

While the ElectroSport Diagnostic charts have served many members well, there has been some room for improvement. So while back I updated the Phase A tests to improve on some limitations as well as clarify the purposes of some of the tests.

At that point the main issue was that the connector voltage drop tests were not being measured under load but rather only at idle (but I digress). I think this is the original thread and contains general information on charging system heath specifically focusing on how your R/R should be connected to maintain good charging performance.

FIX Chronically burning stators with SERIES R/R COMPUFIRE see GS CHARGING_SYSTEM_HEALTH

Here is the update to the Phase A tests.

Revised PHASE A of Stator Pages

And here are some updates with background to the Phase B tests that focus on the stator alone.

Revised PHASE B of Stator Tests

At some point (later) I plan to integrate the whole procedure with the revisions and include the Quick test

Want to just get a feel for how the charging is working?
Do a Quick_Test

The intent is to enhanced GS ownership by furthering the objectives as described in the stator pages :

The object is to present you with a clear testing scheme to determine which component in your charging system is at fault.

here is a reported problem on an Aprilla

you can't test the stator with an Ohm meter...i've got a fried one that reads 0.7,0.7,0.6 Ohms at the yellow wires with visible breaks in the windings!

http://www.apriliaforum.com/forums/showpost.php?p=1945577&postcount=72

[posplayr]

Here is at least one example that I recall where a temperature dependent fault caused alot of confusion and wasted time. The main issue is intermitence related to how insulation fails.

http://www.thegsresources.com/_forum/showthread.php?t=149577&highlight=stator

Psyguy had followed the Stator pages tests but still ended up with ambiuities in the tests results which failed to point out the stator issues.

In the revised tests, I have looked at the various combinations of fault and the principle factors in detecting insulation breakdown. The tests as outlined a simple and straight forward, but should improve the overall result. If you have a troubling problem the dummy load can be used to do further confidence testing of the stator insulation.

[Steve]

I have not had a chance to read it all, but I'm reasonably sure the information is all correct.

It's certainly not any easier, though.

I know, I know, the "easy" way is seldom the right way.

.

[posplayr]

I have not had a chance to read it all, but I'm reasonably sure the information is all correct.

It's certainly not any easier, though.

I know, I know, the "easy" way is seldom the right way.

.

Steve,
All you would do is replace the current stator tests in the manual or the Phase B stator tests (which are essentially the same) with the attached.

These are the same tests Graham suggested although I modified them to follow the description in the manual as well as add the dummy load test.

You can do them without a dummy load or do a dummy load test as well. The same tests the same method of diagnostics; You only need an AC Volt meter for 500 hz signal.

The dummy load test will only drop the voltage about 5% from what you would see open circuit (with a good stator).
Jim

[bakalorz]

Steve,
All you would do is replace the current stator tests in the manual or the Phase B stator tests (which are essentially the same) with the attached.

These are the same tests Graham suggested although I modified them to follow the description in the manual as well as add the dummy load test.

You can do them without a dummy load or do a dummy load test as well. The same tests the same method of diagnostics; You only need an AC Volt meter for 500 hz signal.

The dummy load test will only drop the voltage about 5% from what you would see open circuit (with a good stator).
Jim

Page 2 on your PDF: "open phase" this is not just a case of insulation breakdown, it is a broken wire, It is found with the unloaded tests, two of them will show 0 volts.

Your Excel power chart is incorrect.
the WIKI page uses Vpeak. the 80 VAC is in Vrms. this causes the voltage to be 1.414 times what you say. All the resistor power loads are 2 times what you think they are.

Your heat sink is 1/12 the size called for by the spec sheet (how could you get this wrong, you even circled the right size on the spec sheet; all you had to do was multiply by three)

Your note that "the dummy load will get hot to the touch in 2-3 minutes" is delusional: I have a dummy load for a power supply that has 10 watt resistors being run at 6.25 watts (slightly more than half rated power)
I just tested it against a clock: 10 secs to warm, 20 secs to hot, 35 seconds to "too hot to hold" ... this was a power resistor run at 1/2 rated power, you are running yours at 4 times rated power on a heat sink 1/12 the required size ... it will be burning/melting hot in far less than a minute.

You still need to provide a hot check of the stator. (your dummy load as designed does not draw enough current, nor will it survive long enough to heat the stator) I recommend (as I have since the beginning) take a ride
for 20 minutes with the stator connected as normal, when you return disconnect and check it.

All in all, some excellent highly competent work here Posplayer ... I just can't believe I ever doubted your competence ...

But my main question is still WHY.

After all that, what do you think loaded tests are going to show you that the unloaded ones do not. Assuming you eventually figure out how to build a load that won't actually catch fire (and hopefully not while sitting on your bike) what do you hope to gain.

