Tuning the C20LET - Breathing/Intake

 

Pressure drop galore

We assume that the turbo intake operates at exactly atmospheric pressure, but this is not necessarily the case. We don't have a cone filter stuck on the turbo out in the open (it would be atmospheric then).

In our case it has to breathe through a pipe (restriction) which is connected to the AirFlowMeter(another restriction) which introduces a slight diameter reduction and two wire meshes. This in turn sucks through an airbox containing a panel filter, which in turn sucks air through a snorkel.
It's like breathing through a straw.
At full throttle and high revs such a setup can easily account for a couple of psi pressure drop.
This means that the turbo will have to produce an extra 2psi to reach it's predetermined boost target. In the process it will have to rev faster, reducing its lifespan and shoving extra heat in the system (not good)

That's not the end of it.

We also assume that if we operate off-boost the intake manifold experiences ambient temperature conditions, but this is also false.
At operating engine temps we have to account for:

• heatsoak due to radiated heat from the bay affecting pipes/intake manifold
• heatsoak due to convection (VERY) close to the cylinder head, and the inlet gasket isn't exactly asbestos
• heat build-up due to pressure drops up until the head

They all add up, and on a hot summer day it ain't pretty. And that's before the turbo boosts and unleashes it's own heatwave.

 

Airbox and snorkel

The stock LET snorkel is of not particularly generous proportions, but it's not bad either.

It faces forward, under the headlight, protected by the bumper.
This is an understandable location for an OEM design, as it's well protected from road debris, excessive dust and mainly rain.

The downside is that this is a location of very low air pressure.
In an ideal world we would extend this snorkel straight through the front bumper and have it protrude like a bellmouth about an inch forwards.
Research has shown that this is the best setup to pressurise the airbox. I have read several books/papers/articles on the subject and I have even put my own ideas to the test, mainly on motorbikes (with various degrees of success, I might add...) There is a lot of misunderstanding and a lot of bull on the subject, as there is a distinct lack of research available in the public domain.

Here's an article on how air flows

And here is a PowerPoint document on Air density and performance

 

Here is some RAM-air history and facts

Now here are ideas and implementations for RAM-air for the LET

 

"High Flow" replacement airfilters

Engines are very particular when it comes to breathing dirt. It acts like grit, scoring cylinders, valves and pistons. Turbochargers are even more particular, as the debris has to pass through the compressor that may be spinning at more than 100K rpm. Even fine dust can damage it irreparably. It looks like pitting. Larger particles will simply wreck it. Therefore the quality of filtration is very important:

K&N panel filters are very popular, because they are not filters - nope, they are "filterchargers" and they groom the air as it flows through, maybe even accelerate it. Or so the marketing blurb will let you believe.

In reality their filtering capability is far worse than a paper filter (comparing new with new) and lamentable after a couple of cleaning cycles.

They DO need cleaning after 5 -10 thousand miles, because they get black and filthy. Nowhere near the 50K miles advertised.. Using a bright light behind the freshly cleaned K&N the horror of the gaps in cotton becomes clear - and that's with K&N's own cleaning spray and oil set (click for larger pic) Here is a post from the Porche 944 club showing their horror to discovering the effects of inhaling these large particles If the link is dead, then try a local copy here

Another con is the way the figures are manipulated. Advertising may claim that you only lose less than 2% of filtering ability. The streetfighter thinks : Hey, losing 1.8% is nothing really, compared to that spine-ripping power boost I've always lusted for!

Alas, they typically reach that figure by subtracting 98% (new filtering ability) from 99.8% (stock filter). Do the math and you see that the overpriced "performance" garbage lets in 20 particles (of so many microns) in every 1000, while the stock 'grandma' filter only 2 out of 1000. Twenty versus two, that is ten time worse in my book. Depends how you look at it, really...

Flow-wise the differences are minute. Yes a new K&N may (claim to) flow x% better than a new paper filter, but that doesn't mean that you'll get x% more air in the engine (and ultimately x% more power at the wheels). Oh no, the airbox/airfilter restriction accounts for 12.5% of the total intake restriction (which can reach a couple of psi on a 300bhp LET at full boost). That translates into less than x% of one eighth of 2 psi - not a lot, and keep in mind that in stock cars the total restriction will max at 1 psi. In fact, if you run without filter AT ALL you'd be hard pressed to notice any difference in performance. On a dyno it's almost within the error margins of the runs.

