[d3mon]

Originally Posted by Chethan B G

Update : 46,000 Kms on the ODO - 4,000 Kms after the remap.

Got the Jetta on a 2-wheel DYNO at Race-dynamics. Details are as follows:

Ambient temperature: 32 degrees centigrade.
Bangalore's Altitude: 3000 ft above sea level.

Attachment 1428440
Attachment 1428441


As the DYNO is just for the front wheels, back wheels were not running - MID immediately threw up errors on ESP / XDS .. etc.

Attachment 1428442

Looks like the ECU recognised that the car is being tested and switched modes! Practically, with the cheat software working; 2-wheel DYNO is a waste on any VW Tdi!

Interesting link: http://www.carscoops.com/2015/10/201...n-reveals.html

Thanks to VW - Looks like wait is still ON, until I get the car on a 4-wheel DYNO!


Wish you all a very happy Dasara!!

Wow, these figures are with the remap? These seem to be quite some way off from the expected numbers with the remap. Sounds like the defeat device does curtail emissions (and performance) with a very heavy hand when it detects emissions testing.

I'm wondering what would happen if you dyno tested with the wheel speed sensors (for ABS) disconnected. Would it fail to be able to detect that it's on the dyno and not cut the power?

[Vid6639]

Originally Posted by d3mon

I'm wondering what would happen if you dyno tested with the wheel speed sensors (for ABS) disconnected. Would it fail to be able to detect that it's on the dyno and not cut the power?

Yes. This is what I was going to say. There was a video online which did the comparison with the rear wheel speed sensors disconnected. That way the ECU doesn't go into cheat mode.

But now I am wondering one more thing. Even with a remap why does the system go into cheat mode?

Is the cheat mode a hidden map that the remap does not overwrite?

Ideally you wipe out the entire OE map and flash the new remap onto the ECU after taking a backup. So why will the car switch to cheat mode?

Secondly are the given power/torque outputs after correction?

[d3mon]

Originally Posted by Vid6639

But now I am wondering one more thing. Even with a remap why does the system go into cheat mode?

Is the cheat mode a hidden map that the remap does not overwrite?

Ideally you wipe out the entire OE map and flash the new remap onto the ECU after taking a backup. So why will the car switch to cheat mode?

Secondly are the given power/torque outputs after correction?

ECUs have a bunch of fueling 'maps' that all work together to produce the final amount of diesel that's injected in the cylinders.

Some examples of these maps are:
1) Torque limiter maps - these limit the max torque to protect the gearbox
2) Smoke maps - these limit the max unburnt fuel to limit emissions
3) Driver's wish maps - these maps give the max fueling desired by the driver
...and so on.

The final fueling is the minimum fueling among all such maps.
And no, you don't wipe out the entire map in a remap - you just tweak a few values in these maps to give you more fueling.

Now these maps are tuned by the tuners, but the cheat software was definitely hidden, otherwise tuners must've detected it a long time back.

[Vid6639]

Originally Posted by d3mon

ECUs have a bunch of fueling 'maps' that all work together to produce the final amount of diesel that's injected in the cylinders.

Some examples of these maps are:
1) Torque limiter maps - these limit the max torque to protect the gearbox
2) Smoke maps - these limit the max unburnt fuel to limit emissions
3) Driver's wish maps - these maps give the max fueling desired by the driver
...and so on.

The final fueling is the minimum fueling among all such maps.
And no, you don't wipe out the entire map in a remap - you just tweak a few values in these maps to give you more fueling.

Now these maps are tuned by the tuners, but the cheat software was definitely hidden, otherwise tuners must've detected it a long time back.

I know about the multiple maps and what I meant about the remap was that they must be tweaking multiple maps and not just one.

So when you flash the new map you can't just write one file, you need to write the entire package right? It's not just edit one file and then copy that file into the ECU.

It's more like flashing a new ROM into your phone or a new OS on your laptop.

[d3mon]

Originally Posted by Vid6639

I know about the multiple maps and what I meant about the remap was that they must be tweaking multiple maps and not just one.

So when you flash the new map you can't just write one file, you need to write the entire package right? It's not just edit one file and then copy that file into the ECU.

It's more like flashing a new ROM into your phone or a new OS on your laptop.

Yes, you write the entire package after tweaking only a few maps.

[Chethan B G]

Originally Posted by tikku.dk

Hi chetan, for accurate results, put your car on a shaft dyno, gives accurate results, dynapack of RRP is shaft dyno. The one u tested seems to b inertia dyno.[b]

Hi Tikku,

Checked RRP website: http://www.redroosterperformance.com...controller=cms

Looks like RRP has dynapack only for the front wheels. In case of a shaft DYNO, traction / friction loss is not effected. Whereas, in case of an inertia dyno, on-road experience is simulated.

We can measure power/torque at the crank, wheel-shaft and at the wheel. Measuring at the wheel by inertia, gives us the power/torque delivered at the wheels after all losses.

