Introducing PerfExpert car database

on Wednesday, 16 March 2016. Posted in Announcements, Using PerfExpert

Fill in your car profile faster thanks to our car database

Setting the right values for all your car profile parameters is very important to get realistic results with PerfExpert, especially for Dyno tests. And most of the time, searching the web for all the information to fill in your car profile properly can be tedious. To help you in this task, we used to post car sheets information on our support forum when you requested it. Now, since PerfExpert for Android v2.0.1, you have the possibility to get your car profile completely filled with values coming from our car database, without the need to enter them manually.

Values from our database correspond to 100% standard cars (stock) with most common options. We encourage you to always review all parameters to ensure they match your own car, especially weight, wheel dimensions and pitch.

At the car profile creation, you can select your exact car model from our database. Don't worry if it's not listed : just choose "Other..." and tell us your car model the most accurately possible so we can update our database accordingly.

PerfExpert Car Database

PerfExpert Car Database

PerfExpert Car Database

Once the car profile created, it's pre-filled with values from our database. If some parameters are not yet available from our database, they will be left undefined in your car profile, and it's up to you to enter them manually if you know them, or to ask us for some help (via the "Need help..." button) if you don't know them. At this point, if you requested help from us, we will search for the right information to update our database and then we will update all the undefined parameters of your own car profile. So if you want to have your car profile automatically filled, we suggest you to leave all undefined parameters empty, waiting for us to update them.

PerfExpert - Car profile screen 

Don't forget to review parameters to ensure they match your own car, especially the following ones :

  • Car weight. As it depends on your car options and modifications, it's better to rely on the curb weight mentioned on your vehicle registration document.
  • Wheel dimensions. Simply check the size mentioned on your tires.
  • Pitch. If you've modified your car suspensions, you will have to adjust this parameter accordingly.

Car profile tips part 2 : using gear speeds instead of gear ratios

on Sunday, 05 July 2015. Posted in Using PerfExpert

How to use the new Gearbox feature to set gear specification without knowing their ratio

Car profile tips part 2 : using gear speeds instead of gear ratios

Since PerfExpert for Android version 1.7.0 there is a new Gearbox screen that is intended to help you filling your car profile parameters when you are not able to find any info about your gearbox ratios and your final drive ratio. This provide another way to set the gearbox specifications allowing PerfExpert to calculate the engine rev without any connection to the car. As you already know, PerfExpert calculates the engine rev from the car speed, so why not simply telling to PerfExpert what is the speed of the car when the engine is at a specific rpm value on a given gear ?

Here are some explanation about how to use this feature. Notes that when you choose to edit gears by specifying their related car speed, the Final Drive Ratio parameter and Gear Ratio parameter values are completely ignored by PerfExpert when calculating engine rev.

Case 1 : you know the car speed at 1000rpm

If you know accurately the car speed at 1000 rpm for each gear then you can choose "Edit by speed @ 1000 rpm" and enter the right values. Usually, you will find the speed at 1000 rpm information for your car in repair manuals or on car sheet websites.

Case 2 : you don't know the car speed at 1000rpm

If you can't find neither the speed at 1000 rpm nor the gear ratio value, then you can use a GPS to help you find the right information. The purpose is to measure the exact speed of your car at either 2000 rpm (better for diesel engine) or 3000 rpm (better for gasoline engine) on the given gear.
Let's say we have a gasoline engine and we want to enter the 2nd gear info in PerfExpert. To do so you will have to drive the car at 3000 rpm on the 2nd gear and to stabilize the car speed in order the GPS can provide a fixed speed value for that engine rev. The more accurate the GPS speed will be, the more accurate the calculated engine rev will be. Once you managed to get a fixed speed from the GPS :

  • Go in the Gearbox screen
  • Select "Edit by Speed @ 3000rpm"
  • Touch the 2nd gear speed field. A checkmark button will appear at the top of the screen.
  • Enter the GPS speed value for the 2nd gear
  • Validate this value by touching the "checkmark" button at the top of the screen
  • Validate the new gearbox infos by touching the "OK" button at the top of the screen

Edit gear speed

This technique is an alternative way to have your car profile filled if you can't find info about your car in any documentation. When gear ratio or speed at 1000 rpm values are available, they will give more reliable results than using this GPS technique.

