Julian Edgar, 50, has been writing about car modification and automotive technology for nearly 25 years. He has owned cars with two, three, four, five, six and eight cylinders; single turbo, twin turbo, supercharged, diesel and hybrid electric drivelines. He lists his transport interests as turbocharging, aerodynamics, suspension design and human-powered vehicles.
Being less circumspect
Despite what some might say in discussion groups, everything I ever write for AutoSpeed is carefully considered and thought-through, whatever the consequences of its publication.
So, for example, I was prepared to write content in Part 2 of our water injection series that showed, despite the good hardware, the results in some ways were disappointing.
“What a let-down!” said one discussion group poster.
That may well be so, but I don’t ever want to be in the position of claiming some modification benefit that others cannot duplicate.
In the same way, for our stories on advancing the ignition timing by tweaking the intake air temp sensor (see The 5 Cent Modification) modifying the action of the EGR valve (see Part 1 and Part 2), I was deliberately understated in my description of the potential benefits.
“There’s clearly improved driveability and potentially a little better urban fuel economy,” I said of the first modification.
“…in carefully tested urban conditions, there was a small but measurable improvement of 3 per cent [in fuel economy],” I said of the EGR changes.
“[But] in highway conditions there was no measurable change.”
However, let me be less circumspect in this blog.
I think that the two modifications are absolutely bloody brilliant in fuel economy and driveability outcomes, and certainly have an impact on my car, and in my driving, that is better than I described in the articles.
I also have a gut feeling that they will work even more impressively on other cars…
So, for those who haven’t read those articles, what were the mods?
In short, the first involved tricking the ECU into thinking that the intake air temp is always about 30 degrees C colder than it really is, so causing it to run advanced ignition timing. To compensate for this, I run the car on 98 RON fuel. This combination gives better power, especially at light loads.
The second mod involves tweaking the feedback signal from the EGR valve, so causing the system to run a lot more exhaust gas recirculation than normally occurs. This reduces pumping losses, improving power at low loads and small throttle openings.
The result on my hybrid Honda Insight is exceptionally good – even if the bill had been a thousand dollars (not less than 50 cents!), I would still class it as worthwhile.
Today, I added a very bright LED on the dashboard. It shines whenever EGR is occurring.
Now it gets a bit complicated, but the Honda is working most of the time in three different modes.
1. The first mode is lean burn, where the air/fuel ratio is about 25:1 – exceptionally lean. EGR doesn’t operate in these conditions. This mode occurs after a short delay in constant throttle conditions.
2. The second mode is part-throttle EGR, where EGR is reducing pumping losses and the air/fuel ratio is stoichiometric (ie about 14.7:1). This occurs in light throttle acceleration (amongst other conditions).
3. And the final mode is where the air/fuel ratio is stoichiometric, but EGR isn’t working. This occurs at larger throttle angles than #2.
(There’s also another high load mode, where the air/fuel ratios go much richer.)
In order of best to worst fuel economy, the results largely follow the sequence of these modes. By adding the ‘EGR-working’ LED I could stay more often in mode #2 and less in mode #3.
Today I did a drive that is absolutely typical of any outer suburban Australian city. Some free-flowing urban, some congested stop/start urban, some freeway (then traffic jam on freeway for perhaps 10 minutes), up and down a steep country road hill. Five stops, and all the time keeping up with the traffic. However, the air con wasn’t on (it wasn’t needed) and I deliberately didn’t ever accelerate really hard – so, I was drifting up to the speed limit rather than getting there fast.
I drove in part according to my new dashboard LED, keeping it in mode #2 more often than I would otherwise have done – taking advantage of the new-found torque and low rpm driveability. In fact, when there was no need to accelerate hard, I was changing at 1500 rpm – no big deal for a large engine, but pretty stunning in a naturally aspirated 1 litre!
Fuel economy for the 85-odd kilometres was 3.6 litres/100km – at least 0.2 litres/100km (or 5 per cent) better than I would have normally, pre modification, achieved on this sort of drive.
For imperial non-metric aficionados, that’s 78.5 mpg, and in US gallons that’s 65.4 US mpg.
And that’s a for a real world drive, doing 100 km/h on the freeway, climbing my long steep country road hill on the way home (in fact, shaking off a Mercedes through the winding bit of this road – he was too close behind me!), and being stuck in traffic.
Put it this way: I am confident that any competent driver could achieve very close to this fuel economy figure in this car (no pulse and glide, no dawdling away from the traffic lights, no rolling in neutral gear, etc).
And this improved economy came with better than standard driveability – the car now drifts along beautifully on tiny throttle openings in tall gears…
Would I clamour these gains from the rooftops for all cars? Nope – and in the ‘proper’ AutoSpeed articles, I didn’t.
But do I think anyone interested in fuel economy should look closely at these modifications? Yep, sure do!
on June 5th, 2008 at 6:37 pm
I for one am very interested in fuel economy, and as you suggest have been very interested in the above articles.
An interesting thought occured to me reading this blog. The 50c modification is essentially being done automatically on some later model cars, when the computer progressively searches for the most advanced timing possible and on these cars, you can get the benefit without cutting wires or even pulling things apart. Just run a couple of tanks of 98 octane fuel through and let the computer do the work. If you achieve a measurable increased fuel economy, great. Importantly if your fuel consumption is unchanged, there is a great chance the 50c mod will work!
on June 5th, 2008 at 9:34 pm
Hmmm, I wonder why Honda didn’t use so much ignition timing/EGR on the original car?
Both seam like easy things to change if your tuning the factory ECU.
