MARINE ELECTRICAL SYSTEMS SEMINAR, PART 1
Cameron
By Cameron Clarke "S/V JUPITER"
First short intro:
While cruising the Pacific Coast, Mexico, Central America, and the
Caribbean, I conducted several hands-on seminars for cruising sailors
to learn about and improve their electrical systems. I now present
this same material to others, via a multi-part posting of articles
to help you understand and troubleshoot your boat's electrical system.
This self-directed seminar aims to help you understand the material
without diagrams or graphics. If you feel you need the specific
diagrams and graphics, you may request them. Send
mailto:cam...@unix.infoserve.net and request "diagrams for marine seminar".
For a more complete introduction, see posting titled
"MARINE ELECTRICAL SYSTEMS SEMINAR, INTRODUCTION".
Copyright notice (C) 1995: All rights reserved. No material presented
herein may be copied in whole or part without the express and prior
written consent of the author, myself outside this particular news
group "rec.boats". This will attempt to defeat revised or incomplete
information from being distributed.
If you or your group would like to reprint or forward copies, please
mailto:cam...@unix.infoserve.net and request prior permission. And
withstanding any other matters we embark:
Part 1: ELECTRICAL CONNECTIONS
This part examines proper Wire size for amp loads, use of Ring
Terminals, connections, resistance, and losses in the wiring system.
This is the most important material in this seminar. If there is
nothing else you learn in this material, you will benefit most from
this part. Concentrate on learning this part as it sets up the
understanding for all the following material.
We begin by defining a few items. A ring terminal is a preformed
copper piece, often zinc or solder plated, which is placed over the end
of a prepared (insulation striped back to reveal copper) wire and crimped
with a tool to make a tight connection to the wire. The end of the
terminal has a hole, not unlike a washer, that is used to attach to
something usually by a screw. The screw will keep the terminal captive,
unless of course the screws falls out of the hole. A fork terminal is
similar, except there is a slot in the washer which allows the terminal
to slide over or away from the attaching screw without removal. A ring
terminal will prevent a wire falling off and causing a short. For this
reason, I recommend them over spade (slide on/off) and forks. Fork
terminals sometimes have bent up ends to eliminate this problem and are
a good substitute when you might inadvertently lose the screw in the bilge.
I define a connection as any point two physical metallic conductors
are made to touch, ie "connect". Connections can be "crimped" (squeezed
together via a special tool - please do not use pliers), soldered
(heated with an iron and a lead-tin alloy is melted into the air space),
or by any mechanical means (two or more wires or terminals held together
by a nut and bolt, etc). If two wires are connected in a manner which each
wire has a crimped terminal, then I count 3 connections, one for each
crimped terminal (2) and one for terminal to terminal (1), or 3 total.
Perhaps this seems a minor point, but I shall elaborate as to why it
is important for you to consider this detail. By counting and examining
connections in a wire path, you will solve many problems. Stay with me.
There are many types of wire, many of which have little use in marine
applications. Sometimes it is impossible to buy the right types or better
grades in foreign ports. That is ok, as long as you recognize that and
can deal with other problems later. Anchor Marine makes a very good
wire for marine use and I would always recommend it when available. It
is composed of many fine strands, each tinned (coated with tin to prevent
salt corrosion of the underlying copper) and the insulation resists
oil and saltwater, much better than household or automotive wire. If
you consult a wire resistance table, you will notice there is less
resistance in multi-stranded tinned wire than other types.
Copper Wire Table @ 95 deg F follows:
American Wire Gauge Ohms per foot Max Ampere load (continuous)
00 0.0000811 200
0 0.000102 125
2 0.000162 90
4 0.000253 70
6 0.000403 50
8 0.000641 30
10 0.00102 25
12 0.00162 20
14 0.00258 15
16 0.00409 7
18 0.00651 3
Use the table above to compute resistance of wire runs, include
length both to and from device. If you use 25 feet of #14 two
conductor wire for a cockpit lamp, then you have 50' of #14 wire.
