Ground Loops

[mottyl]

"I really do have a life outside of GS charging but this really Pi$$ed me off."

Glad you're bike is working. You dug out the calculator and went over the whole system to try and find the problem?

I'm either impressed or scared not sure what yet.

Note to self "Don't upset this guy!"

Chris

 

[rustybronco]

Anybody that doesn't, doesn't know and so there is no harm in having a technical discussion in a technical forum.
Jim

I would like to add a comment if I may.

some like myself may have had electrical/ electronic experience. in my case it was over 30 years ago. I took first place in radio servicing in the state (they didn't have national competitions back then). I even played around with television repair. I have designed circuits, built transmitters and even went to school to get a BSEE . but sadly my family got to be a large load and I had to put it on hold, and never finished what was started.

Having these kinds of discussions are not lost on deaf ears, even those that don't understand can and do pick up valuable information. it may even turn on a few lights to those that wondered why the voltage readings they have taken, read a negative value instead of an expected positive value, upon checking the wiring to find an answer to their problems.

Jim, I know it takes away your time to do things like this, but I ask you to keep it up.

some of us enjoy re-learning what we have forgotten...

 

[duaneage]

I used to solder wires where ever possible but was recently told by an engineer friend that crimped connectors are more reliable. No flux acid involved and the heat from the solder tends to make the wires brittle. Makes sense to me.

It's true. The pressure in a properly crimped connection is incredible. The wire forms into the connector. Soldering after crimping is acceptable as long as you don't use acid flux. Rosin flux is the standard these days. I use 63/37 audio grade solder.Also use a low wattage iron, around 25 watts or so.
Soldering does have the advantage of sealing out moisture but without a good mechanical joint it is prone to breakage. When I extend the wires on a Honda RR I wrap them tightly together to form a strong joint. The two wires won't pull apart even before soldering. After soldering I seal with heatshrink and overwrap with tape. So far there have been no reported problems with the connections. The main advantage to connectors is you have access to the device easily without having to resolder again and again.

I used wire wrap on telecom gear years ago. There are wire wrapped connections in use today that were done 60 years ago.

 

[reddirtrider]

I used to solder wires where ever possible but was recently told by an engineer friend that crimped connectors are more reliable. No flux acid involved and the heat from the solder tends to make the wires brittle. Makes sense to me.

The solder makes the wire inflexible and prone to breaking at the joint. Yes, crimping is better, talk to any defense avionics technician.

If you really want to get into this topic check out what methods of wire stripping are acceptable. They have manuals on this stuff.

 

[reddirtrider]

I guess I'm confused by the term 'sneak paths' here in reference grounding schemes.

When engineers speak of sneak paths it's typically of redunant paths that interfere with the normal operation of a system. For instance an alternate path for power to a radio that bypasses the fuse if there is a component failure in another system.

I've gotta be truthful here though, I haven't read this entire thread. I deal with this stuff during the day and have a very short attention span during the evening.

OK, back to your regularly scheduled programming.

 

[mottyl]

"When engineers speak of sneak paths it's typically of redundant paths that interfere with the normal operation of a system. For instance an alternate path for power to a radio that bypasses the fuse if there is a component failure in another system."

plus 1

I"ve heard that phrase before in that context.

Chris

 

[Sandy]

It's true. The pressure in a properly crimped connection is incredible. The wire forms into the connector. Soldering after crimping is acceptable as long as you don't use acid flux. Rosin flux is the standard these days. I use 63/37 audio grade solder.Also use a low wattage iron, around 25 watts or so.
Soldering does have the advantage of sealing out moisture but without a good mechanical joint it is prone to breakage. When I extend the wires on a Honda RR I wrap them tightly together to form a strong joint. The two wires won't pull apart even before soldering. After soldering I seal with heatshrink and overwrap with tape. So far there have been no reported problems with the connections. The main advantage to connectors is you have access to the device easily without having to resolder again and again.

