KatieKat 2004 Cruise Chapter Six
KatieKat's Electrical System, With Many Updates

Click on underlined text to jump to it
GoBackTo 2004 Cruise Chapter Five
15 May 2004 KatieKat Power System
Techie Yachties
Power Consumers
Power Generation
Power Storage
Modifications updated 26 June, 2004
Power Regulation
Shorepower updated 26 June, 2004
Parallel Batteries added 3 August, 2004
Dec. 2006 Upgrade Discussion, updated 27 Feb., 2007
Photos of Batteries, updated Nov., 2007
Electrical System Information Update, updated 4 Aug., 2011
Links to Other Related Webpages:
2006 Panel, Regulator, and Battery Addition Dec. 2006
2007 Electrical Performance Assessment 10 Oct., 2007
GoForwardTo 2004 Cruise Chapter Seven

Click here to go to KatieKat Cruise Index Webpage
Click here to return to KatieKat Home Page

Click on the small photos to see larger-scale images,
then hit your browser BACK button to return to the small photo.

15 May, 2004 -- KatieKat Power System

Aboard KatieKat we are grid-energy independent. On this webpage I'll describe our onboard power consumption, generation, storage, and monitoring/control equipment, as well as modifications I'm presently undertaking. Sorry, I got carried away with my scribblings - unless you're into this stuff, don't bother...

Power Consumers

Our boat's power system is 12v, with a few small portable 110vac 60Hz inverters scattered about the boat. Without being technically strict, the power consumption unit that we are concerned with is the ampere-hour (Ah), with the number of amp-hours per day put back into the storage system hopefully exceeding the number of Ah consumed.

Our power consumers onboard are -

[Refrigerator and Freezer] Our refrigerator and freezer (brand name ICEER), happily chugging along on the solar-powered 12vdc for the past four years.

[Nav Table] Navigation table, as setup during our passage from New Caledonia to Australia. Radar, VHF, HF, Pactor Modem and speaker, autopilot control, computer, and GPS makes for functional but messy wiring.

[Nav Table] Another view, showing the second computer just peeking out behind the first, the GPS, and the boat instrument panel rotated 90-degrees. Everything sucks power during a passage, and the name of the game is to not run down the batteries.

[Portside TV] [Starboard TV]

The portside LCD TV (can double as computer monitor) over the bunk has a DVD player under it and a wonderful digital converter which produces a crystal-clear display using the off-the-air signal (NOT cable or satellite). The right photo shows the TV and boombox in the starboard hull. All this electronics runs directly off 12vdc (or 240vac shorepower). Note the two high-power reading lights, one 12vdc and the other 240vac.

[Inverter] One of a number of small onboard 110vac inverters. This one is in my study and normally powers the printer and scanner, in addition to charging the toothbrush :-)

[Connectors] The boat has about a dozen strategically-located universal cigarette-lighter outlets. Although convenient, I don't like them, as the plugs are often intermittent. For the last 20 years I'd standardized all my 12v connections using inexpensive two-prong car-trailer connectors. No intermittents!

Click here to go back to top of page

Power Generation

Our primary power source is two 120-watt solar panels (each optimally puts out 7.1 amps at 16.9v). Love them! They're mounted horizontally on the aft targa bar. They feed the batteries through a fairly sophisticated voltage regulator (which is a separate topic in itself). These solar panels alone are normally sufficient to keep the boat's power system happy and fully charged and the only improvement I would have liked is to be able to articulate them so they could be aimed into the sun.

Our secondary power source is a wind generator (Aero4Gen) which starts putting out whenever the wind is over ten knots. At 20 knots it generates about 6 amps, and I've seen 15 amps in 40 knots. The AeroGen is absolutely silent, a feature I most appreciate! I have no regulator on the windgen, and simply switch it off (short-circuit its output) when I don't need it (this is preferable to physically immobilizing the blades, since, as I understand it, localized bearing wear can then result as the blades rock ever so slightly). On overcast-sky passages, it's comforting to have. The windgen's best use was when we were sitting to the para-anchor in 30+knot winds and since we had so much power available I lit us up with all the deck lights in addition to running the radar all night long. Normally, the windgen provides a minimal overall contribution as a power source, since our apparent wind is low when we're running and it also gets blanketed when we're hard on the wind on port tack. If I had to do it again, I would perhaps locate it dead center on the targa bar.

[Solar Panels and Windgen] This cockeyed photo shows the two 120-watt solar panels and the Aero4Gen wind-driven generator.

