ADVrider

ADVrider (http://www.advrider.com/forums/index.php)
-   GS Boxers (http://www.advrider.com/forums/forumdisplay.php?f=3)
-   -   Electrical No-No! Have I analyzed/fixed correctly? (http://www.advrider.com/forums/showthread.php?t=832041)

1200gsceej 10-05-2012 08:38 PM

Electrical No-No! Have I analyzed/fixed correctly?
 
Was going to do some things to the electrical system so took off the seat and removed the cover of my Centec AP-2. When I looked at the positive switched connections I noticed ‘gunk’ on top of the tightening screw; then noticed a burnt smell and discoloration on the last two wires. They went to the relay for my Hella DE Fog lights (which has a lighted, push-button switch in the left hand guard). I did not take the pictures below until I had removed and separated the two wires (red power; blue trigger).

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>
The Hella’s and the Centec were installed 4+ years ago. The relay had a twisted copper red wire (not sure of gauge, but probably at least 12) and a small blue relay wire (maybe 18 gauge). I did not want to find and tap into any headlight wires (I was brand new to this bike, and as I guess is obvious, to installing my own fuse block). Since the Centec itself is switched, I figured that I could attach the blue relay trigger wire to one of the positive terminals as well. Since it was only supposed to draw enough to trigger the relay (milli-amps?) I thought I could just “pair it up” with the red wire. I put the blue and red wires side-by-side, wrapped some black electrical tape around them, and inserted the pair into the terminal for fuse #4. It had a 10 amp fuse.

<o:p></o:p>
It has remained this way, even through other additions to the Centec, most recently my Garmin Zumo installation 18 months ago. Certainly the fuse block did not look like this at that time. Also, the lights were still working at the time I discovered this.


http://i32.photobucket.com/albums/d4.../IMG_0630x.jpg



<o:p></o:p>
<o:p></o:p>
<o:p></o:p>

http://i32.photobucket.com/albums/d4.../IMG_0631x.jpg


I removed the wires from the terminal block. I did not have to loosen the screw. Actually, I did not know there was still a screw there until later when I cleaned out the ‘gunk’ in the recess. I removed the black tape and discovered that the insulation had gotten hot enough for the two wires to stick together. The ¼ inch of insulation at the end of both wires was black. I used a utility knife to separate them, leaving some blue insulation on the red wire and exposing some of the blue wire in the process. The two wires themselves had not come in contact. However, the red wire had ‘blistered’ somewhat, and there appeared to be 1 or 2 ‘pin-head’ sized holes that exposed some wire.


http://i32.photobucket.com/albums/d4.../IMG_0634x.jpg
<o:p></o:p>

So what happened here?

<o:p></o:p>
I am assuming that the current in the red wire heated the blue wire as well (duh) since they were both attached to the same 10 amp fuse, probably more than the blue wire should be exposed to safely. I assume that there was some resulting temperature gradient along the blue wire for perhaps a half-inch or so. But it was enough to overheat the insulation and cause the two wires to stick together. Once that happened, the heat just might have gotten worse, or ‘other things electrical’ may have begun to happen as well.
<o:p></o:p>
I’m guessing that I was lucky that I did not have a fire. :eek1

<o:p></o:p>
What I have done is strip the bad insulation from both red and blue wires, attached the red wire to terminal #3 with a 10 amp fuse, and attached the blue wire to terminal # 5 with a 3 amp fuse. Am I OK now?

<o:p></o:p>
Thanks. And all comments related to my stupidity will be read, at least twice.<o:p></o:p>
-ceej<o:p></o:p>

def 10-05-2012 09:03 PM

When I make connections similar to those in your fuse block, I like to tin the wires rather than merely twisting them. It is likely over the years that some of the strands in your #12 red wire became isolated and no longer carried current. As such, fewer strands were sharing the load and overheated even though the circuit was not overloaded. I would tin each wire inserted into the fuse block.

Beezer 10-05-2012 10:28 PM

looks like overheat due to a high resistance connection. as above, tin the wire. clean everything, consider splicing in fresh wire if the strands look less than wonderful

Marki_GSA 10-06-2012 03:16 AM

as above your connection became loose causing it to heat up. As said you can either tin the ends or use a bootstrap crimp.

mikegc 10-06-2012 03:31 AM

Electrical No-No! Have I analyzed/fixed correctly?
 
Quote:

Originally Posted by def (Post 19758878)
When I make connections similar to those in your fuse block, I like to tin the wires rather than merely twisting them. It is likely over the years that some of the strands in your #12 red wire became isolated and no longer carried current. As such, fewer strands were sharing the load and overheated even though the circuit was not overloaded. I would tin each wire inserted into the fuse block.

Like def, I tin the wires going into the Centech. Just before going on the Trail of Tears ride a few weeks ago, I was going over the machine with a torque wrench just checking things out. For some reason, I decided to check the screws on the Centech and almost all of them were loose after three years of riding around.

