independent kiwi spirit of invention.
Research Topic: Fuelstar tin-based fuel combustion catalyst
By Ian Mander BSc, 31 May 2008.
Updated 3 June 2008, minor changes 5 October
2008, updated 3 March 2009, updated 21 April 2010, updated 25 April 2012, updated 15 October 2012 (quote from Jonathan Holmes), updated 3 May 2014, updated 1 April 2015 (emphasised the damage that might occur, especially to diesel engines).
Question: Does the Fuelstar (or Fuelmate, or Broquet, or Carbonflo) tin pellet fuel combustion
catalyst really improve fuel economy by 10-15%, engine power by 5%, and
do that for 5 years or 500,000 km or 12,000 hours?
Answer: No! At best, it probably does absolutely nothing except
lighten the wallets of gullible people, while encouraging them to maintain
their vehicles better and making them more aware of their wasteful driving
habits. At worst, it's a scam with no basis in scientific reality that could conceivably destroy your engine.
Update 25 April 2012. Since my last update of this article, con man Ian Cornelius has mostly ceased operations in New Zealand. His Fuelstar scam products are no longer marketed for cars, and only available for rent for trucks etc.
Update 3 May 2014. Dr Jim Sprott has died, aged 89. It is hoped he will be chiefly remembered for his good work, including demonstrating the innocence of Arthur Allan Thomas; as this NZ Herald article mentions, in 1995 he was made an OBE for services to forensic science and the community.
The main way of marketing the Fuelstar devices is through personal
testimonies that they work. This is known as anecdotal evidence, and
is subjective, not objective like a laboratory test would be.
A close investigation shows that most of the anecdotal accounts have
little to offer. One diesel 4WD owner claimed saving a 27% of his fuel
bills after fitting a Fuelstar device, but on closer investigation it
turned out that most of that saving actually came from removing a roof-mounted
tent from his vehicle! The rest of the apparent improvement could easily
be attributed to maintaining his vehicle better, such as a (by the sound
of it) much needed oil change.
Another example of anecdotal evidence is Fuelstar's claim that its
product is ideal for preventing valve seat regression. For evidence
it presents a single engine which was fine after three years of use.
But according to this
article "FuelStar ... failed to meet Australian government
standards for the prevention of valve seat recession under independent
tests similar to the one described above for Valvemaster (which passed)."
One of the problems with anecdotal evidence is it sounds good but even
if there's a real improvement it might be just bringing the vehicle
back to normal and is quite likely due to other causes. Or an apparently
real protective effect might be due to something else entirely – the
single engine Fuelstar tested may have had hardened valve seats that didn't need protecting
anyway. Without an objective test with a control there's no way to tell,
which makes these anecdotal tests completely useless.
Another happy customer was absolutely certain of improved economy,
but he destroyed any weight his evidence might have by admitting "I
haven't taken a calculator to it yet". This customer might have
convinced himself he didn't waste his money, but his testimony is about
as useful as someone being absolutely sure the Sun rises in the west
"without having taken a compass to it."
The expectation of an improvement could itself lead to a strong impression
of an actual improvement. Without careful record keeping it's basically
meaningless. Conversely, a lack of good record keeping can keep people
from speaking out against it, along with the problem that people don't
want to admit they were gullible enough to spend hundreds of dollars.
As far as evidence for Fuelstar working, these anecdotal claims mean
nothing except to those who already want to believe the thing works.
Remember that a sincere believer can still be sincerely wrong.
Conventional before fringe
Much improvement can be obtained from a combination of:
An engine tune.
A new air filter.
A new fuel filter.
An oil change and new oil filter.
Increased tyre pressures.
New spark plugs (for petrol engines).
New injectors (more expensive than the other items).
All these except the last item should be done as regular maintenance
anyway. Do these first before trying any fringe products. In one of
Fuelstar's favourite tests, California Environmental Engineering did
these things and achieved a fuel saving of 27% (although I note that
no units are given for the fuel used in the test runs, so it could be
an increase of 27%).
"Restorative maintenance was performed
to include changing the oil, and the oil, air and fuel filters. ...
The unit was installed vertically on the engine side of all filters.
The fuel line between the Fuelstar unit and the engine was appropriately
replaced with a new comparable line." All the test shows is
that on the right vehicle, a bit of maintenance goes a long way. The restorative maintenance restored the vehicle's fuel efficiency to normal.
The tests that Fuelstar so proudly shows off lack credibility and are
of no more value than the anecdotal testimonies of their misled customers.
Amount of tin
The amount of tin that would be included with every fuel injection
into an engine cylinder is ridiculously small. How could such a small
amount of tin provide a catalytic effect?
