This article was first published in 2011.
Last issue in Powering-Up the 1.9 litre TDI, Part 1 we looked at the intake system of the Volkswagen 1.9 litre PD turbo diesel fitted to the Skoda Roomster.
Measuring cross-sectional areas of the standard intake showed that the system would likely have major restrictions – at the pictured point, the cross-sectional area of the intake duct suddenly decreases by 80 per cent!
On-road measurement indicated a maximum pressure drop of just under 14 inches of water at 3500 rpm that then held at 12.8 inches of water through 4000 and 4500 rpm.
This pressure drop, and the observed design of the system, indicated that it was likely major improvements could be made to the intake system ahead of the airbox. On the other hand, the airbox, air filter and airbox exit duct all looked pretty good.
The Factory Upgrade
As is the case with a surprisingly large number of cars, a factory upgrade intake exists for this engine. Called the ‘PD160’ air intake and fitted as standard to a higher-powered Seat, the new intake replaces the adaptor out of the back of the air intake mouth (the part that otherwise drops cross-sectional area by 80 per cent!) and also uses a larger flexible duct. It’s a direct bolt-on to the standard system.
The Seat part numbers are:
Trumpet – 6LL 129 621
Pipe – 6LL 129 618
Here in Australia, where Seat dealers are non-existent, the parts are available from Dub Addiction at a cost of AUD$189.
In addition to the high cost (surely the production cost of these items would be only $10 or $15?) my concern about this upgrade was that the inlet to the airbox remains unchanged in size. At around 42 square centimetres, this is smaller than the actual opening in the airbox wall (that’s about 50 square centimetres). So while I am sure the PD160 upgrade provides a much better flowing intake, I thought a new intake should be able to be made that was both cheaper and also worked better than the PD160.
Making a New Intake
Starting at the very beginning of the intake system is this mouthpiece. As described last issue, its opening (red arrow) matches the large cut-out in the over-radiator metal panel – and that’s good. However, at the other end (green arrow) there’s an abrupt change in cross-sectional area. Sudden changes in cross-sectional area should be avoided as they lead to internal turbulence and so flow restriction. There appears to be no reason why the duct is designed in this way.
So it was out with the hacksaw and the rear part of the duct was removed.
To make the new intake duct I decided to use 80mm PVC plastic pipe and fittings. This approach has a number of advantages: different fittings are readily available (eg pre-formed bends) and the inner surface of the pipe has a smooth surface. The pipe can also be heated and then shaped and formed. You would not use this pipe in close proximity to an exhaust manifold or turbo, but situated in cool areas of the engine bay (and with intake air constantly passing through it) the pipe will survive without problems.
PVC plastic pipe comes in different grades. I initially started work with thin-wall PVC pipe and found it incredibly hard to work with – when heated, it just collapsed, etc. I then bought higher quality plastic pipe that had about double the wall thickness and found things progressed much more easily. Don’t pay bottom dollar for the pipe - you’ll get a crap result.
The first task proved to be the most difficult - making a transition piece that would go from the rectangular mouth duct to the round 80mm diameter tube. Unfortunately, no ‘gutter type’ adaptors available off the shelf had the right shape or smooth internal contours. In the past I’ve made lots of adaptors like this by heating the pipe (using a heat gun) until it is soft and that moulding it into shape, either by gloved hands or with hand tools.
However, that proved difficult in this case and so I made a wooden block that replicated the dimensions I needed the tube to take. I then cut the tube off at a slight angle (so increasing the open area), heated the end of the tube and forced the shaped wooden block into the mouth.
This approach formed the correctly shaped opening but I found it very difficult to mould the piece without wrinkles occurring in the walls, as pictured here.
After lots of trial and error I developed a technique that got rid of the wrinkles. I found a wrinkle-free shape could be formed if the pipe was stretched after it had been shaped. I used my engine crane to do the stretching but the required tension isn’t that great – so a rope over a pulley could probably also be used.
The pipe was first heated and forced over the block. It was then nailed to the block using short brads placed close to the end of the tube.
Vicegrips were locked over the upper part of the tube and these were connected to the crane hook. I then heated around the tube just above the wooden block until the plastic softened. Tension was then applied by the crane and the plastic allowed to cool.
After the ends of the adaptor piece were sanded back (getting rid of the nail holes and straightening the end) this was the result: a smooth, nicely-shaped adaptor that fitted neatly over the shortened original mouth piece.
Next I tackled the airbox.