Each outboard has a 10-amp alternator, voltage-regulated to 14.4 volts. Since increasingly we are no longer sailing purists and end up motorsailing often, these alternators do a great job of topping off the batteries.

When we were in New Zealand, I brought an extra solar panel from home which puts out a maximum of 3.5 amps. It's portable so I simply lean it either against the cabintop or foredeck for the optimum right-angle to the sun. Surprisingly, this panel was very much appreciated as we were sailing from New Zealand to Fiji (dead north) since the mainsail often blanketed the main solar panels for much of the day. When not in use (actually, it's almost never used), this panel unobtrusively stands up against the hull in Kathy's boudoir (inside the port forward cabin).

[Auxiliary Solar Panel on Forward Deck] [Ammeter Measuring Solar Input Current]

The backup solar panel put to good use during the New Zealand to Fiji passage. Small ammeters like this are great for measuring relatively low current levels.

Before I bought the windgen, while in a marina we once had a couple of weeks of very overcast skies and our batteries were indeed getting low. I bought an inexpensive 8A automotive battery charger (it has yet to put out more than 6A) which was a temporary solution. I think we used it twice in two years, until we got to spend our winter in Hobart - there, I used it often, but insufficiently to fully charge the batteries, and thus I managed to kill one of the main batteries. The problems with cheap battery chargers are that their dc outputs can contain a significant ac component which is harmful to batteries and there is typically no regulation on the chargers' outputs and they can overcharge batteries and boil off the water if not monitored carefully. Here in Queensland I certainly don't need a shore-based charging source, but I am presently doing my homework as I'll be permanently installing a shorepower-based battery charger onboard for future higher-latitude sailing venues.

Finally, just for fun I had experimented with a home-made water propeller-driven generator (sorry, no photos). It was certainly more trouble than it was worth. I think that a firmly-fixed (like a liftable outboard-motor leg) water-driven charging system is the way to go. Towing a water generator on a towline (even though very productive) seems to me to be another complication I do not wish to deal with on a passage.

As an aside, for a larger catamaran I would definitely consider a battery-powered hybrid power system instead of a couple of large diesels, where the weight of the small motor/generator, batteries, and small genset would be roughly equal to the weight of a couple of diesels and the normal battery bank. The power regeneration while sailing is automatic.

Click here to go back to top of page

Power Storage

Our power storage system simply consists of lead-acid batteries. I had considered Gel-Cell and AGM batteries, but didn't want to mix types, was too stingy to throw away perfectly good lead-acid batteries, and I'm not really convinced that they offer that significant an advantage for our application. The only concern that I have with lead-acid batteries is that if we happen to be upside down then the acid dripping into the saltwater produces a rather toxic gas - since this is a somewhat remote possibility (knock-on-wood), I'm willing to live with it.

When we started off (April 2000), the boat came with three 130 amp-hour deep-cycle batteries, the idea being that two were paralleled (260 amp-hours) and the third was kept in standby as a backup and for emergency engine-starting.

After a couple of years I had destroyed one of the batteries due to inadvertently undercharging it during our winter in Hobart. In July 2002, just before setting off to cross Bass Strait, I replaced my #1 battery with a couple of 225Ah 6v golf-cart (deep cycle) batteries (total is still 225Ah at 12v).

Just before we left New Zealand in August 2003, I replaced my now-weak #2 battery with a very small lightweight automotive starting battery, and brought online the third 130Ah battery (which had been in standby). Saved about 40 lbs doing this.

Last week, after I had added lots of monitoring instrumentation (to be discussed later) I realized that this last 130Ah battery was now doing nothing, so I replaced it with two 6v 105Ah deep-cycle batteries (these I call my secondary battery bank). Two 6v batteries are sure easier to manhandle than a single 12v battery of the same total capacity (and they fit better into KatieKat's storage compartment).

[Battery Two] The two new 105Ah 6v batteries fit nicely out of the way, leaving plenty of storage room in this main saloon compartment under the starboard forward window. The wires on the left are the coiled-up VHF antenna and Radar antenna wiring. That large circular thing on top of the battery terminal is a 100-amp fuse for the microwave oven!

Battery sizing should be such that the number of ampere-hours removed shouldn't typically be more than about 30% of battery capacity; i.e., I don't want to discharge my batteries regularly below about 70%. Practically speaking, when at the dock using shorepower for running the TVs, our overnight power drain is presently between 30 and 40 ampere-hours which translates to less than 20% of my primary 225Ah battery bank. No worries, and I don't even need to run the secondary battery bank in parallel with the primary.

