Thanks for your input but your wrong on why I use to get the shutter.

I've been doing this for several years, gone to PSDS, a racing school and even recently hired a private coach.

What you and pk have written is furthest from anything I have been told, read or experienced on the track.

If it works for you that's great.

But it's not for me.

Good luck

Thanks adias....kind of like turn 12 in thunderbolt @ NJMP but you can't take that one flat out (well I can't at least)

Rodolfo, I can't decide whether you're a troll or just very slow on race tracks. For the sake of civility, let's assume you're sincere and let me explain why your advice is based on an impression you've gathered while traveling quite slow for a track.

On tracks, we threshold brake. That means taking the tires to the limit of deceleration they can produce without locking up. We have a very sophisticated ABS that we'll talk about later, but in less capable cars, what you suggest has the following effect, using your own example.

I'm in fifth gear on some track, not Laguna Seca because I can't go that fast there, but maybe Willow Springs at the big track. Some hypothetical track anyway. If I shifted into fifth while accelerating in fourth and I'm being conservative, shifting at 7000 rpm, then I was going 124 mph at the shift and if I needed to shift rather than hang on near redline until the braking point, I had some room left and the additional acceleration time in fifth will put me at... let's say 140 mph. At least. Now I have a corner approaching that in your example called for... I think you said second.

From 140 mph, or about 205 feet per second, I roll on threshold braking and the car begins to slow at roughly 35 fps per second. This isn't an Auto-Union from the thirties, the brakes are quite powerful enough to brake the car without help from the engine overrun, that is the compression braking that was needed in those old cars. I could in theory just brake without shifting until I reached the target speed for entry into the next corner. Then, trail braking all the way to the apex (since I'm coming off a godawful fast straight segment), I shift into second and accelerate away to the track-out point.

That works if I can hit the correct speed for the engine to match my apex speed on that corner. Let's say my apex speed is fifty or so. That means I have to compute ahead of time the rpm for fifty in second gear and blip the engine to the corresponding speed. In my car, that's about 5,000 rpm and I have to hit it from below.

Why from below? Well, consider for a moment. If I didn't use the power all that way down from 140 mph while threshold braking, then I could have done one of two things. Either I left it in fifth gear with the clutch engaged, or I disengaged the clutch and let the engine return to idle. In fifth gear, the engine will be turning about 2400 rpm at fifty mph. If I let it return to idle, it will be about a thousand rpm for the sake of discussion. Either way, I have to time a blip. Either one that adds 2600 rpm or a bigger one that adds 4000 rpm in the brief interval between taking my foot off the brake and putting it on the throttle. That's not easy. Most of us can't hit that rpm so precisely in a fraction of a second. What's worse, with a road flywheel, the engine accelerates fairly slowly when changing rotation speed by several thousand rpm. Without going outside for a test, I'd guess it takes at least half a second, maybe more like a full second. We won't even worry about the obvious niggles here. At speed on a track it is entirely too easy to overdo that blip and bounce the engine off the limiter.

Partially, we don't want to do that because we sound like a complete novice at such times, with the engine revving but producing no power. More significantly, we don't do it because it's bloody slow and likely to cause a spin. The spin? Well, if the blip isn't right, if we hit 4000 rpm instead of 5000 the engine speed and car speed in second have to be brought into sync very quickly. The synchros and clutch are powerful enough, at least the first hundred times we try that, but the physics of the matter require the car speed to change as well. If the engine is going the speed for only 40 mph when we let out the clutch at 50 mph, the car will lurch as someone tried to explain. If we also are cornering (and we will be at maximum cornering potential at the apex of a corner that completes a fast segment), the rear tires will be asked to suddenly provide additional deceleration thrust. A spike of thrust in fact, and the rear tires will be pushed beyond their threshold of grip. They will break loose while the fronts hang on and bada-bing. We spin.

Okay, I know. You didn't suggest that method. It was an example of where your advice takes us. Let's go back to 140 mph and try it your way. We roll on threshold braking and the car begins to slow at 35 fps every second. We don't want to subject this expensive engine to outrageous stress, so let's pick a target rpm below redline. 7000 rpm perhaps. That is the speed we want the engine going when the clutch is engaged in the next lower gear. You can use redline if you prefer, but I pamper my engine, especially while racing. To finish first, you first must finish, as the saying goes.

