390
Battery conversion cures V8 hot start problem
Bob Own provides a sequel to his recent V8NOTE388 on how he solved a hot start problem. (Oct 08)

The problem
For some time my V8 has been giving some starting problems when hot. I had cured an ignition problem that caused the engine to stop when hot - see V8NOTE388 - but this second problem was a "lazy" starter which gave inadequate cranking speed when the car was very hot, although it gave adequate cranking speed when cold. A typical engine needs to crank at more than 90 rpm to start.

Why should it occur only when hot?
The V8 starter and solenoid are close to the exhaust and so can get very hot if the car is idling or in slow moving traffic, especially in hot weather. A heat shield was fitted by the factory but often this is corroded or missing. Clive Wheatley does an excellent aftermarket RV8 heat shield which can be used on an MGBGTV8. You need to use a cut down fixing to allow you to fit it with the exhaust in place. However, the heat shield is only partly effective in reducing the temperature of the starter/solenoid. When the copper windings get hot the resistance increases and the current passed for a given voltage decreases. The solenoid thrust and the starter torque are both proportional to the current passed. The resistance of copper is proportional to the absolute temperature (0C is 273 absolute or 273K). So in going from 17C - 100C (290K to 373K), the resistance goes up by nearly 30% so the starter torque/thrust drops by 30%. If you compound that with a less than pristine battery that has a cranking voltage that may also be down by 30% then you have only half of the design torque or thrust. You might think that half full power should be enough, but this overlooks the two stage operation of a pre-engaged starter motor.

Pre-engaged starters
When the starter relay feeds current to the starter solenoid as a result of you turning the starter key, two solenoid windings are energised, one in parallel with the battery and one in series with the starter motor. The one in series with the motor passes quite a high current and will cause the motor to rotate slowly while the solenoid operates the lever that pushes the starter pinion into mesh. When the pinion is fully in mesh and the solenoid is at the end of its travel, the main starter contacts are closed causing the battery to connect direct to the starter and the full starter power to be developed. Inadequate current through the solenoid, tightness of the solenoid or damaged solenoid contacts may prevent correct operation of this final stage.


Driver's side battery compartment viewed from rear. The battery isolator is just visible near top of picture; thin fused lead feeds the alarm so that it still operates with the isolator OFF.

Curing the problem
When exhibiting "lazy cranking", with the final stage of connection not being reached, if the V8's batteries were jumped to another car's battery (usually poor Malcolm Sayers when I was in Italy!) the starter would zip away and the car would start straight away. So at least I had identified that a significant part of the problem was insufficient "welly" from the batteries. I resolved to sort this problem when I got home. I would change the batteries first and then, if necessary, do something about the starter/solenoid - not something to be considered lightly on the V8.

Original 6V batteries
The factory MGBGTV8 was originally fitted with two 6V batteries connected in series to give 12V. This is the same arrangement as in the 1.8 MGB of the time. However, the V8 engine requires about 50% more starter torque than the 1.8 model, so starting is likely to be more "iffy", even without the effects of high exhaust temperatures on the starter/solenoid. Whilst preferring originality if possible, I was getting a bit fed up with the 6V batteries. These were less than 3 years old and already tired; and they're not cheap. So I decided to weigh up the pros and cons of various approaches.

6V in series versus 12V in parallel
At first sight there would seem to be little advantage in using two 12V batteries in parallel over two 6V in series. Other things being equal, the same volume of battery would have the same energy storage. However, most development in the last 40 years has been on 12V batteries so the 6V batteries are still essentially as supplied in the 1960s, whereas there is a wide choice of modern more advanced 12V batteries.

Another factor which is often overlooked is the effect of imbalance between the batteries. Batteries are charged by the alternator to a fixed "fully charged" voltage. This corresponds to about 2.4V per cell or 14.4V for the 6 cells of a 12V battery. Each of the cells should have the same capacity and then they will become fully charged after the same amount of current has passed. If some cells become charged before others the higher terminal voltage will reduce the voltage available to the other cells leaving them undercharged. In a single 12V battery all the cells are manufactured together at the same time and are likely to be closely matched. With two 6V batteries, although nominally the same, the batteries may differ by the full tolerance range of the production process.

A usual electronic rule of thumb is that components manufactured together will only have one tenth of the full production spreads. So your two 6V batteries could have significantly different capacities. Connected in series, the low capacity or "poor" one would charge first and prevent the higher capacity or "good" one from ever becoming fully charged. A lead acid battery degrades if it is not kept fully charged, so if you have a situation where one of a pair of batteries is not getting a full charge then that could have a damaging effect on that battery.

Batteries in parallel must have the same terminal voltage, so it doesn't matter if they have unequal capacities: they will both be fully charged when the terminal voltage reaches 14.4V. So it looks like there are good reasons to choose two 12V in parallel rather than two 6V in series.

The Kai Knickman 12V battery conversion
Kai Knickman did a 12V conversion using two 12V batteries in parallel which he wrote up in V8NOTE357. This appealed to me as, unlike the single 12V battery conversions, it involved no modifications to the battery box and it gave more power. Kai used two Bosch 4FN batteries. However, he cautioned that you had to live with the fact that the batteries in the back were designed for use in a Fiat Panda! Kai, it's even worse - my internet searches reveal that this size of battery is also used on the Piaggio Ape (the Italian "Tuk-tuk")! Ah, well; never mind. In fact the battery is used on several Fiats - for example
the Fiat Punto Mk2 1.2,
Cinquecento, Panda 1.1 and 1.2 2004 on, the Seicento 1.1 and Bravo/Brava 1.4. It is also used on the Citroen C1, C2 and C3 1.1, Peugeot 107 and 1007, and Toyota Aygo, so it should be widely available. The batteries are made by several manufacturers and the size is given variously as 002L, 202, 002FS or DIN54059 - I was surprised Kai didn't quote the latter!

