Collected Uncertainties in the Permeability Test

Hey everyone.

After the "Durchzugsrechner" (draft calculator) now includes additional options for wind and inclines, the 2000-4000 test seems even less reliable or unsuitable for determining the absolute engine power.

Variations in tire rolling circumference (deviations from the standard value of +1.5% or -2.5% are permissible for new tires) and the - rarely possible - precise determination of weight make it quite inaccurate to determine engine power based on measured times.

Considering the overall impact, including... Wind and a sloped road surface can reveal subtle "Peanuts-inspired" design elements that are often overlooked.

- Wind speed: 15 km/h
- 0.5% gradient / slope
- Weight deviations of 30 kg or more compared to the value stated in the vehicle registration document.
- The maximum and minimum speeds are 2000/min, due to permissible tire tolerances.

enter the values in such a way that each influencing factor either extends or shortens the time.

In the 110 horsepower Golf 3, these factors (without changing the engine power!) occur in the 3rd gear. The times ranged from approximately 5.6 to 7.0 seconds, which corresponds to a performance variation of roughly +/- 10%.

"That is, the dyno test, without an exact determination of the test conditions regarding absolute power, is as useless as performance test benches, but still provides relatively accurate results for comparative statements after changes to the vehicle (e.g., cleaning the mass airflow sensor, filter replacement, tuning modifications, etc.) under otherwise identical conditions."
Capture errors of just 0.1 seconds are quite achievable with a little practice using a stopwatch, and this corresponds to a performance level of approximately 1.5 horsepower in precision – significantly better than "the roller" with typical tolerances of around 5%.

Now, all that's really needed is an adjustment to the outside temperature.
Does anyone have reliable information about a "standard temperature" for performance specifications, and a rule of thumb for how engine power changes with temperature?

Hi Ulf,

While not directly related to your question about temperature, but regarding the influence of wind and slope: In my opinion, these factors can be quite well accounted for by repeating the experiment exactly in the opposite direction, one after the other. Under otherwise identical measurement conditions (and the same skill in using the watch), the times will only be the same if there is no wind and no incline, or if the effects of wind and incline cancel each other out (i.e., going downhill with a headwind once, and going uphill with a tailwind once).

This should at least reduce that element of uncertainty.

Regarding temperature: Based on my observations using the MFA (Multifunction Display), my car consumes less fuel when it's warm, compared to when it's cold. I've noticed a difference of about 0.5 liters, with the consumption being lower when the temperature is 20 degrees Celsius compared to 0 degrees Celsius.

The likely cause is the increased power demand for the mechanical components. However, this also consumes power and is likely to counteract the effect of the cooler air.

Best regards,

Jan.

Hi Jan,

Quote:

While not directly related to your question about temperature, but regarding the influence of wind and slope: In my opinion, these factors can be quite well accounted for by repeating the experiment exactly in the opposite direction, one after the other. Under otherwise identical measurement conditions (and the same skill in using the watch), the times will only be the same if there is no wind and no incline, or if the effects of wind and incline cancel each other out (i.e., going downhill with a headwind once, and going uphill with a tailwind once). That should at least limit this factor of uncertainty.

I thought so too, but if you're interested, try calculating both directions with a slightly "stronger" scenario, for example, a 30 km/h wind and a 2% incline.

The average of both times certainly does not yield the time without wind and gradient...

Hi Ulf,

"Sorry, there seems to be a misunderstanding. I wasn't referring to the average, but rather to comparing the measured values as averages in order to identify these influences. In other words, to be able to see the "overlooked" 10 km/h wind or 0.5% incline." Just to make sure there wasn't any wind and that it wasn't going uphill.

I think the reason why the average doesn't work out is because the effect depends on the duration, and the uphill portion is longer than the downhill portion. At 5.5 seconds (example), the assistance helps less when going downhill, compared to how much it hinders progress when going uphill in, for example, 7 seconds.

Best regards,

Jan.

Okay, the standard temperature for performance measurements is 20°C (68°F) ambient temperature.

Deviations (when the engine is warm) may occur from the set temperature. Dependence on the density of the air (including humidity and air pressure), and the altered efficiency of the engine...

