[C38] Wiring Question
Les
hlhowell at pacbell.net
Sat May 17 00:32:31 EDT 2008
I guess this was aimed at me. Attached is a diagram of a loop ground
system. The positive I have portrayed as a two branch star, (two wires
distributing to loads. On our boats this would be the jumper across the
switches, and the switches would be between the load boxes A-J and the
respective lead).
If all the loads are equal and distributed as drawn, The branch with
C,D,E,F,G, and H would have 6x the load in watts and the branch A,B, I,
and J would have 4x the load in watts. This would leave the ground
between J and H to balance the remaining wattage as a distributed
current. This wire would not have the effect you might expect. This is
because the sums of the current errors and paths are not equal, but
distributed around the loop at differing intervals. If the length
between J and H is quite short, probably no problem, but the resistance
would be very low, and the current is probably 2 or three times the load
current. As the lenght between J and H increases, the current will
produce more and more heating effects (I squared R equals power, so the
heating is the square of the current times the resistance. If the
resistance is 25milliohms, and the current is 6A, the heating effect is
36*.025, or 900 milliwatts. But remember the size of the wire, and so
if the wire were say a 12G wire, it might rise about 25degrees F. Not
too bad, unless some of the wire breaks, or the insulation is thin, or
some other problem occurs. Then it could be a fire. It depends on the
distribution of that 900mw.
On the other hand, if the current and voltage worked out so
that the voltage at the base of j and the base of g were equal, then
there would be no current flow between those two points. Strangely
enough this will affect the current from the legs of H and I, and they
will not operate as expected. It is this bridging effect that is used
in many forms of measurement and control of currents in DC systems.
I have fallen pray to this myself, when I added Jace's second battery
bank, with a common ground resistor. The charging current to one bank
was exactly equal and opposit to the run current of the refer. The net
effect was that neither the charger nor the reefer worked as expected,
and the battery went flat rather than being held by a charger.
Also we lost some food because the reefer wasn't cooling properly.
Its a magicians trick, but one with strange consequences.
Regards,
Les H
On Fri, 2008-05-16 at 08:10 -0500, Jan Pitesa wrote:
> Could you perhaps include a wiring diagram for us non electrical
> engineers?
>
> Jan Pitesa
> Mysteri
> New Orleans
>
> Inactive hide details for Tom"Tom T." <tdtron at earthlink.net>
>
>
> "Tom T." <tdtron at earthlink.net>
> Sent by: listserve-bounces at catalina38.org
>
> 05/15/2008 07:40 PM
> Please respond to
> tdtron at earthlink.net; Please respond to
> Catalina 38 Listserve <listserve at catalina38.org>
>
>
> To
>
> "hlhowell at pacbell.net, Catalina 38 Listserve" <listserve at catalina38.org>
>
> cc
>
>
>
> Subject
>
> Re: [C38] Wiring
> Question
>
>
>
> It seems to me that loop circuits are illegal in a/c circuits for
> several
> reasons, the most important possibly being a safety issue with
> possible hot
> circuits when the circuits were supposedly cut.
>
> DC circuits don't exhibit inductive reactions like AC circuits do
> unless
> they are inductively coupled with an external AC source. A DC loop
> actually has a lower "Q" making it even less able to exhibit parasitic
> oscillations which could cause RFI than the standard harness. "Q" is
> the
> sensitivity of an antenna in regards to a specific frequency. The
> higher
> the Q, the more the antenna will react to a particular frequency and
> more
> immune to inductive coupling to other frequencies. A low Q antenna
> won't
> perform like a high Q antenna but a low Q antenna will react to a
> broader
> range of frequencies. Every piece of wire will react to some
> frequency,
> our job is to minimize that effect unless it is our antennas and not
> our
> wiring. We want a very low Q on anything but our actual antennas.
