[graaja]

In this thread, I would like to write about an interesting Android App called “PerfExpert” that can be used to do dyno test of cars. From the data I got for my Jetta and Polo, and from the reviews by many other users online, this App is quite accurate in the dyno results.

Pros:

• At INR 650, this is a poor man’s dyno indeed
• Unlimited number of dynos and cars
• For an App that does not even use any data from OBD port, the results are quite accurate
• This App is equivalent to a four wheel dyno and the cheat software of the Jetta did not kick in

Cons:

• The results solely depend on the accuracy of parameters like weight, drag coefficient and gear ratios entered by the user. Any wrong data will result in totally wrong results. Finding these details could sometimes be quite difficult.
• The tests have to be performed under very tight conditions like straight and level road, less wind, consistent starting point, consistent WOT point etc. As it is difficult to have good control on all these parameters there is bound to be variations in results. The best work around is to do multiple runs - 3 or 5, and group the ones that are closest and take the average.

Thanks to:

I would like to thank fellow TBHPians @Chethan_B_G and @d3mon for bringing up this app in Chethan’s Jetta ownership thread.
I would like to thank fellow TBHPian @kat for pointing out my mistake on the final gear ratio which was producing wrong results for my Jetta Dyno

A few disclaimers:

• I do not claim PerfExpert is more accurate than a dyno.
• The theory of operation that I have outlined in the "How It Works" section is purely based on my assumptions based on fundamental laws of Physics. I will not be able to offer any explanation other than what I have already mentioned in this section.
• My conclusion that the App is accurate is based on the measurements I did on my Jetta and Polo GT TSi and how they compare to manufacturer's specifications. I have not tested any other cars with this App. Fellow users are free to share their experience.
• Finally, I do not have any affiliation with the author(s) of PerfExpert App

The Search for a Dyno

My search for a dyno started when I was thinking about remapping the Polo GT TSi. I wanted proper data to quantify the gains of a remap. But the tuners who I spoke to did not have a dyno of their own.

In November 2016, I came to know of a dyno in a place called Spitfire in Coimbatore (which happened to be very close to where I live). I took the GT for a dyno. I must say I was not very happy with the experience at all. First they did not have a printer to provide a hard copy of the dyno charts. Forget a printer for hard copy, they did not even have a USB port on their machine to export and save the report to an external thumb drive. I had to take a picture of their PC screen for the record. Here are a few pictures taken during this dyno.







The Polo GT TSi is rated to develop 105bhp of power and 175Nm of torque. Hower the dyno results showed a peak power of about 130 bhp and a torque of 210 Nm. This definitely is not possible. The guy who did the dyno kept insisting that the dyno is accurate and my car was indeed that powerful. As any argument was futile, I left it there. And I also postponed the plan to remap the car.

Note: After analyzing the results from PerfExpert and how results are dependent on weather data (more on this later), now I am sure the guy who did the dyno did not feed accurate weather data to the dyno.

Then a few months back, the VW software update for the Jetta started. Lots of users reported that there was a significant reduction in power and torque, post the software update and many fellow Jetta owners like Tanveer, iliketurtles and Mission_PGPX got their Jettas remapped at Pete’s.

GTO and Tanveer also tried to do a dyno of the remapped Jetta and a stock Jetta. But as the cheat software was kicking in on the two wheel dyno, the results were inconclusive. This is documented in the below post.

Jetta Dyno Comparision

The only solution was to have a four wheel dyno which was not available in Coimbatore, or some other tool that could measure power and torque. I started experimenting with Torque Pro and OBD adapter. This app does have dials for measuring horse power and torque, but I did not find a way to do a proper dyno plot.

That is when I remembered Chethan’s ownership thread where he had mentioned about using an Android App to measure power output when he did a Pete’s remap (I had not paid attention at that time as there was no need for a Dyno at that time). Chethan had posted the dyno results and had mentioned that the results were as expected. This is documented in the below post.

Chethan's Jetta Dyno Results

Here is the picture of the dyno result of Chethan's car. Picture Credit: Chethan_BG from his Jetta ownership thread.


I immediately decided to give this App a try and downloaded and installed the App from Google Play. The App costs INR 650.

[graaja]

How It Works

If you are bored by Physics skip this section

This is the excerpt from PerfExpert website on how it works.

