Toyotec's how to make your MK3 16v breathe on a budget. Part 2 added 14-12-09

Come on Toyo, I like an interesting factual thread. Get those results written up

 

My work and other commitments have been fighting for my time. The pictures for the second session are loaded up on photobucket, I just have to start putting together the words for it.

 

You know you wanna help us with your testing...
Caammaaaannnnn....!

 

A voice can be heard howling in the wilderness 'eddie..... eDdie... eDDIe... where are yoooouuuu?'

Nothing like keeping a heard of ABF 'idiots' in suspense.

 

16v AIS (Air Induction System) testing Part 2 25-09-09 *Plots to follow*

It has taken a long time to update this and a lot as happened since the thread was started. So let's get cracking.


Introduction.

In the last series of tests, a VW Golf MK3 16v was mounted on Garage Streamline's dyno and analyzed for power and torque depending on the combination of air filtration system that was used.
From these tests it was concluded that a factory airbox with subtle modification to the internal velocity stack in the top half and larger air hole on the bottom section of the airbox resulted in the best power and torque curve, that even superseded a pipercroess cone type filter.
In this series of tests, with a baseline established, the STD corrugated air duct was swapped with a well built homebuilt one, the velocity stack from the top half was deleted and an long air tube was experimented with to see if there would be change to torque through out the rev range. We also had an interesting diversion when establishing the new baseline.
Read on...

NB In these plots, the colour of the line represents the test condtion measurement.
i.e Baseline 25-09-09 = Blue for power, TQ and AFR.

Baseline 25-09-09



Recall that the last test the highest achieved measured power was 151.2bhp@6200rpm


The vehicle was strapped on the rolls and run up to normal operating temperature. The powertrain was in the same condition as the last test on the 9th of September, except the spark plugs had been changed as part of a minor service.
After a few conditioning runs, well actually 10, the most power indicated was 143bhp@6300rpm. This was highly strange as the engine lost torque for the entire rev range. What had shifted?
Dyno input settings were again checked and these were fine, The vehicle was checked for faults and these where also fine. So what was going on. Well a quick call to the owner, eatonMK2, revealed the truth. He had been a naughty boy and cheekily installed that so called resistor modification in series with the IAT (Intake air temp) to effectively make the ECU think the air temp was at 5C deg. Not good when the air in the cell is at 21C the engine coolant is at 95C and the ECU is advancing spark and perhaps injecting extra fuel to where it should not be for the real ambient conditions.


This bad modification hurt drive feel and performance and of course cost 45mins of debugging, when testing should have been well on the way. Stay very well clear.
With this resistor removed and the wiring patched-up, the offending discrete component was binned and the engine instantly increased torque to 130lbft levels again and measured 147ish bhp still a bit down. It did not get to 150bhp on this day but seemed to be happy at the 147ish bhp peak. After the test session a study of the road power to flywheel correlation was conducted and it was observed that the delta in measurements at the peaks were 23-23.5% on the 25-09-09 vs 22% on the 09-09-09. Also on this day the cell temperature s was hotter and the ambient pressure was 101kPa vs 103kPa on the 9th of sept. Another theory may have to do with the method of strapping i.e. how hard was it strapped to the rolls, this could increase driveline friction. This may explain why the actual wheel measurements where peaked at 119whp vs 123whp on the 9th.
Funny thing dyno testing
With the vehicle consistently repeating a tad over146bhp@6300rpm, this result was taken as the new reference for this series of tests.
During the establishment of a baseline measurement, the lambda sensor was unplugged to observed the effect on fuel control. When this was done there was positive shift, rich, by one point. And the closed loop operation during part throttle and sometimes full load ceased.

Here is the plots for the baseline on the 25th of Sept vs 09th Sept.

Notice how the torque at the end of the rpm range starts to drop off much earlier than the previous baseline resulting in a lower peak BHP measurement. Comapred to the last time the vehicle felt the same through the gears.