If I were starting from scratch, I would have two large (around 300Ah) 6V batteries as my primary bank, and simply have a very small lightweight automotive 12v battery as an inactive emergency backup.

A couple of gadgets that I've used over the years are negative-pulse generators which are supposed to prevent battery plate sulphation. I've installed one across my standby battery which is simply trickle-charged from the main power bus and will install the other unit across my now-inactive secondary battery bank.

Finally, for all other active onboard uses I have tried to standardize on AA rechargeable NiMH batteries, with AA alkaline batteries for inactive emergency gear.

Click here to go back to top of page

Monitoring/Control Instrumentation and Equipment

The stock Seawind 1000 comes with a couple of batteries, a battery-selector switch, and a circuit-breaker panel with voltmeter. If you want more capacity, they'll add another battery to parallel the primary and put in a larger pair of solar panels. Quite frankly, for most applications, this is good enough and does the job well, especially with the smart solar-panel voltage regulator.

[Circuit Breaker Panel] The stock Seawind circuit breaker panel with built-in voltmeter.

For a couple of years I resisted the temptation to add instrumentation. A couple of cheap LED voltage-indicator lights was all I put on. When I added the windgen, I rigged a small ammeter in series with its output, and the unobtrusive external windgen wiring and ammeter was a kluge that we lived with in the main saloon for the last three years. When the first battery went out due to undercharging, I installed an old current shunt that I've had for years and simply monitored total I/O current with a $5 DVM. Then I added another $5 DVM to more-precisely monitor the voltage. Both of these were in the main saloon. Even though there is some wonderful special-purpose boat power instrumentation available, I simply couldn't justify the price of this marine stuff.

[Windgen Ammeter] The unsightly ammeter for windgen and the LED voltage monitor worked just fine for the last three years.

[Voltmeter and Ammeter] Two $5 DVMs were sufficient to measure system bus voltage and current for the last couple of years.

As you'll see below, I've just added a bunch of non-marine power monitoring gadgets - five ammeters, a voltmeter, and a multi-function power instrument.

Click here to go back to top of page

Modifications (updated 26 June, 2004)

Well, here we are, four years along and a bunch of klugy wiring which I had added to the neatly-done factory electrical system. Although functional, this looked like discards from a spaghetti-factory. It was time to get serious and clean things up a bit. Since I was going to tear the wiring access channels apart to bury the windgen wiring, I decided to put in some additional dc wires and (gasp) install an AC power distribution system as well.

Anyway, it was time to clean up the wiring and add some nicer power system monitoring instrumentation. Since the main saloon is the center of the boat's action, it was only logical to place the instruments where they'd be visible while sailing, and the space between the two forward windows seems like a great location. Over the last couple of weeks I've added a homemade panel and some instruments that I brought out with me from home, and have done some significant customization of the boat's dc power distribution system. This is still a work-in-progress, and if the temporary panel works out well, then I will build a nice enclosed permanent panel with completely invisible wiring. For the time being, the panel can be put away (thus the wires coming out the battery compartment cover) as I continue the philosophy of being able to completely strip and clear the main saloon.

[New Panel] A better perspective on what's going on. The temporary wire on the right dangling from up above is coming from a portable solar panel as I'm evaluating the new regulator seen on the right.

[New Panel] A work in progress showing the new panel, with a temporary solar current ammeter above it (a similar windgen digital meter is not shown), a temporary analog meter, and the new solar regulator I'm presently evaluating. On the panel are the two ammeters, voltmeter, switched outlets to provide power for the tabletop instruments, a voltage-level LED display, and the Tri-Metric battery monitor.

[New Panel] Update May 24. I incorporated the Solar and Windgen ammeters directly into the panel - they consist of a couple of $10 displays driven by isolated 9v batteries (hence the on/off switches). Makes for a cleaner-appearing panel layout. If this proves successful, I'll build a nice enclosed panel in the future.

[New Panel Labels] Update June 26. Merely updated the panel labels, hopefully improving the appearance a little. Since the switches are not dedicated to a particular function, they will remain unidentified - the switched outputs have been used over the past month for radar, GPS, VHF, inverter power, computer power, table lamp power, and auxiliary solar panel input.

[Ammeters] Update May 24. I now have two spare portable ammeters which work in conjunction with a current shunt that I can carry around with me. Not as handy as a Hall-Effect coil current sensor, but the price is right.

