The Broken Fiat Club

So the ultimate question, the answer to the meaning of it all.....

Which tune is the best?

There is so much noise and real vitriol on the subject, that I'm a little nervous posting about this. I'm posting it here as I think that will keep things calmer, and I can say things like "I don't care what you like; this is about me" and people here seem to be grown up about it.

Honestly my question really is "Which tune is best for me?". To answer this I'm going to do some testing of the tunes I have access to. I have 3 different PCM tunes from 3 very different tuners and I'm going to compare them.

So I think the first part for me is to design the test. I could throw them all on dyno, but that costs money and time and there is the whole dyno setup issue (2 wheels vs 4 wheels, vs 3 wheels) that I don't want to touch. The bigger concern for me though is does a dyno pull translate to what I want out of my car. A dyno is a hugely useful tool to see what the powertrain is doing, I don't think it's reasonable to expect a good tuner not to use one. My plan however is to examine how the tune works for me. I don't live my life one 1/4 mile at a time. I'm not a track day warrior, driving on the track requires patience and mental discipline that I seem to have less and less time for as I get older. So I'm an enthusiastic driver who learned to drive on the mountain roads in Wales and gets the most enjoyment managing a car through the twisty stuff. Probably closest to hill climbing, but without the desire to die.

So what does that translate to for me and this test. Ultimately the goal is to figure out how to shake Sportfan off my tail at the next Twisted Sisters event. Here's my initial thoughts:
- I think I need to look at how the cars pulls from 20mph to 50mph in 2nd gear. That equates to me exiting a slow corner properly.
- I probably need to look at a similar pull in 3rd gear, (going to play with that a little and see what I think is appropriate). I'm hoping this will equate to picking the wrong gear in a slow turn but still wanting to leave Sportfan behind. Also perhaps looking at how the car comes off half throttle in 3rd and goes forth.
- Probably 60 to a speeding ticket in 4th, because I can.

I'm going to datalog with my Kiwi3 and Dashcommand combo (keep it simple stupid), and going to log:
- time
- speed
- IATs and ambient temps
- timing
- boost
- throttle pedal position (if I have that, I need to check)
- throttle body position
- AFR (because I want to make sure my car isn't going to die)

I'll try to run each tune at about the same time of day, and avoid any extreme weather variations (tough to do in the Gulf Coast). I can always rerun a tune if it looks like one didn't get a fair shake. I'll only run one tune each day, so the car will be totally cooled down and I'll run the same warm up route to the testing area. If I end up having to change anything, plugs etc, I'll rerun all the tests for a level playing field.

As I run each tune I'll post the results without identifying the tuner, so that each tuner has a chance to address any feedback and add to the tunes (especially in the new tunes that have never been tested on my car before). I think this will also avoid external bias when you are looking at the data. I'll know but I can't think of a way around that without letting someone else put the tunes on. It will be interesting to see if I can even tell the difference, it is strange how much your brain can invent power increases when they don't exist.

When everything is complete then I'll identify the tuner and their results, as well as announcing my favorite.

My cars current power train configuration is as follows:
Car is a 2013 500c Abarth.
NGen Performance turbo (11 blade)
Ported/polished manifold
Supersprint hi-flow cat and exhaust
C&B cam
Forge WGA & BOV
ATM fenderwell intake
ATM FMIC
EC boost leak fix
93 octane gas

I also have a water/meth kit and Tork MA springs that I might add later just to see if the impacts can be measured.

So, what are your thoughts about the testing? Am I missing anything? I'm doing it anyway, even if you're not interested.

I also thought about picking up a used piggyback for comparison, so let me know if you see anything out there.

I'm going to run the tests with a stock tune tomorrow, as a baseline, assuming the weather cooperates and assuming I install one or more seats today.

My thoughts: get a Draggy to measure performance. They’re not expensive and it’ll be way more accurate than anything else. It’s basically the industry standard today. Plus, there’s probably going to be minute differences between these tunes so you’ll need all the resolution you can get. I was using a draggy for my “can you beat me...” thread.

Also... are you using OTS tunes from these vendors or special versions to match your turbo/cam? If so, im not sure your results will be relevant to anyone but you.

What’s your data rate frequency with that logging combo? I’m currently getting 3hz for 5PIDs in BlueDriver, but will be upgrading to a wifi logger to get that up to 10hz.
I might also suggest logging enough data to input into virtual dyno, to plot the hp/tq in WOT runs.
It would be interesting to see how each tune performs with a rear wheel speed sensor disconnected, this would help answer any dyno limiting questions once and for all (in conjunction with virtual dyno plots)

I don’t think disconnecting them will prove anything or it would have been done years ago.