Intake comparison on 4AGE 20V BlackTop

on Tuesday, 07 February 2012. Posted in Using PerfExpert

Performances difference between the original plenum and open velocity stacks

Here's a little comparison that has been done on Stephane's Toyota AE82 Twin Cam GT : seeing with PerfExpert the difference in torque and power between the original plenum (without the air filter) and open velocity stacks.

To perform reliably this test it was necessary to proceed as following :
- 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...).

 

Toyota Corolla AE82 Twin Cam GT

 

At first, measurements with T3 open stacks

Toyota 4AGE 20V blacktop open stacks

PerfExpert 4AGE 20 blacktop cornets 1

PerfExpert 4AGE 20 blacktop cornets 2

 

Here are the results with this setup :
- 159.6hp at 7300rpm et 131.8Ft.Lb at 5500rpm
- 159.1hp at 7300rpm et 133.2Ft.Lb at 5300rpm

 

 

Lets see now the results with the OEM plenum and OEM stacks. Unfortunately we didn't have a cold air feed to the plenum for the tests. And note that the curved OEM stacks had been cut by the previous car owner...

 

4AGE 20V blacktop plenum 14AGE 20V blacktop plenum

PerfExpert 4AGE 20V blacktop plenum 1

PerfExpert 4AGE 20V blacktop plenum 2

 

Here are the results with this setup :
- 150.4hp at 7200rpm et 128.2Ft.Lb at 5200rpm
- 151.3hp at 7200rpm et 130.4Ft.Lb at 5200rpm


This shows a clear advantage of open stacks in power (more than 5%), and a slight advantage in Torque. But the most interesting is to see the curves comparison. Here is a Google Doc with an overlay made ​​from a fusion of data exports made ​​from the application :

Toyota 4AGE 20v blacktop : open stacks VS plenum

Not enough room for the legend so I'll give more details here :
- P1 & P2 are the 2 power curves in open stacks setup
- T1 & T2 are the 2 torque curves in open stacks setup
- P3 & P4 are the 2 power curves in plenum setup
- T3 & T4 are the 2 torque curves in plenum setup

Note that with the plenum there is a little more torque before 4000rpm and less torque after. So really, even if the open stacks give more torque maximum value, the plenum give a more torquey engine at low rpm.
A pity that we didn't have a 100% stock plenum (whose stacks are not cut) and the original airbox + filter. The comparison would have been even more interesting.

Highlight your PerfExpert vs Chassis Dyno comparison

on Monday, 22 June 2015. Posted in Announcements

Send us your comparisons to better convince the most skeptical ones about PerfExpert accuracy

Highlight your PerfExpert vs Chassis Dyno comparison

Most people find it hard to believe that a smartphone application can measure power and torque with an accuracy very close to chassis dynos, without any connection to the car or without any external measuring equipment. They generally think of a hoax, because they cannot imagine the real potential of the modern phones sensors, especially when they are combined with powerful algorithms and calculations. Thus, providing them the proof with "PerfExpert vs Chassis Dyno" comparisons seems to be the solution.

What could be better than seeing this proof brought by other car enthusiasts who tested PerfExpert by themselves?

If you carried out this kind of tests you can contact us to transmit us your results, by sending us the following elements :

  • A copy of the full report of the chassis dyno result, in other words, the result sheets, perfectly readable, scanned or photographed
  • An Image export of PerfExpert measurement curves, and, if possible, with two PerfExpert tests carried out under the same conditions (same road, same direction, same starting point, same day)
  • The links towards your tests on the Web PerfExpert Network
  • And for the fun, if you agree, a beautiful picture of your car, outside and/or on the chassis dyno! :) We will hide the license plates if they appear.