The first step of the wiring cleanup consisted of adding five (!) current shunts. A shunt is a calibrated very low resistance put in series with power wiring, the voltage drop across the resistor (in millivolts) being a measure of the current. The shunts are: Battery #1, Battery #2, Total System, Solar Panels, and Windgen. The reason for measuring each of the battery currents is that I want to see how both the charging and discharging current is being shared when they are paralleled - that's a whole separate techie topic.

[Joe Rewiring Boat] [Battery Compartment] [Battery Compartment]

Any project on a small boat results in an instant mess for the duration of the project. In addition to all the rewiring, I threw in a smoke detector inside the battery compartment. That small lightweight "Maintenance Free" (ha!) Battery 3 is for emergency engine starting only. You can just see the large 225Ah 6V batteries in the right photo.

Now, in addition, there are gadgets on the market which measure not only amps and voltage but also calculate ampere-hours and present the information in a variety of ways. I picked up a relatively inexpensive non-marinized unit (the Tri-Metric, designed for home systems) which keeps this engineer happy by providing: volts, amps, % full, amp-hours from full, days since last charged, days since equalized, cumulative amp-hours consumed, highest battery voltage, lowest battery voltage, and a number of different programming options relating to battery capacity, charging efficiency, and battery fullness. One cool use it to hook it up directly to any onboard power consumer or generator and directly compile the number of ampere-hours used or generated. The only problem with mounting the Tri-Metric in the main saloon is that its display is rather dim for full daylight-viewing, but I guess that I could mount it down below next to the existing power panel (but then I couldn't conveniently play with it, so I won't). That's right, these are all techietoys :-)

Click here to go back to top of page

Power Regulation

A few words about the solar panel voltage regulator... the unit that came with the Seawind is labelled Plasmatronics (previously I believe called a Smart Regulator from Solar Generation), and it has worked out very well as it measures time-related parameters that it stores in its memory and optimizes charging based on those parameters in addition to normal voltage/current sensing. This older unit has a two-stage regulator (Bulk and Float) and has a battery temperature sensing option, which is certainly sufficient. I've added a boost switch so I can manually purposely overcharge the batteries to equalize them occasionally. Now (June 12), some friends on a neighboring catamaran (Sean and Faye Thomson on Thomcat) just gave me a new Plasmatronics unit that has a damaged output on it, and I can't believe the features this little baby has! In addition to being a very smart full four-stage regulator, it has such a variety of accessible features as to exceed the needs of all but the most geeky engineer - love this stuff! The damaged output isn't needed for KatieKat's application, so I'll see if I can obtain a remote control panel and install this onboard.

[Solar Regulator] The solar power regulator is mounted in the cabinet behind the circuit breaker panel. Unfortunately, need to open the cabinet door in order to see the regulator's status lights.

[Rear View CktBrkr Panel] Looking at the rear of the circuit breaker panel you can see the three Schottky diodes and resistors I added (as a temporary measure three years ago :-) which provide trickle charge currents to each of the batteries, irrespective of whichever one is active. On the list is relocating this assembly to the battery compartment and having the paths be switch-selectable as well.

Another regulator that I'm presently evaluating is very technically creative. Solar panels are basically constant-current devices, putting out their peak current at some voltage higher than that required to charge batteries. What one outfit (RV Products) did is take this current at the higher output voltage, run it through a highly-efficient dc-dc converter, and output the battery charging voltage with a current higher than that put out by the solar panel. Way cool, as typically, the power thus salvaged is normally lost in the voltage drop of the series regulator. Like I said, I'm still evaluating it...

The whole topic of battery charging is a lively one amongst cruisers and battery and charger suppliers, as the need for the fancy four-stage chargers (bulk, absorption, float, and equalize) whilst living aboard is open to debate.

Click here to go back to top of page

Shorepower (updated 26 June, 2004)

In recent years, while in marinas, I've been reducing the battery loads by running the TVs on 240vac. The wire to the boat was simply a long extension cord from the dock and down the hatch. We've discovered the popular-here (and in Europe) 240vac water kettle which boils water very quickly (twice as fast as the rarely-seen ones running on 110vac at home). In fact, I liked this kettle so much that I took one home to California with me and am happily running it off the stove's 220vac in my kitchen there.

[Kettle] Our 240vac kettle, used while in a marina, sure saves on propane!

The final phase of KatieKat's wiring upgrade is the addition of a built-in AC system. It consists of 110vac panel and North American outlets, and I'm presently using only a portion of the panel for the Australian 240vac.

[AC Panel] Update June 26. I'm temporarily using one of the circuit breakers in this 110vac panel for my 240vac input, being fed into the 110vac connector inside the companionway. When it's time to convert over to 110vac I'll mount a proper weatherproof connector outside, thus eliminating this single klugy inside wire.

