Testing alternator.

Discussion in 'General Motoring' started by Matt Ion, May 22, 2005.

  1. Matt Ion

    Matt Ion Guest

    Bought an '87 Accord hatchback recently (2.0l, 2bbl, 5spd)... owner was
    selling it because the alternator "didn't work". Seemed to me that it
    worked, but not 100%. In any case, I swapped in the alternator from my
    identical '87 sedan that had just bit the dust, and all was happy until
    a couple nights ago: charge light flickering, dashlights dim, except
    when revving high. Typical indication of bad brushes. Figured I was
    doing okay trying to get home when suddenly the charge light came on
    solid and wouldn't go out for anything. JUST made it home, engine
    sputtering every time I hit the brakes.

    So today I pulled apart the original alt. and checked it out: diodes all
    test fine, field windings seem okay, no opens or shorts anywhere.
    Brushes weren't too badly worn, so not really sure what was wrong with it.

    Got the replacement alt. out of the car (finally - some Japanese SOB
    needs a serious bitchslapping over the layout of that engine) an opened
    it up... brushes were REALLY worn, but still making contact - barely.
    Diodes and windings checked okay as well.

    Now the only thing I can't test is the voltage regulator. Or can I?
    With a stanard multimeter, that is, since I don't have a nice
    super-duper alternator/starter bench tester sitting on my kitchen
    table...?

    I took a few readings across various terminals of both regulators, DMM
    set to Diode Check, and got several readings that differed between the
    two... for example, testing from "terminal A" (named arbitrarily for the
    sake of argument) to terminals B and C on one regulator showed infinite
    resistance, but on the other, they showed about 1200Mohms aa 1800Mohms,
    respectively. Reversing the leads, the first regulator showed 1200 and
    1500 (B-to-A and C-to-A), but the other, I got 1500M and infinite,
    respectively. Obviously I don't know which one is "right" reading, and
    it's not enough to make an intelligent diagnosis.

    So... does anyone know of any way I can test the regulator? Or better
    yet, know how I can hook up the alternator on the bench (ie. what goes
    to what connections) and test it there, maybe spin it up with my
    cordless drill?

    TIA...
     
    Matt Ion, May 22, 2005
    #1
  2. There isn't any great way to test it on the bench with common tools - it is
    a lot easier to test it in service with a DMM. However, most major auto
    parts stores have a test jig, where they can spin the alternator with an
    electric motor and measure the output under various loads... and they will
    do it for free.

    Mike
     
    Michael Pardee, May 22, 2005
    #2
  3. There isn't any great way to test it on the bench with common tools - it is
    a lot easier to test it in service with a DMM. However, most major auto
    parts stores have a test jig, where they can spin the alternator with an
    electric motor and measure the output under various loads... and they will
    do it for free.

    Mike
     
    Michael Pardee, May 22, 2005
    #3
  4. Matt Ion

    Matt Ion Guest

    There must be some way to test the regulator module itself - supply
    juice to the inputs, measure the outputs? As I say, all the other
    components are fairly "testable" on their own and seem to be fine...
     
    Matt Ion, May 23, 2005
    #4
  5. Matt Ion

    Matt Ion Guest

    There must be some way to test the regulator module itself - supply
    juice to the inputs, measure the outputs? As I say, all the other
    components are fairly "testable" on their own and seem to be fine...
     
    Matt Ion, May 23, 2005
    #5
  6. Maybe, but I'm not sure. Some old voltage regulators used analog drivers,
    but I think all of them now are switch mode, pulsing the inductance of the
    field. You could connect the regulator to the alternator and measure current
    drain as you turned the input voltage up - the current should drop rapidly
    as you exceed the regulator point (somewhere around 14 volts).

    Mike
     
    Michael Pardee, May 23, 2005
    #6
  7. Maybe, but I'm not sure. Some old voltage regulators used analog drivers,
    but I think all of them now are switch mode, pulsing the inductance of the
    field. You could connect the regulator to the alternator and measure current
    drain as you turned the input voltage up - the current should drop rapidly
    as you exceed the regulator point (somewhere around 14 volts).

    Mike
     
    Michael Pardee, May 23, 2005
    #7
  8. Matt Ion

    Matt Ion Guest

    Cool, except it still has to be in the car for that, which in itself is
    a nightmare, and at that point is wholly inaccessible :) Assuming one
    determines which connections are which on the thing, too.

    I guess that's what I'm really hoping, is to find the pinouts for the
    module; given those, I may be able to at least tell, without putting it
    all back together and into the car, whether either of the two regulators
    I have are *obviously* faulty.

