Tiger Sport 1050 Forum

Please login or register.

News: Warm welcome to our new members !


Author [ES] [MY] [PL] [PT] [IT] [DE] [FR] [NL] [DK] [NO] [GR] [TR] Topic: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner  (Read 21009 times)

0 Members and 1 Guest are viewing this topic.

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
There is a lot of speculation regarding the best tunes for different configurations. Unfortunately, with a few rare exceptions, there has never been a systematic test of performance on any bike, and no repeatable test of fuel efficiency.  The tools to do so have until recently been cost prohibitive. 

I recently purchased a Bluetooth capable OBDII scanner.  This scanner links up with a Bluetooth capable device (in my case my cell phone) to transmit and record all OBDII data. Since it also uses a cell phone, you also get GPS and accelerometer data as well as the capacity to record video with data overlays.  The smart phone is relatively expensive, but was already purchased.  The scanner is $20, the program is $5.   

Some of this data is more accurate than others, but it is all very precise.  For instance, I have been recording MPG vs. Speed.  The Speed is very accurate (it's GPS and gives a correct speed) and precise (the speed says 65 everytime it actually is 65).  However, the MPG is precise, but inaccurate.  It's off by a factor. The program allows you to account for that factor, but since my intent is to measure power and efficiency of tunes relative to each other, I don't need the mpg to be accurate.  In other words, if the MPG says that A is 20 MPG while B is 21 MPG, that is fine as I only need to know that B is more efficient even if the true consumption is 45 MPG. 

I plan on doing 6 tests over 7 configurations. 
Test 1:Average Minimum Fuel Consumption    //a simple measure of fuel consumption
Test 2:Speed of Minimum Fuel Consumption    //an indirect measure of changes to brake specific fuel consumption
Test 3:Calculated Horsepower         //to determine any changes in high RPM power (not actual horsepower, a calculation based on acceleration through the rev range)
Test 4:60-80 6th gear, about 3400-5000 rpm   //to measure changes in lower RPM power
Test 5:Short Term Fuel Trim          //a measure how accurate the tune is to the demands of everyday driving
Test 6:Subjective impressions         //just what it sounds like

My 2011 1050 SE has a S3 cut snorkel, FUEL brand exhaust (straight through with no baffle), CAT INSTALLED w/SAI intact, K&N filter coming soon, Chevron 91 AKI fuel, 10W-40 AMSOIL oil, stock sprockets, PJ1 chain lube, Michelin PR3's with 8000 miles on them.

Configuration 1: Stock air filter on my bike with TOR tune 20549
Configuration 2: K&N air filter on my bike with TOR tune 20549
Configuration 3: same with Indy's Tune on base 20550 (created and used with Indy's permission)
Configuration 4: same with Indy/AK hybrid tune on original base
Configuration 5: same with Stock Tune 20548
Configuration 6: same with Dave's Tune V2
Configuration 7: same with Arrow Race Tune 20546

I expect each round of testing to take about a week, and I've already started on Config 1, so stay tuned!
« Last Edit: 03 March, 2014, 05:06:26 AM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #1 on: 22 January, 2014, 07:18:24 AM »
Configuration 1, TOR tune with stock air filter:

Test 1: Over 3 different tests the fuel mileage peaked at 30.31.
Test 2: The peak at 30.31 was at 61 mph +1-0.
Test 3: Over 3 tests I got max of 80.8, avg of 79.8, and a standard deviation of 1.7.
Test 4: 3.0s was pretty consistent over 4 attempts. Of note, I did do the same test in 5th 4300-5700 and 4th gear 4800-6700, but they were both very close to 3s as well.
Test 5: Over 3216 non-zero samples, the average short term fuel trim was -8.26% with a standard deviation of 2.36. The one other fuel trim PID was similar. I removed the 0's because though some may indicate an accurate VE map, others are simply due to being in open loop.  Choosing non-0 values removes the noise.   
Test 6: I had the Arrow race tune on before the TOR.  It was a pretty noticeable difference and as I've said before I liked the Arrow better. Somehow the engine seemed to have a throatier sound, but I really don't understand how the tune would cause that.  The arrow tune seemed to have a bit better response, and I think it had better mileage too, but I guess we'll see.  There isn't too much to say yet, as this is really the base tune that others will be compared to. 