We will gather them in our Google + collection of PerfExpert vs Chassis Dyno comparisons.

Speaking about PerfExpert around you and showing that it is a reliable application when it is correctly used allows us to keep on improving it and creating new features! As a user, you now know how to answer those who ask you about the application's accuracy and reliability compared to chassis dyno's. In fact, this is often the first question people ask when they discover that the PerfExpert application allows them to measure your car engine’s power and torque using only your smartphone!

Here is an comparison example from our collection :

PerfExpert Car Dyno=> Power : 301cv / Torque : 370Nm

Cartec Chassis Dyno => Power : 300.5cv / Torque : 371Nm

Nissan 350Z

Nissan 350Z chassi dyno result

Nissan 350Z PFX result 

 

Join PerfExpert

on Saturday, 10 January 2015. Posted in Announcements

Contribute with us in building things that matter

PerfExpert is a startup that provides solutions for car performance measurement and analysis. We build products that help car enthusiasts and professional tuners to better understand cars behavior and we are looking for bright people to help us in this task. We are located in the beautiful Martinique island. You are a car enthusiast ? You like to write code ? You feel good when you solve problems ? Then contact us.

UX & UI Designer

To work on our Android and iOS native apps and on our PerfExpert Network web app.

Job Responsibilities

  • Reinforce and maintain the visual identity of PerfExpert.
  • Improve the UX of all our apps.
  • Design specific UI by following different platforms guidelines (Android, iOS and web).
  • Ability to work in a very small team.

Job Requirements

  • Strong vectorial and bitmap design skills.
  • Experience in designing native mobile apps and web apps.
  • Create reusable widgets
  • Startup mindset.
  • We are also looking for interns!

Mobile App Developer

To work on our Android and iOS native apps.

Job Responsibilities

  • Create new awesome features .
  • Integrate UI and UX.
  • Ability to work in a very small team.

Job Requirements

  • Strong Android or iOS programming and debugging skills.
  • Experience in developing native mobile apps.
  • Startup mindset.
  • We are also looking for interns!

Full Stack Web App Developer

To work on the PerfExpert Network web app.

Job Responsibilities

  • Create new awesome features.
  • Integrate UX and UI.
  • Ability to work in a very small team.

Job Requirements

  • Strong Javascript programming and debugging skills.
  • Strong HTML5/CSS skills.
  • Experience with Node.js, Angular.js or Parse.com would be appreciated but not required
  • Experience in developing web apps.
  • Startup mindset.
  • We are also looking for interns!

Support & Community Manager

To help our users to have the best possible experience with our apps.

Job Responsibilities

  • Provide customer support via Email and via our Support Forum.
  • Help users to better setup and use our apps.
  • Provide car performance related content on our Blog.
  • Interact with the PerfExpert community on social networks.

Job Requirements

  • Car enthusiast.
  • Good general knowledge about cars and their technical specifications.
  • Ability to write in proper English.
  • Know what is a gear ratio.
  • Know what is a torque curve.
  • We are also looking for interns!

Interested ? Contact us !

Get your tests featured on the PerfExpert Network

on Friday, 23 January 2015. Posted in Using PerfExpert

Have a chance to be featured by sharing your car performance with the PerfExpert community

Get your tests featured on the PerfExpert Network

Some of you may noticed that their test results has been displayed on the Featured Test Results page of the PerfExpert Network. Indeed, we regularly pick out some users tests that we consider well done and reliable, to show them, as true using examples, to the rest of the PerfExpert community.

We can select your test result only if it meet the following criteria :

  • You have set an avatar for your car
  • Your car description is complete and correct. This mean the make, model and year matches the car we can see on the avatar photo.
  • You exhaustively described the Setup associated to the test. This mean you described all the modifications of your car if it's not stock.
  • Test results values are realistic. This mean they are in adequacy with the described car and setup.
  • Graph curves are smooth and their shape is realistic. This mean that you used a good phone mounting and that the road was smooth, flat and plane.