In addition, there are also differences in the heating of the intake manifold, which results in different intake air temperatures even with the same ambient temperature.

However, depending on the outside temperature, there's also the increased air resistance caused by denser air.

So, it's not that simple, everything .

Simply put, 10°C cooler intake air results in a 2% power increase... up to a certain temperature limit (3°C for gasoline engines, for diesel engines). Below that temperature, it's worse; above that temperature, it's also worse.

Starting from that, it would be possible to determine more accurate values using data from the 'Dichter der Luft' project... I could provide data on the temperature dependence of the air... if there is a need. Would also help with calculations, but first, a 'concept' needs to be established.
(what to consider, what to disregard, etc.)

Best regards,

Hi,
I once worked on a project involving non-dynamometer performance measurement for large diesel engines.
All temperatures affect performance differently at varying speeds. Since your test doesn't exceed the rated power, it will likely be difficult.
The oil temperature also has a significant impact.
At low engine speeds, the maximum power output is achieved with an oil temperature of 60°C.
At high speeds, the temperature increases. Up to 80°C.
At high rotational speeds, the flow rate is approximately 15% higher at 80°C compared to 20°C.
The higher the speed, the more relevant the temperature becomes.
Interestingly, at even higher temperatures (above 80°C), the performance decreases again. I don't know if this is because the fuel in the engine block is also being heated (which I couldn't measure).

The air temperature, on the other hand, follows a linear trend.
My diesel engines were limited to 2800 RPM.
At 2500 rpm, a power difference of approximately 3% was observed for every 10°C change.
At 1000 rpm, a power difference of approximately 0.4% was observed for each 10°C.
The higher the engine speed, the more relevant the air temperature becomes.
I have always normalized the values to 20°C.
I'm not sure how we can rely on these data to predict an AAZ. Many test runs, varying only one factor at a time, would be necessary. It is hardly possible without a stationary testing rig.

Hi Ulf,

At home, there's a calibration chart from Bosch; on the back, there are correction charts for temperature and air pressure.
I'll send it to you when I'm home again.

Best regards, Rainer.

Docter wrote:

So, the standard temperature for performance measurements is 20°C ambient temperature... Deviations (when the engine is warm) may occur from the set temperature. Dependence on the density of the air (including humidity and air pressure), and the altered efficiency of the engine... In addition, there are also differences in the heating of the intake manifold, which results in different intake air temperatures even with the same ambient temperature. However, depending on the outside temperature, there's also the increased air resistance caused by denser air. So, it's not that simple, everything :)

.

Well, I think the TDI engine generates its own air pressure (as long as the turbocharger is working) and is likely less sensitive in that regard than naturally aspirated engines.
Based on general forum experience, it seems that temperature has the biggest impact on engine performance, with the rule of thumb being "cold makes it stronger."

And the increased air resistance in cold air is unlikely to significantly distort the approximately 50-100 test results, because air resistance doesn't play a dominant role at those speeds.

While it's true that..., one could also factor in the influence of higher air resistance, but whether one wants to or not is a matter of opinion... but in my opinion, it could be neglected at least initially.
The external air pressure isn't *that* important for turbocharged engines, but it still has some influence... The air being drawn in is thinner at higher altitudes.

perhaps... but one could also factor in the influence of higher air resistance, it's a matter of whether one wants to or not... but in my opinion, one could initially disregard it.
The external pressure isn't *that* important for turbocharged engines, but it still has some influence... the air being drawn in is thinner at higher altitudes...

Hi,
'This raises the question of what the reference pressure is. If it's the ambient pressure, that could provide an explanation for the engine's sensitivity to weather conditions (high and low pressure areas).' Then, in the mountains, he would also experience less absolute pressure, and consequently, less power.
If that were the case, one could potentially use a hose to redirect the back pressure in front of the vehicle to the pressure gauge, thereby manipulating it to show an increase in boost pressure. Does anyone know anything about that?
Regards,

Eike.

Hi,

My Golf is an AHU automatic transmission. Now, I wanted to take a closer look at the Ulf's ventilation test, and as an automatic driver, I've encountered a few inconsistencies.

