I first found out about Fuelstar about a year ago, I had received a flyer in the mailbox that proclaimed 12% savings on fuel consumption for my car. At the time petrol prices had really started going through the roof so this was a deal that seemed too good to be true, I thought I could already see where this was going but decided to investigate further to determine if there was any validity to the claims. After and hour or so online I had satisfied myself that the only fuel savings I would get would be those that came with my wallet becoming lighter I forgot about the whole thing. Fast forward to earlier this week when a colleague pointed out a half page ad in the paper and says “You should look in to that”, well after I had finished giving him the low down and climbed back off my high horse I figured I should cover it here too.
So what is Fuelstar? Essentially it is a metal canister containing tin pellets that is fitted to your fuel line before it enters your engine. Ostensibly the idea is that the pellets release micro particles of tin that flow with the petrol into the engine and helps the fuel burn more efficiently. The following claims are made on the back of this: That your fuel consumption will go down, your power output will go up and the device will clean your engine. Now, do these claims hold up? Fuelstar would say the 180,000 happy customers says “Yes”, OK lets consider these customers. First, if someone decides to install a fuel saving device in their car it stands to reason that they have fuel economy on their mind, in which case their driving habits may change, they are more likely to pay more attention to how the car is running, whether the tyres have the correct pressure etc. If so, and they started to be more observant about their fuel consumption around the time of installing the Fuelstar device they are likely to attribute any gains they see to it.
In addition, would these customers have gotten the device fitted by an experienced mechanic?, who would likely have also given the car a tune up at the same time, which has an effect on fuel consumption. Also how many of the customers would have carefully tracked the amount of fuel used and under what conditions for significant periods of time both before and after installing the device? These are just some considerations and don’t include simple psychological effects such as expectation bias, confirmation bias or any of a host of similar congnitive biases. My point here is that no matter how many happy customers there are they are not equivalent to a rigorous test of the technology, also they are a self selected sample; only those who believe the technology will work are going to get it installed in the first place.
OK enough of this psych stuff, what about the cold hard evidence? Well, looking at the Fuelstar website there are a number of listings in the “Formal Tests” category of the Test Results page. Of these three are case study type that purportedly show a reduction in X from vehicle Y and are quite light on testing procedures and supporting documentation, interesting but hardly good evidence I don’t feel bad dismissing these out of hand. Of the three that are left one is a kind of meta-analysis that consists of a chemist associated with Fuelstar saying “It works, ok and those tests that say it doesn’t are wrong” again, no guilt about flushing that one. Another study, the most promising by the look of the description, is entirely in Taiwanese, sigh. All of this leaves us with a single test performed by a seemingly reputable facility on the USA, however although the test did compare engine performance both before and after installation of the Fuelstar device the report is ambiguous on the exact timeline of events. In particular the Pre-conditioning step included performing “restorative” maintenance on the engine at the same time the device was installed, does this mean the system that was tested with the device was different to that without the device? It is unclear from the report and I have to file this under “Inconclusive”.
Finally the EPA (Environmental Protection Agency for those not in the know) tested a substantially similar device in 1999 and concluded that “When the fuel economy data are analyzed (using the
Student’s t test) as a whole, there is no statistical difference in the fuel economy results as a function of use of the [Device]“. This along with the generally poor quality of the supporting evidence leads me to think that the claims made for the system are mostly hype and there is no reason to suppose that installing it will have any significant effect. Darn, I could have used the extra cash to pimp my ride.