I'll take a look at Draggy.

I can't find the max frequency of the Kiwi, it just says (unhelpfully) 4x faster than the nearest competitor. Dashcommand documentation is equally unhelpful, so I'll play with it and see what the options look like and what comes out. I have torque on an Android somewhere I can try too, and can always try to figure out Multiscanecu or AlfaoBD if necessary. Maybe the key thing here is to get a good sent of data points with the stock tune and then go from there ;and of course I don't want to make something crazy complicated to setup (because I'll get it wrong).

The existing tune is modified to the cam, and has had revisions to it. I've given the details to the other tuners; I do not expect them to get it perfect the first time which is why I don't feel it's fair to name them on the first round. And yes, of course, this is all about me.

I've had the potential offer of dyno time not too far away from me, so we might go that route too. It would be good to have different eyes on the tunes and perhaps we can unhook the rear of the car and see which tunes don't like that. Although that might spoil the blind test

retrophit wrote: ↑Fri Jun 19, 2020 11:36 am I don’t think disconnecting them will prove anything or it would have been done years ago.

How so? If someone’s already done this then I’ve missed it. If it’s a standard Hall effect sensor without signal voltage, it’ll be an on off switch, disconnecting it would be the same thing to the ecu as the wheels not spinning (like on a 2 wheel dyno), tcm, abs etc will fault. If there is a signal voltage, removing the sensor from the hub would accomplish the same thing.

If the car behaves/logs differently with the sensor unplugged, it’ll be obvious, especially with a virtual dyno plot. There’s only 2 sensors that read rear wheel speed, it’s really the only way to verify without a 4wd dyno.

Regardless I don’t mean to derail, i still plan on testing for my own curiosity, but it would be interesting to see if the different tuners found a way around it.

texanbrit wrote: ↑Fri Jun 19, 2020 11:40 am I'll take a look at Draggy.

I can't find the max frequency of the Kiwi, it just says (unhelpfully) 4x faster than the nearest competitor. Dashcommand documentation is equally unhelpful, so I'll play with it and see what the options look like and what comes out. I have torque on an Android somewhere I can try too, and can always try to figure out Multiscanecu or AlfaoBD if necessary. Maybe the key thing here is to get a good sent of data points with the stock tune and then go from there ;and of course I don't want to make something crazy complicated to setup (because I'll get it wrong).

2-3hz with a few PIDs is common with low energy Bluetooth adapters, It’ll get you the information you need, but for logging I find it lacking fidelity.
10hz would be an ideal minimum, I know the wifi kiwi’s were up for it, I’m not sure the new Bluetooth ones are.

Around 20 time stamps @3hz for a 2nd gear run vs. 80-100 at 10hz, which will tell you exactly what happened when and where in the rpm range

This is a good resource for different data rates for existing combos:
http://forum.gps-laptimer.de/viewtopic.php?t=2009

Romanoaf wrote: ↑Fri Jun 19, 2020 12:33 pm

texanbrit wrote: ↑Fri Jun 19, 2020 11:40 am I'll take a look at Draggy.

I can't find the max frequency of the Kiwi, it just says (unhelpfully) 4x faster than the nearest competitor. Dashcommand documentation is equally unhelpful, so I'll play with it and see what the options look like and what comes out. I have torque on an Android somewhere I can try too, and can always try to figure out Multiscanecu or AlfaoBD if necessary. Maybe the key thing here is to get a good sent of data points with the stock tune and then go from there ;and of course I don't want to make something crazy complicated to setup (because I'll get it wrong).

2-3hz with a few PIDs is common with low energy Bluetooth adapters, It’ll get you the information you need, but for logging I find it lacking fidelity.
10hz would be an ideal minimum, I know the wifi kiwi’s were up for it, I’m not sure the new Bluetooth ones are.

Around 20 time stamps @3hz for a 2nd gear run vs. 80-100 at 10hz, which will tell you exactly what happened when and where in the rpm range

This is a good resource for different data rates for existing combos:
http://forum.gps-laptimer.de/viewtopic.php?t=2009

I know on my OBDLink LX, I can get some pretty quick polling rates, but it highly depends on the software and bluetooth stack on the receiving end. Torque Pro is set up to expect a slower polling rate, and getting it also to actually work on a higher poll rate is a tricky combination of settings. The result is sometimes I get good polling rates, sometimes I don't. I guess if I spent more time, I could nail down what the factor is that keeps it from always having a high poll rate.