Here are some questions:

Does anyone know the final drive ratio for the Golf 3 AHU automatic transmission?
Does anyone know the final drive ratio for the Golf 3 AHU automatic transmission?
And in which gear is the engine speed of 2000 RPM referring to (can I force the automatic transmission into specific gears?)?
Is it even possible to take this test with an automatic transmission?
Where is someone supposed to find out exactly what performance gains a tuning box provides?


I hope that wasn't too many questions at once.


Regards

eike wrote:

"This raises the question of what the reference pressure is. If it's the ambient pressure, that could provide an explanation for the engine's sensitivity to weather conditions (high and low pressure areas)." Then, in the mountains, he would also experience less absolute pressure, and consequently, less power. If that were the case, one could potentially use a hose to redirect the back pressure in front of the vehicle to the pressure gauge, thereby manipulating it to show an increase in boost pressure. Does anyone know anything about that?

Hi,

I recall reading in a documentary (I think it was about industrial engines, but I'm not sure) that the TDI engine uses the ambient air pressure to prevent the turbocharger from over-spinning at very high altitudes where the air pressure is significantly lower.

Consequently, the control system can precisely regulate the boost pressure (meaning it always delivers the same amount of air to the cylinders, regardless of the pressure, as long as it's not too low).

"A blockage" at that location is likely to have a significant impact on the boost pressure, because standard measurement techniques measure relative to the environment, so the ambient pressure must be added to obtain the absolute value. With a low external pressure, the difference between the target pressure and the external pressure is greater, so the charger has to work harder. A similar effect might occur (I think!) if you simulate a lower external pressure.

Okay, so it's the exact opposite, not increasing pressure (Eike's suggestion), but decreasing it - but that would probably be more effective through electronics in the sensor lines .

Of course, simply increasing the turbocharger boost pressure alone won't achieve much with a TDI engine.

Best regards,

Jan.

Okay, I would approach the issue of air resistance differently. I'm actually in the process of creating a table for that. After the acceleration phase, it needs to return to the initial speed, just like in a real testing setup. Since you know the mass of the vehicle, you can calculate how much energy is dissipated in the process. Based on the time it takes, you can calculate the power. Essentially, it's the sum of all the forces acting as resistance, including rolling resistance and air resistance.
The engine also had to overcome this resistance during acceleration. You can add that to it right now.
Therefore, the slope and wind are not relevant, as long as they do not change during the test. They are included in the loss factor.
With this method, it's possible to drive at higher speeds (in higher gears) because friction is compensated for. This means you have more time. Therefore, you can operate closer to the rated power output and don't need to create such a large difference (delta). The empirical values for the curvature of the curve become less interesting at that point, as one is approaching the end of the curve.

BruderTom wrote:

Hi, Does anyone know the final drive ratio for the Golf 3 AHU automatic transmission? Does anyone know the final drive ratio for the Golf 3 AHU automatic transmission? And in which gear is the engine speed of 2000 RPM referring to (can I force the automatic transmission into specific gears?)? Is it even possible to take this test with an automatic transmission? Where is someone supposed to know exactly what extra power the tuning box provides?

I am not familiar with the gear ratios.

With a manual transmission, you can theoretically drive in any gear. The table automatically adjusts based on the corresponding speed entered at 2000 RPM.

"Usually, the third one." The approach taken is that, since the final speed of 100 km/h can still be reached at many locations, and the stopping time, which is approximately between 6 and 8 seconds, is not so short that small measurement errors can completely invalidate the results.

With an automatic transmission, the test is generally possible if a torque converter bypass coupling is present and is active during the test. For the evaluation, the overall translation and the speed at 2000 should obviously be known.

The performance improvement can be estimated by comparing the "before" and "after" values of the measured time: first, adjust the table with 0% performance increase and decrease by varying the weight input, so that the actual measured time in the standard condition is calculated.
Then, using an input for "additional performance," aim for the desired tuned time. If the result is perfectly accurate, the percentage of additional output will be displayed at the top.

Hi,

In my opinion, all modern automatic transmissions bypass the torque converter once they are "engaged." I see the problem elsewhere.

How do you prevent an automatic transmission from downshifting? When you're at 2000 RPM and give it full throttle in third gear, it downshifts at least one gear and then revs it all the way up before the next gear engages...

Best regards,

Jan, who really doesn't like driving automatic cars.