I think most people's common sense would dictate against paralleling a 9.6v pack with a 7.2v pack but it is still a problem with two nominal 7.2v packs - note I use the word 'nominal'. In fact the voltages of each pack will be slightly different - representing manufacturing variations, battery condition, cell imbalance and slightly different states of charge. When you connect them in parallel, the higher voltage one of the pair will attempt to 'charge' the lower voltage one with only the very low internal resistances of the two batteries to limit the current. This is the same internal resistance that scarcely make the battery voltage drop when you take 30A out of it! So an extremely high charging current results and both batteries get cooked.

Lets illustrate this with a few numbers. Assume one charged 7.2v pack reads 8.3v and the other reads 8.1v and that the internal resistance of each pack is 0.005 ohm (that's 5 milliohms - comparable with the typical resistance of a good speed controller). So if we parallel the batteries we have a voltage difference of 8.3v-8.1v = 0.2v impressed across 2x 0.005 ohm and Ohm's Law tells us that 20A will flow between them! Remember they will get warm when charged at a tenth of this rate.

Doubtless, all of us will have heard about someone who has actually got away with doing this without ill effect, but what makes you think you should be so lucky? (It may be that they put the paralleled pair under immediate discharge and ran them pretty flat without stopping)

So, what to do? Well, if you have an Electronize speed controller the answer is easy - to double the running duration of one battery, connect a second one in SERIES with it. WARNING This only applies if the doubled battery voltage remains within the 24v rating of the Electronize ESC !!!!

How it works is like this:- for example, to supply a motor with 3.6v, an ESC would switch a 7.2v battery on and off very rapidly (anything from 50 times a second to several thousand times a second) so that the battery was on for half the time and off for half the time - so the motor would see, on average, half of the 7.2v - ie the 3.6v required. Now if you use two batteries in series, and want 3.6v on the motor, then the ESC will turn the batteries on for a quarter of the time and off for three quarters of the time and a quarter of 14.4v is, surprise surprise, 3.6v! Clearly if doubling the battery voltage results in the batteries only being on for half the time to supply the same motor voltage then their duration is doubled. (note for nit-pickers - yes I know this is not strictly true as the switching losses are higher, but lets not over-complicate the issue)

IMPORTANT NOTE : Of course, this solution has the potential to run the motor at twice the normal operating speed (which may or may not be disastrous depending on the boat set up) but you may guard against that by setting the Electronize ESC's 'speed range' adjuster pot to half its previous value.

However, if you have a Mtroniks Viper type ESC, it may not be so easy. They are only rated for 12v (8 cells actually, so that means a nominal 9.6v battery) so unless you are running on 4 cells at present, forget it. That is not to say I haven't recently seen a set up where a Viper was running on 12 cells. OK, the power FETs inside it may, in themselves, be rated for 40 to 60v, but the manufacturer's voltage/cell limit is undoubtedly based on you not cooking the unit with a stalled motor - the higher the voltage the worse the situation and the Vipers have a very small heatsink. I noted that the reverse had already failed on that particular unit - and I'm willing to bet the failed direction used to be 'forward'.

Also, I am not sufficiently familiar with the Viper set-up to know whether it can be programmed to reduce the effective speed range.

But if your ESC is voltage limited and you really must double your running duration (I'm neglecting the obvious idea of taking a break at half time to swap between two batteries!) there is still a solution, though far less than perfect. Common the battery negative leads and feed each battery positive to the load via a diode. The diodes will block any reverse (ie 'charging') current into whichever is the lowest voltage battery and ensure that they both 'share' the load throughout their discharge. This would be true for 3, 4 or as many batteries as you wish to 'pseudo-parallel'. It would also work for two mismatched battery cell counts, but the boat would then operate so as to exhaust the higher voltage pack first before automatically cutting in to the lower voltage pack - and incidentally holding the first discharged battery at that lower level so a gross mismatch of cell count is not recommended.

The price to be paid for this is that each diode will drop a small voltage so some of your battery energy will be wasted as heat (which you also have to get rid of) so whilst your duration will be increased it won't be doubled.

The voltage drop of a normal silicon diode has two components - about 0.7v regardless of the current plus a further resistive drop dependent on the diode's rating. The best diodes to use are Schottky types where the fundamental drop is about 0.3v as opposed to 0.7v. A 30A rated Schottky diode will drop, all told, about 0.7v at 30A - but that's 21 watts wasted so in this example a BIG heatsink would be required. But for a modest tugboat, taking just a couple of amps, this technique might be workable.

Those who wish to explore this possibility further are invited to contact me and I can suggest parts, estimate voltage losses and heat dissipation (ie required heatsink size) for your supplied operating current, and draw up a wiring diagram.