Fuelstar claims its devices last 500,000 km, and improving fuel economy
a "Typical ... 8% to 12%" (or in South Africa always more than 10% and usually 15%), and increasing power by 5%. It has formerly had a guaranteed 10% fuel saving, down from an earlier guaranteed figure of 12%. The implicit
claim is that the devices will still be saving 10-15% of fuel (down from a claim
of 20%, incidentally) at the end of 500,000 km, which means there
must still be enough tin left in the thing to supply it to the fuel.
I've seen only two possible (not definite) figures for the amount of
tin in a Fuelstar car-appropriate device: 200 grams and 100 grams. Let's
run with the 200 gram figure and to assure an even supply of tin throughout
its lifetime assume there would still be 50% left after 500,000 km,
meaning 100 g of tin has been used over the 500,000 km. Average tin
economy is thus 5,000 km/g.
If the car it's fitted to uses fuel at an
average of 10 L/100 km then the fuel used per 5,000 km would be 500
L. At petrol's density of 737.22 g/L, 500 L is 369,000 g of petrol.
1 g tin per 369,000 g fuel means a catalyst comprising just
0.00027% of the tinned fuel is credited with a guaranteed 10%
(formerly 12%) fuel saving. That's a huge claimed benefit for
a really really small amount of substance. (I once described it as a
homeopathic amount, but that's not strictly accurate on average.)
This is not in any way reasonable and is a very good indication
that it is a con.
Compare the figure to that of Valvemaster (needed by some cars to avoid
valve seat recession after leaded fuel was taken off the market), which
had a concentration of a chemically significant 0.1% of the fuel. The
chemistry behind Valvemaster was also clear.
Update 3 March 2009, 25 April 2012: In response to a couple of emails I calculated the
volume (instead of mass) of tin and petrol in each injection. Skipping
the calculation details, because tin is about ten times as dense as
petrol the result came out as roughly ten times the above figure, or
a ratio of 1:3,700,000. In each injection there will be about 0.000 000 009 cm3 tin to 0.033 mL petrol. (Note that 1 cm3 = 1 mL.)
is no way this is reasonable. That volume of tin simply doesn't
have enough surface area to catalyse several million times the volume
of petrol for any measurable fuel saving (and especially not in less than a hundredth of a second). Whatever the tin particle sizes, each particle of tin is required to catalyse a volume of petrol with a diameter 154 times its own diameter (and much more if the particles are not perfectly evenly spaced; it's very unlikely they would be). That ratio is equivalent to a tennis ball catalysing a whole 25 metre swimming pool, and in just one hundredth of a second.
Since the tin is being consumed (and possibly converted to tin oxide
and other compounds) it could be argued that the tin cannot be viewed
as a catalyst, because a catalyst is not consumed in a reaction.
However, Dr Jim Sprott has claimed that tin suboxide (chemical symbol
SnO) is the real catalyst, not tin metal. I wanted to find out more
about tin suboxide and its role as a hydrocarbon catalyst, so naturally
enough I tried "tin suboxide" catalyst (with the quotes)
in Google. I got just 8 hits. Yes, eight. Compare that to the 102 results
for fuelstar scam. (Update 25 April 2012: "tin suboxide" catalyst has 50 results, the first one is this page; fuelstar scam has 4,850 results, this page is 3rd.)
Of the 8 original results, two were forums quoting a report by Dr Sprott (promoted as being "without
doubt, the foremost authority in New Zealand in fuel oxidation chemistry"),
one was the report itself, one was a misleading advertising complaint
regarding Fuelstar's emissions claims, one was simply a mention in a
very long list of chemical names as an appendix to a long paper on solar
absorbers (ie, nothing to do with hydrocarbons), one was a technical
paper mentioning how tin suboxide is a p-type semiconductor, etc.
None of them were explanations or references to explanations
of how tin suboxide might act as a hydrocarbon combustion catalyst.
This startling lack of information is a good indication that the
catalytic effect is fictitious.
Fuelmate's explanation for their own version of the catalytic devices
goes further and claims that after several thousand kilometres of using
their devices the engine is further enhanced by a layer of tin oxide
(not "suboxide", although the formula is the same) coating
the combustion surfaces. The problem with that is that combustion occurs
in the middle of the cylinder, not on the edges, and yet somehow, magically,
"This condition creates an even more controlled burn rate across
the combustion chamber." How? Catalysts do not affect reactants
at a distance by some sort of magical conduction effect, so how can
a catalyst on the chamber walls affect the burn rate in the centre?
Furthermore, SnO (tin(II) oxide) is unstable at high temperatures. Perhaps
they meant tin dioxide (SnO2), or perhaps they don't know
what they're talking about.
The lack of scientifically sustainable explanations is a good indication
the products are bunkum.