    Recall that the original alternator in the car worked intermittantly,
    and upon disassembly and testing, the brushes seem sufficient and
    everything else I CAN test (windings, diodes) seems good, so that leaves
    the regulator as "questionable".

    The behaviour of the second one worries me a little, the way it was
    doing the typical worn-brush thing for only a little while before
    packing it in completely: again, the brushes are obviously gone, and
    everything else tests fine, and I don't want to put it all back together
    and into the car only to discover the regulator is pooched. A simple
    works/doesn't-works test is all I'm looking for.
     
    Matt Ion, May 23, 2005
    #8
  9. Matt Ion

    Matt Ion Guest

    Cool, except it still has to be in the car for that, which in itself is
    a nightmare, and at that point is wholly inaccessible :) Assuming one
    determines which connections are which on the thing, too.

    I guess that's what I'm really hoping, is to find the pinouts for the
    module; given those, I may be able to at least tell, without putting it
    all back together and into the car, whether either of the two regulators
    I have are *obviously* faulty.

    Recall that the original alternator in the car worked intermittantly,
    and upon disassembly and testing, the brushes seem sufficient and
    everything else I CAN test (windings, diodes) seems good, so that leaves
    the regulator as "questionable".

    The behaviour of the second one worries me a little, the way it was
    doing the typical worn-brush thing for only a little while before
    packing it in completely: again, the brushes are obviously gone, and
    everything else tests fine, and I don't want to put it all back together
    and into the car only to discover the regulator is pooched. A simple
    works/doesn't-works test is all I'm looking for.
     
    Matt Ion, May 23, 2005
    #9
  10. Matt Ion

    twillmon Guest

    Is there anything wrong with taking your alternator to a parts store
    for testing on their machine? Next town over has a Checker parts
    store with at least one real competent guy who would test yours.


    Tom Willmon
    near Mountainair, (mid) New Mexico, USA

    Sure it's funny! Now beam my clothes down Scotty!

    Net-Tamer V 1.12.0 - Registered
     
    twillmon, May 24, 2005
    #10
  11. Matt Ion

    twillmon Guest

    Is there anything wrong with taking your alternator to a parts store
    for testing on their machine? Next town over has a Checker parts
    store with at least one real competent guy who would test yours.


    Tom Willmon
    near Mountainair, (mid) New Mexico, USA

    Sure it's funny! Now beam my clothes down Scotty!

    Net-Tamer V 1.12.0 - Registered
     
    twillmon, May 24, 2005
    #11
  12. Matt Ion

    Matt Ion Guest

    Did that. Tested fine. Put it back in the car. Now it works
    intermittantly. It'll run 13V at idle, up to 14.1 at speed, but if I
    start turning stuff on (lights, blower) it'll drop below 12V; ie. not
    charging at all. Turning everything off again, it takes upwards of 30s
    of revving to start charging again.

    I'm wondering if a brush isn't in quite right, as there's a 'ticking'
    sound coming from it now that changes with engine speed.
     
    Matt Ion, May 24, 2005
    #12
  13. Matt Ion

    Matt Ion Guest

    Did that. Tested fine. Put it back in the car. Now it works
    intermittantly. It'll run 13V at idle, up to 14.1 at speed, but if I
    start turning stuff on (lights, blower) it'll drop below 12V; ie. not
    charging at all. Turning everything off again, it takes upwards of 30s
    of revving to start charging again.

    I'm wondering if a brush isn't in quite right, as there's a 'ticking'
    sound coming from it now that changes with engine speed.
     
    Matt Ion, May 24, 2005
    #13
  14. Matt Ion

    Jason Guest

    Unless you know how to rebuilt it yourself, I advise you to buy a new
    alternator. Another option is to visit a car junk yard. If the manager or
    owner is on a database of thousands of junk yard, he can order you one
    even if he does not have one in stock. The owner or manager may even
    provide a warranty for it in case it does not work. I once found a wrecked
    car just like the one I owned at that time and got lots of parts off of it
    for almost nothing.
     
    Jason, May 24, 2005
    #14
  15. Matt Ion

    Jason Guest

    Unless you know how to rebuilt it yourself, I advise you to buy a new
    alternator. Another option is to visit a car junk yard. If the manager or
    owner is on a database of thousands of junk yard, he can order you one
    even if he does not have one in stock. The owner or manager may even
    provide a warranty for it in case it does not work. I once found a wrecked
    car just like the one I owned at that time and got lots of parts off of it
    for almost nothing.
     
    Jason, May 24, 2005
    #15
Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.