[AC Outlet] Update June 26. This is one of four unobtrusive and convenient 110vac outlets which serve our 240vac needs just fine, using simple connector adapters. Amazing how most of the electronics is multi-voltage and multi-frequency nowadays.

So, there you have it - all you didn't want to know about our onboard power. I'll try to keep updating this webpage as I make mods to the boat's electrical system.

Click here to go back to top of page

Parallel Batteries (added 3 August, 2004)

I've received so many emails relating to my initial question about paralleling batteries (none of which have answered the question to my satisfaction) that I've now added this as a separate topic on this webpage. What follows is my initial question followed by my analysis - since I'm a firm believer that a little knowledge is dangerous, please feel free to shoot holes in my arguments and enlighten me.

Question About Paralleling Batteries

I recognize that different types of batteries should never be mixed. For example, the charging requirements for Gel-Cells are different than for AGMs which are also slightly different than for deep-cycle lead-acid batteries. No issue there.

What I wish to talk about is the paralleling of two different lead-acid deep-cycle batteries. All available literature says that one should NEVER parallel non-identical batteries, even of the same type. Discussions with battery representative yield different reasons for saying this, with some conceding that it's ok to parallel dissimilar batteries when only discharging while others say its ok to parallel them when charging but not discharging. My questions pertain to the paralleling of two lead-acid deep-cycle batteries of different brands, different capacities, and different ages:

  1. When discharging, why don't two dissimilar lead-acid deep-cycle batteries simply share their current proportionally to their internal series resistance?
  2. Why should paralleled lead-acid batteries be damaged if charged from a source which limits the maximum charging voltage to, say, 14.4 volts at room ambient temperature - doesn't the internal impedance of the battery simply limit the current once the battery is charged, thus allowing the other battery to continue charging? [We're talking charging currents on the order of 20 amps into the combined 300Ah banks.]

If someone could answer these questions, I'd appreciate it. I do not like paralleling batteries - not because their power draw or charging is at different rates , but because if one battery happens to get a shorted cell (rare in deep-cycle batteries, but common in automotive batteries) then the other (good) battery gets drained by the bad guy. See what happened on another boat when a paralleled battery blew up!

[Shattered Battery Case] Thought you'd like to see another photo of that battery that had blown up on a neighboring catamaran.

Added the following on 3 August 2004:

Re: Exploding Battery

I am now convinced that it was indeed a hydrogen explosion on the other boat, and not simply a gradual increase in pressure which caused the case to blow out. I've talked with a number of battery reps who confirm that a fire rarely results from such an explosion (but what a mess!). Starting batteries have thin plates and are susceptible to shorting, especially if the fluid has boiled off. It is very rare that deep-cycle batteries experience the same problem, with their failure mode being a break in the plate-to-terminal connection. In either of these cases, the fluid must be boiled off in order to generate the spark to cause the explosion.

Re: Paralleling Batteries

We're talking about parallelling deep-cycle lead-acid batteries of different brands and/or capacities. Wherever we use two 6v batteries in series to generate 12v, then indeed they must be identical, but that's not the issue nor the question on the table. The question is simple: WHAT IS WRONG WITH PARALLELING DEEP CYCLE LEAD-ACID BATTERIES OF DIFFERENT CAPACITIES, AGES, AND/OR MANUFACTURERS? What follows is my take on this subject, and discusses the four phases or charge modes of modern microprocessor-controlled regulators and my perception of the effects of having two batteries in parallel.

Before I begin, let me once again explain the scenario: we are talking about two or more sets of same-type (e.g, lead-acid deep-cycle, but of different brands, capacities, and age) batteries that have been connected in parallel for a long time.

BULK (sometimes called BOOST) charging merely applies all the current a charger can deliver, the only thing usually being monitored being voltage. The current between the batteries is simply shared in inverse proportion to their individual internal impedances. When the voltage reaches a pre-set point (around 14.4v) the charger changes to the Absorption Phase. SO WHY DOES IT MATTER IF TWO DISSIMILAR BATTERIES ARE PARALLELED?