Problem here is the cheat software, masking performance. Only a four-wheel Dyno can solve the issue and make a difference.

Originally Posted by d3mon

Wow, these figures are with the remap? These seem to be quite some way off from the expected numbers with the remap. Sounds like the defeat device does curtail emissions (and performance) with a very heavy hand when it detects emissions testing.

Yes. I was surprised and for a moment, thought that I had wasted my money on the remap! Looks like the torque/power curve (spread) is also different for an actual run - Not just the peak values.

Originally Posted by d3mon

I'm wondering what would happen if you dyno tested with the wheel speed sensors (for ABS) disconnected. Would it fail to be able to detect that it's on the dyno and not cut the power?

Originally Posted by Vid6639

Yes. This is what I was going to say. There was a video online which did the comparison with the rear wheel speed sensors disconnected. That way the ECU doesn't go into cheat mode.

Once the car was on the DYNO, most lights pertaining to ABS started blinking and the engine would not rev beyond 3000 RPM or so. Then Karthik (RD) asked me, if there was a switch to turn off ESP. There is no switch and also, I had not carried my VCDS. Obviously, this was not anticipated.

Then suddenly, all those lights stopped blinking and were permanently lit - After this the engine revved all the way till the red line. We thought that the ECU had recognised the car being tested, switched-off the ESP (ABS?) and changed the MODE.

Not sure what will happen if I turn-off ABS/EBD/EDL/ESP through VCDS, before DYNO. I can still do it; but the problem is the DYNO costs 3k after all discounts! I therefore thought that I will wait until I get a 4-wheel DYNO and try it.

Damn it! It is very interesting and tempting!!

Originally Posted by Vid6639

Secondly are the given power/torque outputs after correction?

Yes. Both Temperature (32 degrees centigrade) and Altitude (3000 ft) were taken in to account. Correction factor was roughly around 1.13

Originally Posted by Vid6639

Is the cheat mode a hidden map that the remap does not overwrite?

Ideally you wipe out the entire OE map and flash the new remap onto the ECU after taking a backup. So why will the car switch to cheat mode?

Hi Viddy, This is from my understanding:

Service / Manufacturing / DEMO / Testing MODE parameters can be part of the software. These parameters can be permanently frozen, as it need not take user requirement in to account. This can be hard-coded like any other programme sequence.

Parameters for actual driving can be stored as look-up tables. There can be different table for fuel intake / air intake / EGR valve position / torque limiter .... etc. Some of these tables can be programmed to learn and adapt to the driving pattern of an individual. These tables are generally referred-to as MAPS by the programmer.

In case of a ECU remap; Entire ECU software is first down-loaded. Only the MAPS are tweaked and then, the entire software is flashed on to the ECU.

MODE is untouched as it is part of the software / programme.

Please correct me if I am wrong.

Originally Posted by d3mon

Yes, you write the entire package after tweaking only a few maps.

You are right.

I see 3 steps.

1. Cheat mode: Peak power / torque is slightly limited compared to the stock rating. Torque band is considerably reduced.

2. Stock rating: Peak power / torque may be slightly better than the cheat-mode. But the power/torque spread more than makes up for the loss.

3. Remapped rating: This practically replaces the second point. Peak power / torque is boosted and the spread can also be altered to enhance driveability.

[Chethan B G]

Originally Posted by d3mon

I'm wondering what would happen if you dyno tested with the wheel speed sensors (for ABS) disconnected. Would it fail to be able to detect that it's on the dyno and not cut the power?

Hi d3mon,

Any idea on the type of signals that are received from the wheel speed sensors?

Logically it has to be a varying feedback rather than binary? It has to continuously sense the variation in speed.

If it is a varying feedback, disconnecting will not help as the ABS module would be comparing the signals from all the four wheels simultaneously.

If it is binary; lot depends on what the "healthy condition" is. To avoid problems with snap-out wires, "healthy condition" would have been designed to be the higher voltage. Therefore, It may not work if the connector is removed.

One way to check this is to detach the connectors and run the car for about a furlong - to check if the ABS lights come-on. Will try to do it during the weekend.

Other options is to turn it off through the VCDS. If I do this, obviously the errors won't pop-up as the CAN-Gateway stops communicating with the module. There-fore, logically ECU should not realise that the car is being tested and run the normal map.



Plan: I have spoken to Race-dynamics and requested them for an other session. Hope they do not charge me the same amount. But right now, their DYNO is down and is under repair. Will take about 3 weeks for it to function.


I have also spoken with Pete's for the GPS - VBOX - May get it in a couple of weeks time. Will check the following:

• 0-100 timing
• 30-100 in 3rd gear
• 40-100 in 4th gear

Not interested to check the top-speed, as it is practically un-necessary for the Jetta.

Will update once done.