Multiply 50 feet times 0.00258 ohms per foot from table above to
determine 0.129 Ohms of wire resistance, without allowance for
any connection or terminal resistance, just wire itself. I would
use at least #14 wire for a bilge pump, if the run were over 100'
I would consider using #12.
But the major concern is connections! If you possess a very
accurate ohmmeter you can directly measure the electrical resistance
of each connection, expressed in ohms. Don't worry about this for
now. My analysis over the years says the average connection (even
crimped after it is used six months) measures a mere 0.03 ohms.
The average connection that was crimped and soldered measures 0.01
ohms. What does this mean?
Take our two preceding pieces of wire. There were three connections,
right? That is what I counted. I would estimate the resistance in that
wire after six months service to be about 0.9 ohms (3 x 0.03). Again
not much, right? It would be much higher if corroded by salt water.
Ok, now lets examine a typical circuit in a boat. How about the bilge
pump? A typical 12 volt pump would draw about 5.4 amps pumping water.
What comprises the wiring circuit? Lets see, there is a battery, a
battery switch, a + terminal distribution block, a fuse, an on/off switch,
a float switch, the pump, a - terminal distribution block, back to the
battery. Agree? Oh, plus some wire. Now I will examine this again and
indicate in parenthesis (#) the number of connections.
The positive post of the battery is connected (1) to a battery lug, which
is connected (1) to wire, connected (2 wire to lug & lug to switch) to
battery switch (1 for switch contacts too), connected (2) to a wire
connected (2) to pos. distribution post. From post (2) to fuse (2 + 2
for removable fuse) via wire, and to on/off switch (5) each wire to lug
plus switch itself), connected to float switch (1), then wire (1) to
+lead of bilge pump, -lead of pump (1) neg. distribution post (2), and
neg. distribution post (2) to negative battery post (2) via wire and
terminals. Did you follow that? That's 29 connections! 29 connections
times 0.03 ohms each results in 0.87 ohms over all. That does not account
for any wire losses, just connections. How did we get so many connections
in the first place? Can we eliminate any?
Let's cheat a little to make our example a little easier to understand,
at least the math, OK? Lets say there were only 20 connections, each 0.03
ohms. The result, 20 x 0.03 is 0.6 ohms. We will also assume no wire
resistance loss, which of course is impossible. A typical battery voltage
in use would be 12.5 volts to run our pump. What then is the voltage the
pump actually sees? Does it receive the full 12.5 volts? No. The voltage
is reduced at the pump in proportion to the resistance of the wire path
(the sum of connection resistance and wire resistance) and the current
draw. In other words, more resistance or more current draw reduce voltage.
Ohm's law states current in amperes times resistance in ohms equals the
voltage drop in volts. So 5.4 amperes times 0.6 ohms equals 3.24 volts.
What this means is that the wiring and connections, (the wiring network)
consumes 3.24 volts, so that the bilge pump receives only 9.26 of the
original 12.5 battery volts. Power measurement is the product of amperes
times voltage and is expressed in watts. In summary:
Battery Wire Network Bilge Pump
Volts 12.5 3.24 9.26
Amps x 5.4 x 5.4 x 5.4
Watts 67.5 17.5 50.0
Note: Amps remain the same in each item as the system is wired totally
in series. The full amount of current must flow through each and every
component, wire, terminal, and connection.
In this example the Battery must provide 67.5 watts to provide the pump
with 50 watts, as 17.5 watts are consumed in the wiring network and lost
as heat. That is a 26% Loss of power!
This is the point in the class we get our voltmeters out and actually
measure the voltage drop from connection to connection. Try it yourself!
Take the positive lead of your voltmeter (set to measure 15 volts or so)
and attach it to the positive battery post with a clip. Then put the
negative lead on any connection down the wire path of something drawing
current. You can measure the drop in each wire and connection as you
continue down the path. (Note: you may have to adjust measurement scale.
Always work your way down the scale to avoid damaging you meter). Got
a lamp on? Try measuring the voltage at the bulb, then at the battery.