Note: I did change test 4, there was just no way to safely do even the 1/8 mile test, and it was a duplicate test anyway.  The 60-80 in 6th gear is a very realistic and repeatable test, though I admit that I'm afraid that it won't have the resolution to determine any differences. 
« Last Edit: 17 February, 2014, 06:06:15 PM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #2 on: 22 January, 2014, 07:18:37 AM »
Configuration 2, TOR tune with K&N air filter installed:

Test 1: Over 3 different tests the fuel mileage peaked at 30.27.
Test 2: The peak of 30.27  was at 63 mph +1-0.
Test 3: Over 3 tests I got max of 88.8, avg of 87.6, and a standard deviation of 0.98.
Test 4: 3.0s was pretty consistent over 7 attempts.
Test 5: Over 13221 non-zero samples, the average short term fuel trim was -8.67% with a standard deviation of 2.48.
Test 6: I really noticed no functional difference with the K&N filter in, with the exception of the sticker that is now on my top case, next to the Race Tech one.  I was impressed with the additional top end numbers though.  The rev limiter did seem to come noticably quicker once in the higher rpm range, and the numbers back my feeling up. However, there is a caveat that is beyond the scope of this post. 

Note: I adjusted the way I figured the mileage. I went back and changed the first configuration results accordingly. Since the peak was always in the same range (55-65), I gathered all the Instant MPG figures for each MPH, found the mean, average, and standard deviation. This changed the peak figure (which was never actually the "peak", because the peak comes at high speeds with the clutch pulled in) and showed a slightly different speed result. 
« Last Edit: 17 February, 2014, 06:07:14 PM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #3 on: 22 January, 2014, 07:18:52 AM »
Configuration 3, Indy's Tune:

Test 1: Over 4 different tests the fuel mileage peaked at 29.98.
Test 2: The peak of 29.98  was at 57 mph +1-0.
Test 3: Over 7 tests I got max of 94.78, avg of 85.7, and a standard deviation of 6.93.
Test 4: Over 10 tests I got a average of 3.46 seconds with a standard deviation of .53 seconds.  Minimum 2.42 maximum 3.87.
Test 5: Over 23616 non-zero samples, the average short term fuel trim was -8.64% with a standard deviation of 3.86.
Test 6: The first thing that I notice is that it was really, really hard to get consistent results with this tune. I ended up using it for 6 days spanning 600 miles and took over 1,371,684 data points both commuting 30 miles a day and on a backroads mountain ride spanning 280 miles.  I am not a fan of the tune, but it's hard to pinpoint why. I think part of my dissatisfaction with the tune was a slow pick up from idle.  I stalled it once, and otherwise found a lack of throttle precision at 0-6% TPS. I've decided to go from this tune to the stock tune to hopefully get a better feel for the differences between Indy's and the Stock tune. It is very powerful on the top end, for sure. The front wheel has a tendency to lift off the ground during the HP runs, but not every time. I got excited when I did an unofficial 60-80 and saw 2.5 seconds, but repeated tests failed to materialize any repeatable figures. It does seem to get decent mileage, especially considering the power. The ride I went on got no worse mileage than is typical for less rich tunes. It's just really hard to pinpoint why the power numbers have so much inconsistency compared to the previous two configurations.  While messing around with the numbers, the Fuel Trims correlate with the RPM's in general, but the correlation is much stronger with the TOR tune. 