Here is a condensed internet version of a jap magazine test between "performance" airfilters (for jap twinturbos, but the principles are the same).

Here is the original jap test. Remember that most of the gains are from the elimination of the airbox, NOT the 'freeflowing' elements. Unsurprisingly, they all flow the same.

Tests on a Calibra setup have shown that the K&N panel filter (freshly cleaned once) improves the pressure drop by LESS than 1 cm (water) out of 80cm (total intake drop). This was repeated 3 times, by the way.

The stock snorkel created 18.75% of the overall restriction and the AFM 68.75%.

Measurements with a modified snorkel (shorter, wider and bigger bellmouth) showed slightly lower overall pressure drops, (72cm instead of 80) and the distribution was predictable:

• snorkel:10%
• airbox/filter:14%
• AFM:76%

It's clear that the contribution of the filter itself in airflow restriction is minimal, hence the "filtercharger" improvement is negligible. Definitely not worth the risk of inhaling larger dust particles.

Stock paper filters cost £3 (if one cares to shop around), that's a TENTH of the price of a 'performance' replacement. No contest then.

Here's another independent test showing the superiority of paper filters over K&N and the like (local copy here)

Here's yet another (local copy 1 2 3 4 5 )

On a 300bhp 350z, a cold air intake only showed 3-5bhp gains on back-to-back dyno runs. Despite the reviewers' ravings, that's perilously close to the dyno/temp error margins, and shows how good stock airboxes can be.

Here are the results (in flow restriction and dirt accumulation) from another enthusiast's tests (loca copy here). Note how the stock filter (AC Delco) outperforms all the 'performance' alternatives, including K&N - sometimes by a big margin.

 

Hey, you're wrong! I know of people who got over 10% gains with airfilters - on a dyno, too!

If only!

Nothing will stop people believing that they can get that sort of power just by swapping the most accessible bit. The magazines and the traders who support them all back this mentality, since they all make money selling overpriced rubbish.

Here is "proof" from Revs magazine of the massive gains from aftermarket filters. They claim 12% gains just from drilling holes in the box and fitting a K&N panel filter

Of course the 'test' was not very scientific as there was no run without any filter at all, neither do we have any proof of the tests running with the bonnet closed. The engine air temp is not mentioned (vital!) neither is the corresponding engine temp.

The common trick is to run the stock setup at the end of the test, when the car is very hot indeed, so power is down anyway. Remember that for every 4C increase in air temp, power goes down by about 1%, so 40C increase will result in 10% power loss. This is not an extravagant figure - just stick a temp probe under the bonnet and in a 20C garage it can easily exceed 60C, especially after the car has had a few dyno runs.

Here is a quick page from the excellent Autospeed website. The article is called Don't bother changing the factory filter!

Here are the dyno comparisons from another enthusiast from PlanetSoarer. The stock airbox/filter proved to be the best performance-wise. (local copy here)

 

Cone filters

Impressive claims are being made by their manufacturers regarding power increases. These are all overstated, and while there will be a few bhp improvement with any cone filter, that is not because the filter flows better but because the snorkel/airbox combo have been ditched (that's a fourth of the intake pressure drops gone). The downside (unfortunately) is that these 'dyno tests' are usually done with the bonnet fully open. I fail to recall the last time I was driving down the motorway with my bonnet open - I tend to drive with it closed, and the heat in the bay gets really bad, especially in the summer.

They do provide some 'airducts' to feed the cone with fresh air, but this doesn't stop the cone from breathing a lot of pre-heated air as well. If the average air temp around the cone is 20C above ambient (very typical, it's usually worse) that's 5% of engine power lost. So we gain 2% due to decreased pressure losses and lose 5% (or more in heavy traffic and a black car in July). You do the math.

A cone filter can only be an improvement when a good shield protects it from the engine bay heat, and it seals airtight as the bonnet shuts. It's not hard to fabricate, and a cold airpipe would then complete a proper setup.

