Announcement

Collapse
No announcement yet.

FIRST and SECOND LAW of Stator testing

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    FIRST and SECOND LAW of Stator testing

    I have posted the following update GS Charging health relating to various tests of the actual stator v.s. the whole set of charging system tests: The points made below was one of the major factors in developing the "Quick Test".

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

    A revised test is at this link; The actual test is on Page 6 of 9. We added the leg to ground AC voltage test as this helps isolate insulation breakdowns to ground using the relatively high 60-80 VAC stator voltage when it is open loop.

    Link to Revised PHASE B of Stator Pages with discussion of testing methods:
    http://www.keepandshare.com/doc/3977...4-pm-649k?da=y

    Reading just this limited amount of information, there should be ample information to conclude that the testing of a stator amounts to the testing of the insulation. Insulation tests can not be done conclusively with a low voltage ohm meter (like measuring sub ohm resistance with a 9V battery).

    Standard engineering insulation tests that have been developed are based on much higher voltage breakdown. Unless there is a high voltage consistent with the expected operational range of use, then any positive test results will be inconclusive

    So when I say a conventional stator test (non - MEGGER) can only unambiguously proves a fault, it is because it those tests do not fully test the stator to the level required in operational use, then concluding that the stator will pass in operational use leaves a large gap in performance and therefore an apparent ambiguity in the results.

    The stator test is only valid is it fails, as if it fails at a low level you can be assured it will fail at a higher level. I can state this in the following RULES for testing stators.

    This brings us to the following LAWS for Stator testing (i.e. those related to Phase B of the stator pages)


    FIRST LAW OF STATOR TESTS:
    ANY POSITIVE RESULT of a STATOR TEST HAS AN AMBIGUITY PROPORTIONAL TO THE UNTESTED OPERATIONAL STRESS.

    SECOND LAW OF STATOR TESTS:
    ANY NEGATIVE RESULT of a STATOR TEST IS A NECESSARY AND SUFFICIENT CONDITION TO DECLARE A STATOR BAD.

    What is positive and what is negative you might ask? You will have to refer to the state ranges of the test and deduce make your decision based on the cost of making such a decision. The costs will vary depending upon your situation.

    For example, the stator voltage tests say to have between 65-80 volts at 5000 RPM. Say you have two stator at 75 VAC and one at 65 VAC. Is that good or bad? Well it depends. There is a clear indication that there is a degradation in the single low measurement (there is imbalance), It is still good enough withing the stated ambiguity of of the First Law. So the question of cost is, are you willing to forgo the maintenance and increase your probability of inopportune failure? Or do you want to incur the expense to change something that is still operational but appears to be on the way out. This decision has nothing to do with the test or whether something is good or bad, but instead determines the optimal decision for maintenance based on the likelihood of failure and cost.
    Last edited by posplayr; 06-11-2014, 02:43 PM.

    #2
    Related Links



    Comment


      #3
      Don't understand a word of it. Fortunately I'm pretty good at following your instructions and don't need to.....
      Old age and treachery will beat youth and skill every time
      Originally Posted by Schweisshund I mean, sure, guns were used in some of these mass shootings, but not all of them
      1983 GS 750

      Comment


        #4
        Originally posted by bccap View Post
        Don't understand a word of it. Fortunately I'm pretty good at following your instructions and don't need to.....
        Stator tests only say when something is bad not when something is good. So a positive test result does not mean it is not failed.

        Comment


          #5
          One quick question. I was performing the first test on the below linked page where it says start and never got to Part B. It was before I knew about your tests. I had boiled away my new battery and my quick charge tests showed I was more than 18 volts at 5000rpm. Can't remember the exact number. No matter what I did the test sent me in an endless circle as my number was always higher in the last check to be below .2 v. I was pretty sure it was the r/r as I had jumped it with a running car so I ended up putting in a new one and problem solved. Was I doing something wrong ?
          Old age and treachery will beat youth and skill every time
          Originally Posted by Schweisshund I mean, sure, guns were used in some of these mass shootings, but not all of them
          1983 GS 750

          Comment


            #6
            How much voltage does a megger use? I know it's quite a bit, I got zapped once.


            Life is too short to ride an L.

            Comment


              #7
              Agreed. No component should be deemed as fully functional until all tests have been completed. This is a basic premise of test strategy. There is also the concept of probability of failure and confidence intervals. You can conduct the majority of the provided tests and have a confidence factor of say 75% or more that the component is likely functional based on the number and quality of the completed tests. It wont guarantee that it's good but might help you to identify the most likely conclusion. Of course I would recommend performing all tests whenever possible. Some testing is simply out of the realm of some or are to restrictive. Equipment constraints and understanding of the procedure are usually the source of those restrictions.

              1980 GS1000GT (Daily rider with a 1983 1100G engine)
              1998 Honda ST1100 (Daily long distance rider)
              1982 GS850GLZ (Daily rider when the weather is crap)

              Darn, with so many daily riders it's hard to decide which one to jump on next.