>
> The stock lighting harness on a Catalina 38 is quite long between the
> panel
> and last light over the ice box, possibly around 35'. The natural
> resonate
> frequency of such a long length of harness would have a tuned
> resonance
> very low, too low to make much impact on any modern electronics. The
> branch circuits going from the harness to individual lights would have
> a
> much greater chance to make a parasitic resonance but those circuits
> would
> not be effected by the type of harness we employ in any case.
>
> By making the shunt across the engine compartment, the effective
> resonate
> frequency of the harness would be raised but the resonant "Q" of the
> harness would be greatly lowered making parasitic oscillations much
> less
> likely. A loop antenna has a very low "Q" by nature and this closed
> loop
> harness would actually have an even lower "Q" than a loop antenna
> because
> unlike an antenna, our harness is composed of two conductors with one
> being
> grounded to the boat battery making a ground for any parasitic
> voltages.
>
> One thing I did do on the Renata that I failed to mention was that I
> extended the wiring harness to the sail locker before returning
> through the
> loop shunt back to the breaker panel. I attached an auxiliary ground
> at the
> galley and ran this ground to my house batteries which gave me even
> less
> resistance in the circuit. I expect this additional ground also acts
> as a
> parasitic oscillation killer also.
>
> I have had this low resistance DC harness modification for several
> years
> and I have never experienced any hum or any other type of induced
> radiation, either on the AM radio or SSB radio. I would not be afraid
> of
> this loop circuit on a DC circuit. I totally agree about the dangers
> of a
> loop AC circuit but I think the benefits of lower loss on the DC
> circuit is
> a great plus.
>
> If you do decide to make your harness into a loop as I did, turn on
> your
> cabin lights before you attach the last loop connection and watch how
> much
> your lights improve when the final connection is attached. You will
> be
> amazed at how much brighter your cabin will be.
>
> Tom Troncalli
> Renata #95
> St. Pete, Fla
>
>
>
>
> > [Original Message]
> > From: Les <hlhowell at pacbell.net>
> > To: Catalina 38 Listserve <listserve at catalina38.org>
> > Date: 5/15/2008 1:28:01 PM
> > Subject: Re: [C38] Wiring Question
> >
> > There is another issue that is not well understood (nor is it easy
> to
> > explain, probably why it is not well understood).
> >
> > If you redraw your loop and attachments, you will see that it forms
> a
> > ladder like structure around the loop, something like two hoops of
> wire
> > connected by loads from one to another. This is not bad, IF all the
> > loads are almost equal. However this structure is also known as a
> > Bridge structure. The current that balances the ladder legs is the
> > bridge current. It becomes a complex sum of all the currents
> leading to
> > that particular segment of the bridge. When this is in operation,
> one
> > of the effects of a bridge is to allow a small current at one point
> to
> > prevent current flow. That is the current through legs on one side
> of a
> > segment are unbalanced to exactly the point that the legs on the
> other
> > side are, to the same degree and polarity. This has the strange
> effect
> > that no current flow occurs through that segment. Worse, the lack
> of
> > that path will unbalance the other legs in bizarre ways, leading to
> some
> > circuits getting lower voltages, others higher voltages, and the
> balance
> > points across the reaches of the ladder (your ground bus in this
> case)
> > can end up with say 3v across a length of wire. Calculate that
> current
> > and you begin to see the problem.
> >
> > It sounds weird, but it is true. Circular wiring produces higher
> > probability of fires because of this problem. That is why it is not
> an
> > approved means of wiring.
> >
> > And yet another problem can occur that under changing loads a
> segment
> > of the circle can end up with 0 current flow. For example say one
> > segment has 1A flowing, and produces a differential from the battery
> > of .025v. On another leg, some number of links away a similar load
> > produces the same condition. That means the wire between those two
> > points has zero voltage differential. No voltage, no current. The
> > other points between those two might not even work at all. You
> could
> > chase opens and shorts forever, but never isolate the problem
> because of
> > the effects of a bridge.
