Find the smallest starting battery for you

[Rick]

The smallest "size" starting battery which will reliably start your diesel engine is one SAE rated with a CCA (cold-cranking-Amps) value equal to or exceeding the stall current rating of the starter motor. I've not seen this concept promulgated within the marine industry so here it is. If you know the power rating of your starter motor and not the stall current rating divide the power, in Watts, by 5 (for nominal 12V systems) or 10 (for nominal 24V systems). A 5kW rated starter motor is 1500 Watts, thus the stall current will be close to 1000A and needs a battery with a CCA rating of 1000 or higher.

If possible choose the battery having the lowest internal resistance value and, in no case, choose one having an internal resistance higher than 10 milli-Ohms (10mOhm). Neglect ALL other ratings like reserve capacity. Because we are attempting to minimize the size and expenditure never consider using a deep-discharge flooded electrolyte battery because of the higher internal resistance size for size compared to a start-only-start battery. If possible use the technically advanced cylindrical construction batteries such as Optima or Hawker due to the low internal resistance and higher specific gravity (which directly translates to higher voltage under load) than flooded batteries. In the case of these SLA (sealed-lead-acid) batteries they may be rated to serve as deep-discharge use as well yet because of their intrinsically low internal resistance they work better than the "wet" ones, pound-for-pound.

All starter motors exhibit an output power versus current curve shaped like a parabola or inverted chain catenary. What this means to you is that when operating at any power out, other than exact maximum, there are two values of starter current which will deliver the SAME output of the motor. The high values on the right side of the power curve will, in general, be due to operating at sagging battery voltages (high internal battery resistance and cable and connector resistances. Using inferior batteries (like flooded deep-discharge ones) that are small WILL likely cause operation in this region creating unnecessary heating and wear and tear on the starter motor even though the engine may start. If the voltage to the motor sags enough then the starter motor will stall and you will be operating at the extreme right point of the curve with no engine rotation. However, if the battery fails to deliver this stall current you might have NO chance to get the engine rotating with cold oil and freezing ambient temperatures.

In short, it is only motor CURRENT which causes the starting torque. It is starter terminal VOLTAGE which allows a peak power to be developed with that current which translates ultimatly into rotation speed. The faster the starter motor rotates the higher the compression of the engine and the faster the engine will start.

I had the occasion to measure the effects of using a 2000 A-hr rated Rolls battery (great for operating lights for long periods of time) which would only slowly grind the engine barely starting it (sagging voltage caused by high internal resistance). Switching to a much smaller 1400CCA start-only-start battery (no good for running lights without ruining the battery) caused the engine to start almost instantly.

A note about reliability: Flooded-cell constructed start-only-start batteries should never be trusted as being reliable if they have been left standing too long without a float voltage applied or if ever deeply discharged. Shallow discharging (on the order of a 3% capacity discharge) is also usually deliterious in the long term unless specifically designed to do so.

Hope this clears up some of the mis-information floating amoungst the marine community.

Regards,
Rick Young

[Rick]

OOps! Obviously the 5kW value in the example is equal to 5000Watts, not the 1500Watt value that I stated, sorry!

[GordMay]

Rick:
I’ve been trying to ‘get my head round’ your interesting post on Starter Motors and CCA - I know you’re on to something here, but I have a few questions:
1. How do you derive the specific denominator “5" for the Numerator “Watts” (for 12V) in determining required CCA ?
2. Where do you locate battery specifications that describe internal resistance (Ri) ?
3. I’d be interested in your elaboration on the concept of operating at the Maximum on the Output Power Parabola ...
Ie: Aren’t we interested in the torque Required to Start (a function of the Engine inertia & rotational speed required), rather than the Available Torque (Stall Torque) of the starter motor?
Thanks for the excellent information,
Gord

[jim lee]

Are we talking about an "emergancy use it once cause the house bank is dead" kinda' application or something that we "always" use as a starting battery?

I run into this delema all the time with the Jetboat boys.