Mike

aschnell 10-06-2012 04:19 AM

Actually tinning the wire isn't good unless you have spring terminals. The solder will cold flow from the force and will loosen up the screw terminal connection. You probably had the small blue wire loose since it is a smaller gauge wire. The screw terminal touched the larger gauge wire first and you didn't get any compression on the blue wire. That wire connection had high resistance and heated.
Get wire ferrules to keep the wire ends neat instead of solder and be careful putting different gauges in the same spot. If you can stack them vertical they will both get compressed, but side by side won't work.

roger 04 rt 10-06-2012 04:34 AM

Quote:

Originally Posted by mikegc (Post 19759679)
Like def, I tin the wires going into the Centech. Just before going on the Trail of Tears ride a few weeks ago, I was going over the machine with a torque wrench just checking things out. For some reason, I decided to check the screws on the Centech and almost all of them were loose after three years of riding around.

Mike

+1, this is a compression fitting that came loose over time. When you don't tin the wires, vibration can cause some strands to escape the screw's compression, lessening the compression of the rest. Once the resistance starts to build up the result is what you see. Personally I'm not a fan of that type of connection in applications where there's vibration.

If you tin the leads which is better, there is a weak spot where the tinning stops, avoid sharp bends like below. Make sure the wires are tie-wrapped zso they don't move.

Also, only one wire per terminal.

http://i32.photobucket.com/albums/d4.../IMG_0631x.jpg

ScEd 10-06-2012 05:10 AM

The type of wire you use can also make a difference. I use wire that has a high number of strands and comes pre-tinned. The number 12AWG has 60 strands for example. Assembling and wiring various equipment for the aerospace industry (back when it existed) we called this type of conductor switchboard wire. I think the designation is SISX wire.

ragtoplvr 10-06-2012 07:40 AM

years ago the US military conducted many tests on making reliable connections. Tinning the wires is bad for 2 reasons.

The tin will cold flow as previously stated. The tinning ends at some point, making a transition from stiff to flexible.

You need to replace the connector or the entire fuse box. Then the burned wires need replaced.

In each cavity of the connector, put in a dab of dielectric silicone.

Correctly strip the wires, do not nick strands.

Put in connector and tighten screws. The dielectric silicone will exclude oxygen, and prevent corrosion.

Check them every so often, they will come loose. I suggest after first month, then every election year.

I bet some connections on your home breaker box are loose too. I tighten mine a few years back during an epic power outage. About 10% were loose.

Rod

1200gsceej 10-06-2012 10:51 AM

Thanks everyone! Straightforward responses with information I (mostly) understand. But I have a few comments and follow-up questions:

First, if I read correctly, several of you indicated that the red wire probably overheated, either due to fewer strands carrying load (why? Aren’t they all touching?) and/or that the terminal connection became loose (that did seem to be the case since I was able to pull the wires out w/o loosening the terminal screw). Can someone explain how these cause the resistance to rise? Is it that the effective wire size gets larger (i.e. the diameter gets smaller)? I am not an EE, but I can follow you if you write slowly. :D
<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>
<o:p></o:p>
Quote:

Originally Posted by def (Post 19758878)
{snip} It is likely over the years that some of the strands in your #12 red wire became isolated and no longer carried current. As such, fewer strands were sharing the load and overheated even though the circuit was not overloaded. I would tin each wire inserted into the fuse block.

Quote:

Originally Posted by Marki_GSA (Post 19759662)
as above your connection became loose causing it to heat up. As said you can either tin the ends or use a bootstrap crimp.

Quote:

Originally Posted by roger 04 rt (Post 19759824)
+1, this is a compression fitting that came loose over time. When you don't tin the wires, vibration can cause some strands to escape the screw's compression, lessening the compression of the rest. Once the resistance starts to build up the result is what you see. Personally I'm not a fan of that type of connection in applications where there's vibration.<o:p></o:p>

Quote:

Originally Posted by aschnell (Post 19759791)
{snip} You probably had the small blue wire loose since it is a smaller gauge wire. The screw terminal touched the larger gauge wire first and you didn't get any compression on the blue wire. That wire connection had high resistance and heated. <o:p></o:p>



<o:p></o:p>
<o:p></o:p>
Next, I did some research and found references to the issues about tinning mentioned below. So the alternative is to crimp a wire ferrule to the end of the stranded wire? The crimp compression is better? because it is less likely to loosen over time?
Quote:

Originally Posted by aschnell (Post 19759791)
Actually tinning the wire isn't good unless you have spring terminals. The solder will cold flow from the force and will loosen up the screw terminal connection.

