Honda Insight VTi

Time to settle the question for good; is Honda's hybrid hatch as thirsty as our February economy comparison indicated?

Honda Insight VTi
Economy Run

The Honda Insight didn't cover itself in glory the last occasion the Carsales Network drove the petrol/electric hatch.

At the time we were puzzled by the Honda's average fuel consumption of 14.5L/100km for an economy comparison test. But our bemusement was as nothing compared with Honda's collective concern. How could a 1.3-litre petrol/electric hybrid hatch -- weighing just 1205kg -- return such a figure?

Honda's National Public Relations Manager, Mark Higgins, told the Carsales Network that he thought he could have achieved that sort of figure "driving a [Mercedes-Benz] C 63"...

Honda has subsequently arranged its own economy run for the Insight, circumnavigating Tasmania on one tank of fuel. Fuel consumed for that run totalled 39.6 litres and, based on the 821km distance travelled, the average consumption rate was 4.82L/100km.

So it seemed time to grant the Insight a rematch.

The method
A day was set aside for the Insight to be tested in accordance with the original parameters of the comparo, although weather and traffic conditions would vary of course. As with the comparison, the Insight's economy run would be conducted in an environment as close to a 'combined-cycle' test as we could find, running over a series of loops through Melbourne's inner suburbs, along freeway, arterial road and through high-density shopping strips.

How serious would this test be? Ten laps of the same drive program, each one measuring 12.3km. Five of the laps would be driven in a style that was a mix of economical and 'normal', with the aim to keep up with traffic and not hold up anyone behind. The other five laps would be driven in a way that did the Insight no favours whatsoever -- turn off any fuel-saving feature and drive it as hard as possible within the allowances left by the traffic. It was hoped that these two methods would bracket the 90th percentile of driving styles and provide a realistic score.

At the end of the ten laps, the Insight would return to the service station where it was initially filled, to be refilled from the same pump, to two clicks. The car's tripmeter reading was to be manually divided by the number of litres pumped into the tank on the refill and the subtotal divided by 100.

The driver would also monitor the car's average consumption readout from the trip computer and keep notes of how the average figure varied at a milestone point on each lap. This waypoint was the exit from the Eastern Freeway to Hoddle Street.

The route
We settled on using the same drive loop worked out for the comparison back in February. Starting off from Carsales' HQ in the Melbourne suburb of Hawthorn, the Insight motored up Power Street to Kew Junction, from where the driver veered left onto Princess Street, left again at the Chandler Highway and immediately left onto the Eastern Freeway.

Exiting the freeway, the Insight turned onto Hoddle Street, a major arterial thoroughfare in each direction, with traffic lights every few hundred metres. From Hoddle Street, the Insight turned into trendy, boutique-store-lined Bridge Road, which is congested, tram-infested and riddled with 40km/h speed limit zones for most of its length. Where the trams finally turn off, Bridge Road becomes Burwood Road, which takes us back to the start of the loop at the Power Street intersection.

Total length of the route: 12.3km.

The preliminaries
The Insight was provided by Honda on the usual seven-day test basis and the test was scheduled for the day before the car's return. This provided a chance to get a feel for the Insight and establish benchmarks. The tripmeter readouts were reset from the moment we picked up the car from Honda's head office facility at Tullamarine, in Melbourne's north.

On the drive from Tullamarine to the Carsales office in Hawthorn, the Insight's trip computer showed the average fuel consumption tumble from the first observed figure of 7.0L/100km to 4.6L/100km, but that included an initial cruise down the Tullamarine Freeway and some gentler driving on arterial roads through the inner suburbs.

In total, the Insight covered about 290km (of which 120km was freeway driving), leading up to the test. Average fuel consumption, according to the trip computer, was 5.4L/100km. Driving style was 'normal' -- which meant occasionally using more throttle than necessary. Usually the car was doing just enough to keep up with traffic or cruising at open-road speeds.

The test
On the day of the economy test, the Insight's tank was filled to the second click with 91 RON unleaded -- 15.38 litres in all. From the reset, the trip computer initially jumped to 14.8L/100km before dropping back to as low as 3.2L/100km on a downhill section of arterial road on the way to the start of the economy run loops planned for the test.

The test began at around 11:00 am on the Anzac Day public holiday. Traffic was lighter than that encountered during the original comparison, although it gathered mass as AFL followers wended their way to the MCG along Hoddle Street for the Geelong v. Hawthorn game on the day. Two SUVs in a nose-to-tail on Hoddle Street added to the chaos around lunchtime, but traffic was much lighter than during peak hour on a typical work day. It was still heavier than during the comparison test however.