[Gannu_1]

Originally Posted by Chethan B G

I have also spoken with Pete's for the GPS - VBOX - May get it in a couple of weeks time. Will check the following:

• 0-100 timing

This can be done using VCDS itself.



Ross Tech: Acceleration Measurement

[Chethan B G]

Originally Posted by Gannu_1

This can be done using VCDS itself.

Attachment 1429500

Ross Tech: Acceleration Measurement

Thanks!

This may have the speedo-error. Never the less, it will be a comparison between stock and remap with the same set of instruments.

Can the in-gear acceleration also be measured through VCDS?

[d3mon]

Originally Posted by Chethan B G

Hi d3mon,

Any idea on the type of signals that are received from the wheel speed sensors?

Logically it has to be a varying feedback rather than binary? It has to continuously sense the variation in speed.

If it is a varying feedback, disconnecting will not help as the ABS module would be comparing the signals from all the four wheels simultaneously.

If it is binary; lot depends on what the "healthy condition" is. To avoid problems with snap-out wires, "healthy condition" would have been designed to be the higher voltage. Therefore, It may not work if the connector is removed.

They can't be binary signals. They should measure individual wheel speed and hence should be analogue.

I have another idea - you can use an android app for measuring your real world horsepower and torque. I've used it quite a lot and it's eerily accurate. Posting a couple of torque curves that I've measured for my Ford Figo aspire and a swift.
I live in bangalore too, so if you are somewhere close by, we could meet up and perform this test together using my phone.

[Chethan B G]

Originally Posted by d3mon

I live in bangalore too, so if you are somewhere close by, we could meet up and perform this test together using my phone.

Thats excellent!

Will ping you right away! We can meet at your free time.


Does this app measure acceleration also? Is this the BT-OBD connector setup?

[d3mon]

Yes, it measures acceleration (0-100, quarter mile etc) too. It doesn't use the OBD - just the smartphone's accelerator and some fancy algorithms

[Chethan B G]

Update: PerfExpert Dyno and VCDS Real-Time-Log of Turbo-boost and Fuel-rail pressure.

PerfExpert is an android application which relies on the accelerometer of the smartphone, to measure the actual acceleration of the car. Based on the car data / atmospheric data provided by the user and the acceleration; Torque / Power delivered by the car is calculated.

Power and Torque at wheels - measured in 2nd Gear: Close to the figures claimed by Pete's. WIDE Power bandwidth!! Air filters are approximately 6K Kms old and were not cleaned before the test. With a clean air filter, these figures could go up by another 5%.




Real Time values of the Boost Pressure and the Fuel Rail Pressure, were logged through the VCDS in 3rd Gear. Could not find EGT on the VCDS, to obtain a real time log.





Note: I sincerely thank D3mon for accompanying me during these tests. His knowledge and inputs were of great help.

[hubolt]

Did you also have to also provide the estimated car weight in the app ?

[Chethan B G]

Originally Posted by hubolt

Did you also have to also provide the estimated car weight in the app ?

Yep. Car weight (Stock), Additional weight (driver + co-passenger if any), Car specifications, correction norm, units preferred and Weather (on-line), are the details to be entered in the app before the Dyno. Also, the results are with AC. Ideally, a Dyno is run with fresh / clean air filters and with AC switched off. This makes a difference of 5-10% in peak figures.

If done correctly; Though the peak power / torque can be correct within tolerance of + 2% (?) and the spread / band may be relayed on, accelerometer takes time to respond. This can be seen by the initial trend. IMO, in an actual Dyno run, the entire graph would be shifted by at-least 500 RPM towards the beginning.

This does not replace an actual Dyno, but results are supposed to be fairly comparable. This will do until I come across a 4-wheel Dyno!

Originally Posted by Chethan B G

Update: VCDS Real-Time-Log of Turbo-boost and Fuel-rail pressure.

Turbo-boost graph looks normal. Fuel-rail pressure raises immediately after 1000 RPM. Obviously, more fuel is injected before the turbo kicks-in. This could be the reason for the drop in FE in B2B traffic.

Fuel+Air mixture logic as I see; Inject more fuel. And then, provide more air to the extent that the fuel is completely burnt and the emission is minimised.

I do not have the graphs on stock power to compare. My inference is partially based on the experience in driving the car, before and after remap.


Dyno Map: Stock and Remap.





Main difference is with the way the power is delivered. As torque is mainly important to gain initial momentum, it is the power spread which gives constant acceleration once the car starts moving. IMO, this is the main reason the car drives so differently. It is not just about those extra horses, car drives much better than earlier. By saying this, I do not mean that the drive was bad earlier, by any standard!


0-100 timing was attempted, but did not succeed as I could not "launch" the car the right way - took 2+ secs for the launch itself. I did not want to extensively use WOT and get in to any problems - Hence, dropped the idea.

Stock 0-100 is rated at 9.7 Secs. After remap, based on the way it drives, timing would have reduced by 0.5 secs max? Not sure.