-jim lee

[Rick]

First answering Jim Lee's question: One implication of my post is that if you cannot guarantee starter operation in the "left-hand" efficient side of the power-current curve using a house bank then you should have a dedicated start-only-start battery. This technical position helps to derive a battery bank design philosophy for cruising. It IS possible to have a single bank system, however, it would be prudent to have a backup like a dry-charged battery and bottle of acid properly stored. Another option is to have a minimum-sized SLA battery always on float (when a charge source is available. Note: with the advent of affordable properly designed battery monitors single house battery banks are the most cost effective and energy efficient manner of configuring your house bank. Alternating two banks of batteries is only life-time effective if the "resting" battery is left fully charged while using the other bank...a caveat not bandied within the marine community yet known in alternative energy applications. This proper method of two bank usage is not practical on a cruising boat which has limited space and budget.

Gord: The number "5" derives from dividing the Power (watts) by 10 Volts (12 V would be O.K only if one could guarantee that the manufacturer's power spec is rated at that number and not for a more practical sagging voltage situation) to achieve the peak load current. We then multiply the peak rated load current by two to reach the locked-rotor (stall) current value. Observing the parabolic curve it is easy to note that the locked-rotor current is always twice the peak power current.

Most batteriey specifications (for batteries intended for use other than consumer) include internal resistance because the buyers are more technically astute than consumer users or battery distributors. GNB, Optima, Hawker, Prevailer, and others include the Ri of their batteries...sometimes you have to search for the inclusive spec sheets to get the values. Some manufacturers (like Prevailer) publish Ri versus state of charge which is VERY nice to have. These numbers vary with temperature, of course.

If you are fortunate enough to have a battery monitor, a battery, a high power inverter, and a heavy load such as a 1500W heater or microwave oven (1200W input power or so) it is convenient to measure Ri by starting with a fully charged battery and appling the heavy heater load (or quart of water in the microwave) and logging the battery terminal voltage and output current for 10 minutes. Calculate Ri as the voltage divided by the simultaneous current value. You will notice that the Ri starts very low then climbs to a value and flattens out (if the battery bank size is designed to carry the load and is in good condition). The average value of Ri in the "flat" part of the curve is the real initial value to use. This value is good to log down and compare to future values in order to determine battery condition and wiring system integrity.

The starter output power parabola does NOT infer that it is desirable to operate at the peak power at all times. It merely means that the peak value is the most that you can get for a given (rated) terminal voltage. What I am pointing out is that it is desirable to operate ONLY in the left-hand portion of the curve and proper battery choice helps to guarantee this. Obviously when ambient temperatures are high and the battery is larger than a minimum size you will be able to operate closer to the left-hand origin of the curve thereby minimizing cranking times and unnecessary power loss.

Yes, we are interested in the torque required to start yet if we do not pay attention to the phenomenon of possible operation in the "right-hand" mirror image of the power-current curve we might fall into the trap of using an insufficiently sized power source (the battery). I've heard people say, "what difference does it make, the engine starts anyway"? Well when their starter fails permaturely or when very cold conditions conspire to prevent the engine from starting at all then, they will be more interested in this concept! Remember that the stall current value produces the instant cranking torque at zero engine speed under the most limiting drastic starting conditions...if you can't break through the starting stiction to get the engine rotating it just woun't turn. It is that value which is the beginning point for determining a starter size by an engine manufacturer engineer.

Hope this helps,
Regards,
Rick Young

[Talbot]

Optima is a great battery , IIRC it is a carbon fibre reinforced plate construction so could be called a dual purpose design in that it is much more capable of deep discharge and recovery, and also of fast charge, than a normal engine start battery. However, in UK this sort of battery is VERY expensive by comparison to a standard car start battery . It hasd advantages in weight and size, as well as the fast charge, but you need to be sure that the extra cost is worth it to you.

[David W]

While I will admit that this information is fascinating, what is a simple sailor with a 15 year old greasy inboard with non-existent starter motor specs to do when the time comes for a new starting battery?

Me? I will fit the largest CCA battery I can fit into my available space... unless there is an easier way of determining what might be more suitable...