PerfExpert relies on the accelerometer of your phone to measure the actual acceleration of your car. This requires that you always take off the car from complete halt (0Mph), which allows the application to compute very accurately the instantaneous speed and distance. PerfExpert can compute not only car performances in acceleration times (0-60Mph, 1/4mile...) but also horsepower and torque of your engine thanks to the information you previously provided in your car profile (total vehicle weight, gear box ratios, final drive ratios...).

Resulting from long research and development, techniques used by PerfExpert to handle acceleration and compute engine horsepower give reliable and realistic results.

Based on the above and the Physics we studied at high school, here is my own hypothesis (with help from my daughter to refresh my knowledge of Fundamentals of Physics)

Consider the following equations for Power and Force

Power = Force x Velocity
Force = Mass x Acceleration

Therefore
Power = Mass x Acceleration x Velocity

From the above, if I know the mass of an object and can measure its acceleration and velocity, the power at any given instance is the product of its mass, acceleration and velocity.

• The app measures acceleration using the accelerometer in the cell phone. Velocity can be easily calculated using same acceleration data. With these values, the App first calculates horse power.
• RPM is calculated by using the velocity and gear ratios.
• Torque is calculated by formula: Torque = Horse Power * 5252 / RPM

There will be other factors that will have to be factored in the calculations, like the rolling resistance offered by the road, drag, temperature, humidity etc. I guess the app estimates these from the tire size, drag coefficient numbers and weather data.

The horse power measured would be the actual horse power at wheels. The app would then apply correction factors for the loss in drive train to derive power and torque at the crank. I think even the regular dynos measure the power and torque at wheel and apply correction factors to derive the numbers at crank.

I decided to verify if my understanding of the App. If my understanding is correct, for the same dyno run, changing the weight, drag coefficient and weather data should result in change in horse power. The App allows modifying vehicle parameters after a dyno run and recalculates all the data. I took one of my Jetta dyno runs and edited weight, drag coefficient, temperature, pressure, humidity and gearbox ratios tabulated the final results for the change. Below tables show the result. The first column is with the results with accurate parameters.



As can be seen, weight, drag coefficient, temperature, atmospheric pressure and gearbox ratio can change the final results. Humidity did not have any effect.

This is the reason why accurate data is a must for accurate results.

• Car data like kerb weight and gearbox ratios should be available in the car manufacturer's website. Make sure you download the correct specifications and enter these values. At the time of the dyno, the app also asks for additional details like weight of the passengers on board, weight of the luggage, and the % of fuel in the tank. Provide accurate details in this step.
• The App downloads weather data from a weather site. When I did my Jetta test run, I found that the temperature value was not accurate, and I manually entered the temperature value from the car's MID. So always cross check the temperature fetched by the app with the actual atmospheric temperature. Most cars provide this value on their MID. The only other relevant weather data is atmospheric pressure. As getting a barometer could be difficult, I think we have to assume that the weather site provides a reasonably accurate atmospheric pressure value.

Once all the above data are provided accurately, another variable that affects the final accuracy is the accuracy of the accelerometer in the cell phone. With the improvement in technology, the accelerometers in modern cell phones are very accurate.

In summary, there are so many parameters that need to be entered to get accurate results. But with some searching all these data are available. With accurate data fed in, I found the App to be pretty accurate in the results with both my Jetta and Polo GT TSi, both tested in different locations and conditions.

[graaja]

Using The App

Installation of the App is quite straightforward. Download it from Google Play and complete the installation. On opening, there is a registration process to register at their website. This is useful to store car information etc. so that when you install the App in another phone with the same account, the settings get carried over to the other cell phone as well.

Before the App can be used, we need to enter the details of the car and do a calibration. Multiple cars can be added.

Setting Up the Car

This is a very important step. The accuracy of the results depend on the accuracy of the data entered in this step. Below are the parameters that need to be entered.

• Weight of the car
• Engine displacement
• Fuel
• Tank capacity
• Wheel dimension
• Cd.A (Drag Coefficient)
• Transmission
• Final drive ratio
• Pitch
• 2nd, 3rd and 4th Gearbox ratios (Alternately, speed at specific RPM may also be entered)

When we select the manufacturer, make and model, some of the information like weight, fuel, tank capacity etc are automatically populated from the App's vehicle data base. However, parameters like Cd.A, pitch and transmission ratios have to be entered manually (atleast I had to do this manually for both the Jetta and the Polo GT TSi).

This is where some searching will be required to get the specifications. I got the below specifications from VW UK website on the Jetta.