I used wire wrap on telecom gear years ago. There are wire wrapped connections in use today that were done 60 years ago.

LOL. When first read the title I thought I was going see a really great picture of Jim doing huge smoking donuts on his streetfighter!!!

Same here. If you're only going to use one method use crimping, you need that mechanical bond. Simply soldering connectors without first crimping could lead to the wire popping out if the wire gets too hot. I also wrap the wire or connector with whatever is necessary for the application. To duplicate something like the factory bullet connector covers I first put a small piece of shrink tube on the connector at the wire joint to secure the wire then cover that with a larger piece of clear that will more than cover both connectors and then only shrink one end. I also somewhat secure it to something nearby such a frame strut or other wiring, don't secure too tightly or you'll have the connections breaking.

 

[posplayr]

Connections

Connections

The solder makes the wire inflexible and prone to breaking at the joint. Yes, crimping is better, talk to any defense avionics technician.

If you really want to get into this topic check out what methods of wire stripping are acceptable. They have manuals on this stuff.

Yes it is generally accepted that crimping is more reliable that a solder joint. The primary reasons are:

1.) Crimping in this context means using a $500 crimper, and it not the one you get at autozone.

2.) Solder where the solder wicks way up the wire will make the wire stiff and prone to breakage under vibration. (Don't do that).

3.) A solder joint when it gets hot will melt the solder and without other mechanical restraint fall out.

However, if you do the following I prefer a crimp and solder approach for home motor applications.

a.) clean the wire and if it is old soak in some acid metal prep.

b.) crimp the end of terminal to the wire

c.) If you can see some of the metal of the connector and have a find soldering iron flow some solder into the crimp to securely join it and keep out corrosion.

d.) Put dielectric grease on the spade/block connectors/contacts.

and for belt and braces block connectors:

e.) strap a zip tied lengthwise around the mail and female connectors (from wire end to wire end)and then shrink wrap (using 1" or larger from Harbor freight; it is pretty inexpensive ) the entire block pair.

 

[posplayr]

Sandy

Sandy

LOL. When first read the title I thought I was going see a really great picture of Jim doing huge smoking donuts on his street fighter!!!

.

Sounds like that would be fun. :lol:

Gotta get the ED running again for the Escondido trip. Are u going to the Mt Shasta event?

Jim

 

[posplayr]

Sneak path

Sneak path

I guess I'm confused by the term 'sneak paths' here in reference grounding schemes.

When engineers speak of sneak paths it's typically of redunant paths that interfere with the normal operation of a system. For instance an alternate path for power to a radio that bypasses the fuse if there is a component failure in another system.

I've gotta be truthful here though, I haven't read this entire thread. I deal with this stuff during the day and have a very short attention span during the evening.

OK, back to your regularly scheduled programming.

The sneak path is the ground path between the motor and the frame.

All loads have to terminate at the R/R (-) to return to the stator.

So in this context the only sneak path would be if the R/R (-) were not well frame grounded. Then the load current in the frame, would have to return through motor mounts then through the ground strap and then from the negative side of the battery to the R/R (-) and finally back to the stator.

The primary path should/would be any frame grounded load simply return to the same R/R (-) directly via a solid frame ground at the R/R (-).

Pos

 

[posplayr]

Rusty

Rusty

Hey Rusty,
Thanks;
I am trying to keep it pretty simple and understandable but there are some details that have to be brought out to understand the issues. I think this circuit diagram is a pretty good simplification of all the GS charging systems and helps understand how power and grounding works.

I know i had tried to draw something similar before but not knowing the relative currents made for too many unknowns to draw any conclusions. This simplified physical schematic model seems to work.

I thought you had some EE courses under your belt. Most of my BSEE schooling was in the late 70's. I have done some electrical design over the years but I've mainly studied Modern (it was back then) Control and Estimation Theory in grad school; as a systems engineer this is not my main line of work. But I have resources to draw on to basically check my work. I have been double checking this stuff with a other real H/W guy at work.