On that straight, the car was moving at 205 fps, which is 6700 rpm in fifth gear. Now let's figure out how fast we'll be going at a shift that puts us in the next lower gear at 7000 rpm. In the right column, is the time into braking that gets us to that speed:

5:4; 124 mph; 0.7 sec into braking sequence.
4:3; 101 mph; 1.6 sec
3:2; 68 mph; 2.3 sec

The intervals there are 0.7 seconds, 0.9 seconds and 0.7 seconds. (Actually, that last interval would be extended by trail braking, but to hell with it.) What you're suggesting is we just slam the shifts home in that interval and count on the synchronizers and clutch to match engine speed to road speed in the target gear, that is the lower gear into which we're shifting.

Well, you might get the rhythm close enough. Good times are all about finding the rhythm of the car and timing events to that rhythm. You might, but let's consider now how much rpm differential the synchronizers and clutch are absorbing. When we complete the shift, the engine will be going 7000 rpm of course. At least if we time it as stated above. What will the engine speed be just before the clutch engages? Well, it will be whatever speed the higher gear called for just before we disengaged the clutch less however many rpm the engine lost while coasting down with no overrun torque holding up the rpm. For simplicity, let's assume the coast-down slowing is the same as the speed decrease from braking. (Those who don't see what I mean, trust the rest of us. It's a simplifying assumption that is conservative in the sense that it gives the benefit of the doubt to the proposed method.) So the engine speed as the clutch engages on those downshifts will be as shown and the differential, the sudden speed-up of the engine, is shown to the right:

5:4 124 mph, 5150 rpm in 5th, 1850 jump for 4th gear;
4:3 101 mph, 4825 rpm, 2175 jump;
3:2 68 mph, 4260 rpm, 2740 jump

Those are pretty big differentials the synchronizers must overcome. They can, at least for a while, but what bothers the rest of us is the jolt to the driveline. The energy to achieve that jump in engine speed comes through the synchronizers but it has to originate at the tires. Furthermore, those are [not] leisurely shifts. You imply we can allow a little clutch slip to let it happen more slowly, but really we can't. The times involved don't permit that. We're making one shift every second roughly, so we have to do it that way. Roughly, that is. Slam, bam, you're engaged, Ma'am. Three times in succession without providing any blip to help accelerate the engine rotation to the higher rpm needed in each case. I don't like to treat my driveline that way, but assume I have a very wealthy sponsor and consider just the physics.

The car will be hit with a jolt to the rear tires once a second and the last one will come in the middle of trail braking. Any ordinary car will spin at one of those shifts. That is, the torque demand on the rear tires will push past the threshold of performance, they will be at a higher slip rate than is optimum and will begin to outgas violently. The car spins if we take no action. If the driver anticipates the torque jolt, he will back off the brakes momentarily so the tires can return to nominal rolling speed, then he will roll back on full braking. Doing that three times in a braking sequence will be giving up something more than a second. We don't do that to avoid a blip of the throttle that we could time at the same second instead of coming off the brakes.

Alright, suppose it isn't a race car, but a modern road car like our Carreras. It has sophisticated ABS and even PSM to deal with that final shift during trail braking. Will that save us? Darned if I know. Letting a clutch 'pop' like that comes very close to what we call a "square-wave" event. Real world systems don't like square waves. They always get rounded off by some physical phenomenon. In this case, clutch slip I suppose. Will that "rounding off" give the ABS time to reduce pressure in the brake lines, will the ABS be quick enough to spot the incipient lock-up to keep it from happening? If it does, will it overshoot and underbrake the car for a few crucial tenths of a second? What about PSM? When I pop the clutch that last trime in the trail-braking sequence we're creating an rpm jump of nearly 3000 rpm. Will the stability management kick in fast enough to keep the car balanced? Again, darned if I know. Sounds like a good question for a long test day. With lots of instrumentation. And someone with enough vague curiosity to pay for all that.