Jetta Technical Specifications.pdf

From this specification sheet:

Cd.A (Drag coefficient) = 0.30
2nd gear ratio = 2.05
3rd gear ratio = 1.30
4th gear ratio = 0.90
Final I = 4.12
Final II = 3.04

A note about Final I and Final II. DSG gearboxes have two final ratios the first ratio apply to a set of gears (1st to 4th) and the 2nd to the rest of the gears. I made a mistake with this and got some wrong results (more on this later).

A few screenshots of configuring the car.

Provide a name, and select make, year and model.


Notification that some of the parameters have been updated from database.


Parameters that have already been updated.


Parameters that need to be updated.


Edit gear ratios.


If gearbox ratio is not known, then there is an option to set the speed vs. RPM information as well. However, if the speed is not entered accurately, this will add error to the dyno results.


Select Pitch according to the suspension setup. I selected 3 for both the Jetta and Polo GT TSi.


The completed car profile.



Calibrating the Sensors

The App then asks to calibrate the sensors. This is a straightforward procedure. Below YouTube video shows the calibration.

https://youtu.be/CnLtzwCbVc4

Once the calibration is done, we are ready to do a dyno or a timed run (0-100 time, 1/8 mile time, 1/4 mile time)

[graaja]

Doing the Dyno

From PerfExpert website:

For reliable test results:
- Always launch the car from the same mark on the road
- Always WOT at the same mark on the road
- Take measurements at the same direction of travel
- Be sure to follow the PerfExpert basic measurement procedures (phone calibration performed, impeccable phone fixation, little wind, free road, lift windows...)
- If the car remained parked in a while then drive a few minutes at medium speed to cool the engine bay in order to always make the measurements under similar engine bay temperature conditions
- For each setup (plenum or open stacks), perform at least two measurements having less than 2% deviation to confirm their reliability. Indeed non-repeatable measurements (differences of more than 2%) reflect a poor phone fixation or poor conditions (road, wind...)

The first most important thing to take care is mounting the cell phone on the car. The mount should hold the cell phone firmly and there should be no vibrations. This is very important as any vibration in the phone will affect accelerometer data and hence the final result. Here is a picture of how I mounted the phone on my car.



This picture is just for reference. I did not do the dyno run on this road

Once the phone is mounted firmly, open the App, select the car, and a screen with option to do a dyno or a timed run shows up.

Selecting the dyno walks you through four more steps

1. Select the configuration. Each car can have multiple configurations (like Stock, Stage 1 remap, Stage 1 remap v2 etc.)
2. Select gear to run the dyno and minimum rev to start the log
3. Setup additional weight like weight of passengers, luggage and tank capacity
4. Weather data – we can either enter this manually or the app gets this from some weather website. During one of the runs I did for Jetta, I found this to have wrong temperature and I entered temperature manually from the temperature displayed on the MID.

Screen shots of the complete dyno run.

Select test mode.


Select tuning setup.


Select gear and minimum rev.


Additional Weights.


Weather data.


Once all the configuration is entered, the dyno screen shows up. Press the start button.


The App shows wait.
Important Note: The car should be in absolute stand still till the status changes to “Go”. Do not let the car creep forward when it says “Wait”.


Go.


Once the App shows "Go" Engage first gear and once on the move, go to second gear and go WOT (wide open throttle) by flooring the accelerator pedal. Once the RPM builds to red line release the throttle, slow down and stop. The app finishes the dyno and displays the summary of dyno result with peak power and torque numbers and asks if you want to discard or accept. Once you accept, the dyno results are displayed.

Dyno results page


Touch the graph in the results page to open the detailed graph.


Move your finger over the graph to see the data points.

Very Important Notes:

1. Safety is paramount.
2. Do not do this test on a road that has traffic. Do this on a straight road which is totally devoid of traffic.
3. While doing the run, do not look at the phone screen, but only the road

Clicking on the "Share" icon provides options to create an image of the test and share it using email, WhatsApp etc, or to share a web link of the results, or to export the data.

Below is the image that was created and emailed.


Below is the web link the App created for the test.
https://network.perfexpert-app.com/results/K6VpvUUuSH

Selecting "Data" creates a text file. As I could not attach the text file here, I opened it in Excel and saved as Excel spreadsheet. Attached below.
Jetta 2.0 TDI DSG - 2017.02.17 - 04.21.23.xls

Problems Faced and Solutions:

While testing the App, I faced two issues.

