Re: (ET) E-15 Reverse relay Arcing

[Dave & Debbie Barden <daveb seanet com>]

This is a long one and my apologies to anyone who may think I'm flogging a 
dead 
horse  :^)

Here is a more complete description of how my friend and I tackled the 
reversing relay problem on my E12S.  We haven't solved the problem and if 
anyone can see why after wading thru the following we'd appreciate 
comments/thoughts.  This note ends with what we'll try next when we get 
time to 
devote to it again.  It's been about 6 months or more since we last worked 
on 
it.  Hopefully there's something here that may help someone else just 
experiencing the problem.

dave


My friend Craig's commentary starts here:
 
 The problem is definitely the big EMF collapsing when the motor goes from
 fwd to neutral. (And even bigger from fwd to reverse.)
 The theory I was following was as follows:
 * The field winding supply V goes thru the small reversing relay. Under
 normal running conditions, it draws 2 amps. The DPDT relay (Hosfelt) was
 rated at something like 15A. And we added a 10A fuse inline. The field
 voltage turns on when the ETrak key turns on.
 * In neutral and fwd, this relay remains in the 'fwd' direction, and the
 field winding is energized at all times.
 * When switching from neutral into reverse, the reversing relay coil
 energiezes and reverses the polarity to the motor field.
 * The motor armature (lots of amps) always flows in one direction. The 
+36V
 to the armature gets switched off when in neutral, and switches on in 
either
 fwd or rev.

 Ok. Now for the problem...
 When the ETrak goes from fwd to neutral, the motor EMF needs to go
 somewhere. As the motor spins down and the ETrak coasts to a hault, it's
 generating EMF energy. Since the armature is now an open circuit there's 
no
 path to ground there. But the field still provides a path. And this is
 obviously where it's going. Thus the big pile of previously burned out
 relays and now 10A fuses.

 Our attempt was to connect a shunt power diode from the motor armature (+)
 to ground. The theory was that when the armature circuit opens, the
 generated EMF voltage (which is a reverse voltage of normal motor volts) 
now
 has a path to ground.

 The only good choice on the field/reversing relay circuit is a MOV
 (varistor) which acts as a shunt diode above a fixed DC voltage. One 
problem
 with a MOV is that it is absorbing a very substantial spike of energy and
 they deteriorate over time. The more spikes, the faster they wear out. And
 this one is getting lots of spikes. So a new one might solve the problem 
for
 awhile.

 My next experiment would be to use a low ohm power diode to diagnose the
 problem. (I've got a couple .01 ohm resistors.) Hook it in series with the
 field circuit. Then hook up the VOM across to measure the voltage. Use the
 VOM's 'max' setting and it should tell you how big the voltage drop is
 across the resistor. Actually, this isn't going to tell you much new since
 we already know it's more than 10A. (It regularly blows the 10A fuse.) In
 fact, just remove the 10A fuse and hook the power resistor across those 
two
 fuse contacts. (And be ready to replace the relay if it fries.) Using ohms
 law, you can then compute the amps being generated by the motor. I suspect
 you'll see small spikes when the tractor coasts to a stop (on pavement), 
and
 big spikes when it comes to an abrupt stop (mud or thick grass or on an
 uphill slope). This is pretty much what we saw. When the wheels were off 
the
 ground, we never blew any fuses.

 This test will also tell you whether the MOV is doing its job. With an
 operable MOV, you should never blow the 10A fuse. But you might eventually
 melt some wires. The MOV has to be on the motor side of the
 battery/relay/motor path, and preferrably on the 'motor' side of that
 connector. That was the original problem when you first got the ETrak. The
 MOV sucked up a big spike and melted a pin in the connector.

 The next test I would do is then move the power resistor in series with 
the
 shunt power diode that we added. I don't think the MOV is working that 
well,
 at least not for very long. Its voltage rating (the breakdown V before it
 shorts) may be too high, or it's just getting worn out really fast. The 
far
 better solution is to shunt this high energy using the high amp wires in 
the
 armature. But from our attempt, I'm not sure if that works. There might 
not
 be enough of this EMF being transfered from the outer motor windings into
 the armature. But it would be nice to quantify the EMF spikes in these two
 circuits, to see how much the shunt diode is handling and also how much 
the
 MOV is absorbing. You might need a big stack of MOVs.

