[MichaelDelaney]

since someone brought back the "what is a powerband?" thread for discussion, I thought this would be a good time to also bring back another related concept at the same time for you to discuss.

I recently saw this decent explanation of engine transients (enter transient in the articles search and in the search window here to get up to speed):

Tuned engines use resonating pulses that enable an engine to have a high volumetric efficiency (VE) [TI Moderator Note: search under resonance or ram theory or read the IM article and also search under "sequential pairing" in the headers for the background to these pulses ].

These resonating pulses can only work if they have a pulse before them to follow. After an engine has been jerked down in RPM from a shift, this pulse timing is messed up.

Transient responce is the amount of delay before the pulses have timing that can be benifical to VE again.

Its like the engine getting back on its feet and ready to pull 100% again after getting knocked on its ass from the shift.

Compromises are made in flow, hp, tq all sorts of things to enable this delay to be lessened. Also, if you're using a glide (auto tranny), you will not need to sweat transient response as much as someone with at short ratio 6spd. Its about compromises.

I heard that in formula atlantic, they ended up dropping a ton of time with a header that had its tq peak 1500rpm below the lowest engine speed that the engine would ever see in the race. They dyno numbers were crappier everywhere you looked, but the car was faster around the track because it would make power after each shift with less delay. This is why we need to not see the dyno as our king, and tune deeper then numbers can quantify.

Something we try to teach you here with those Quizes...

BTW the next "Quiz 4: Which 1 is the Fastest?" is coming in March. Look for it and participate.

If you haven't participated in one, check out the old ones linked in the thread entitled "Intense Learning" at the top of this forum. They're fun.

 

[HiddenDemons]

Something we try to teach you here with those Quizes...

BTW the next "Quiz 4: Which 1 is the Fastest?" is coming in March. Look for it and participate.

If you haven't participated in one, check out the old ones linked in the thread entitled "Intense Learning" at the top of this forum. They're fun.

It's funny that I just asked about when it was gonna come out!!

And I highlighted a good reason why I wanted to see another quiz.

It's also interesting the things that you find when you go digging in the older posts trying to do more research for an engine build. Butt loads of good info, that's what the search is for!!!
68 pages!!! Well only 60 left to sift through.

 

[MichaelDelaney]

I think once you've done the search thing enough times you learn little tricks to reduce the number of threads being identified and try to get more relevant threads instead.

the "sensitivity" of your search will improve if you use the advanced search option. you can sometimes choose 1 forum instead of the entire site. you can sometimes choose a "phrase" option with 2 words instead of using 1 term or 2 terms with AND.

you'll get these tricks down once you start using it often. I appreciate the fact that you are at least trying to gain from the wisdom of the older threads by using the search in the first place...most of the time they provide more info than what is on the forums currently.

I was going to release quiz 4 in March ( with StyleTeg and I running the rapid fire replies to your clue questions like we did in quiz 3 ) but in March and April there was this onslaught of new members who just didn't want to use the site the way it was intended.

There was no way they could have gotten anything out of a quiz because they just hadn't even learned the individual ideas here yet...you can't connect the dots if you haven't even learned the dots yet.

these quizes are meant to confirm to yourself whether you "got it" or not after reading thru the ideas presented here. they are more of a way to prove to yourself that you understand the more advanced concepts and can continue to build a foundation of correct knowledge from them. it makes you a more sophisticated enthusiast with more depth. But with the "what's the best exhaust?" invasion in those months from the "tuner yo!" crowd from the other boards, I felt it was entirely inappropriate to have a quiz since it would be way over their heads to be beneficial.

btw those quizes take time to prepare,...downloading the pics and facts in preparation and then discussing amongst the mods how we present it and what ideas we want to convey as the take home message. We're talking hours of our own time to have these run. I'm not going to waste my time, if people aren't going to invest in themselves here to do the necessary preliminary legwork reading.

 

[MichaelDelaney]

"What is a transient dynanometer?" thread in the closed MD Tech forum.

there's also a nice explanation in Peter Wright's book "Formula 1 Technology" (Society of Automotive Engineers, Inc. Press, 2001) Chapter 13 Test Rigs, page 202.

there's a difference between steady state vs. transient recovery time for engine performance that must include or take into account the gearing effect when looking at the engine-tranny relationship for acceleration performance, for road racing especially (deceleration and re-acceleration).

 

[descartesfool]

I am still not a big believer in transient response chasing. In a spec engine series like Formula Atlantic or Formula Ford or 2000, etc, etc, I can see it. All these quasi identical motors have been tuned for years and years by dozens and dozens of expert engine builders. All engines have virtually the same horspower and torque curves, up to a point. That torque curve will get you down the straight in given amount of time. Now if you can improve how fast the engine responds to a throttle "event" such as rapid power on after lifting for an upshift without losing too much of the WOT torque curve, your speed down the straight might be a tad faster because the initial pickup after each shift is very slightly faster.

