Attention! Alternatorless modders. Get your voltage up

[orange4boy]

Bringing voltage up from 12 to 18 made a large difference in FE on my alternatorless Previa.

I have finally had a chance to try running the Golden Egg with 18V. I have been given a set of used Trojan T-105s to play around with.

Disclaimer: 18V is actually too high so don't try this yourselves as I don't want to be held responsible for frying anyone's ECU. Cars are designed to operate in over-voltage conditions in the event of a regulator failure so I risked it. First casualty: the headlights. Everything else seems fine though.

The first results are impressive but I have not done a full tank yet, this comes only from MPGuino relative readings. I would love it if someone who commutes could try this. My observations:

1) Idle speed GPH while idling has gone from about .75 to .29. I think this is because the ECU was keeping the revs up to charge the battery as it was showing only 11.5 - 12.5V. CPU is probably programmed to keep 13.5-14.5

2) MPG while coasting/idling at highway speed. This has gone up from about 100 to 160

3) Highway cruise mpg has gone up too. roughly from 30 to 35. (more testing to be done here)

4) Idle is smoother and throttle response is better.

I always thought that the MPG seems to slowly drop with battery voltage but I didn't think it made such a big difference. What's exciting about this is that there are greater gains to be made from alternator removal that I had first thought.

The best way to go here would be either 2 8V batteries in series or a couple of 25A 12V-16V DC to DC up-converters in parallel and a couple of 12V deep cycle batts to run them.

Has anyone else tried running alternatorless at higher voltages? I can't do A-B-A tests myself but would love to see this done.

[dcb]

Originally Posted by orange4boy

The first results are impressive but I have not done a full tank yet, this comes only from MPGuino relative readings. I would love it if someone who commutes could try this. My observations:

Interesting, I would definately wait for confirmation with a few tankfuls though, because it occurs to me that it is possible that your injectors are opening faster and that injdelayUS might need to be tweaked downward to compensate.

[orange4boy]

I was wondering about that. What is the ratio between the delay and the open time during one cycle? In other words how much difference would, say, a 10% change in delay affect gph or gpmicrosecond? Sorry, I don't remember what the right term is.

[Ryland]

Your car is not designed to run at 12 volts, it's designed to run between 13.2 to 14.5 volts when the engine is running, so it's hard to say what the voltage of your 18 volt battery is after all the connections and discharge of the batteries but I would say that your lights and ECU are getting close to their proper voltage, a tad high but not by much.

[orange4boy]

so it's hard to say what the voltage of your 18 volt battery is after all the connections and discharge of the batteries

High enough to fry my headlights in 1 second. The fully charged resting voltage of three 6 volters is about 19V. I was thinking I could use the DRL headlight resistor to lower the voltage as a cheap fix just for now.
The relays on the dash run pretty hot too.

[blueflame]

Is the measuring device, the mpguino, running at the higher voltage too?

It may read incorrectly at 19v?

In so, tank readings or lower mpguino voltage would be needed for accuracy?

When I fitted my engine cutout switch on my shifter, I used too small a size cable and had a 1.5 volt drop to my coil, and got bad FE. I upsized the wire in the end.

[dcb]

Originally Posted by blueflame

Is the measuring device, the mpguino, running at the higher voltage too?

It may read incorrectly at 19v?

Shouldn't matter, it has an onboard voltage regulator and runs @ 5 volts regardless. But applying more voltage to the injectors *should* make them open faster, which could throw things off.

[Coyote X]

From the Megasquirt manual

# Injector Opening Time (ms) (InjOpen) is the amount of time required for the injector to go from a fully closed state to a fully opened state when a 13.2 volt signal is applied. Since fuel injectors are electro-mechanical devices with mass, they have latency between the time a signal is applied and the time they are in steady-state spraying mode. Typically, this value is close to 1.0 milliseconds. Note that the closing time should theoretically be subtracted from the opening time, but the closing time is generally very small compared to the opening time.
# Battery Voltage Correction (ms/V) (BatFac) is the number of milliseconds that MegaSquirt-II adds to each fuel injection pulse to compensate for the slower opening of the injectors with lower supply voltages. Generally 0.10 ms/V or 0.2 ms/V is about right. So, with a 0.20 Battery voltage correction factor and supply voltage of 14.5 Volts, a 1.0 millisecond 'opening time' is adjusted to 1.0 - (14.5-13.2)*0.20 = 1.0 - 0.26 = 0.7 milliseconds.

Conversely, If you had injector specs that stated the opening time was 0.7 milliseconds at 14.5 Volts, and your battery correction was 0.20, then you should enter 1.0 as the 'Injector Opening Time'.

[dcb]

ayup, thanks for the confirmation.

[SVOboy]

How much weight is added when going to 18 volts?