 If anyone would like to try this, here's the test setup:

 Testing the relay/battery side of the MOV:
 This should show low 'max' millivolts across the .01 resistor.
         |--VOM---|
         |        |
 (+)-------\/\/\/-----------|
           .01 ohm  |      [motor
                    - MOV   field
                    |       windings]
 (-)------------------------|


 Testing the motor side of the MOV:
 This should show high 'max' millivolts across the .01 resistor.
              |--VOM---|
              |        |
 (+)------------\/\/\/------|
       |       .01 ohm     [motor
       - MOV                field
       |                    windings]
 (-)------------------------|

 And just for completeness...
 If you use a .01 ohm resistor, the VOM should read 20mV across the 
resistor
 with the nominal 2A that the field draws.
 If you set the VOM to 'max', and do the fwd to neutral thing, a 10A spike
 would give a reading of 100mV.
> 
> Check out my schematic for the proper placement of freewheel diodes and
> transzorbs.
> http://www.qsl.net/w8rnh/etrakh/ELECTRAK.PDF
> (the server seems to be down right now though?)
> 
> This should eliminate most sparking of the relay IF you come to a 
> complete
> stop before
> switching directions.
> When I wrote software for a compound wound forklift drive motor I had a
> similar problem.
> When you switch direction of an electric forklift you typically
> electronically brake by 'plugging' the motor.
> You could quickly switch between plugging to opposite direction on a 
> series
> wound motor.
> The compound wound motor required a 100 mS delay to allow the shunt field
> current to decay otherwise
> a huge surge current would flow.
> This is similar to what can happen when switching direction on the ET 
> while
> it's still moving in one
> direction and you reverse without stopping.
> Rod Hower - Senior Project Engineer
> Ametek Technical Products Engineering
> 627 Lake Street
> Kent, OH 44240-2660
> (330) 677-3875
> (330) 677-3306 Fax
> rod hower ametek com
> USA
> www.ametektmd.com
> 
> 
>                                                                          
>      
            
                   
>                       Dave & Debbie Barden                               
>      
            
                   
>                       <daveb seanet com>               To:       
<ssawtelle fcc net>,     
                   
>                       Sent by:                          
<elec-trak cosmos phy tufts edu>  
                   
>                       owner-elec-trak cosmos ph        cc:               
>      
            
                   
>                       y.tufts.edu                      Subject:  Re: 
> (ET) 
E-15 Reverse
relay Arcing           
>                                                                          
>      
            
                   
>                                                                          
>      
            
                   
>                       05/02/03 03:40 AM                                  
>      
            
                   
>                                                                          
>      
            
                   
>                                                                          
>      
            
                   
> 
> 
> 
> 
> I have an E12 with this same problem.  Admittedly still unsolved.  after
> repairing the relay several times it finally fused together beyond 
> repair.
> 
> before putting in the Bill Gunn version ($20) I came across the Hosfelt
> catalog
> that had  some surplus relays of similar specs for $2.00each. I bought 6
> (they
> were subsequently stolen along with my van but that's another story).  I
> burned
> up one of those before putting in a fuse in line between the relay and 
> the
> field.  now I just blow fuses but all the time.  On the E12 the varistor
> ($4
> from Bill) is behind the lower panel cover jumping the field wires in the
> connector.  I replaced it.  A friend of mine thought putting in a diode 
> to
> block the electrical surge of the collasping field when the motor changes
> direction would help.  It didn't.  I suspect in my case something else is
> amiss
> but I haven't found it yet.  good luck with yours!
> 
> dave
> E12S in Seattle
> 
> 
> > - Yep, the schematic shows one. I'm not sure offhand where it is 
> > located,
> > but I imagine it's on the reverse relay or on a small card nearby.
> >
> > Also  the E12S, E15 manual (Elec-Trak.Org) has reverse relay
> troubleshooting
> > and a service bulletin at the very end.
> >
> >
> > >
> > > There might be a varistor that goes across the relay to absorb the
> > > sparks, or a capacitor. If so I'd bet it's blown.
> > >
> >
> >
> >
> >
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
>