What is the difference in value? Where did the voltage go? It was
dissipated as heat in connections and wire. Ever notice a wire get
warm or hot? Ever feel the heat in the battery wires after starting
your diesel? Want to make it start easier? Reduce the resistance in
the connections. Want to know which connections need repair? Feel them
for heat.
What can be done? It would be hard to have fewer connections. Many
installations will have even more connections than this. What we can do
is keep the connection resistance low. I suggest crimp and solder all
new connection made for items drawing more than 3 amps. In the example
above, if we reduce the per-connection resistance to 0.01 ohms, the
power loss would be reduced from 26% to about 8% and the pump more
water per minute, reducing the on time, thus conserving battery energy.
That means you can charge for less time.
Have you ever heard of a HAM radio 'FM'ing? It is generally due to poor
connections resulting in a decrease of voltage (and power) available
to the transmitter as these typically draw 20 amps during transmit.
Also, a VHF radio will have more power out, if it has more voltage in.
With very bad connections, a VHF will cut in and out very rapidly, a
term called "motorboating", making it difficult or unable to understand.
Many items will perform better and last longer, like motors, fluorescent
lamps, inverters with less resistance in the wiring network. Incandescent
lamps will be brighter and burn out sooner.
Now before you go and tear your electrical system apart, keep in mind
which items will benefit the most. Don't take everything apart. Start
at the Battery terminals. Are they clean and mechanically tight? Are
the battery wires properly terminated into the proper sized connectors?
The what electrical items are used more often or have longer on times?
Decreasing electrical resistance to electric refrigeration, inverters,
12 volt appliances like TV's and VCR's, HAM radios, older Radar units,
VHF radios, and any cycling pumps will produce dramatic increases in
system efficiency for little work invested.
Once I was asked, "I have a strange problem. I have a cabin light that
does not work. The bulb is good. When I take the bulb out, I measure 12
volts at the socket, but it won't light when in the socket. What can be
wrong?" When I measured the voltage across the bulb, I observed only
a quarter volt or so, not enough to light the bulb. A connection in the
wire path was so badly corroded, the high electrical resistance did not
allow enough current to light the lamp, yet a voltmeter needs very little
amperage to measure voltage. The result, the wire net was current limited
and would not light the lamp. The cure; isolate and repair the connection.
I had seen this in another boat where the wires ran through a small hole
drilled in a bulkhead. The insulation was damaged at the hole and salt
water oxidized the copper to green dust (a form of copper oxide). It was
hard to locate, as the actual fault was in the middle of the bulkhead.
Can you think of other reasons to count the number of connections and
clean up the ones with heavier current loads? We will return to this
again and again. Believe me!
Submissions:
If you have a specific problem you wish to have addressed, please
send it in. Include your vessel type and size, what area you sail in,
as well as any pertinent information to the problem.
To obtain related graphics for all parts, or submit comments, send
mailto:cam...@unix.infoserve.net
End "MARINE ELECTRICAL SYSTEMS SEMINAR, PART 1"
Cameron
By Cameron Clarke "S/V JUPITER"
Copyright notice (C) 1995: All rights reserved. No material presented
herein may be copied in whole or part without the express and prior
written consent of the author, myself outside this particular news
group "rec.boats". This will attempt to defeat revised or incomplete
information from being distributed.
If you or your group would like to reprint or forward copies, please
mailto:cam...@unix.infoserve.net and request prior permission. And
withstanding any other matters we embark:
Addendum to Part 1: ELECTRICAL CONNECTIONS
When I talk of ring terminals, I like to pass around some examples
of wire I removed from boats. Some have crimp terminals that were
squeezed by a pair of pliers, not the proper tool. Some are not
squeezed enough and are mechanically loose, meaning you can push
or pull and move the wire in the connector. Others are squeezed
beyond recognition. You have seen them. They are in every boat. Some
were in mine too. I have examples of properly crimped connectors
on good, tinned multi-strand wire, and on non-tinned copper. I have
crimped and soldered terminals. On each of these, I place a tag
indicating the resistance in ohms of each. The values range from a
low of 0.01 ohms (did you guess the crimped & soldered terminal?) to
well over 8 ohms (a loose crimp on untinned copper). Many are in the
0.5 to 2.5 ohm range, as the wire has oxidized inside, after the
crimp was made. The poorest connections result from using untinned
wire. Next come crimps made without the proper tool. Can you find
any terminals so described in your vessel? Invest $6 to $10, buy
a good crimp tool, and replace those loose or improperly crimped
terminals with new ones well made. It will save you energy and
future trouble shooting time.