Fuel filters and the size
One reason given by Dr Jim Sprott that the AA-commissioned test was
invalid is that there was a fuel filter between the Fuelstar device
and the engine. In his report he says "there must be no physical
impediment between the Fuelstar unit and the engine. Consequently there
must not be a fuel filter in the line between the Fuelstar unit and
the [engine]." (He actually starts talking about the tubing there,
instead of mentioning the engine.)
How much tin will there be in the fuel and would a fuel filter block
it? At a steady 100 km/h it would take 50 hours at about 2,500 rpm
to travel the estimated 5,000 km to use a single gram of tin. With
a 4 cylinder 4 stroke engine that's 2,500 x 60 x 50 x 2 injections. So 1
g of tin is spread over 15 million injections, and each injection has
0.000 000 07 g of tiny tin particles spread through it. So
with the required low concentration of tin and at the size of tin particle
which would have to be involved (the 0.000 000 07 g of tin
would have to be reasonably evenly distributed through the fuel injected),
it seems quite unreasonable that any normal fuel filter would be fine
enough to stop any of the tin particles, even if all the tin for each
injection was in a single lump. The fuel filter in the AA test would
have had no noticeable affect.
Also, we have Fuelstar itself contradicting their much touted expert
regarding the placement of the fuel filter with common rail diesel engines.
Their web site recommends the Fuelstar device is installed before
the fuel filter, not after as Dr Jim Sprott says. (25 April 2012 – this has been removed from their web site.) It seems to me a common rail diesel engine would sustain
the most damage if large particles of tin did enter the fuel line. A
cynical view of this is that it seems like a very sensible way to cover
themselves if their non-effective product malfunctions and releases
a comparatively large chunk of tin into the fuel line.
Fuelstar is also clearly mistaken or disingenuous when they say the
tin particles are visible with a microscope, as they would be so fine
that an electron microscope would be required to see them. Dr Sprott
points out that the particles are less than 100 nanometres across –
much less than the wavelength of visible light, and thus invisible under
any visual magnification.
The contradictions between Fuelstar and its expert is a good indication
that they don't know what the real situation is, or that they're trying
to hide the real situation from their customers.
Update 21 April 2010: Fuelstar apparently no longer says they can
be seen with a microscope. This incidentally just makes it harder
to confirm that their devices are actually working, and releasing any
tin. Not only can the tin not be seen, but if they work as claimed there
would be so little tin in the fuel (just 0.000 000 07 g of
tin in each injection) that using an electron microscope to find it
would be a complete waste of time. A spectroscope would probably be
needed instead. Where are Fuelstar's lab reports that there is actually
tin in the fuel?
Update 25 April 2012: Scam company Wizleck Fuel Technologies which sells Fuelstar in South Africa says on their web site "These particles are invisible to the naked eye but can be readily seen in the test laboratory using a microscope." What size would they be? As mentioned above, each injection has 0.000 000 009 cm3 of tin. The volume of a sphere is 4/3πr3 so if the tin was just one roughly spherical particle it would be 0.0026 cm or 26 µm across. That could be readily seen with a microscope. While experienced observers using good optical microscopes with oil immersion can resolve down to 200nm, it's not quite true that objects that size "can be readily seen". If the particles are readily seen it would be reasonable they are at least 400 nm, or even 500 nm across. That's 4 or 5 times the size Dr Jim Sprott said they are. (Incidentally, those particle sizes would mean a maximum of 270,000 or 138,000 particles of tin in each injection – still useless for catalysing.)
The price of tin
The current price of tin is US$21,055 per tonne (as per the NZ Herald
this morning ). If a typical Fuelstar device for a car contains 200 grams
of tin that would be NZ$5.40 tin. The actual price of the unit is perhaps
40 or 50 times as much.
The price of the device being so much more than the price of tin
is a good indication that it is a con.
Too good to be true
There's an old adage that if something is too good to be true, it probably
is. Fuelstaris claimed to work – and work well – on any engine
size, whether small or large; on any fuel type, whether petrol, diesel,
or LPG; with any carburetor or fuel injection system on a turbo or
naturally aspirated engine; and in any application, whether large diesel
ship, locomotive or truck, or LPG van, turbo petrol car, motorcycle,
or small two stroke petrol generator.
The claims of such universal good performance over such a wide range
of applications is a strong warning sign that it is a con.
Not in cars
Dr Sprott: The use of tin as a combustion catalyst has been known
for about 70 years.
So the car manufacturers have certainly had enough time to test it
and figure out if it works or if it's quackery. What have they concluded?
There is no car manufacturer which includes a tin catalyst in
their vehicles as standard equipment.
There are no car manufacturers which offer it as an option.