Auto Upshift Issue

The app requires the full run to be performed in one gear without upshifting. I used the manual mode in both the Jetta and the GT TSi to perform the dyno runs. Though in manual mode, the DSG upshifts when the RPM hits redline. When I first performed the test on my Jetta, I always went to this threshold and the DSG used to upshift to third gear. Below is the resulting graph.



It took me a few trials before I was able to let go the accelerator in time to avoid the upshift.


Wrong Gear Ratio

When I first did the dyno on the Jetta, I had used the Gear ratio of 3.04 which was listed as the Final II gear ratio in the specifications. Following was the dyno result with this ratio.



The torque was shown as 417Nm, which is not possible. Jetta is rated at a torque of 320Nm. TBHPian kat found the mistake. He rightly pointed out that the gear ratio was 4.12, not 3.04. Then I did some reading and found that the DSG6 has two final gear ratios. Final I which is 4.12 is for gears 1, 2, 3 and 4. Final II which is 3.04 is for gears 5 and 6.

The app allows to edit the parameters of a dyno run even after the dyno is completed. Once I updated the final gear ratio to 4.12, the result was accurate as expected.

[graaja]

Jetta 2.0TDi DSG



I did three test runs for the Jetta. I did this test on the Nanjangud - Chamrajnagar road early morning 4:20AM. The road was wide, flat and absolutely devoid of traffic. I took two results which were closest and found the average. Results are in the below table.



The exported picture with the test result.


Polo GT TSi



I did three test runs for the Polo GT TSi. Did these tests on an empty stretch of road near my house in Coimbatore. However, the road was not as straight as the Nanjangud - Chamrajnagar road where I did Jetta tests. And there was more variance in the results. Results are shown below.



The exported picture with the test result.



Conclusion:

After testing the App on my Jetta and Polo GT TSi, I consider this to be pretty accurate and a very helpful tool for enthusiasts who would like to measure the performance of their cars.

If you are looking to remap your car and would like to have accurate data to measure gains post remap (not the seat of pants dyno) and/or if you do not have access to a good dyno, then this App is for you. Even if you have access to a good dyno, doing multiple tests on a dyno could be very expensive affair especially when you are experimenting with the engine tune. You can use this App for unlimited number of "Work In Progress" measurements.

But before you download this App, to set your expectations right:

• The accuracy of the dyno results from this App depends on the accuracy of the data that is fed to the App. You need to feed in details like car weight, drag coefficient, gear box ratios accurately. If there is an error in this data, the dyno results will not be accurate.
• Weather data has to be fed accurately. Though the App takes weather data from some weather website, you will have to verify that the fetched weather data is accurate. If not, you will have to manually enter weather data.
• The dyno runs should be done under very tight conditions - straight flat road, little or no wind, consistent starting point, consistent WOT point etc. If any of these conditions are not met there is bound to be lot of variance in the readings.

Thats all folks, I have on the PerfExpert poor man's dyno.

I request other users who have used the App, or are going to use the app, to share data on Car Profile and test results in this thread so that this may be useful to everyone.

[SDP]

Thread moved out from the Assembly Line. Thanks for sharing!

[mayankk]

Lovely. Someone really sat outside the box for a think here. Great use of an existing tech, and even if it is not a hundred percent accurate, it does provide gaugable benchmarking and comparison. Thanks for this!

[Dr.Naren]

Wow, the results are pretty accurate. I had also tried using the app few months back. But I could never calibrate my phone ( the alignment thing). After trying few times, I gave up. Now I have got a new phone, would definitely try it again

[d3mon]

I've been using this for many years now and have done countless runs using it. I can completely vouch for its accuracy.

Just one note - Be very careful of the weather corrections - in your case, it's showing a correction factor of 1.08, which means that the entire curve is uplifted by 8%. The correction factor is valid for NA cars, but it's very dicey for turbocharged ones, as they can increase the target boost levels to bring back the power levels to a value that's very close to the power at sea level.

The solution to getting more consistent (& humbling!) measurements is to just disable weather corrections altogether.

[graaja]

Originally Posted by Dr.Naren

Wow, the results are pretty accurate. I had also tried using the app few months back. But I could never calibrate my phone ( the alignment thing). After trying few times, I gave up. Now I have got a new phone, would definitely try it again

Please do try and share your results. I am sure you will find it very useful for your 1.6

Originally Posted by d3mon

I've been using this for many years now and have done countless runs using it. I can completely vouch for its accuracy.