Often short pieces of wire, 1" to 3" in length, with a ring or fork
terminal on each end is used to interconnect other wires, attempting
to establish all interconnected wires at the same voltage potential,
but in a series string of connections. For some reason, boats made in
Taiwan use tons of these along with barrier strips in place of bus
bars. Ever look inside one of those really fancy yachts at the boat
show and open the electrical panel? You will see, row after row of
barrier strips (black phenolic strips with screws to attach wires),
and many short pieces of wire to interconnect the wires. Look hard,
as the bundling of wires can make it hard to spot. Electrically it
looks something like this:
_____ _____ _____ _____ _____ _____
/ \/ \/ \/ \/ \/ \
a b c d e f g
Each link will add about 0.09 ohms of resistance if well made. I have
shown 6 links to get from point a to point g. That results in 0.54 ohms
between points a and g. We call this a series string. To make the panel
look nice, the feed point would be point a. It would look messy if it were
the middle, point d, so point a is often the feed. If the circuit breaker
or fuse of your bilge pump were connected to point g, then 5.4 amperes
would flow through each link in the series string. Right? Yes! And 0.54 ohms
times 5.4 amps result in a 2.9 Volt drop just across the series link! Now
add that to the 3.24 volt drop from the wiring network and see if the pump
works very well! A series string is bad because of all the added connections.
There are preformed strips, one piece of metal with many fork like
terminals, that are made to use in place of the wire links. Remove the
links where possible and use the preformed strips. Any questions?
Another method, is to change either the feed point, or the load
point (where along the strip, point a to g, a wire is taken to a
fuse or circuit breaker) for larger amperage devices, i.e. Ham radio,
VHF radio, etc. You can increase time between battery charges just
by moving a connection or two, to reduce wiring network losses. Does
this make sense? Sure it does, if you can reduce your wiring loses by
8 to 10 percentage points, its the same as adding capacity to your
batteries. In addition the range of your radios will be increased
and most likely last longer.
A few boats have bus bars, copper bars with many screws for securing
ring terminals. These are much better electrically than barrier strips.
Bus bars are designed to make many connections at one voltage potential
and very little interconnection resistance. They do that very well. Barrier
strips were made to make many independent, non-interconnected connections
appear neat. They do that very well. Some builders have not realized
this.
If I could tell boat manufacturers something, I would tell them to
use a combination of bus bars and barrier strips to make up their
electrical panels (often pre-made outside the boat) and solder each crimped
terminal for lowest possible resistance and longevity. The panel should
be designed with strong posts to attach heavy gauge positive and negative
wires to the battery.
I really like the use of distribution panels, because they are neat
and place all the electrical distribution in a common location. However
they add a considerable number of extra connections in the wiring path
and are often the source of many loose or poorly made crimp terminals.
Extra care need be taken to make the panel a true asset.
To obtain related graphics for all parts, or submit comments, send
mailto:cam...@unix.infoserve.net
End "Addendum to PART 1"
Cameron
Copyright notice (C) 1995: All rights reserved. No material presented
herein may be copied in whole or part without the express and prior
written consent of the author, myself outside this particular news
group "rec.boats". This will attempt to defeat revised or incomplete
information from being distributed.
If you or your group would like to reprint or forward copies, please
mailto:cam...@unix.infoserve.net and request prior permission. And
withstanding any other matters we embark:
Addendum to Part 1: ELECTRICAL CONNECTIONS
To obtain related graphics for all parts, or submit comments, send
mailto:cam...@unix.infoserve.net
End "Addendum to PART 1"