Prices vary from about £30 each for the cheapest brands to around £60 for the dearest. With batteries you largely get what you pay for. The dearer batteries tend to have a three or four year warranty. I opted for Yuasa "Pro" 202s which my local motor factors sold to me at £83 for the pair - that was with a discount for two! A comparison between the original 6V and Yuasa batteries is given in the table below.


Parameter
Original
6V
YUASA
202
Variance
Size
length (mm)
width (mm)
height (mm)

175
175
200

175
175
175
height is
25mm
lower
or -8%
Volume (litres)
6.125
5.36
-12.5%
Weight (kg)
11.5
10
-9%
Capacity
Total Ah
56
70
+25%
Energy density
Wh/litre
110
157
+42%
CCA* total
270
480
+78%
Warranty (yrs)
1
3
+200%
Total cost
£124**
£83
-33%

CCA = Cold cranking Amps, roughly the current the battery can deliver for 1 minute at sub zero temperatures before its terminal voltage falls below 1.4V per cell (approximately 8.5V for a 12V battery). At low temperatures the current generating chemical reactions within the battery proceed more slowly. The CCA is the key measure for starting ability - more useful than straight capacity. Capacity is more useful, for example, in working out how long the battery would run your lights.
** Moss price, excluding carriage.

The table indicates that the two 12V Yuasas vastly outperform the original two 6V batteries arrangement showing their superior technology. What is more they do this at a cost which is a third lower and have a longer warranty. No contest really . . .


View of new batteries from the front. The isolator switch is lower left; new earth strap is upper right of the driver's side battery compartment. Original earth strap used in the nearside compartment.

Installation notes
Remember that for safety the first thing you do is remove the earth lead from the old batteries and the last thing you do is connect the earth leads to the new batteries.

Many years ago I fitted a battery isolator switch to the bulkhead in front of the driver's side battery, so my installation was a little different to that on Kai Knickmann's V8. Like Kai, I concluded that the best layout was with connectors at the back with an additional earth strap to the existing hole at the back of the driver's side battery compartment. When making that connection, remember to get down to clean metal - I used a cup type rotary wire brush in an electric drill. However, I decided to take my second positive battery lead from the isolator so this lead was to be terminated by a large crimp ring connector at the isolator end and the battery clamp at the other. Lacking a suitable crimp tool, I stripped the red heavy duty cable and inserted it in the crimp ring sleeve and then "crimped" it by using the body of a 5mm drill laid across the sleeve at right angles and then squeezing them together in a vice. You can test this by putting the ring end in the vice and seeing if you can pull the cable out. The cable is then cut to a generous length and a suitable rounded pen top is pushed over the cut end to facilitate pushing it through the corrugated conduit linking the two battery compartments. Then cut to length, strip the end and connect the battery clamp. You can use the same "crimping" technique on the earth strap if you are not using a pre-made one. A pre-made cable cannot be used for the positive link as it will not go through the cross tunnel conduit.

To fix the batteries I adopted Kai's suggestion of cutting down the J bolts and using a spacer to allow for the reduced battery height. I used sawn off lengths of 10mm dia 1.6mm aluminium tubing (from B&Q). The J bolts now fix on the front side of the batteries.

Before and after
With the original batteries and a hot engine, the solenoid would fail to switch to full power after engagement and the resulting slow crank was insufficient to start the engine. Even at lower temperatures the slow initial crank was often evident before relief at the onset of the fast crank at which the car started. With the Kai Knickman twin 12V in parallel set up, the fast crank appeared to be immediate and the car starts with no problems at all. If after a long rest, there is the usual V8 short period of cranking as the inlet manifold fills with fuel air mixture. Not having an instant start is not a problem as this helps oil pressure to build prior to fast running. When starting the car after hot from recent running, the start is almost immediate. My existing starter/solenoid is performing just fine with the new battery set-up so the need to cross this bridge with replacements is now sometime into the future.

A recent email from Kai confirms that his new 4 litre high compression engine he has fitted also starts easily even in the coldest weather. A few measurements show why there is such a big improvement. The actual steady state cranking current from the Yuasas was measured at 180A at 20C ambient with a cold engine (king lead removed to prevent ignition). At this current there was 0.4V drop in the starter feed cable. The original batteries produced an initial 8.8V at the starter under these conditions, so were clearly down from their specified CCA. The Yuasa batteries produced 10.8V at the starter. So, from an initial fully charged off load voltage of 12.8V the original batteries dropped 3.6V under starter load whereas the Yuasa batteries dropped just 1.6V - a dramatic improvement. The 23% higher voltage at the starter makes a big difference; 23% higher voltage gives 50% more power.

Conclusion
I am delighted with this conversion and would strongly recommend it. It gives much better starting performance, higher capacity, lower cost and likely longer service life and involves no fundamental modifications to the car. These modifications could easily be returned to the original set up should exact originality become an issue. It has no downside.
Copyright reserved by the V8 Register