How does the dummy load provide "further confidence testing of the stator insulation"

Explain it to me because I'm apparently not competent enough to see it.

what readings could I hope to see with this test that won't show on the unloaded test, and why do they show up under your test.
(and don't just lamely say "cause its loaded now" ... loaded tests are better in some cases for actual reasons ... what are the actual reasons in THIS case to make the loaded tests better)

Also, please explain how the need for loaded tests squares with Grahams statement that "out of over 300 stators tested in a 5 years span not 1 passed the tests above and failed to work as manufactured." (referring to just unloaded phase to phase and phase to ground (presumably both cold and hot if required))

Why do we need to build a dummy load if the unloaded tests have a perfect record (unloaded as described above, not current stator papers)

And even if the unloaded tests weren't perfect how are your tests better ... explain it in detail, not vague generalizations.

[Graham]

Page 2 on your PDF: "open phase" this is not just a case of insulation breakdown, it is a broken wire, It is found with the unloaded tests, two of them will show 0 volts.

Your Excel power chart is incorrect.
the WIKI page uses Vpeak. the 80 VAC is in Vrms. this causes the voltage to be 1.414 times what you say. All the resistor power loads are 2 times what you think they are.

Your heat sink is 1/12 the size called for by the spec sheet (how could you get this wrong, you even circled the right size on the spec sheet; all you had to do was multiply by three)

Your note that "the dummy load will get hot to the touch in 2-3 minutes" is delusional: I have a dummy load for a power supply that has 10 watt resistors being run at 6.25 watts (slightly more than half rated power)
I just tested it against a clock: 10 secs to warm, 20 secs to hot, 35 seconds to "too hot to hold" ... this was a power resistor run at 1/2 rated power, you are running yours at 4 times rated power on a heat sink 1/12 the required size ... it will be burning/melting hot in far less than a minute.

You still need to provide a hot check of the stator. (your dummy load as designed does not draw enough current, nor will it survive long enough to heat the stator) I recommend (as I have since the beginning) take a ride
for 20 minutes with the stator connected as normal, when you return disconnect and check it.

All in all, some excellent highly competent work here Posplayer ... I just can't believe I ever doubted your competence ...

But my main question is still WHY.

After all that, what do you think loaded tests are going to show you that the unloaded ones do not. Assuming you eventually figure out how to build a load that won't actually catch fire (and hopefully not while sitting on your bike) what do you hope to gain.

How does the dummy load provide "further confidence testing of the stator insulation"

Explain it to me because I'm apparently not competent enough to see it.

what readings could I hope to see with this test that won't show on the unloaded test, and why do they show up under your test.
(and don't just lamely say "cause its loaded now" ... loaded tests are better in some cases for actual reasons ... what are the actual reasons in THIS case to make the loaded tests better)

Also, please explain how the need for loaded tests squares with Grahams statement that "out of over 300 stators tested in a 5 years span not 1 passed the tests above and failed to work as manufactured." (referring to just unloaded phase to phase and phase to ground (presumably both cold and hot if required))

Why do we need to build a dummy load if the unloaded tests have a perfect record (unloaded as described above, not current stator papers)

And even if the unloaded tests weren't perfect how are your tests better ... explain it in detail, not vague generalizations.

Glad you understood the math because I have no idea how he came up with his dummy load. Now who will be the first to risk a stator? Now the stator papers complicated. I don’t know why he couldn’t use the factory regulator for the load. Someone asked that in this thread.
http://www.thegsresources.com/_forum...=1#post1394292

[bakalorz]

Glad you understood the math because I have no idea how he came up with his dummy load.

Heh, posplayr didn't understand the math either, not really. If he would have actually been able to follow the derivation he would have never missed that they used peak voltages rather than RMS voltages.
I bet he just googled until he found something that looked vaguely like what he needed and then plugged and chugged in some numbers.
To be honest I'm surprised he got the phase voltages right this time, he screwed it up in the other thread.

The really curious (to me) thing is that his excel power/resistor chart does not quite follow the equations that were on the wiki page.
I tried various permutations of his load resistors and the stator resistances, and none of them quite matched up with his calculated powers.
So his actual calculations are slightly different ... don't know why, and in any case they are wrong by roughly a factor of two.

Now who will be the first to risk a stator?

His dummy load won't actually hurt the stator, but as described in this post the dummy load will get really really really hot pretty quick.
If you suspend it in the air with wire, or put it on an asbestos blanket or something it might work (or it might burn itself out, I read the datasheet for the resistors after I posted ... they say you can do a 500% overload for only 5 seconds, posplayer is overloading by more than 400 percent. His heatsink is also 12 times too small)

But you just know that some poor guy is going to follow his advice, make the thing and place it on the gas tank or the wiring harness and fire it up. THAT is going to end in at least damage, if not a fire. And thats why I say Posplayer is DANGEROUSLY INCOMPETENT.