              JTGS850GL aka Julius

              GS Resource Greetings

              Comment


                #8
                Originally posted by tkent02 View Post
                How much voltage does a megger use? I know it's quite a bit, I got zapped once.
                I believe they have multiple, 300v 750v and 1000v rings a bell. You should test using an appropriate level for the insulate you're testing so you don't fry it during the test.
                It's smoke that make electronic components work.
                Every time I've let the smoke out by mistake, they never work again.
                '80 GS250T... long gone... And back!
                '86 Honda Bol D'Or... very sadly long gone
                '82 GS1000SZ
                '82 GS1100GL
                '01 Honda CBR1100XX BlackBird

                Comment


                  #9
                  Originally posted by tkent02 View Post
                  How much voltage does a megger use? I know it's quite a bit, I got zapped once.
                  Depends on the insulation factor of the component being tested Usually in the 250, 500 to 1000v range but I've seen tests conducted as high as 15KV. It's a high voltage low current test that yields measurements typically in the gigohms range.

                  1980 GS1000GT (Daily rider with a 1983 1100G engine)
                  1998 Honda ST1100 (Daily long distance rider)
                  1982 GS850GLZ (Daily rider when the weather is crap)

                  Darn, with so many daily riders it's hard to decide which one to jump on next.

                  JTGS850GL aka Julius

                  GS Resource Greetings

                  Comment


                    #10
                    Originally posted by tkent02 View Post
                    How much voltage does a megger use? I know it's quite a bit, I got zapped once.
                    They can vary. Some put out thousands of volts, as they are designed to test the breakdown voltage of wire insulation or the integrity of an earth ground.

                    I think we would need to know what the stator manufacturer wanted in terms of breakdown voltage of the varnish, whilst operating in an oil bath.

                    EDIT: - what they said....

                    Comment


                      #11
                      Originally posted by bccap View Post
                      One quick question. I was performing the first test on the below linked page where it says start and never got to Part B. It was before I knew about your tests. I had boiled away my new battery and my quick charge tests showed I was more than 18 volts at 5000rpm. Can't remember the exact number. No matter what I did the test sent me in an endless circle as my number was always higher in the last check to be below .2 v. I was pretty sure it was the r/r as I had jumped it with a running car so I ended up putting in a new one and problem solved. Was I doing something wrong ?
                      http://www.thegsresources.com/garage/gs_statorfault.htm
                      The stator pages are a 20+ year old diagnostic procedure by a skilled designer of electronics, not someone well versed in optimal diagnostics strategies. It is much better than a sharp stick in the eye, but not perfect.

                      My tests have attempted to simplify things and modify things based on the audience and the measured performance of the community as a whole in performing diagnostics.

                      I chose to not rewrite the entire set of stator tests as they were so well known and recommended. I just tried to update them with a minimum set of changes.

                      In the end I did the Quick test as a feeder to more detailed tests in the stator pages. Some of the stator pages was just flawed:
                      • like not testing voltage drops at 5000 RPM
                      • not doing leg to ground tests,
                      • some of it is based on the original Suzuki tests which are the worst (measuring stator resistance is close to useless although not entirely useless)
                      • R/R tests that are only applicable to Shunt style devices with full wave rectifiers.


                      If you read the GS stators link in my signature, there is an early relationship of the GS Resources with Electrosport and so the Original Stator Pages will not be taken down from the GSR, and they are not being changed to reflect the best knowledge available (despite attempts to inform them)

                      So the original will continue to mislead people as long as other members here keep pointing those new members to the old stator pages.

                      I am still amazed at how many long term GS members don't even know what a Quick Test is???? There are a few people that, recognize the value and actively recommend the test to get to the bottom of the charging system problems of the people that come knocking.

                      I know there are many more GSR members that feel compelled to retype out a outline of a test indicating what to do when a full description and diagnostic section already exists in "Quick Test". Any recommendations or modifications to the test are welcome just post to that thread, but in the mean time it represents the best knowledge on performing an initial assessment that exists and steering anybody to anything else is a disservice to that newcomer.

                      Comment


                        #12
                        Originally posted by JTGS850GL View Post
                        Agreed. No component should be deemed as fully functional until all tests have been completed. This is a basic premise of test strategy. There is also the concept of probability of failure and confidence intervals. You can conduct the majority of the provided tests and have a confidence factor of say 75% or more that the component is likely functional based on the number and quality of the completed tests. It wont guarantee that it's good but might help you to identify the most likely conclusion. Of course I would recommend performing all tests whenever possible. Some testing is simply out of the realm of some or are to restrictive. Equipment constraints and understanding of the procedure are usually the source of those restrictions.
                        Those are all very general notions of theoretical test. Most people don't understand test theory , confidence intervals, measurement error, decision theoretics, estimation theory or any other form of optimization and especially not for diagnostics procedures and systems. These Laws are specific to stator testing but also entirely consistent with the test theory you outline.

                        • A failed test is the strongest evidence of a bad stator
                        • A passed test is ambiguous.


                        The asymmetry of the test result is what I am trying to highlight. It is not obvious to many, and only came to me on analysing the best test procedure for the GS Charging systems.

                        Comment


                          #13
                          Relevant updates to this thread.


                          Comment

                          Working...
                          X