Most of this stuff I can figure out once I understand how all the components work (like SCR or body Diode ).

Pos

I would like to add a comment if I may.

some like myself may have had electrical/ electronic experience. in my case it was over 30 years ago. I took first place in radio servicing in the state (they didn't have national competitions back then). I even played around with television repair. I have designed circuits, built transmitters and even went to school to get a BSEE . but sadly my family got to be a large load and I had to put it on hold, and never finished what was started.

Having these kinds of discussions are not lost on deaf ears, even those that don't understand can and do pick up valuable information. it may even turn on a few lights to those that wondered why the voltage readings they have taken, read a negative value instead of an expected positive value, upon checking the wiring to find an answer to their problems.

Jim, I know it takes away your time to do things like this, but I ask you to keep it up.

some of us enjoy re-learning what we have forgotten...

 

[BassCliff]

OK, Mr. posplayer and Mr. duaneage, where should we be grounding our regulator/rectifier? Or does it matter if you are using one of Mr. duaneage's 6-wire Honda units?

I have a turn signal flasher and a coil relay mod both grounded to the frame. My r/r is grounded directly to the negative terminal of the battery. I have recently cleaned every electrical connection and ground that I could find, including the ground strap to the engine. Should I change anything?


Thank you for your indulgence,

BassCliff

 

[Guest]

GS Power and Grounding Scheme​

Here is an analysis of a preferred GS power and grounding scheme. This probably will diverge from your original GS wiring as well the conventionally wisdom.

The first step is to re-look at what the shunt R/R is really doing. Initially my original thought was that the R/R is shunting current to ground. We it is sort of but not how I originally thought. I was struck by this when I was reading the manual charging system description. Then it because obvious that because of the combo Rectifier and Regulator the R/R does shunt current, but there is a continual load current that it has to sink through its’s negative (black lead). Sinking current means the R/R has to absorb this current direct it back to the stator through the lower legs of the diode bridge. Well big surprise this changes the ball game as far as establishing a proper single point ground.

Analysis:

Here is a summary of the elements of the physically schematic:

a.) I used the current measurements I previously measured along with a simplified version of a typical GS charging system to illustrate where the proper power and grounding connections should be.
b.) Generally at 3k RPM the GS charging system puts out about 12 amps with about 2 of those amps going to charge the battery. The various loads receive the remaining 10 amps.
c.) In the physical schematic, the stator produces A/C power, that is rectified in the R/R producing DC (12 amps) out of the red R/R wire. That power is split between the battery and the various loads (10 amps for load; 2 amps for battery) in the fuse box (I ignored the ignition switch which provides a detour to switch the various loads prior to the fuse box.).
d.) Another interesting feature is that most of the load current return paths are through the frame. In fact some GS schematics I looked at did not even have anything other than the large starter ground strap connected to the negative side of the battery.
e.) I also showed conventional direct connections between:
i.) the R/R + (through the fuse box) to the plus side of the R/R as well as
ii.) a direct connection between negative leg and the battery.
iii.) R/R (-) tied to its’ own case at the mounting bolt.

f.) I also added a R/R ground to the frame. As it turns out this is very important to reducing sensitivity to connector corrosion and high connector resistance. Since the load return current has to come back into the negative side of the R/R to return to the stator, the primary return path should be back through the frame to the frame grounded R/R !!!!

To be clear, the R/R is not shunting current to ground (i.e. to the frame or the battery), it is absorbing load return current from the path of least resistance. That should be the frame ground attached to the R/R. If that R/R negative leg is not grounded, then the only return path for the load current is either
a.) a direct connection of a ground wire from the harness to the negative side of the battery and then back to the R/R or
b.) a frame ground path back through the motor mounts to the battery via the starter ground strap and then finally top the R/R on the R/R ground to the battery.