Wanna share your car performance with the entire community ? Ok, then let's see how you must proceed :

1 - Share your test results using the Web method

The first step is to share your test, from our Android or iOS Dyno & Timed Run app, to your favorite Forums, WhatsApp/Facebook groups, Google+ community, to Twitter or whatever social network you use, by using the Web method :

Golf R32 Turbo dyno curves perfexpert

Golf R32 Turbo dyno test curves share perfexpert

This will upload your test result to the PerfExpert Network, and will allow those with whom you shared the test to view it.

2 - Allow everyone that browse the PerfExpert Network to see your test

By default your test is visible only to those with whom you share its web link. To allow everyone that browse the Explore part of the PerfExpert Network to see your test, you have to make it Explorable. To do so :

Golf R32 turbo perfexpert dyno test results

Golf R32 turbo perfexpert dyno test results Explorable

Once you set your result to Explorable, it will be immediately viewable on your car page, but will need to be reviewed by us to appear on the Featured page.

Introducing Dyno Tests Comparison

on Friday, 09 January 2015. Posted in Using PerfExpert

Make comparisons of your test results using the PerfExpert Analysis Tool

Introducing Dyno Tests Comparison

You requested it and it's finally there : you can now see several dyno test results overlaid on the same graph. This is very useful when you want to find which of your car's setups produce best performance, as we seen on the Corolla AE82 20V intakes comparison and the Mercedes 190E Turbo sizes comparison. This is possible thanks to our new Analysis Tool by using your mobile or desktop internet browser. This tool is available only for results that are online on the PerfExpert Network.

To compare your results :

1 - Log in to the PerfExpert Network and go to 'My PerfExpert'

If you don't have a PerfExpert Network account, you have to register by using our mobile Dyno & Timed Run app : go to 'Account' in the Test screen menu and follow instructions.

2 - Open the desired car to see its test results

Here, you will only see results that have been uploaded from the mobile app, either by clicking on 'Open web link' on the Result screen of the the mobile app or by sharing via the 'Web' method.

Results

3 - Select the results you want to compare and click the 'Compare' button that appears at the bottom-right of the screen

You can select up to 3 results for comparison. For now, only dyno test results with the same units are comparable.

Results - selection 

4 - See results comparison at each point of the curves

Comparison

If you have questions or feedbacks, feel free to contact us.

 

Car profile tips part 1 : weigh your car for more accurate dyno results

on Friday, 03 October 2014. Posted in Using PerfExpert

How to better fill your car profile to obtain realistic results

Car profile tips part 1 : weigh your car for more accurate dyno results

Although PerfExpert can be used to evaluate the mechanical changes of a car by making comparison of measurements before and after modification, PerfExpert can also be used to just check the actual engine horsepower realistically. To do so, not only you must ensure that you perform your tests in the required conditions, but you must also ensure that the car profile parameters you entered are right.

The weight of your car is one of the most important data PerfExpert takes into account when computing dyno test results, as it has a direct effect on the displayed horsepower and torque. In this blog post, we will see how to fill and manage both the 'Car weight' and the 'Additional weight' parameters in the car profile, and how to weigh your car for more accurate dyno results. The other car profile parameters will be covered in future blog posts.

That car used as example here is a Toyota GT86, checkout its PerfExpert Network page here.

Use the 'Car weight' parameter as a contant value

The best way to use this parameter is to consider it's a constant value. That mean you will not alter this value each time you have to test your car.

You don't know exactly how much your car weigh

Ok, don't worry about that, you will be able to start using PerfExpert by considering the advertised curb weight of your car as the 'Car weight' parameter, and later update your test results after weighing up the car.