> >
> > To combat these effects the Wiring Codes specify a "star
> connection".
> > That is from the source to a distribution point, and from that
> > distribution point to the various circuits, just like you have in
> your
> > home.
> >
> > The reason boats (and metal framed aircraft) don't have this problem
> is
> > that they use a star distribution and the shell of the vehicle for
> the
> > return path (except I think all aluminum skinned planes are
> prohibited
> > from this due to electrolysis issues) which has so little resistance
> and
> > so many paths that the bridge effects are too weak due to multipath
> > effects. That means the return circuit has lots of ways to get
> back,
> > all of very low resistance, reducing the issue of heating or balance
> > canceling.
> >
> > Regards,
> > Les H
> > On Thu, 2008-05-15 at 08:51 -0400, Duff, Russ (R.W.) wrote:
> > > The only problem with your loop is that it will be more
> susceptible to
> > > electronic noise; both transmitting and receiving. The loop will
> act
> like a
> > > large antenna and pick up (and transmit) all kinds of noise that
> could
> > > effect any sensitive electronics (VHF, radar, autopilot....). I
> know it
> is
> > > only the lighting circuit (you hope), but that circuit does
> connect
> back to
> > > your main power bus with no filtering. You may want to
> re-consider.
> > >
> > > Sincerely,
> > > Russ Duff
> > > RDUFF at VISTEON.COM
> > > -----Original Message-----
> > > [mailto:listserve-bounces at catalina38.org] On Behalf Of Tom T.
> > > Sent: Wednesday, May 14, 2008 9:51 PM
> > > To: hlhowell at pacbell.net,Catalina 38 Listserve
> > > Subject: Re: [C38] Wiring Question
> > >
> > > The original lighting harness on the Catalina 38 was with either
> crimped
> > > connected 16 ga or 18 ga. wires that started at the fuse panel and
> circled
> > > counter-clockwise around the hull at the chine. The lighting
> harness
> > > dropped under the quarter berth and continued around the top of
> the
> hull to
> > > the galley area.
> > >
> > > That is a lot of length for 16 or 18 ga wire. I rewired my boat
> with
> marine
> > > grade 12 ga wire and soldered joints. I also ran a jumper wire
> from the
> > > fuse panel through the engine compartment to connect to the
> terminus of
> the
> > > lighting harness at the galley to make a closed loop harness. This
> gives
> > > less resistance to the entire circuit, especially the lights on
> the port
> > > side. By making a loop harness, all lights get about the same
> voltage.
> > >
> > > A loop circuit harness is against wiring codes for a/c circuits
> but I
> don't
> > > know if it's illegal for DC circuits. By making the circuit a
> loop, you
> > > reduce the wiring resistance greatly. Use your own judgement, I
> just
> know
> > > it works great. You could always remove the jumper if you ever
> sell
> the
> > > boat or have an electrician work on it.
> > >
> > > Another energy saving idea is to use a bungee cord to create a
> neutral
> helm
> > > when using an autohelm. If you neutralize the helm, it will take
> very
> > > little current to steer the boat because the autohelm won't have
> to
> > > overcome the efforts of weather helm allowing the servo motor to
> have a
> > > minimal load on it.
> > >
> > > Previously, on a windy day, you could have burned yourself on my
> autohelm
> > > servo motor but with a neutralized helm using bungee cords, the
> motor
> now
> > > stays cool to the touch and the battery stays hot much longer.
> > >
> > > Tom Troncalli
> > > The Renata, hull #95
> > > St. Pete, Fla
> > >
> > >
> > >
> > > > [Original Message]
> > > > From: Les <hlhowell at pacbell.net>
> > > > To: Catalina 38 Listserve <listserve at catalina38.org>
> > > > Date: 5/14/2008 8:56:09 AM
> > > > Subject: Re: [C38] Wiring Question
> > > >
> > > > Hi, Steve,
> > > > The master switch on the power panel routes the starting current
> to
> > > the
> > > > engine. Peak this is about 300A.