I will admit to occasionally briefly overloading power resistors. But not by that much, not at that high a power level, and I certainly wouldn't recommend others do it unless they have a very clear idea of the risks. As written, his test is absolutely irresponsible, perhaps even criminally negligent.

Now the stator papers complicated.

And all for nothing ... he has given no real reason his test is desirable other than his say so based on "unloaded bad, loaded better" but with no real reason.

But he's too incompetent to realize this, and to stubborn to give up on it. He's going to needlessly complicate the stator pages because of his ego. Hopefully he won't do it dangerously when all is said and done.

Here's a hint for you posplayer, when you fix this, the resistors must be at least 200 watts each, (stressed cause this is the second time I had to tell you this) preferably 250 (assuming 10 ohm resistors). And even then you need a warning that they will get VERY hot, hot enough that they must be kept away from the bike, all other wiring, and anything flammable.

I don’t know why he couldn’t use the factory regulator for the load.

He was originally suggesting just that; unfortunately it won't work right with an OEM R/R which only regulates 1 or 2 phases ... it would make a good stator look bad (unbalanced)

Someone asked that in this thread.

http://www.thegsresources.com/_forum...=1#post1394292

[tkent02]

I recommend (as I have since the beginning) take a ride
for 20 minutes with the stator connected as normal, when you return disconnect and check it.

This won't work either. The stator is oil cooled, it takes a lot longer than 20 minutes for the oil to get up to operating temperature, at least on any GS with an oil temp gauge. Maybe in Arizona in summer that would be long enough.

All interesting reading, I'll stay out of the electronics part of it. I can follow what you guys say but It's been twenty years since I did much more electrical work than fixing broken things. I have forgotten a lot from disuse. Keep it going, great discussion.

[bakalorz]

This won't work either. The stator is oil cooled, it takes a lot longer than 20 minutes for the oil to get up to operating temperature, at least on any GS with an oil temp gauge. Maybe in Arizona in summer that would be long enough.

All interesting reading, I'll stay out of the electronics part of it. I can follow what you guys say but It's been twenty years since I did much more electrical work than fixing broken things. I have forgotten a lot from disuse. Keep it going, great discussion.

Ok, I stand corrected, thats good information to have. how long a ride would you have to take. Would idling without moving work better since there is less aircooling then, or would the less load on the engine make it take longer.

[tkent02]

Ok, I stand corrected, thats good information to have. how long a ride would you have to take. Would idling without moving work better since there is less aircooling then, or would the less load on the engine make it take longer.

I don't know, I've only ridden GSes with oil temp gauges a little bit. The temp seems to creep up a long time when riding, maybe almost an hour to stabilize.
Some on this forum that have the newer bikes can offer more help here.

[posplayr]

This won't work either. The stator is oil cooled, it takes a lot longer than 20 minutes for the oil to get up to operating temperature, at least on any GS with an oil temp gauge. Maybe in Arizona in summer that would be long enough.

All interesting reading, I'll stay out of the electronics part of it. I can follow what you guys say but It's been twenty years since I did much more electrical work than fixing broken things. I have forgotten a lot from disuse. Keep it going, great discussion.

The point of the dummy load is that it is a passive device, not subject to any particular failure modes from just setting in a parts bin. ESD will not affect it and unless it looks burned it is OK. You can always check it with an ohm meter as it is a 1% tolerance device and should spec out as indicated.

The charging system performance is really a function of four primary elements
a.) Connections
b.) Battery
c.) Stator
d.) R/R

After doing a Phase A set of tests, having a "good" test for the stator not only helps in narrowing the diagnosis but will also help make the correct decision when it comes time to decide whether to swap out the stator or not. Given the stator is generally more expensive than a SHUNT R/R and alot harder to get out, a conclusive test is warranted. In the past , the existing tests have been shown to be problematic.

The VAC stator tests are designed to be accomplished by a single person with a VAC meter; the test should not take more than 10 minutes total. Generally the bike should not be at 5K RPM for more than about 30 seconds for each of 6 tests.

If you have a working R/R it will heat up the stator faster than this dummy load. There is about 10 amps 0-peak running in a normal R/R at the point it starts to regulate (2500 RPM). This dummy load will produce less than 1/5 of that at 5000 RPM.

A working SHUNT R/R will really heat up a stator because it shorts the stator (albiet with a diode) and it gets much hotter. I have previously measured 17 VAC RMS (equivalent of 25 amps 0-peak) stator leg current with a SHUNT R/R at 5K RPM. The dummy load produces less than 5 amps 0-peak (also at 5K RPM).