Just one note - Be very careful of the weather corrections - in your case, it's showing a correction factor of 1.08, which means that the entire curve is uplifted by 8%. The correction factor is valid for NA cars, but it's very dicey for turbocharged ones, as they can increase the target boost levels to bring back the power levels to a value that's very close to the power at sea level.

The solution to getting more consistent (& humbling!) measurements is to just disable weather corrections altogether.

Thanks. I did not know about the weather correction. I will do some more runs with weather correction disabled and share the results.

[Dr.Naren]

Originally Posted by graaja

Please do try and share your results. I am sure you will find it very useful for your 1.6

I tried the App again yesterday, but with my new phone. I had no issues with calibration this time . I found the gear ratios for S-Cross in Autocar magazine and got the CD from an Australian review. I disabled the weather correction and did a WOT run in 2nd gear ( dyno measuring from 2000 RPM).

I am not very sure about results. Power figures are more than my expectation and torque figures lower. I will try to do few more runs to get a better idea. As you all know, my 1.6 S-Cross is remapped and also running with Blackworks Decat downpipe

[graaja]

Originally Posted by Dr.Naren

I am not very sure about results. Power figures are more than my expectation and torque figures lower. I will try to do few more runs to get a better idea. As you all know, my 1.6 S-Cross is remapped

Good to hear that your phone calibrates fine now and you were able to test the App. To rule out accuracy issues, my suggestion is to do multiple runs and take the results that closely match (within 2% to 3%. Also, follow the conditions like consistent start point, consist WOT point, flat road etc. I am sure the readings will be accurate.

And to understand the actual gains from the remap, if you can go back to your stock map, do 5 dyno runs, put back the remap and do 5 dyno runs, you will get a better picture. But that may not be practical at this time.

Before I go for a remap of either the GT or the Jetta, I have decided to do at least 5 measurements of dyno before and after remap under same test conditions. If I get a safe road, I may also do 0-100 runs before and after remap.

PS: Regarding the difference between expectations and measured value, once you have proved that the data from the App is accurate, it is quite possible that the actual value is not as per your expectations

[d3mon]

Originally Posted by Dr.Naren

I am not very sure about results. Power figures are more than my expectation and torque figures lower. I will try to do few more runs to get a better idea. As you all know, my 1.6 S-Cross is remapped and also running with Blackworks Decat downpipe

Wow, very impressive. 164 BHP & 360 NM of Torque. What more were you expecting?
If this run was on a flat road, then everything looks good to me. BTW if you want better torque numbers, do a run in 3rd gear. I'm sure you'll see >380NM easily.

Almost 4.2 m/s2 peak acceleration in 2nd gear. To put that in perspective,

Swift DDIS - 3.2 m/s2
Ciaz DDIS - 3.3 m/s2
Figo aspire in P2 - 3.75m/s2
Rapid 3.8 m/s2
Jetta - 3.8 m/s2
Remapped jetta - 4.1 m/s2

Next time, please update the starting RPM to 1000. You can shift to 2nd at 10-12 KMPH and then just go flat out from there.

[Dr.Naren]

Originally Posted by d3mon

Wow, very impressive. 164 BHP & 360 NM of Torque. What more were you expecting?
If this run was on a flat road, then everything looks good to me. BTW if you want better torque numbers, do a run in 3rd gear. I'm sure you'll see >380NM easily.

Almost 4.2 m/s2 peak acceleration in 2nd gear. To put that in perspective,

Swift DDIS - 3.2 m/s2
Ciaz DDIS - 3.3 m/s2
Figo aspire in P2 - 3.75m/s2
Rapid 3.8 m/s2
Jetta - 3.8 m/s2
Remapped jetta - 4.1 m/s2

Next time, please update the starting RPM to 1000. You can shift to 2nd at 10-12 KMPH and then just go flat out from there.

Thanks for all the information. My remapped S-Cross figures look very impressing now . As I said the figures are not as per my expectation.

Expected figures : 145-150 HP / 380-390 NM
PerfExpert Dyno : 164 HP / 360 NM

I think the downpipe has made good difference and torque figures, I will check again in 3 Rd gear.

The run was done on this flat straight road in 2nd gear!!

[graaja]

Originally Posted by Dr.Naren

I will check again in 3 Rd gear.

IMHO, for a dyno test that is done on open road, 3rd gear would not be safe. My suggestion would be stick to 2nd gear which will have a safe max speed. If you have a totally closed road, then it should be ok though.