Driving techniques - Heel Toe Downshift

[kennythebomb]

Anyway, I got in the Probe just a minute ago to take it around the block (not good to let something sit for weeks at a time, so despite the ache, I drive it every couple of weekends. Anyway, my foot position is NOT pigeon-toed in this car, it's quite naturally slightly right of upright, and I operate the gas with my toes. Apologies to Kennythebomb for *lying/misstating/uttering falsehood*. You should be very slightly NE in a Probe.

Hmm interesting. I'll have to try it some more, but when I tried NE my foot seemed like it was gonna slip off the brakes if I didn't hit it just right... not a good habit for me to start off with. I'll test this some more though.

In other news I have all but mastered the non-braking rev-matching... surprisingly I have to give the most gas when going from 5th to 4th, then from say 3rd to 2nd. I did heel and toe one time on the street today or yesterday and did it f'ing perfect I was amazed.

I can't possibly downshift to first unless I'm going like 3 miles per hour, I've found.

I also set a new high speed record in my car today. Surprisingly my car is more stable at 100+ than at 50, it really tightens up as it goes faster.

 

[Scaff]

If you're looking at the tach, you're not looking at the upcoming situation (curve, apex, intersection, traffic, etc.) The only thing a tach is good for is setting the idle (and if you don't have a rev limiter, finding the redline.) Everything you do in the car is by sound.

Quite agree, the tacho is a very useful tool, just like the speedo, fuel gauge, etc; buy you should not be looking at them constantly. Eyes need to be on the road and looking as far ahead as needed. You glance at the instruments as needed, and it should be only as often as needed. Keep in mind that to look from the road to the dash and back again means that you have to re-focus your eyes twice and in all this time you are not able to observe the road ahead. I actually have a GPS speedo built into my Speed Camera detector and as that's mounted on the very top of the dash its far more handy than the actual cars speedo, as its quicker to look at (less distance and less of a refocus) and has nice big numbers that are quick and easy to read. The other advantage of it, is accuracy, car speedos can legally over-read by as much as 10%, the GPS unit is acurate to +/- 1mph.

I do use the tach, but when heel and toeing I certainly don't stare at it, noise is good indicator, but look for other factors as well. My Celica sends a very distinctive buzzing feel through the throttle pedal when its about 500 rpm short of the redline, very handy little gear-change indicator. I'm sure many average drivers would complain about it, but for me its great.

Regards

Scaff

 

[///M-Spec]

I know downshifting to 1st is an evil, stupid, and nearly pointless thing to do to a car, but hey, I never broke it.)

Actually, I've always downshifted into 1st when crawling around parking lots or 'crabbing' up into a sharp driveway. It seems more 'humane' to do that than lug it in 2nd at walking speeds. I even blip the throttle, double-clutch and heel-toe the whole shebang. If you can execute it smoothly, there should be no undue wear on the drivetrain.

When autocrossing, I often downshifted into 1st on tighter sections that warranted it--usually without double clutching-- the 1st gear sychros never seemed to mind. Of course, some cars have stronger sychros than others.


M

 

[Duke]

Okay, one side question...How does one heal-toe without a tach?

Neosporin and a band-aid. Boom *tish*!

My Neon usually just required two presses of the clutch pedal for smooth downshifts at 3rd-2nd gears... However, every attempt to use the heel-toe method failed for me when changing to 4th, because I had to listen to the engine. Any tips on this?

Frankly, I've never looked at the tach in any of my cars except for reference. In day to day driving I would never miss it. I drive by ear 98% of the time. I do tend to keep half a weather eye on the 325's tach, because the engine (particularly in low gears) will happily charge right into the rev limiter without sounding distressed at all. But even so I've gotten used to the sound and I hardly do that any more.

(By the way, after 180,000 miles, the Neon never needed a new clutch or throwout bearing, although I went through one clutch release cable early on.

That's funny - me too! Somewhere around 45,000 miles? The clutch cable on my '95 ACR broke as I was backing it out of the short-term lot of the Atlanta airport, getting ready to take the seller home and start my way back up the east coast. We managed to get it fixed, but before I left his house, it inhaled the steel retaining washer and locked up hard. So I had to fly back home instead of drive. Luckily he dropped the starter and was able to remove it with no damage to anything, so I flew back down and got it the next weekend.

At 99,000 miles the trans and clutch on that car were still tight as a drum. Other than the clutch cable and some poly shifter bushings, it never needed any maintenance, and I drove the wheels off of it.