The standard intake was cut off with a hacksaw and…
…the resulting hole was enlarged.
A standard 90mm right-angle plumbing adaptor could then be placed through the hole in the airbox. To seal the gap between the bend and the airbox I used a flexible neoprene collar that was cut from…
…an old stubby holder. (More use for stubby holders in a moment.) The use of the neoprene collar on the bend allowed it to be rotated when the duct was being fitted but still seal when the duct was in place. (The bend needs to be rotated because the duct needs to be assembled in situ – it can’t be fitted into place if glued together outside of the engine bay.)
With each end of the system done, the connecting pipe could then be shaped. This required various indents to provide clearances; these were achieved by heating the pipe until it was soft and then using gloved hands to wriggle it in the required position, the wriggling denting the pipe and so giving the required clearances. One side of the pipe was done at a time.
As implied above, the new intake needs to be assembled in position, rather than assembled outside of the engine bay and then inserted. So rather than gluing the shaped pipe to the original intake mouth, I decided to use another neoprene collar cut from a stubby holder. This seals the new duct to the original and also helps hold it in place.
The new duct could then be inserted, with at this stage the joins held together with just tape. The arrow points to a slight bend that is used between the moulded adaptor and the straight length of pipe. In 80mm plumbing a range of bends is available, including these very small angle designs.
With paint and gluing still to come, it was time to do some testing – the details of which follow.
But to maintain the sequence showing the building of the new air duct, here is how it looks with paint. This is the front section of the new duct showing the indents required for clearance and the adaptor piece designed to mate with the original front section.
The duct fits between the battery and the engine.
The ex-stubby holder neoprene rubber collar seals the duct to the original and also helps hold it in place.
Within the view of the whole engine bay, the duct remains inconspicuous.
Testing was then carried out. Pressure drop testing was undertaken as for the standard duct – at full throttle in second gear, with the pressure drop read from a Magnehelic gauge at 500 rpm increments, and with the tapping point located in the exit side of the airbox.
This graph shows the measured results, with the blue line showing the standard pressure drop and the red line showing the pressure drop achieved with the new intake.
As can be seen, there is a massive reduction in the restriction that occurs. Whereas in standard form the airflow restriction peaked at just under 14 inches of water, now the maximum is just 4 inches of water! This is a decrease by as much as 69 per cent. Also able to be seen is that an improvement has been achieved throughout the rev range – although as the amount of airflow increases, so does the improvement.
As indicated in the main text, the standard paper air filter looks good, with a large area and very deep pleats. Plus, in every car intake I have ever measured, if the car is not radically modified for increased power, the factory filter poses very little restriction indeed.
And that’s the case with the Roomster, where the air filter caused a measured maximum of just 1 inch of water pressure drop. Don’t bother changing the filter for a “high performance” one – the standard filter already is!
The new intake duct clearly flows bar better than the original, allowing the engine to breathe more air. So did performance also improve? Nope – it went backwards!
Before we look at that idea in more detail, let’s compare the actual on-road acceleration results. As with the standard car, the acceleration was measured with an on-board accelerometer and the readings were the results of a two-way average. The testing was again done in second gear at 500 rpm intervals.
The modified / standard tests were also done in quick succession as, being a two Roomster household, I could easily fit the dead standard intake and then quickly swap over to the modified intake.
Here are the actual on-road acceleration results – the blue line showing the standard acceleration through the rev range and the red line showing the acceleration achieved with the modified intake. As can be seen, except for very low and very high in the rev range, the standard car accelerates harder!
So is this a disaster or not? If I had intended fitting just the new intake – yes, well it would be a disaster. However, a custom reflash of the engine management system has always been part of the modification program – and taking that into account, then the story looks very interesting.
So what is going on? Put simply, it looks as if the engine management system does not 'like' the changes. Given that the amount of fuel injected is everything to a diesel's performance, it seems as if the mass of injected fuel has actually been reduced!
Despite being widely fitted to 1.9 VW diesels, few people can detect an improvement from the larger-than-standard PD160 factory hi-po intake described above. While I haven’t tested the flow of such an intake, it’s almost certain that it does flow much better than the standard design – so perhaps the lack of any improved performance comes for just the same reasons?
So if all this is correct, any breathing improvements, if undertaken without engine management changes, will result in performance declining. However the corollary is this: when the correct fuelling is provided, performance is likely to be far better than achievable with just a reflash and no breathing mods!
Interesting and more interesting!
In Part 3 of this series: free-flowing the exhaust