ABSORPTION (also called ACCEPTANCE or CONSTANT VOLTAGE) holds the voltage constant (typically around 14.4v at 20degC) and each battery decreases its incoming current based on its own increasing impedance as it charges and its fluid specific gravity increases. Most microprocessor controlled regulators are a combination of smart and dumb - the dumb being simply a preset time that the charger sits in this phase and "smart" (my definition) being monitoring the charger output current and switching to Float when the current drops to a pre-set limit. These are often also somewhat dumb because many have no way of changing this current limit, which DOES MATTER because this limit is different for different-capacity banks. SO WHY DOES IT MATTER IF TWO DISSIMILAR BATTERIES ARE PARALLELED, because one battery just sits there fully charged while the other one is still absorbing? This voltage is still below the point where rapid fluid boiloff would occur.

FLOAT holds the voltage constant (say, around 13.2v) or, as with my smart solar regulator, a slow oscillation with hysterisis whereby the charger waits until the voltage drops to 13.2v then applies full current until the voltage rises to 13.8v and then cuts off completely. The voltage then slowly decays down to the 13.2v trip point. This takes maybe a minute in practice with a light load on the battery. SO WHY DOES IT MATTER IF TWO DISSIMILAR BATTERIES ARE PARALLELED?

EQUALIZATION applies a high constant voltage (say, 15.5v) and is intended to "boil" (burble or stir-up) the fluid and desulphate the plates and equalize the specific gravity in each of the cells. This has to be for a relatively short time (1-3 hours I've been told) or else the fluid boils off. Since the voltage is constant, the problem with paralleled batteries is that the smaller battery may boil off its fluid before the larger one is "equalized", and thus I agree that for this Equalization Phase that paralleling dissimilar batteries is not good. This phase is performed only once every few months and my understanding is that the latest thinking by battery manufacturers is that it is unnecessary if a smart bulk-aborption-float charger is being used. On KatieKat my older controller does not have an Equalize Mode and I've never bothered to manually circumvent it in order to equalize the batteries, with my negative-pulse generators apparently doing a good job at keeping the plates desulphated (they all look very clean and the fluid specific gravity is well-balanced).

The algorithms used in microprocessor-controlled chargers are, in most cases, very simple. Happily, some are nicely programmable (e.g., the Plasmatronics solar charge controller), but I am disappointed that many others do not publish EXACTLY what they do. Certainly, battery temperature compensation (either via a sensor or a switch) is very necessary to do a proper job. Some microprocessor controllers are dumb and can't distinguish between a voltage drop due to a temporary load and a longer-term battery voltage drop, and thus go back into a mindless Absorption phase for their preset time, unnecessarily cooking the batteries. As I mentioned above, a smart controller should also be programmable to measure the current during Absorpion mode and shut off when that current drops to a given percentage of total ampere-hour capacity (recognizing dissimilar units).

A case can be made that a battery charger does not need the Absorption phase unless one is on an ocean passage and wants to stuff the largest amount of current back into the batteries in the shortest period of time while running the engines briefly. In a real-life liveaboard situation, with sufficient solar power available, there is rarely such a time constraint, so why not let the batteries just gradually top-up in Float Mode?

DISCHARGING paralleled batteries doesn't matter, in my opinion, since the current is being shared by the two, once again in inverse proportion to their internal impedance. The battery with the lower internal impedance merely supplies more current than the other, but does not "put current into" the other one (as some have told me). The ampere-hours out of the battery bank are still roughly the total of the capacities of the two batteries. In a no-load situation with two paralleled batteries, the battery with the higher self-discharge current will indeed draw a slight amount of current from the other, but this is negligible in an everyday setting where solar panel recharging takes place. Once again, for the everyday usage scenario, WHY DOES IT MATTER IF TWO DISSIMILAR BATTERIES ARE PARALLELED?

As I mentioned above, a dominant argument regarding paralleling batteries has nothing to do with whether they are identical or not: from a safety perspective, if one battery develops an shorted cell then the other one will boil itself trying to bring the first one's voltage up.

A battery combiner is essentially a voltage-sensitive relay (switch) which connects a second battery to the primary one (in parallel) once the voltage exceeds a certain value (e.g., 13.3v). Now, these are widely used to kick in, say, a starting battery and top it up. Nowhere in the catalog descriptions for these gadgets do they mention that similar-construction batteries shouldn't be thus paralleled. Recall, lead-acid starting batteries are composed of many thin plates (for high peak current loads) while deep-cycle batteries are made from fewer but much thicker plates, but since they are both lead-acid then their charging characteristics are presumably identical. The only reason I'm pointing this out is that this popular setup inherently forces the batteries to be paralleled and the starting battery undergoes the charging routine I discussed above even though it is probably already almost fully charged.

In the overall scheme of things, batteries are very tolerant creatures - after all, millions of troublefree cars are on the world's roads, most with fairly crude voltage regulators.