It is important to note that the R/R attempts to regulate based on voltage it measures internally which the best we can tell is the voltage between the red and black leads on the R/R. Any current flow between the battery and the R/R tends to reduce the voltage that the battery sees because of voltage drops coming from the positive side or from the negative side. Minimizing the current flow from R/R (+) to the battery or from the battery back to the R/R (-) helps keep the sensed voltage more accurate.

You can reduce the return current from the battery by providing a good frame ground so that the bulk if not all of the load return current come through the frame to the negative side of the R/R.

On the positive side of the R/R , the full 12 amps goes to the fuse box. Any resistance there will cause a voltage drop (from 12A) and erroneous regulation. If you can not keep that resistance low, then put in a separate fused wire directly from the battery to the + R/R. It is this resistance on the positive side that can cause the fold back where your full R/R current is going through a relatively small resistance and causing voltage drop off at the battery at higher RPM.

Summary:

So what this analysis suggests is that:

1.) The R/R (-) should be SOLIDLY connected to frame ground.
This avoids current sharing between teh 2A charging current returning from teh battery and the 10A load returns. By having the R/R grounded the voltage sensetivity to resistance between the R/R (-) and the Battery (-) is 1/6 (i.e. 2A v.s. 12A)

2.) The battery should NOT be connected to frame ground.
This causes current sharing as described above in 1.)

3.) Other than the starter/solenoid connections (pos and neg), the battery + should only connect to the positive side of the R/R and the battery negative should ONLY connect to the negative side of the R/R.
minimizing the current between the battery and the R/R on both (+) and (-) side reduces teh errors described in 1.)


4.) If the resistance is low enough the R/R (+) leg can go through the fuse box (to teh battery +) but be careful of getting voltage drops from the full 12 amp supply current.
If you share the 12A coming right from teh R/R (+) to the Fuse box keep any resistance to a minimum becuase it will cause more drop than just 2A of charge current alone.

5.) If there are spare load ground lines they should be tied to the R/R ground or to a frame ground. This follows from a single point ground philosoph where all currents shoudl return directly to the R/R (-). The frame is god enough or go direct to teh R/R (-) if it is closer than a frame ground.

Comments welcome.
Pos

Ground the R/R to both the frame and the battery.

 

[posplayr]

Cliff

Cliff

Sounds like the PRIMARY thing that you want is to ground the R/R (-) to the frame at a closest point.

Mr. duaneage explained he typically did the same and that was the conclusion of the analysis.

All return currents have to flow into this R/R (-) and so any load current that is grounded to the frame has a closer path with the Frame grounded R/R (-). This avoids upsetting charging currents and extra drop between battery and R/R.

If your negative battery terminal has only a.) the starter ground strap to the engine and B.) the R/R (-) tied to iit and C.) everything works you are good to go.

If you have anything else tied to the minus terminal of the battery that might bear some consideration.

I think you can remove it and it will improve charging by tying it to frame ground.

Posplayr

 

[BassCliff]

Thanks Jim, Bill.

Sounds like the PRIMARY thing that you want is to ground the R/R (-) to the frame at a closest point.

Mr. duaneage explained he typically did the same and that was the conclusion of the analysis.

Ever since I joined this fine forum, the conventional wisdom was to run the ground wire from the r/r directly to the negative terminal of the battery. That was the first modification I made to my bike. Perhaps this should be considered as a temporary workaround, an alternative to the original ground connections to the battery box that are found corroded on many of these "barn finds"?

If your negative battery terminal has only a.) the starter ground strap to the engine and B.) the R/R (-) tied to iit and C.) everything works you are good to go.

This describes my current wiring setup. Everything works just fine.

If you have anything else tied to the minus terminal of the battery that might bear some consideration.

No, nothing else is connected to the negative battery terminal.

I think you can remove it and it will improve charging by tying it to frame ground.

Ground the R/R to both the frame and the battery.