The curb weight is the "total weight of a vehicle with standard equipment, all necessary operating consumables (e.g., motor oil and coolant), a full tank of fuel, while not loaded with either passengers or cargo" as explained in the Wikipedia definition. This value is provided by the car manufacturer and can be found on the car manual, on car database like ZePerfs.com, and on the PerfExpert support forum, in the dedicated section.

As mentioned in the Wikipedia article, some countries may have slightly different definitions considering that the fuel tank is filled at 90% for example, or that this weight also include a 75Kg passenger. Also, some manufacturer just advertise a base curb weight, that mean the curb weight of the vehicle with only the base trim features installed, not the optional equipment (sunroof for example), as explained in this interesting post

As you can see, although the advertised curb weight is a good value to start with, it's still uncertain and this affects your test result accuracy. So if you want to get the actual engine horsepower realistically, you will need to weigh your car at some point. Then you will be able to modify the 'car weight' on each of your previously performed test, so PerfExpert can re-compute the horsepower/torque and update them.

You have the opportunity to weigh your car

That's the way to go the get more accurate results ! This can be done at a motorsport garage/club using specific weighing scales.

Black Toyota GT86 weighing

You must ensure that you meet certain conditions when weighing your car :

  • The car must not be loaded with either passengers or cargo (not even the driver)
  • All liquids must be level (engine oil, coolant, brake fluid...)
  • Fuel tank must be full

In other words your car must be in the conditions of a "curb weighing" considering the first definition that we saw above.

Black Toyota GT86 weighing

The measured weight must be entered in the PerfExpert car profile as the 'Car weight' and must not be altered until the car is modified.

Toyota GT86 weight

Here we get 1238Kg for a french 2013 Toyota GT86 that has the following modification :

  • Milltek catback resonated exhaust
  • Cosworth air filter in OEM air filter box

Use the 'Additional weight' parameter as a variable value

Each time you will perform some dyno test, you will have to adjust the total weight of the car at the time of the measurement. To do so, you just have to modify the 'Additional weight' accordingly.

Additional weight = (passengers weight) + (cargos weight) - (missing fuel weight)

Here we are considering that the car has not been modified since the time you entered the 'Car weight' parameter, so no need to alter it. Also we are considering that the engine oil, coolant and other fluids are level.

You have to calculate the missing fuel weight from the fuel tank capacity, the actual amount of fuel at the time of the test and the fuel weight per liter :

  • Gasoline : 0.755 Kg / L
  • Diesel : 0.845 Kg / L

For example, if at the time of the test of a gasoline car that have a fuel tank capacity of 50 Liters, we have inside the car :

  • The driver (75Kg)
  • A passenger (82Kg)
  • A tool box in the boot (15Kg)
  • Fuel tank filled at 45% (so it misses 55% of fuel to fill the fuel tank)

Then Additional weight = 75 + 82 + 15 - (55%*50*0.755) = 151Kg

Hope all these tips will help you get more accurate results !

Turbo sizes comparison on Mercedes 190E 3.0 24v Turbo

on Wednesday, 26 February 2014. Posted in Using PerfExpert

Turbo spooling up differences

Turbo sizes comparison on Mercedes 190E 3.0 24v Turbo

One of the main goals of PerfExpert is to help you see the impact of your mods on performance when doing engine tuning, as you can see in the 4AGE 20V intake comparison performed on a Corolla GT AE82.

Here it's about another PerfExpert user's car : Georgi's impressive Mercedes 190E. It has a 3.0L 24V engine swap converted to turbo at 0.9bar boost, managed by a MegaSquirt standalone ECU with 660cc/min injectors.

We will see how he used PerfExpert to compare different turbo sizes on his car, when he was looking for the one that will provide the best spooling while maximizing power and torque.

Note that in order to perform accurate comparison, Georgi has carefully followed PerfExpert mounting requirements and measurement conditions. Check out the DIY phone mounting article for more information.