Quote:

Originally Posted by ragtoplvr (Post 19760555)
years ago the US military conducted many tests on making reliable connections. Tinning the wires is bad for 2 reasons. The tin will cold flow as previously stated. The tinning ends at some point, making a transition from stiff to flexible. {snip}<o:p></o:p>
Rod


<o:p></o:p>
<o:p></o:p>
<o:p></o:p>
Finally, no one said that my connecting the blue wire with the red wire was wrong, or that it was the (direct) cause of the problem. So it is OK to attach relay-triggering wires to the switched terminals of the Centec? But aren’t you supposed to “fuse the wire”? Meaning that the fuse should fail before the wire does? Or do these draw so little current that it is not important?
Quote:

Originally Posted by aschnell (Post 19759791)
Get wire ferrules to keep the wire ends neat instead of solder and be careful putting different gauges in the same spot. If you can stack them vertical they will both get compressed, but side by side won't work.

<o:p></o:p>

roger 04 rt 10-06-2012 01:03 PM

Quote:

Originally Posted by 1200gsceej (Post 19761478)
Thanks everyone! Straightforward responses with information I (mostly) understand. But I have a few comments and follow-up questions:

First, if I read correctly, several of you indicated that the red wire probably overheated, either due to fewer strands carrying load (why? Aren’t they all touching?) and/or that the terminal connection became loose (that did seem to be the case since I was able to pull the wires out w/o loosening the terminal screw). Can someone explain how these cause the resistance to rise? Is it that the effective wire size gets larger (i.e. the diameter gets smaller)? I am not an EE, but I can follow you if you write slowly. :D
<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>
<o:p></o:p>


<o:p></o:p>
<o:p></o:p>
Next, I did some research and found references to the issues about tinning mentioned below. So the alternative is to crimp a wire ferrule to the end of the stranded wire? The crimp compression is better? because it is less likely to loosen over time?

<o:p></o:p>
<o:p></o:p>
<o:p></o:p>
Finally, no one said that my connecting the blue wire with the red wire was wrong, or that it was the (direct) cause of the problem. So it is OK to attach relay-triggering wires to the switched terminals of the Centec? But aren’t you supposed to “fuse the wire”? Meaning that the fuse should fail before the wire does? Or do these draw so little current that it is not important?

<o:p></o:p>

As you might imagine, these type of connections use compression to make the connection. With no compression, the resistance is high (maybe even open circuit). As you add compression, the resistance quickly drops. With enough compression the resistance is effectively zero. The problem arises when there is only enough compression to connect with a few ohms of resistance. Then the current can flow but the heat comes from the equation: current times current times resistance. So it is the partly-loose situation which causes heating.

If it were me I wouldn't put two wires into one connection unless the wires were soldered together. Inevitably one wire shifts due to vibration, allowing the compression of the joint to lessen. Then you know what happens.

I would connect the wire to a fused connection. The problem you're trying to prevent is not the normal current flow, but the current demand that can occur if something shorts out at the other end of the wire.

Although these compression connections aren't my favorite since they can loosen, even with the problems of tinning, I would do it, and dress/route the wire so there was no strain put on the tinned to untinned point.

kevin g 10-06-2012 04:25 PM

I Like MArine Grade Wire
 
To expand on what ScEd said about pre-tinned wire, that is what i use. It can be gotten easily at boat supply stores like West Marine. Military spec wire is always tinned and has many small strands as opposed to fewer large strands in lower grade wire.

A likely cause of the heating in your connection is that the wire was bare copper and oxidized - quickly. Bare copper sucks for corrosion and the oxidized wire has a high resistant at the connection. Copper oxide can grow thick enough to be and insulator.

Even though tinning the wire with solder has its drawbacks I think it is better than bare copper. Crimp connections to bare copper have the same corrosion problems.

If you want to make the best repair replace all of the wiring with marine grade tinned wire.

1200gsceej 10-06-2012 07:36 PM

Don't always build from scratch
 
I'm familiar with West Marine (although the store that was a mile away closed). I'll remember that when getting new/replacement wire.

Part of the problem is that in many circumstances one will get some form of wiring harness with a purchase (e.g. horn, lights, etc). It often includes a pre-wired relay, shrink-wrapped connectors, enclosed, whatever ... For some (such as me), connecting pre-wired components is doable; assembling from scratch is not (yet).

But I've learned a lot about what I can do to make those connections better.
Thanks again (to all)
-ceej



Quote:

Originally Posted by kevin g (Post 19763040)
To expand on what ScEd said about pre-tinned wire, that is what i use. It can be gotten easily at boat supply stores like West Marine. Military spec wire is always tinned and has many small strands as opposed to fewer large strands in lower grade wire.

A likely cause of the heating in your connection is that the wire was bare copper and oxidized - quickly. Bare copper sucks for corrosion and the oxidized wire has a high resistant at the connection. Copper oxide can grow thick enough to be and insulator.

Even though tinning the wire with solder has its drawbacks I think it is better than bare copper. Crimp connections to bare copper have the same corrosion problems.

If you want to make the best repair replace all of the wiring with marine grade tinned wire.



Times are GMT -7.   It's 12:03 PM.

Powered by vBulletin® Version 3.8.5
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ADVrider 2011-2014