But in the real world, the time to get down the straight is going to be far more dependant on getting the right gearing (easy with a Hewland box, very hard with a Honda box) to match the torque curve to the track. In those spec Formula series, gearing is adjusted for every track. So virtually everything is optimized and tuning for better throttle or transient response could get you some gains over your competitors.

In a Honda product where you are still trying to improve the torque curve and are nowhere near the ultimate limit, I would think you will see vastly more benefit in lap times by increasing the thrust after the shift point by increasing the torque in the appropriate RPM, even if you have to wait an extra fraction of a second for it to come on because you have not fully optimized transient response. But if you are not even changing the diff ratio for various tracks, let alone gears, you are just not making the gross changes needed to improve speed. Worrying about transient repnse when you have not long ago reached the limit of the torque curve wthin a given set of rules would seem to be a waste of effort. When I step on the throttle up the Mosport straight, I can tell you what I am missing, and that is a whole lot more torque. What happens for the small moment after the shift is of little concern compared to the endless time spent at WOT waiting for the RPM's to go up in 4 th gear. Give me torque first as my RPM's go up, and then I'll start thinking about transient response.

 

[MichaelDelaney]

I can't tell you the quantitative contribution of this nonsteady state manipulating of how an engine responds to a change in rpm on your sector times but obviously it's deemed important enough to measure and Formula 1 isn't considered an equal spec engine series (although from the latest rumblings to save costs this may be something they are considering).

Our cars do depend on maintaining momentum instead of sheer grunt.


I was looking at both Ilmor's explanation in SAE's Design of Racing & High Performance Engines 1999-2003 and Peter Wright's F1 Technology book as to what exactly are they trying to accomplish by tweaking on a transient dynanometer. It's how they compensate against inertial and resistance forces which increase exponentially as your speed increases (aero drag, rolling resistance, gradient changes and recovery after braking).

I also was looking at a racing line simulation program which analyses how you take the line through the 4 phases of a corner (braking, turn in, apex, exiting) and how throttling affects your exit speeds and initial acceleration (expressed as a %age of your maximum acceleration). You have to input info on the characteristics of the track & car accurately (number of degrees of each turn and it's width and if there's any banking). The program not only determines what the best line to take through each corner but also calculates when best to start each phase, including when is the best time to begin throttling. This throttling initiation point is expressed in terms of the % yaw of the car (100% yaw would be the full number of degrees of the turn). The program is called The Driving Simulator and was featured in March's Racecar Engineering.

I'll have to see if there's a website for it and link it on here later.

It's interesting though that on some corners, not having the maximum lateral cornering g and roll at the apex doesn't necessarily mean you are doomed to be slower in that sector of the track. This is very counter-intuitive to most of us who believe that you must pull the highest cornering g's to be maintaining speed through the corner by the time you apex. In fact, sometimes a late apex isn't the best guarantee of a fastest exit speed. The simulator showed that if you could lengthen the exit phase and get on the throttle before this lengthened exit phase, you would have the fastest sector times due to a faster initial acceleration (% of max acceleration) at the corner exit. So on some corners we try to shorten the turn in phase and lengthen the exit phase which starts with throttling.

I would think that having a faster engine recovery after coasting steady state through an apex should affect your initial acceleration out of a corner.

 

[avenger]

Do you know if thermodynamics plays into transient response? I was also wondering if (not meant to sound flaming) transient response would have much to do with our relatively low revving engines (compared to F1), because it makes sense to me that an engine that revs close to 20,000 RPM would have alot more resonation (sp?) than our imports. Even though the amount of transient response that would affect our engines may not be negligible, I am just wondering if there is really much that can be done to discover when the ideal shifting points would be? The racing simulation program sounds pretty sweet, do you know if they would use it on the track during a race?

 

[MichaelDelaney]

I'm sure that all of the engine's aspects affects how fast it recovers from a sudden change in engine speed, including combustion efficiency but it seems that liverforphysic's explanation tended to focus on resonance tuning and cylinder filling rather than burn since that's something you can adjust and tune more easily.

yes, the program is used to help drivers improve their sector times at the track.

the true art of driving is in the trail-braking phase from turn-in to apex and how you balance and blend the proportions of braking & cornering at the same time. This is what separates the great drivers from the good drivers. If you give too much cornering and not enough braking, the car enters the apex at too much speed and you can't correct that. If you give too much braking and not enough cornering you can't tighten the turning circle and you either miss your apex or go too slow.

In the end, you want to have the car in control (hit your apex at the optimal speed) that will allow you to lengthen the throttle phase and maintai your highest momentum through the corner relative to your competition.

Race engineers can show a driver that according to telemetry there is a better line or brake point or throttle on point. Sometimes the faster way isn't according to how the butt registers it and counter-intuitive. It's funny though that in the end the compromise that all race engineers concede is that they will give up convincing the driver to take a line or throttle on point if his confidence is taken away. Inspiring confidence in the way the car behaves appears to take precedent over physics.