Baseline 090909
Peak power was 151.28bhp@6200rpm, peak torque was 132.8lbft@4550rpm
Baseline 250909
Peak power was 146.30bhp@6300rpm, peak torque was 130.6lbft@4500rpm

Here is the plot for that mod that was sneaked into testing.


Here we see a total drop in torque for the entire rev range. Runs richer too as you would expect.
Vehicle did not feel as urgent.

Baseline 250909
Pwr 146.30bhp@6300rpm, TQ 130.6lbft@4500
IAT 'mod'
Pwr 143.1bhp@6315rpm, TQ 128.1lbft@4548rpm

Baseline 25-09-09 as reference vs no airfliter vs EssThree's custom made air duct.

In my quest to established why the vehicle seemed to be down on both torque and power the airbox factory filter was removed to see it was the cause of any issues but there was not really noticeable cause for concern. Next EssThree's air duct was swapped with the factory item and tested a few times. Once again this test was line on line with the factory system both with filter and w/o.

Here is a plot that represent these tests.

Strange is the slight increase in lower end torque without the filter. This trend continued for the rest of test iterations even with the filter.

Baseline250909
Pwr 146.3bhp@6300rpm, TQ 130.6lbft@4500rpm
Essthree duct
Pwr 146.2bhp@6200rpm, TQ 131.1lbft@4550rpm
Baseline no airfilter
Pwr 146.94bhp@6514rpm, TQ 131.7lbft@4550rpm

[Baseline 25-09-09 as reference vs STD top half with vel stack removed and long air duct ending at headlight.



Back in my day of spannering I would tinker with air boxes as it was observed they had lots of restrictions. I clearly remember carrying out some stealth modifications on Garfay Liu's MK3 16v when he managed pranged a couple of valves on a 53K mile motor. The modification did have an impact on drive feel that I told him to take the 30quid ebay chip out and run on the factory chip. Well it is time to recreate that experience on this vehicle. My reasons for this are do you have to reinsert the vel stack to increase smoothen airflow?
What happens it if just simply removed. What happens if the air duct was increased so that it would end beyond the headlight. Perhaps the engine is not flowing sufficient enough air where that it matters but with the velstack removed here is plot of what happened.
There was no real benefit over the system with the bell mouthed vel stack. The comment applied to lengthened aiducting.
Here are the plots.


Baseline250909
Pwr 146.3bhp@6300rpm, TQ 130.6lbft@4500rpm
Baseline no vel stack
Pwr 146.75bhp@6243rpm, TQ 131.7lbft@4313rpm
Long intake duct from H/L
Pwr 146.1bhp@6243rpm, TQ 131.8lbft@4313rpm

Very very small increase in low end torque shown on this plot for the long inlet duct. May be atributed to test to test variablity. Too small to be felt though. Top half w/o eatonMk2's vel stack showed no loss in torque and made no difference.



My conclusions to AIS testing.

ABF MK3 16v air paths seem to benefit from airfliter inlet crossectional area increases to the top half by simply removing the factory air horn/noise restrictor and increasing bottom aperture to equal or slighty exceed to crossection area of the top section. The larger the hole on the bottom means more induction roar. The roar is much muted to open pod filters on a 16v. I still believed that the air horn is the correct way of finishing this modification but cannot see any benefits to engine torque for this level of engine. Maybe as the engine gains more bolt on components to increase airflow, then there might be a benefit...we will see.
When making extra holes in the bottom be careful not to bore the airbox toward the alternator drive/cambelt or at the rear by the manifold heater hose. This will induce engine bay heat into the airpath. Use locations towards headlamp or towards front of airbox towards fender.
For this road engine, the OE airbox to throttle duct, with its corrugated section, is fine and does not hurt performance over larger diameter aftermarket silicone or metal parts. Any turbulence created in the corrugated section does not seem to have an impact on the actual maximun air delivered to the throttle/engine as the smoothed airduct from EssThree made no difference to the torque curve during these tests.
It would appear that any other investigations and optimisation on the AIS system would result in a "polishing the turd" condition i.e. there where no more benefits to be had from optimising this system. Further modification for increased torque would need to come from elsewhere e.g. engine control parameters.
There seems to be a tendency for the mixture to tract to stoich up to 4.5K then run into some type of open loop/ component protection lambda. This was more pronounced in tests run on the 09-09-09 although not consistent.