Other hot spots around Bridge Road and Kew Junction progressively worsened closer to lunch, but cleared considerably after 2:30 pm, by which time eight of the ten planned laps had been completed. Trams and holiday traffic on Bridge Road held back the average speed and reduced the prospects of getting through each intersection without stopping for a red light. Power Street ran up a gentle but constant grade. However, Princess Street, on the other side of Kew Junction, descended steeply to offset that, typically knocking 0.1L off the Insight's average consumption.

From Princess Street, the drive program took the Insight along the Chandler Highway and on to the Eastern Freeway for a 3.0km run along the slightly downhill incline, knocking another 0.1L off the average consumption figure.

After completing five laps driving economically, but not unsafely, it was time to complete five laps as uneconomically as possible. This was achieved through turning off the car's Econ mode and slotting the transmission lever into S, rather than D (and thus disabling Auto-Stop). To really play mischief with the car's fuel efficiency, the driver used the Insight's sequential shift paddles to change the CVT's programmed steps manually and applied full throttle wherever possible.

The difference was immediately apparent on the next checkpoint (at the Eastern Freeway exit to Hoddle Street), where the average had 'slumped' from 4.8 to 5.4L/100km. For each subsequent 12.3km lap, the average continued to worsen, before settling at a plateau of 6.4L/100km on the final (10th) lap. By that point, traffic was improving considerably and the Insight's posted fuel consumption average was unlikely to vary much beyond that. If anything, it might have begun to improve again.

Returning to the same service station, the driver refilled the Insight. The fuel pump clicked a second time at 8.86 litres. Based on the distance travelled for the test -- 136.5km -- the average fuel consumption figure for the test was 6.49L/100km. That was pretty close to the trip computer's figure of 6.4L/100km; the trip computer readout only providing a figure significant to one decimal place. Total travel time for the test was four hours, 38 minutes, for an average speed of 30km/h.

The conclusions?
The Insight's trip computer was pretty accurate, based on the final fill and assuming the tripmeter readout was also on the money. The trip computer's average had briefly hit 6.5L/100km before finishing at 6.4, so it was plainly on the cusp of the two figures.

There was a difference of 1.4 litres per hundred km in consumption between driving carefully and carelessly (without Auto-Stop operational) in similar traffic environments (4.8 versus 6.2L/100km) on different 'laps'. If there's a happy hunting ground for the Insight, it seems to be on flowing, arterial roads, where an average speed of up to 80km/h can be maintained. It's possible to sneak in under 5.0L/100km on freeways too, but the added torque from the Insight's electric motor seems to provide more benefit once the CVT ratio is in a higher band, but aerodynamics and weight play a lesser role.

There is one 'easy' way to achieve an SUV-like figure of 14.5L/100km average consumption in the Insight. Leave it creep forward in Drive for 100km -- staying in the lowest ratio of the CVT and letting the engine run at idle. But consider how long it would take to complete a trip of 100km using that sort of driving style… got a spare five to 10 hours?

Other than that, there appears to be no simple, easily repeatable 'method' for using large quantities of fuel in the Insight. No matter how badly the Insight was driven during this test, its average fuel consumption never exceeded 7.0L/100km -- less than half the figure we recorded back in February.

The data
The trip computer's average consumption figure was noted at the same waypoint on each lap, each of the first five ('gentle') laps showing a gradual upward trend as traffic congestion built up, but the average remained below 5.0L/100km. Subsequently, the Insight was driven with less care and the auto-stop disabled. On this lap ('6'), the average consumption jumped, but remained lower than it might have been, due to the previous five laps skewing the average. The average trended upward by increments of 0.4L for the first three laps of harder driving, but then began to plateau as the traffic thinned out again. If the Insight had been driven hard for the full ten laps, it might well have returned a figure in excess of 7.0L/100km -- but a figure as high as 14.5L/100km, such as our comparison testers saw in February would have been practically impossible.

An afterword
We couldn't explain why the Insight performed so poorly in the first test, in what should have been its element. It was the only vehicle of the four tested that offered auto-stop and should certainly have produced a much better result than it did.

Various theories have been floated. First was the driving style of the volunteers behind the wheel. At the time we mentioned that the Insight was enjoyable to drive and, once drivers were accustomed to the continuously variable transmission (CVT), they were more than ready to use the car's available performance in a straight line -- to a fault. Our camera crew in a diesel HiLux had difficulty keeping up with the hard-charging Insight along Melbourne's Hoddle Street and Eastern Freeway. But that in itself couldn't explain the unusually high measured consumption.

Other suggestions included a vapour lock in the tank during the preliminary fill or the prevailing weather conditions -- very hot on the day. There's even the possibility of miscalculation, although the figures were checked and double-checked. Nothing really explains the discrepancy, but we're happy to observe that this time around the Insight has performed to specification.

Read the latest Carsales Network news and reviews on your mobile, iPhone or PDA at the carsales mobile site