 

[Keef]

My Sol's clutch is getting long in the tooth at 115,000 miles. Every now and then I'll shift and not let my foot all the way off the pedal and drag the revs out something nasty. It rarely happens, but when it does I regret my loud exhaust. I hate it when everyone can hear your mistakes. Also, when I shift hard the clutch refuses to hold back the massive 100 lb-ft of torque available at about 5800 rpm for about a second, but it finally reels the engine back down. That's a boon.

 

[Perfect Balance]

I always wondered, what is the point of double clutching? You can still rev match on down shifts without pressing the clutch twice, All it takes is a blip of the throttle between shifts. What difference does releasing the clutch pedal and holding the clutch make if the car is in neutral anyways? Releasing the clutch and pressing it again seems like a pointless thing to do if you can just skip that step and end up with the same result, and much faster too.

 

[Keef]

When you double clutch the syncronizer cones are sped up closer to the speed they need to be going to allow a gear to be engaged. If you try it you'll notice that the stick slides into gear with almost no effort what so ever. That's because the syncros don't need any time to speed up, since they're already spinning at (basically) the right speed.

Other than smoother engagement there isn't really any performance benefit that I can decipher.

 

[backspace]

I always wondered, what is the point of double clutching? You can still rev match on down shifts without pressing the clutch twice, All it takes is a blip of the throttle between shifts. What difference does releasing the clutch pedal and holding the clutch make if the car is in neutral anyways? Releasing the clutch and pressing it again seems like a pointless thing to do if you can just skip that step and end up with the same result, and much faster too.

A simple explanation is that engaging the clutch in neutral helps match the input speed of the transmission with the output, rather then just rev matching the engine RPM to the speed of the car.

 

[Duke]

Double-clutching is a relic of the old days of non-synchronized transmissions. As backspace and Keef have mentioned, the point was to get the input shaft and the output shaft of the transmission running at approximately the same speed to let the gears mesh easily. In a modern synchromesh trans, it slightly reduces the wear on the synchronizers. However, this technology has been well-refined and at this point double-clutching is mostly a waste of effort.

Unless you drive really old MGs. Or farm trucks.

 

[Perfect Balance]

According to wikipedia, synchronizers were invented in the 1920's. Seems to me like all cars made in the past 20 years should have a synchronizer. Maybe some people just don't have enough to do, so they practice pointless driving techniques.

 

[Swift]

According to wikipedia, synchronizers were invented in the 1920's. Seems to me like all cars made in the past 20 years should have a synchronizer. Maybe some people just don't have enough to do, so they practice pointless driving techniques.

Actually, I find it fun and in some cases it can really help with matching revs. Especially on non-sports cars. On my Spec V, this is not the case as the gear ratios are very close. But on say, a cavalier, it's very helpful in timing because the throw of the shifter is so long and the gear ratios are a bit longer as well.

So, it does have its uses, but generally speaking for newer sport-like cars, there's no point. As you said.

 

[Scaff]

According to wikipedia, synchronizers were invented in the 1920's. Seems to me like all cars made in the past 20 years should have a synchronizer. Maybe some people just don't have enough to do, so they practice pointless driving techniques.

In the current day and age, in road cars its simply a practice some people like to use. Sometimes for the rhythm of it and just sheer pleasure when you get it just right, but strictly speeking in the majority of cars its not a requirement.

However, even if syncro's were invented in the '20's does not mean they were commonly introduced at that time. Up until the '60's a good number of cars did not have syncro's on 1st or reverse. Some such as the series 1 Land Rover did not have syncro's at all. In addition the manufacture of gearboxes in the past was not as well developed, tolerances were a great deal larger and so even with a syncro double-declutching could help significantly. However you are right that in the last twenty years roads cars have been fully syncro'd.

As for its use in racing, well some racing transmissions do not use syncro's and a good number use straight-cut teeth rather than road car Helical Gears. Not only do straight cut gears give a very characteristic (and loud) whine, but they can also harder to engage smoothly, once again double de-clutching can help here.

Regards

Scaff

 

[Keef]

Indeed, all cars I know of today have fully synchronized forward gears, but some take it a step further. A big point about the 2008 Dodge Viper's new transmission is that there are three synchro cones per gear in 1st through 3rd, and 2 per gear from 4th to sixth. There are even some cars with a synchronized reverse gear, though I've never had any experience with any myself. It's rare, but some cars to have it.

I'm not sure if a helical gear is necessary for synchronization. Is it? The vast majority of cars use a spur-cut (straight cut) gear for reverse. Reverse isn't sompletley necessary for locomotion, and you don't get where you're going by going backwards, so they save weight, space, and money by leaving it straight and without a synchro.

 

[Perfect Balance]

Actually, I find it fun and in some cases it can really help with matching revs. Especially on non-sports cars. On my Spec V, this is not the case as the gear ratios are very close. But on say, a cavalier, it's very helpful in timing because the throw of the shifter is so long and the gear ratios are a bit longer as well.