Apexi claim air-grooming and smoothing fantasies for their own high-tech, space age designs. Nice graphics, but shamefully airflow doesn't work like that. (the shame is on the air molecules of course)

Here is a bloke from PlanetSoarer doing dyno comparisons between stock and K&N pod. Not good results for the 'performance' item... (local copy here)

The same stands for foam filters, which have similarly low dust retention as the K&N. If they're not properly oiled, they're much worse. Wire mesh filters don't flow much better, despite their pitiful filtration capabilities. Filters using oil as a dust-retention medium need to be cleaned an re-oiled regularly. If there's not enough oil they filter even worse. If there's too much, the excess will end up on the AFM's hot wire. I'd give them a miss.

Here are some creative uses a Supra owner found for his HKS mushroom filter:

Door stop	Tool container	Garden gnome chair

The Pipercross Viper seems to be a move in the right direction as it uses carbon fibre to protect the filter from the heat and it incorporates a RamAir intake as well. Unfortunately it won't fit the LET, as the orientation is 180degrees wrong. Costing £250 wouldn't help either, especially when pressurising the stock airbox is so much cheaper.

 

The 'airbox mod'

Lets not forget to mention a well-publicised (yet moronic) airbox 'mod' I've come across quite often: It involves drilling holes to the lower part of the airbox. Lots of them. Claimed benefits include more noise and more power (in that order).

I fail to see how letting an engine breathe warmer air can lead to an increase in power. The airbox intakes are not that restrictive, either.

Another claimed benefit is that the 'mod' is free. Again, I fail to see how it's free when you have to go and buy a replacement airbox later on (when the novelty wears off).

Air Mass Meter

There is a slight diameter reduction, but it isn't practical to improve that. The hot wire itself presents minimal disruption. Not much can be done there, short of dumping the whole thing for a MAF (needing programmable management in the process)

However, we can dump the two wire meshes, one before and one after the hot wire:

Opinions vary regarding the effectiveness of this move. According to Mr. A. Graham Bell it could give an improvement worth a couple of percentage points at full bhp. That's 5-6bhp on a 300bhp LET. Every little counts.

However, Ben Watson in his book How to build and Modify Intake and Exhaust systems has figures from modifications of a similar Bosch AMM for the injected Chevvy V8 (of a larger diameter probably, but identical design). Amazingly they found the stock sensor to be able to flow 529cfm and without the two screens 711cfm, an increase of 182cfm. That's a massive improvement at full throttle of a modified engine, where it could make or break a bottleneck.

They also found another 39 cfm from trimming off the cooling fins, which are there to protect the electronics during extremely hot days. Legend has it that the electronic modules failed due to overheating during stress tests in Death Valley. If you live in a cold country, they can go, or at least be trimmed back (and rounded off again of course!)

After the liposuction, the AMM will look like this:

Nice, smooth and unrestricted. There is this theory that the meshes perform a 'swirling' action and removing them will 'upset' the meter. It won't.

I've never understood the rationale behind the fitting of these meshes. The one before the hot wire is there (apparently) to protect the fragile hot wire from thick particles (a few millimetres wide). But how are these particles expected to pass through the airfilter? OEM designs do have to err on the safe side, fair enough.

Then there's the other mesh, after the hot wire: what is that for? Do they expect the AFM to disintegrate and are trying to protect the compressor from eating the junk? For one, if the fine wire breaks off, that mesh won't stop it even for a second. The only reason I can think of is to protect the fine wire from accidental damage during servicing. In that case, fair enough...

Some people like to fit the AMM after the intercooler, usually because they have ditched the airbox and use a different turbo as well. They report no problems with the new location. Beware though, if the post-intercooler temps are high, they risk the AMM reaching its limits. By design it passes varying amounts of high current through the platinum wire so that its temperature is kept at around 100F above that of the incoming air. In the standard location the incoming air will never be much warmer than ambient, but after the intercooler it will most certainly be, especially during full boost runs where the intercooler is heatsoaked. These few moments will happen to be the most critical for the sensor's output, so beware of this limitation.

Here's an article on a smart Helmholtz resonator - strictly not forced induction material, but interesting.

A basic primer into induction theory by GrapeApeRacing (local copy here)

NOS - Injection of Nitrous Oxide

A more drastic way of adding oxygen molecules, without increasing boost pressures while actually lowering intake temperatures

 

Here is a handy calculator for air density based on altitude, temperature and humidity

 

Credits:

I'd like to thank our friend Jo for the intake pressure measurements (jochen_herrmann@trace.ch)

The Helmholtz stuff is courtesy of Animal

 

...On to the Air Pipes