In the app, he created three Setups : one for each turbo. This allows better organization by grouping together results that belong to a specific turbo. Also, he took care of using exactly the same boost pressure, and not changing other parts in the engine or in the car : only the turbo. This ensures that the changes in power and torque are only due to the turbo swapping.

Dyno result with the T66

The first setup was with a T66 (A/R 1.15) turbo :

Mercedes 190E 3.0 24v turbo T66 DynoTest 

Max power : 451hp @ 6300rpm
Max torque : 520Nm @ 5900rpm

Not satisfyed because of the turbo lag, Georgi decided to replace this big T66 turbo with a smaller one. The purpose was to get full boost at lower rev. In other terms, he wanted to improve turbo spooling.

Dyno result with the smaller turbo

The 2nd setup was with a smaller turbo, that is a hybrid turbo built using Holset turbine and compressor wheels inside Garret housings (A/R .82) :

Mercedes 190E 3.0 24v turbo Hybrid Dyno test

Max power : 452hp @ 6300rpm
Max torque : 537Nm @ 4600rpm

As you can see, with the hybrid turbo, max power value and max power rev are still exactly the same, but max torque is acheived 1300rpm earlier and is a bit higher than with the T66.

Dyno result with the T6682

Unfortunaletly the hybrid turbo didn't last long and failed after a few weeks, so it was time to fit back the T66, but this time with a A/R .82 turbine :

Mercedes 190E 3.0 24v turbo T6682 Dyno test

Max power : 479hp @ 6200rpm
Max torque : 581Nm @ 4700rpm

Results are very good here : same spooling as with the small turbo (hybrid one) but with more power and torque! With still 0.9bar boost!

T66 vs Hybrid vs T6682 comparison

Now let's have a more detailed look at these different dyno results :

Mercedes 190E 3.0 24v turbo T66 vs Hybrid vs T6682 Dyno comparison

The T6682 setup is definitely the best one through the entire engine rev range.

This comparison has been made using OpenOffice spreadsheet desktop application from data files exported by PerfExpert mobile app (Tabulation Separated Values files).

The PerfExpert comparison feature is still in development and will be available in a few weeks on the PerfExpert Network. It will allow a user to overlay several dyno or timed run results on the same graph inside a web page.

For more details about Georgi's car, check out its topic on the PerfExpert Forum.

Mercedes 190E 3.0 24v turbo

DIY phone mount

on Friday, 25 April 2014. Posted in Using PerfExpert

How to perform consistent measurements easily using the cheapest phone mount solution

DIY phone mount

To perform consistent, accurate and reliable Dyno and Timed Run tests with PerfExpert, you must ensure that all the measurement conditions are met.

One of these conditions is that the phone must be one with the car during the test : its sensors must perfectly "feel" the car's movements, and no other parasite movement or vibration. So you must pay close attention to the phone mount you use, and check that the phone position reach the PerfExpert mounting requirements. In the dyno test video tutorial, the phone is secured by a very strong windshield mount, equipped with a large suction cup and a very short, solid arm.

There is a much cheaper way to have your phone correctly secured in your car, and that allow you to immediatly perform tests : using the original box of your phone, the passenger seat, and a heavy book!

Remove the cover of your phone's box and all unnecessary packaging

DIY Phone mount 1DIY Phone mount 2DIY Phone mount 3DIY Phone mount 4DIY Phone mount 5

Put the phone inside the box and shake it sideways to check if the phone can move

DIY Phone mount 6DIY Phone mount 7

Prevent any phone movement by wedging it inside the box

DIY Phone mount 8DIY Phone mount 9DIY Phone mount 10

Make the power button accessible if needed

DIY Phone mount 11DIY Phone mount 12

Put the box on the passenger seat, and wedge it against its backrest

DIY Phone mount 13

If needed, you can stall the box with heavy objects located on its sides and/or in front of it

DIY Phone mount 14

Ok! You're ready to test your car!