So, it does have its uses, but generally speaking for newer sport-like cars, there's no point. As you said.

Can you not rev match while pressing the clutch? I think blipping the throttle and getting a perfect match is more fun when done in one simple move in half the time.

 

[Swift]

Can you not rev match while pressing the clutch? I think blipping the throttle and getting a perfect match is more fun when done in one simple move in half the time.

I can see you didn't read my entire post. Or didn't quite understand it.

I said with my current car, there isn't any real point. Because the throw and clutch position actually sync up pretty well(after a shortshifter ). But in my old car it was very helpful on UPSHIFTS because it would make the timing better because of the long throw.

If you want to do it on downshifts or not is a totally personally preference I think.

Also, you don't get it done in "half" the time. When I do double clutch on downshifts, I never stop moving the shifter. It's a smooth motion. It may take a half of a tenth of a second longer, but if that's comfortable for me, what's the problem?

 

[Scaff]

Indeed, all cars I know of today have fully synchronized forward gears, but some take it a step further. A big point about the 2008 Dodge Viper's new transmission is that there are three synchro cones per gear in 1st through 3rd, and 2 per gear from 4th to sixth. There are even some cars with a synchronized reverse gear, though I've never had any experience with any myself. It's rare, but some cars to have it.

Spot on, hence the reason double de-clutching has become less and less of a requirement for smooth changes on modern cars. The lack of syncro in reverse is also, as you say, still quite common. Hence the reason its easier to crunch the gear when you engage reverse.

I'm not sure if a helical gear is necessary for synchronization. Is it? The vast majority of cars use a spur-cut (straight cut) gear for reverse. Reverse isn't sompletley necessary for locomotion, and you don't get where you're going by going backwards, so they save weight, space, and money by leaving it straight and without a synchro.

Helical gears are not a requirement for syncro's as far as I am aware, by the arrangement of the teeth on a helical cog does help matters significantly. You've also ID'd the main reason why a number of racing gearboxes don't use syncro's, it saves weight and space (money is less of an issue), but its also one (or more) less components to go wrong.

Regards

Scaff

 

[///M-Spec]

Maybe some people just don't have enough to do, so they practice pointless driving techniques.

In certain gears on certain cars, double clutching DOWN (like going from 3rd to 1st on a BMW Getrag 6 speed) is slightly smoother. Unnecessary in almost any modern car, yes. But like Scaff said in his earlier post, sometimes the enjoyment is simply from being able to do it. In other words: people do it because it's fun.

So on the notion of doing useless things for the fun of it.. with the advent of quick shifting 6, 7 and even 8 speed automatic transmissions, the manual gearbox itself is becoming something of a relic. So you could even say driving stick itself is a 'pointless driving technique'. But people still do it because there's something to enjoy about it.


M

 

[Swift]

So you could even say driving stick itself is a 'pointless driving technique'. But people still do it because there's something to enjoy about it.


M

Not to mention the more direct and complete control you have of the car.

 

[Perfect Balance]

I still don't see any reason to do it. As long as you get the revs right, the gear will go in smoothly, whether you double clutched or not. The transmission is always moving because the wheels are rolling, so all you have to do is line up the engine speed with it. If done right, it will be very smooth, we tried it on my friends Miata today a couple times. If done right, you don't feel the shifts at all. Sure, if you don't have synchronizers, then I see why, but I'm sure all the people who do it on here have synchronizers in their cars. I'm either overlooking something here, or there truly is no point. The synchronizers barely have enough time to slow down on a downshift anyways, and unless you have a really wide gear ratio, it should only be a difference of 1,500 rpms or so.(which isn't a lot) Even less since you are slowing down between the downshifts too, letting the engine drop a couple hundred rpm's in between.

EDIT: Is it so desirable to be a little smoother(from almost perfectly smooth) to go to the length of learning a complete new driving technique?

 

[///M-Spec]

I still don't see any reason to do it.

Do you like driving stick?


M

 

[Perfect Balance]

Do you like driving stick?


M

Yes, but that is different. With a manual transmission, you have full control of the car, and you can use it to brake and provide more power when needed. You can't achieve the same thing on an automatic as with a manual transmission, but you can achieve the same result as double clutching, without doing it.

EDIT: Also, driving a manual car is fun. How is double clutching any more fun than a regular heel-toe style downshift? All you are doing is moving your foot an extra time.

 

[backspace]

EDIT: Also, driving a manual car is fun. How is double clutching any more fun than a regular heel-toe style downshift? All you are doing is moving your foot an extra time.

Double de-clutching and heel toe are two different techniques for different situations. The heel toe technique is only used in braking situations.