Given less than 5 amps in each leg of the dummy load, the total power in the load is about 300 watts, but each leg is only 100 watts. With a decent heat sink those resistors are rated to dissapate 100 watts continuous. The tests do not need that as you will probably only run for less than 5 minutes at a 30-50% duty cycle. When the bike is idling the current will drop way off and will not heat the dummy load to anything near the same extent as when it is at 5000 RPM or above.

[posplayr]

This is not intended as a sophomore course in Electrical Engineering but since Graham "called me out" (I guess that is a challenge to produce actual equations) and nobody else seems to know how to calculate the power in a dummy load, I have summarized the calculations for a 10 ohm power resistor below.

The equations are a simple and straight forward application of balanced, non-reactive, 3 phase power equations. This is typically covered in a 1st semester junior year in a EE curriculum (I probably did it in an undergraduate lab back in the late 70's). It is probably covered much earlier in a trade school. There are many readily available "How To" tutorials all over the web for this stuff. If you don't know how to do it, maybe you should go pick up a reference and figure it out.
There is nothing particularly complicated and with a minimum sense of mathematical and technical awareness of electrical circuits should know how to apply the equations in the referenced link.

It would be very unusual for someone with an EE degree to not remember at least doing these calculation and they would be fully expected to be able to figure it out even if they forgot something from many years ago. It certainly would not even cross my mind to suggest someone with a EE could NOT go back and figure out these equations. So when someone suggests I don't know how to do it, I take it as an insult. It would be equivalent of me claiming that Graham doesn't know how to change the brake pads on a bike. That would be a pretty ludicrous statement, because assuming he does work as a mechanic (not sure why he would lie about that), he is bound to have found himself in a situation where he needed to change pads. It is expected.


This thread makes me question why I even bother to post analysis of GS electrical systems and trouble shooting when what happens is that it draws fire from apparently insecure idiots hiding behind the internet (I'm specifically referring to Balzar and Graham at the moment).

I try to keep the material relevant at a MINIMUM technical level to convey the issues for generally non technical members. I'm not paid to do this, so it is more of a way of "paying it forward" for the help I have received.

For those individuals that can't figure out these simple calculations , but nevertheless claim that they are wrong, for that I will bestow the highest honor I can; display it with pride :

Stator resistance :
R_stator – measured in ohms
R_stator _leg = 0.5 (spec is 0.4-0.5 ohms)

Phase Resistor:
R_dummy leg – stator leg resistor value in ohms
R_dummy_leg selected for a 100 watt thermal constraint
R_dummy_leg = 10 ohms

Phase to Phase Voltage:
VAC_pp – ac voltage 0-peak phase to phase
VAC_pp – 80 VAC at 5000 RPM (measured)

Phase to Neutral Voltage:
VAC_pn – ac voltage 0-peak phase to neutral
VAC_pn = VAC_pp/sqrt(3) = 46.2 VAC 0-peak

Power _Total – total power dissipated in resistive loads (ignores reactance)
Power_Total = 3/2*VAC_pn^2/(R_stator_leg+R_dummy_leg/2)
Power_Total = 312.2 watts

Power per resistor – power per dummy leg
Power per resistor – Power Total/3
Power per resistor = 104.1 Watts

Amps per Phase – current through each leg
Amps per Phase – 2*Power per resistor/VAC_pn
Amps per phase – 4.5 Amp 0-peak

[Skateguy50]

This should answer all your questions:

[duaneage]

I like Skateguy's stator paper the best. Definitely saved that one off.

[duaneage]

This thread makes me question why I even bother to post analysis of GS electrical systems and trouble shooting when what happens is that it draws fire from apparently insecure idiots hiding behind the internet (I'm specifically referring to Balzar and Graham at the moment).

I told you years ago, you're overanalyzing this with solutions in search of the problem. We all know what the problem is, crappy RRs and under designed charging systems. All the average GSer has to do is swap the RR out and slap a new stator in and call it a day.

There are thousands of GS bikes out there and the vast majority run every day with that silly shunt regulator and a regular old stator. Every once in a while a RR gives up a diode or SCR and it breaks. Who cares about power tables, dummy loads, phase relationships, etc. Remove, Replace, Ride. It's really not a factory defect if millions of cars, bikes and boats use the same regulation method mile after mile every day. Put that in perspective.

I've read your threads on the mods to your bikes, you've got real talent and it shows. I'm perplexed why you've made GS charging systems your primary interest, how about the ignitors or electronic turn signal controllers? Design a digital shift indicator or an electronic speedo. That would be something!

Can we call all this off before it gets ugly and Frank has to step in and call a time out?