I don't say it can't be done your way. It just seems like a lot of demand to put on the systems that were designed for someone willing to blip the throttle. If I want to let the car manage those things, I'd go to a PDK, which does blip the throttle at each downshift. No kidding. I'm not being sarcastic. Blip your throttle. Learn how to do it while braking at the limit. It's really not that hard.

Gary

Wow Gary! What a masterpiece! I hope some 911s will be better for your effort.

classic.

rep points for the 'quote of the day'

Thanks Gary for having the time and patience to explain in depth what I was trying to get Rodolfo to understand. Rep up to you, sir, and amazing post!

Edit: and one thing to add, I don't know of anyone who does heel-toe shifting "just to show off," and that includes the sound caused by said action. The only showing off exhaust sounds I've ever seen are people at stop lights, or idiots on freeways. In both cases they are actually revving their engine to gain speed, and not braking as is the case in heel-toe.

This is not with the intention of bothering anyone,

I am just saying you dont have to rev match every downshift, everytime you brake in a straight line.

If you downshift while just letting off the throttle or mid corner, then do rev match.

Just try it in your car, and take it out of your system.

During hard braking, downshift through all the gears, and you wont feel any jerking of the transmission

And remember, just try it, dont try to do this all in your notebook and play with numbers, just go out on the car, go up to fifth or fourth, and brake hard, and start going down the shifts, one by one, without letting of the brake, and the car will not suffer or loose traction.

Just to be clear, Rodolfo, you're talking to people who have tried doing it the wrong way. Quite a lot of us have been driving sports cars for several decades and we learned what does and does not work long ago. Cars change, and sometimes new technology offers new ways to go fast as well as new equipment tolerance of abuse, but this isn't one of those cases.

If you really are doing what you describe, the only reason you aren't seeing dramatic problems is your comparative slow speeds. The effect won't be visible in lap times if you're already far below the car's capability. It simply hurts longevity of the car when you abuse the driveline like that.

Gary

+1 Gary, I think we should make your post a "sticky"!!

It could read straight out of one of Bondurant's books!

Ok, I am slow

Continue heel and toe-ing every single downshift

Welcome back to the 50`

I'm curious Rodolfo: what part of the engine or transmission's mechanicals does the rev matching when you down shift from say, 4000rpm in 4th to 6500rpm in 3rd, when you aren't heel toeing?

On somewhat of a tangent, I've heard that big rig drivers do what Rodolfo is talking about, but their rev range is extremely limited and engine internals and shifting mechanics are completely different from any car. Maybe I'm wrong, as it's just something I've heard, and I've never driven a big rig.

You stated tracking at Laguna Seca...

Do you mind telling us which car you were tracking?

At laguna I was in a Skip Barber Formula car on a practice session.

I have also raced formula vee, clio cup, shifter go karts.

On race bikes I ride on the 125cc and 600cc expert class.

And in normal street cars, several cars including a 911 at track days, and I can assure you, heel and toe is not necesary in 95% of the downshifts, except when you are:

- DOWNSHIFTING MIDCORNER ON A DOBLE APEX CORNER
- MADE A MISTAKE AND NEED TO DOWNSHIFT MIDCORNER
- OR IF YOU WANT TO SKIp A SHIFT DOWN, LIKE GOING FROM 4TH TO 2ND, WITHOUT GOING THRU 3RD.

Syncros have come a long way since the 60 or 70, when heel and toe was a total must. Now it is a good skill to know, but doesnt have to be used in every single downshift,

That was just my point, NOT EVERY SINGLE DOWNSHIFT

In some special situations is a good help, but going down the straight and breaking in a straight line, really have no need to do this, with proper timing, and the help of the sycro transmision and clutch.

Really, dont want to bother anyone, just giving my opinion on the subject, based on self experience and talks with instructors, that actualize this teknic to the current tecnology. (that indeed has come a long way since the 60`and 70`, enven if some doubt to believe it)

can you explain how the synchros work so much better now and why it's not necessary for abnormal wear and tear and such? I'm not trying to be condescending, I honestly want to know what facts you can pull these statements from. I understand you get your "feel" of no need to heel toe(ie no lurching) from your experience, but where is the data to back up that it doesn't affect your drivetrain negatively?

Rodolfo is correct, you don't actually have to heel-toe... oh, but wait this isn't the 370Z forum.