Further work.

The AFR control does not seem consistent with the vehicle at WOT and this inconsistency seems to be reflected in the torque plot. In order to test this statement the vehicle will be put through a series of tests with aftermarket access to engine control parameters of fuel and spark. This custom calibration will be referenced to popular eproms and the results will be reported in another post.
With ECU control access there are also plans in the pipeline for testing exhaust manifolds typical 4-1, 4-2-1, LHD VW ABF 4-2-1 on STD downpipe.

Stay tuned

The vehicle was tested by Toyotec.
Data generated and complied by Toyotec.

With many thanks to,

Garage Streamline and the team including Will and Theo for dyno development time and assistance in swapping components very quickly.
EssThree for supplying airduct of his own car and a MS V3 x ABF PnP ECU by Basil Duham.
Mike H for supplying an unmolested ABF MK3 airbox,
eatonmk2 for lending "N7NDO" as the development vehicle.


Regards
Toyotec.

 

Read that one, any timelines on the next episode of "tune your ABF"...?

Good work Toyo

 

The suspense is now over .

 

Time, components, money, vehicle, and dyno use use all have to line up to make the next post possible.

 

nice one buddy... thanks for putting the time and effort in. Always thought the velocity stack was restrictive... put it back in a few times and it always feels slightly less responsive.

What about drilled airboxes?

Next thing should be an investigation into the effectiveness of aftermarket exhaust systems.

For the money people spend on them and the who-doo surround whether there are actually any benefits it would be soooo worth while

i would be willing to lend my ridiculously expensive supersprint magnum for the cause

 

Please read my conclusions and see how they relate to increasing bottom section croessectional area with respect to the top half.

It is a good thought although at 2/1/4 inches roughly, I believe that the STD system is more than up to the job. But who knows if after the rest of tests we could just do that.
Takes time to execute these tests though.

 

Off topic a bit , but to replicate as close as possible the mods you did on lower powered cars , like mk2 8vs , would if help the power curve? Especially on cars with a fast road cam , or a P+P head ?

 

Great update toyotec, look forward to seeing these updates. would it be benificial if you were to take the standard manifold and downpipe and port them?

 

If I have the time I'll port out a std manifold over xmas and throw that into the mix.

 

This will be interesting vs a STD unit vs a VW ABF 4-2-1 vs an aftermarket 4-2-1 vs a 4-1 system. I know that EssThree did port the exhaust manifold on his MK3 16v.

Cheers. Before you start, I will come around to inspect and see which areas can be improved.

 

It can be done, but would take time and money to start with a good base car and add components from there. I would expect the results to take a similar direction to what we are doing here.

 

Thanks for the reply . Just I'm rebuilding a Digi MK2 and think now I'll follow your guide to helping the OE induction work to its full potential.

 

What did you do with the hot air feed you disconnected. Did you leave it venting to atmoshphere or block that off too?

 

Just turned the pipe to face down so that the hot air is vent'd away from the airfilter.

 

I can't remember off hand and it's too cold and dark to go and look, where does the hot air come from? Any greater danger of foreign bodies/moisture getting into the engine when it's not running? Could the pipe also be blocked off?

Just curious.

 

The hot air feed, is just a bit of metal hose that comes off the exhaust manifold heat-shield.

It will only ever be used if the vacuum from the inlet dictates it so, perhaps if the filter was part wet, or frozen...where the hot air would help dry it out maybe?
At least that's how mine was set up...and it never opened.

I binned the bit of duct, the vac hose from the TB (or was it from the upper section of the airbox?) and the internal flapper mechanism...and blanked off the hot air feed hole.