Some people have asked for Indy's tune laid on top of the 20550 instead of the older base tune, and as I have already gotten permission I'll post it here. 
« Last Edit: 17 February, 2014, 06:07:50 PM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #4 on: 22 January, 2014, 07:19:09 AM »
Configuration 4, Hybrid AK/Indy's tune:

Test 1: Over 10 different tests the fuel mileage peaked at 29.54.
Test 2: The peak of 29.54  was at 56 mph +1-0.
Test 3: Over 3 tests I got max of 87.64, avg of 86.77, and a standard deviation of 1.31.
Test 4: Over 4 tests I got a average of 3.03 seconds with a standard deviation of .05 seconds.  Minimum 2.99 maximum 3.10.
Test 5: Over 50448 non-zero samples, the average short term fuel trim was -5.14% with a standard deviation of 2.78.
Test 6: This tune seemed a lot easier to handle than Indy's, and actually brought out some of what I didn't like about that tune into the light and I think will allow me to express it a little better.  This tune has better off-idle response, but seems to have a heck of a surge at low speeds, where below 20 mph or so the fuelling is on/off and it feels like I'm riding a rocking horse.  It seemed worse the first day, but it was still quite noticeable on the last day I tested it. It was more consistent, where I was able to get repeatable numbers for the acceleration tests a lot easier. It seemed to guzzle gasoline, but it only showed up in the higher speeds, and I think that it may be due to the way the older version of the base map handles the closed loop mode. Unlike Indy's tune, this tune stayed in closed loop a lot easier. The data shows that it is the lowest mileage I've tested so far, but it went through gas a lot more and showed much lower MPG on the clock than the data shows.  I spent so much time using this tune in part trying to make sure what I was experiencing was accurate in terms of fuel consumption.  So, in short, I liked the smoothness of it, but there are critical flaws in the tune which I believe are inherent in the base tune but not necessarily in the modifications to the base tune. 
« Last Edit: 28 February, 2014, 02:40:55 AM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #5 on: 22 January, 2014, 07:19:22 AM »
Configuration 5, Stock Tune:

Test 1: Over 4 different tests the fuel mileage peaked at 30.84.
Test 2: The peak of 30.84  was at 64 mph +1-0.
Test 3: Over 7 tests I got max of 88.97, avg of 84.5, and a standard deviation of 4.36.
Test 4: Over 9 tests I got a average of 3.60 seconds with a standard deviation of .35 seconds.  Minimum 2.98 maximum 3.90.
Test 5: Over 26167 non-zero samples, the average short term fuel trim was -0.714% with a standard deviation of 2.70.
Test 6: This tune was noticeably less powerful, especially coming from Indy's tune, which so far had the best top end. It had a lot less overrun popping too. I didn't have the throttle issues anymore, it was much quicker to rev up from idle, and as such it was a lot more fun to ride. I really had to nurse the throttle with Indy's tune.  I tried to watch the Closed Loop operation a bit better this time.  The closed loop acts much differently than with Indy's tune. It stays in closed loop a lot more, and has really easy transition between open and closed loop.  I suspect that this has something to do with the fact that the STFT is really close to 0, so when it transitions to and from closed loop there isn't a sudden jump in fueling.  It seems like it sounds different...quieter almost.  I like the throatier sound of the richer tunes, but I do like the way this tune drives around town.  It's a dog on the throttle though.  It seemed to get slightly better mileage at the point in my commute where I slow down to 30 mph and start filtering through traffic.  It was a lot easier to get consistent numbers this time around. I'm not sure why the tune would affect that, but I was relieved that I wasn't doing something to cause all that variability. 
« Last Edit: 17 February, 2014, 06:08:49 PM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #6 on: 22 January, 2014, 07:19:44 AM »
Configuration 6, Dave's Version 2 Tune:

Test 1: The fuel mileage peaked at 29.33.
Test 2: The peak of 29.33 was at 60 mph +1-0.
Test 3: Over 10 tests I got max of 85.96, avg of 75.53, and a standard deviation of 5.20.
Test 4: Over 7 tests I got a average of 2.998 seconds with a standard deviation of .057 seconds.  Minimum 2.918 maximum 3.03.
Test 5: Over 49707 non-zero samples, the average short term fuel trim was -1.073% with a standard deviation of 3.34.
Test 6: Ok, so I know I've said before that I don't really experience throttle snatch, but after riding Dave's tune V2 I think I may really know what smooth low RPM throttle really means! There didn't seem to be a lot of overrun popping, which I thought was strange, and there also seemed to not be a lot of engine braking, as if it was fueling a bit more under higher vacuums and thus not creating as much of a vacuum.  It seemed hard to start too, the mornings in San Diego are pretty consistent, +/- 5 degrees pretty much every day, so I can't account for hard starting based on temps. It's not super hard, but it takes noticeably longer to get get it up.  It seemed really smooth on the freeway too, and the 60-80 times back that up. Overall, this was a really good tune for the commuting I generally do, and definitely on the top of the list. 
« Last Edit: 08 January, 2015, 04:05:40 AM by day7a1 »

Offline day7a1

  • Tiger Master
  • ****
  • Posts: 786
  • C'mon!!!
  • Bike/Model: 11 Tiger 1050 SE
  • City / Town: San Diego
  • Country: United States
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #7 on: 22 January, 2014, 07:19:59 AM »
Configuration 7, Arrow Race w/ Sai Tune:

Test 1: The fuel mileage peaked at 29.64.
Test 2: The peak of 29.64 was at 63 mph +1-0.
Test 3: Over 7 tests I got max of 86.54, avg of 81.89, and a standard deviation of 3.55.
Test 4: Over 3 tests I got a average of 3.01 seconds with a standard deviation of .054 seconds.  Minimum 2.95 maximum 3.05.
Test 5: Over 50362 non-zero samples, the average short term fuel trim was -4.076% with a standard deviation of 2.64.
Test 6: It's nice to remember how much I liked this tune, and have that backed up after testing other tunes. It's not the fastest, the most fuel efficient, nor the smoothest.  The Arrow Race tune is like the Tiger, it's darn good all around.  The decel popping came back along with the engine braking too.  It was a bit snatchier than other tunes, but nothing that I couldn't deal with with practice.  It's also decent on gas mileage, especially considering that I've been letting the stress of work get relieved on the ride home....:) It did act a bit funny with the closed loop operation, but I'm not sure if it's the tune or if I need to do some maintenance on the O2 sensor.  The sound and feel of this tune makes it my favorite, though it could be tuned up for some improvement a bit I think. 
« Last Edit: 18 April, 2014, 03:58:58 AM by day7a1 »

Offline indytiger1050

  • Supreme Tiger
  • ******
  • Posts: 21523
  • Bike/Model: 2007 White Tiger
  • City / Town: Indianapolis
  • Country: USA
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #8 on: 22 January, 2014, 10:36:50 AM »
 Sounds interesting, should be fun.

Do post up the adapter and program your using, since what works on your setup the results may not be the same with another Tigers configuration.
Your approach should help give some hard data to help one decide best tune for setup. Nice.  :460:
Support the forum.   :031:

Offline Zeebad

  • Tiger Master
  • ****
  • Posts: 866
  • Bike/Model: Ex Tiger 1050 10 years
  • City / Town: Stourbridge
  • Country: United Kingdom
Re: Quantitative Testing of 7 Tiger Tunes Using Bluetooth OBDII Scanner
« Reply #9 on: 22 January, 2014, 10:42:22 AM »
Good stuff! Any info on the scanner you are using? I wouldn't mind one myself for that money. Same for the app.

Edit: nvm, google is my friend! :)
« Last Edit: 22 January, 2014, 10:44:01 AM by Zeebad »
Danny

I haven't been everywhere, but it's on the list.

My mods https://www.tiger1050.com/index.php/topic,30477.0.html