I've now had ammeters in series with each of my two paralleled sets of 6v batteries (each 6v set is identical) for three months. The older (two years) 225Ah "heavy-duty" deep-cycle USA-made battery bank supplies (and absorbs) roughly three times as much current as the brand-new 105Ah Taiwanese-made bank. On the face of it, I would have thought this number would be roughly 2X and not 3X. Anyway, both banks are happily providing the boat's needs, with noticeably less overnight voltage drop from this paralleled hookup than if running only one bank by itself, which tells me that current sharing is indeed taking place.

You should see the fur fly after a few drinks when techie yachties get into debates over this subject. I'm hoping that a battery manufacturer's techrep will one day address the issue, as the question is still on the table: WHY SHOULD DISSIMILAR BATTERIES OF THE SAME TYPE NOT BE PARALLELED? I'm just now beginning to suspect that the issue may not be technical: I've seen yachties spend thousands of dollars replacing their entire battery banks because of one bad battery, and would hope that the reason or answer is not sales-driven. Conspiracy or collusion by the battery manufacturers or dealers? I don't think so, but I'm afraid that the older I get, the more cynical I become... please prove me wrong!

Paralleling Batteries and 2006 Electrical System Upgrade, December 2006 (updated 27 February 2007)

It's been 2-1/2 years and I haven't received any technical refutation of my above contentions - pity, as I was hoping this would spawn howls of protest from knowledgeable individuals. Well, once again I put my money where my mouth is: when it was time to increase the overall battery storage capacity of KatieKat's electrical system, I paralleled a 12v 4D 198AHr AGM battery with the existing bank of two 6v 225AHr Flooded Lead-Acid batteries! I did this after further research into the subject and discussions with engineers at some of the name-brand battery manufacturers (forget the marketeers). Click here for a discussion of the September 2006 upgraded electrical system, including batteries, solar panels, and solar regulators.

The ideal AGM charge cycle (by the way, this varies with manufacturer) is close enough to flooded lead-acid that the solar panel voltage regulators can be programmed to a compromise which keeps both batteries happy. The only issue with AGM is its dislike for overvoltage when charging, and for this reason I've added an overvoltage alarm and have actually had to physically disconnect the fully-charged AGM battery a few times when motoring because the Yamaha unadjustable voltage regulators are set too high - this is something to consider if AGM batteries are used with outboard motors or, for that matter, any motors or charging sources not configured for AGM batteries. I am obviating the need for a periodic overvoltage flooded-lead-acid desulfation cycle by using a negative-pulse generator across that battery bank.

Note that paralleling a Gel-Cell battery with a flooded lead-acid battery would be inappropriate because the Gel-Cell charging regimen requires significantly lower voltages than the other. Note also that battery temperature sensing is a necessary ingredient for a proper charging regimen.

The results so far are very encouraging, with very good (but unequal) current sharing taking place - this unequal (sometimes greater for AGM, at other times greater for flooded-lead-acid) sharing (unequal both charging and discharging) seems to be a function of system voltage at any given moment. I hope to start taking data in order to rigorously see what's happening. Incidentally, I am continuing to monitor the currents in each battery bank separately and am convinced that any large paralleled battery bank should have such a monitoring system (preferably automated with alarms/disconnects) in order to assess the health of each individual battery and also hopefully prevent an explosion (such as the one discussed above) from occurring.

8 December 2006: I forgot to mention one more simple safety feature which I have not yet implemented: a large (100A - 150A) circuit breaker in the leg of each battery bank - if one battery gets a shorted cell then one or more circuit breakers will hopefully open up as the other batteries try to feed it. Thank you Steve on Adagio for reminding me of this.

Update 27 February, 2007. At the end of December I eliminated the two large "1-2-Both" rotary switches and replaced them with two circuit breakers - a 120A breaker in the positive leg of each of the two battery banks, and a switch for the third small starter battery, which is now a tiny AGM.
Update 4 August 2011. Replaced the 120A circuit breakers with 70A breakers (see below)

[Photo of two circuit breakers and a switch] The two circuit breakers and switch replace the two large rotary switches. Battery 2 breaker is shown in the open position. The Battery 3 switch is always OFF, as that is only for my emergency starting battery which is now a tiny 11AHr Odyssey AGM typically used in motorcycles.

[Photo of two circuit breakers] This photo shows a couple of other new circuit breakers - one for the two new solar panels and one for the watermaker circuit. I also added a fuse dedicated to the shorepower charger. Hard to keep the wiring neat, as it's an evolving mixture of old and new.