I have still not heard any pro-Fuelstar explanations why this should
be, if the Fuelstar products work as claimed. Don't make the mistake of thinking the car manufacturers
wouldn't want to know about it. Car companies are so keen on improving
fuel economy they're even going to the point of using expensive systems
to automatically turn off the engine while the vehicle is stationary
– just to boost fuel efficiency slightly. Because tin catalyst devices are less
expensive than those complicated start-stop systems, even if supposed tin catalyst
products only offered a similar predictable 2% improvement it would be used
in at least some models, and probably most models since it's so much simpler. That it isn't used in any production car is a very serious
problem with Fuelstar actually working.
This lack of industry adoption – especially these days – is a clear sign
there's absolutely no value in it.
Other claims and queries
NZ Herald motoring editor Alistair
Quotes from Alistair Sloane are in italics.
None of those who rubbish it have tried it. I was hoping for at
least one person to say they fitted it and it's all smoke and mirrors.
... Those who say it works can't say why. Those who say it doesn't work
haven't tried it.
This is a very sad misconception. One doesn't need to try something
to understand it doesn't work, or to be able to disprove claims made
for it. This attitude he has that the anecdotal evidence for Fuelstar
(and Mr Sloane knows it's all anecdotal) must be countered by more anecdotal
evidence opposing the claims shows his lack of reasoning on the issue.
Alistair Sloane says he is not pro-Fuelstar but has repeatedly written pro-Fuelstar
articles, full of anecdotal evidence. A quote from Jonathan Holmes (host of Media Watch in Australia) is relevant: "To put it bluntly, there's evidence, and there's bulldust. It's a journalist's job to distinguish between them, not to sit on the fence and bleat 'balance'."
There is also the problem that people who would do a proper test on a Fuelstar product are unlikely to be gullible enough to buy one, and those who do buy Fuelstar-type devices and realise they don't work are not likely to want to admit they were duped into paying for a scam product.
These claimed deficiencies [with the AA test]: what are they? All
the scientific community will say is that there is no basis in science
for Cornelius' claims. It's turning into a he said/she said circus.
The scientific community is right – there is no scientific basis
for Fuelstar's claims. I've explained several very good reasons in the article above. Be that as it may, Dr Sprott had four main complaints
regarding the test, which are clearly outlined in his report. Even a
poor reporter should have been able to find that report, even if he
didn't understand what was being said. (And a good reporter would find someone who could explain it clearly.)
Dr Jim Sprott
Dr Sprott had four main complaints regarding the AA's test:
There was a fuel filter between the unit and the engine.
I've addressed that above, concluding the tin particles would be so
small they'd go straight through the filter. However, Dr Sprott goes
further and claims that the colloidal tin particles would have an
electric charge, and would thus be attracted to the filter. His wording
implies the charge is because the particles are colloidal.
While colloidal particles are often charged, this is not always the
case, and these particles are (it is claimed) being produced by tin
pellets being shaken together, not by an electrical process,
which is how colloidal
silver, for example, is made. Thus I find it hard to understand
why charged tin particles would be a problem. How would they become
charged? Where would the filter get an opposing charge to attract
and hold the particles? Or if the filter were neutral thanks to it
being grounded, the tin particles would just lose their charge passing
through the filter. So the complaint either doesn't make sense or
it's a non-issue.
The test was an accelerated one, with the engine run at wide
open throttle. If the product works, it should still produce a
clear result at high revs. If the product doesn't work, no rev range
will show any difference. Get over it.
The device was not installed properly and the test didn't duplicate
on road conditions. What this failure was is not elaborated on,
but apparently it doomed the test to failure. Again, if the product
really works so well, it should still produce a clear result in the
laboratory. The exception would be if the unit was installed so that
an important component of normal operation was avoided, such as mounting
it on the floor instead of on the engine itself, so vibrations could
allegedly shake tin particles loose. Ian Cornelius, the Australian con-artist,
er, inventor of Fuelstar makes this claim, but Dr Sprott doesn't.
I haven't seen the test report so unfortunately cannot comment further,
other than to say it seems strange that a test would be so designed
if vibration was critical to correct functioning. (However, if vibration
is critical to correct functioning, how is it that the device
is claimed to work on any sort of engine, with whatever sort of vibration
the engine happens to produce? It's not credible.)
The test should have been run after 50 hours of engine conditioning.
This implies all the tin is not swept straight out of the cylinders,
exhausted with the combusted fuel. One can only speculate as to what
it is doing. There is no realistic science that explains how
the stuff might work, so it makes it very hard to design a test for
anything other than realistic usage. However, I also found a report
that a test commissioned by Fuelstar specified 20 hours of conditioning
and I've also seen claims that benefits can be seen instantly. The lack of consistency indicates a lack of any real measurable effect.
Fuelstar, Fuelmate, Broquet, Carbonflo and other similar devices/companies are a scam and just make gullible
people poorer. Better to spend your money on vehicle maintenance.
independent kiwi spirit of invention.