I can certainly try either and both, take a few measurements and observe any changes. Gentlemen, I appreciate all this discussion and the practical application. It's been a while since my electronics courses in college, but I'm following most of it. Let me know when I have to get out my trusty old TI-35.

Thank you for your indulgence,

BassCliff

 

[posplayr]

Cliff

Cliff

Having been here at GSR for about 18 months, (maybe be it was just me ) but I was under the distinct impression that current was being sinked from the R/R (-) to the battery (-). But this just simply is not the case.

The R/R (-) is the sink for all load/charging currents and since most things ground to the frame the R/R (-) needs to be tied there as well. Sounds like that is the only thing you are missing.

If your connections are all clean, you may not notice a difference in charging. However, if you are careful to measure the voltage difference between battery (-) and R/R (-) it should drop with the R/R (-) frame ground in place.

As Chef said, connect the R/R (-) to the frame to pickup all of the load currents being grounded to the frame, (strictly speaking this is the only ground) and connect the R/R (-) to the battery (-) so that you have a return for any charging current going to the battery (+). Keeping these separate improves regulation and you will notice it the most if you dont have perfect connections at either end.

Pos

 

[reddirtrider]

The sneak path is the ground path between the motor and the frame.

All loads have to terminate at the R/R (-) to return to the stator.

So in this context the only sneak path would be if the R/R (-) were not well frame grounded. Then the load current in the frame, would have to return through motor mounts then through the ground strap and then from the negative side of the battery to the R/R (-) and finally back to the stator.

The primary path should/would be any frame grounded load simply return to the same R/R (-) directly via a solid frame ground at the R/R (-).

Pos

I suppose I'll just have to disagree with your supposition that the ground path between the motor and frame is a problem. Return currents will take the path of least resistance to the battery negative terminal. All other points are referenced to it.

 

[reddirtrider]

Let's look at this from a manufacturers viewpoint. They are not interested in creating the 'best' ground paths, they are interested in creating the easiest to manufacture ground paths that do the job.

In order to understand the voltage drops associated with the various grounds, take a look at this link and check out the resistance per foot of various AWG wire. http://www.radiolocman.com/shem/shem-cache.html?di=18899 18 guage wire has a resistance of 7.51 milliohms/foot. That means you need 8 amps of current to drop 0.5 volt. You'll need 10 amps to drop 0.5 volt with AWG 16.

Tying everthing to the frame is probably the best method, but it's costly from a manufacturers viewpoint. It's also unreliable long term as each tie point can corrode. This means they all must be properly sealed from the elements -> more $$.

How much current are you going to use? Determine that, then either make your ground wires larger, or shorter, or both.

 

[duaneage]

RRs and Grounding is turning into the new Oil Thread it seems.

 

[posplayr]

Redrider

Redrider

I suppose I'll just have to disagree with your supposition that the ground path between the motor and frame is a problem. Return currents will take the path of least resistance to the battery negative terminal. All other points are referenced to it.

I think you missed the primary "Eureka" moment.

All load and charging currents do not return to the battery, they return to R/R (-).

The rest of the discussion is pointless if you dont agree with this.

If the currents do not return then the Stator is not providing power to the system and you are running off battery power in which case load currents return to the battery.

I just looked at every GS750 schematic from 77-82 in my Clymer manual. It shows that:

A.) every R/R or Rectifier is grounded to the frame.

b.) It also shows that there is nothing other than the starter ground strap to the engine connected to the negative post of the battery.

C.) All load and charging paths return via the frame ;there is no other way to get the R/R (-)


This is how Suzuki delivered the bike. My preferred ground scheme did not develop anything new just supports how the system was originally delivered.

My point though is that with the ground strap modification between the R/R (-) and the battery (-) that is now so popular here at GS resources and has been the conventional wisdom, you need to also make sure that the R/R(-) stays well grounded because all of the currents still have to get there.

If you don't do that then the added wire will carry more current than it should and at the very least result in an excessive voltage drop between the R/R and battery terminal resulting in undercharging.

Pos