[Battery compartment wiring diagram] The never-ending struggle to keep up with all the changes has resulted in a pretty gawdawful drawing (to think I worked my way through school on a drafting table). One of these days when I'm bored silly I should properly redraw this diagram, but at least it gives you an idea what's inside.

End-February 2007 - Parallel Battery Status. This non-standard implementation of paralleling a 4D 198AHr AGM battery with two 6V 225AHr Flooded batteries continues to work wonderfully, with the two paralleled dissimilar battery banks happily sharing their charging and discharging currents since September 2006. No problems so far.

15 May, 2007 - Regarding Paralleling Batteries, I've been vindicated! Came across this paper: www.battcon.com/PapersFinal2002/McDowallPaper2002.pdf.
My dissimilar paralleled battery set is still very happy.

October 2007 - Click here for a discussion of the performance results of the 2006 upgraded electrical system.

Photos of Batteries, November 2007

After all the talking about batteries, I realized that I forgot to show the latest locations of our three batteries. These photos answer some of the email questions I've been receiving. This is my battery configuration since September 2006.

[Photo of battery one] This is Battery One, consisting of two 6v 225AHr flooded lead acid golf-cart batteries (yielding 12v at 225AHr), installed when we were leaving Hobart in July 2002. Still going strong, but I am now adding about 10cc of distilled water about once a month to each cell. Note the blue Australian desulphator sitting on top of the battery.

[Photo of battery two] This photo is looking forward under the settee at Battery Two, the large 198AHr 4D Deka 12v AGM, which I installed in September 2006. This is excellent utilization of an otherwise inaccessible location. The foam blocks jammed in on the side and top of the battery are hopefully sufficient to immobilize it. As discussed previously on this page as well as here, this AGM battery is operating happily paralleled with the above bank of two flooded-lead-acid 6v batteries.

[Photo of battery three] This is Battery Three, a tiny 14AHr Odyssey PC545 AGM battery used in Harley Davindsons (it has a short-circuit current of over 1200 amps!) which is my emergency motor starting battery and is isolated from the rest of the system with a switch. I keep it charged via a resistor and Schottky diode off the main battery bank. I've had this battery for a few years... it just sits there doing nothing, but I occasionally test it to make sure it can still start the Yamahas. Note the smoke alarm inside this battery and wiring compartment.

Update, 4 August, 2011 - Circuit Breakers, Optima Batteries, Shunt Voltage Regulator

Made a number of changes to the setups described above, the most important of which is that I've replaced the individual battery 120A circuit breakers with 70A circuit breakers. This is just above the worst-case individual current draws which are either microwave oven or motor starters. These circuit breakers not only function as individual on/off battery switches but are meant to protect good batteries and keep bad batteries from burning up should any battery short out when paralleled with others. Have no idea how they would work in practice (i.e., whether a failed battery short-circuit resistance will be low enough to draw enough current to pop the breaker), as I've never had an onboard battery short out.

[Photo of Optima yellow-top battery] I finally got tired of adding water to the 6v flooded lead-acid batteries (which we installed in Hobart in 2002) and replaced them with a couple of spent 55AHr Optima AGMs from my electric cars. Although no longer suitable for the electric cars (which can draw hundreds of amps continuously), these Optimas still have plenty of life left in them for use onboard the boat. I have a 70A circuit breaker between these two, in addition to the (now) 70A breaker to the system bus. After nine years onboard KatieKat the flooded batteries were still in good shape and I donated them to a museum for use in an old electric car.

[Photo of shunt regulator with cooling fan] When running the outboards at the marina with the batteries already fully charged from the solar panels, I've seen overvoltage in the system (up to 15.5v), despite the outboard alternators having voltage regulators supposedly limiting the voltage to 14.4v. Although sustainable by the Optimas, this voltage is way too high for the Deka AGM battery, which has a 14.4v max rating at 70degF. After replacing both voltage regulators with new ones and still not being able to identify the cause of the problem, I've implemented a temporary brute-force approach using a shunt regulator cooled by a small computer fan. I originally picked up this regulator off a cruiser in New Zealand for use with the Aero4Gen wind generator, but never installed it. When living onboard and cruising, having the batteries go into overvoltage is simply never an issue because there are always some loads applied to the system which then seems to trigger the alternator voltage regulators into working properly. Go figure. Remember, batteries don't die: they're murdered!

Click here to go back to top of page
Click here to go to next chapter
Click here to go to previous chapter
Click here to return to KatieKat Home Page