> > > > I think I used #2 for that run from the batteries to the switch
> and
> > > > from the switch back to the engine. If you have the 12V Bible,
> they
> > > > have some nice calculations in there for that.
> > > >
> > > > The rest of your calculations seem OK. I am running 12ga for
> the
> > > > lighting when I redo mine, and should be quite sufficient for
> the runs
> > > > which average about 18' one way. I would also caution you to
> use only
> > > > marine grade wire, because standard copper will fail in short
> order
> with
> > > > corrosion eating it away at each connection. Avoid those
> riveted 3x
> > > > terminal thingies like the plague. They invariably corrode and
> cause
> > > > poor connections.
> > > >
> > > > The reefer might like a 1% drop, because each bit of drop to a
> > > > motorized device will cause current to go up approximately 2% to
> sustain
> > > > the load. This is rule of thumb, but you can look it up in some
> of
> the
> > > > books on electrical wiring if you want precise numbers.
> > > >
> > > > Also use the terminals that heatshrink with heat glue to seal
> the
> > > > connections. It will save you beaucoup problems later.
> > > >
> > > > Regards,
> > > > Les H
> > > > On Wed, 2008-05-14 at 00:07 -0700, Steve Smolinske wrote:
> > > > > I am getting ready to replace some of my wiring and my
> electrical
> > > > > panels. I have been drawing up my schematics and would love
> some
> > > > > input on my calculations and assumptions. I have based my
> assumptions
> > > > > on an all night sailing scenario with continous use of systems
> that
> > > > > would quickly draw down my batteries but conservation is
> another
> topic
> > > > > at this time Im more interested in determining proper wire
> size for
> > > > > peak usage. Ive assumed that I would be using: 3 cabin
> lights
> (4.5
> > > > > amps), Running Lights (2.5 amps) VHF (2 amps), Depth and Knot
> (2
> > > > > amps), Wind (2 amps), Auto Pilot (1 Amp) Radar (8 amps) GPS (3
> amps)
> > > > > Heater (8 amps) and Refrigeration (5 amps) for a total of 38
> Amps
> > > > > multiplied by postive and negative wire run length of 15 feet
> (Battery
> > > > > Cable to the batt switch then from there AWG 6-8 to the pos.
> bus and
> > > > > from the neg. bus back to the battery) which equals 570
> Famps.
> Based
> > > > > on the wire charts for a 3% drop this puts me in the 6-8 AWG
> wire
> size
> > > > > to and from the batteries. Any comments on any of these
> numbers is
> > > > > greatly appreciated. Refrigeration, Heater and lights are
> variables
> > > > > calculated on the high side just to be safe.
> > > > >
> > > > > Thanks all,
> > > > >
> > > > > Steve
> > > > > Peregrine #312
> >
> >
> >
> > _______________________________________________
> > Listserve mailing list
> > Listserve at catalina38.org
> > http://catalina38.org/mailman/listinfo/listserve_catalina38.org
> >
> >
> > --
> > No virus found in this incoming message.
> > Checked by AVG.
> > Version: 7.5.524 / Virus Database: 269.23.16/1433 - Release Date:
> 5/14/2008 4:44 PM
>
>
>
> _______________________________________________
> Listserve mailing list
> Listserve at catalina38.org
> http://catalina38.org/mailman/listinfo/listserve_catalina38.org
>
>
>
> _______________________________________________
> Listserve mailing list
> Listserve at catalina38.org
> http://catalina38.org/mailman/listinfo/listserve_catalina38.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Loop.pdf
Type: application/pdf
Size: 10007 bytes
Desc: not available
URL: <http://catalina38.org/pipermail/listserve_catalina38.org/attachments/20080516/645fa4fe/attachment-0001.pdf>
More information about the Listserve
mailing list