Have you had an emissions test? I wonder if it is much affected. In other countries (EU) they have emission tests every year. Lots of aftermarket mods like tricking the O2 sensors and the “5cent resistor mod” might get better fuel consumption but ruin emissions. NOx is a much worse greenhouse gas than CO2.
on June 6th, 2008 at 12:41 pm
It would be interesting to see these mods performed on cars that are not engineered from the ground up to be so frugal. If these mods were tested on non hybrid cars, eg. Falcons, Commodores, Lasers, Corollas, or other “mainstream” cars, would the results be more pronounced? Julian, have you tried tweaking the EGR on your Lexus?
on June 6th, 2008 at 2:04 pm
Wayne, the ECU will adjust ignition timing only within the constraints of the expected fuel octane.
Henry, I run my car on higher octane fuel than specified. I also tuned the EGR very much to suit my individual car – no generic factory setting could do that. NOx is likely to be lower than standard with the increased EGR, and probably a little higher than standard with the increased ignition timing. I don’t have access to a proper drive cycle emissions test but I’d be pretty confident it would pass in current form. Certainly, on the basis of my previous experience, it would pass local roadside sniffer tests.
Nick, I don’t own a Lexus any more – you have probably been reading repeated articles.
on June 6th, 2008 at 4:26 pm
I kind of agree with Nick above. How much more can you honestly expect to squeeze out of the Insight? You could go to a lot of trouble for 0.1 L/100, but that might translate to 1-2L/100 on any other car.
To be honest, after all the build-up, I found the water injection series a bit light-on. It would have been interesting to try the water injection in conjunction with lower-octane petrol, for a $$$ rather than absolute consumption saving. Especially if the factory ECU only has a certain amount of extra advance available to it. I’m exploring this avenue myself, although adjusting the timing on my car costs less than 5c 😉
on June 6th, 2008 at 4:51 pm
I would have thought that a modification that gained ANY fuel economy improvement out of the Insight was a bloody stunning modification – and I have now done it with three separate mods.
Of these modifications, none (to my knowledge) have ever been done on that model of car, and two (to my knowledge) have never been covered in any aftermarket magazine in the world – being performed on ANY car.
But each to their own – if you think they’d be more impressive on a Commodore, then that’s fine.
Re the water injection – again, use the presented technology on any car you want. The point of the stories is the presented technologies! Again, if you can find a better DIY water injection system being presented in any aftermarket modification publication in the world, please show me the detail.
Those suggesting these mods should be done on other cars – get on with it! None of the mods costs much, all have been shown in detail, and all are straightforward to perform.
on June 7th, 2008 at 4:45 pm
Stock factory delivered cars will allmost never come with absolutely optimal tuneing in regards to power/fuel econemy, they must leave a slight margin for error and for the many circumstances the car will face, by tweaking a car as an individual machine gains should be expected.
on June 7th, 2008 at 9:52 pm
I was always under the impression that having more advanced ignition timing, without detonation, resulted in lower exhaust gas temperatures and so should result in lower NOx.
on June 8th, 2008 at 11:53 am
Exhaust gas temperatures will go down with increased ignition advance, however these temperatures are generally to low to effect N0x emmisions.
As at low temperatures N2 is a fairly stable it only starts to become part of the chemichal reaction of combustion above 1800-2000 Degress C, which is more in the area of peak cylinder temps, and therefore within in certain ranges will increase with more advanced ignition timing, resulting in higher N0x emsiions
With large diesel engines (i.e 25-100Megawatts) we have allready reached the point where the most efficient running of the engine cannot be used and a tune with better exhaust emisions is used instead, which uses more fuel.
on June 10th, 2008 at 8:54 pm
I think what u achevied is brilliant!
Congrats.
Well all drive around trying to find the cheapest fuel, so why not 50c teaks?
Anyway J i will supply u my 03/08 as in almost new LPG Falcon ute and please perform these changes on it.
I am 100% focused on more MPG
Cheers
M
on June 10th, 2008 at 9:28 pm
Hey Julian, Some interesting info for some readers.
With the aid of a ” Launch ” OBDII Reader, I found just how effective the 5 cent mod is on VY2 V6 Commodore (with the higher octane fuel LPG). It did achieve the change in the intake air temperature the computer thought it was injesting, however no more than 10 minutes into driving, the computer had learnt its way around it as you suggested may happen with some cars. Taking it off ( i took the time to get an old plug and socket and soldered a resistor between ) again taking 10 mins to correct itself. At all points in time (before fitting, while on, after removed, after self correcting), the ignition advance was continually being advanced until a knock condition was sensed. It was sensed at differant times on each of the 6 cylinders. Under Light throttle ( eg decelerating at part throttle ) up to 40 degrees was indicted. Under light throttle accelerating, mid 30 to low 20 degrees were indicated.
low to mid 30 at cruise depending on the speed. Incidentally at 80km/h up a 10% gradient, I noticed the EGR was more than 90% open, being in open loop, the AFR was aparently 16.8:1 at that time.
Cheers
on June 12th, 2008 at 11:18 am
http://www.urbandictionary.com/define.php?term=hypermiling&defid=2350578
and now there is a name for it!
on August 11th, 2008 at 9:54 pm
Hi Julian, I have the little 5.7 LS1 ( greedy people have them in size 6 ) in a Commodore and an Efilive tuning package. This lets me play with lean cruise and to add in timing specific to the AFR selected. My problem is what Air to Fuel Ratio is best for economy?
I am building an instantaneous MPG meter at the moment to help with the development of the ‘tune’.
Holden run these things at up to 17.6:1 in some models.
Anyone care to place a bet?
Cheers, Steve