My car's 3rd gear synchros are pretty well shot, so when I need to quickly downshift from 4th to 3rd I always double clutch.

 

[Perfect Balance]

Double de-clutching and heel toe are two different techniques for different situations. The heel toe technique is only used in braking situations.

My car's 3rd gear synchros are pretty well shot, so when I need to quickly downshift from 4th to 3rd I always double clutch.

I tried to imply "not while braking" by saying heel-toe type of downshift, instead of the actual thing. I mean just blipping a throttle a little while downshifting. And again, you do it because your synchros are shot, not just for the heck of extreme smoothness.

 

[///M-Spec]

Yes, but that is different. With a manual transmission, you have full control of the car, and you can use it to brake and provide more power when needed. You can't achieve the same thing on an automatic as with a manual transmission, but you can achieve the same result as double clutching, without doing it.

EDIT: Also, driving a manual car is fun. How is double clutching any more fun than a regular heel-toe style downshift? All you are doing is moving your foot an extra time.

A manual doesn't help you brake any faster. With modern braking systems, the tire is the no.1 limiting factor. Nor does it create extra power. It just allows you direct control over what portion of the torque curve the car should use.

You DO have more control over the car, but for 98% of normal driving conditions, a manual offers no exceptional advantage over an automatic other than it is more efficient at transmitting power than a slushbox.

As for the double-clutch smoothness issue, I've been driving stick for 12+ years. In my experience, the shifter slides into gear more smoothly when you do it. Don't take this to mean I think the entire process is smoother --just the shifting.

The real issue seems to be a question of effort vs. reward. You think the effort of double clutching has no reward. And that's fine. I'm not trying to change your mind.

I am trying to illustrate that different people will have different perceptions of what is worth their while and what isn't. Many people feel the downsides of driving a stick at all outweigh the benefits. Is it worth all the work just to get that extra 2-3% of performance? Most people would say no. Why not save a whooping 50% of effort (by keeping 1 arm and 1 leg perfectly still) and drive an auto?

You and I may say otherwise, and it really just comes down to personal preference.

I sometimes double clutch for no particular reason other than it's fun. The same way I sometimes play a video game for no particular reason other than I get to meet interesting people on the Internet and shoot them full of holes. My wife thinks it is a colossal waste of time and from her point of view, she's right. But it ain't her time I'm wasting (though she thinks otherwise)...


M

 

[Perfect Balance]

A manual doesn't help you brake any faster.

I never said it does. Although in theory it could, since my friends car is RWD, using the engine to brake keeps the car from tilting forward as it would hard under braking. So you can use the engine and use less brakes to keep the car's weight more evenly distributed. I'm sure its just a foot or 2 of difference, maybe not even enough to measure it, but it keeps the brakes from fading on mountain roads.

The real issue seems to be a question of effort vs. reward. You think the effort of double clutching has no reward. And that's fine. I'm not trying to change your mind.

You and I may say otherwise, and it really just comes down to personal preference.



M

That's a good way to put it/sum it up. +rep

 

[Scaff]

I never said it does. Although in theory it could, since my friends car is RWD, using the engine to brake keeps the car from tilting forward as it would hard under braking. So you can use the engine and use less brakes to keep the car's weight more evenly distributed. I'm sure its just a foot or 2 of difference, maybe not even enough to measure it, but it keeps the brakes from fading on mountain roads.

Sorry but could you clarify exactly what you are describing here. Are you simply talking about braking to slow the car in gear or using engine braking in addition to the brakes when down-shifting?

BTW - The method of braking does nothing at all to change the amount of load that is actually transferred to the front of the car, regardless of the degree of dive you actually see or feel.

Under deceleration load transfers to the front. Contrary to popular belief, you can't change the amount of load transferred by adjusting shocks, springs, anti-roll bars, or by varying anti-dive or anti-squat geometry. The amount of load transferred is determined by three things; 1) the height of the car's centre of gravity (a car with a high CoG will transfer more load than one with a low CoG), 2) the wheel base of the car (a shorter wheelbase will transfer more load than a longer wheelbase), and 3) how fast the car is decelerating (a car decelerating at 3g's will transfer more load than one decelerating at 1g).

In-gear engine braking combined with the actual brakes is pretty much the norm, and if that is what is being described here, would have very little effect on the actual load transfer itself. As the only alternative is in-gear braking with the clutch engaged, I can only assume you are not talking about this. I am also going to ignore using engine braking only, as while this is a common technique off-road or on very steep descents, you mentioned using engine braking combined with the actual brakes themselves.

However in both of these scenario's the actual level of load transfer is still determined by the three factors above, not the method by which the degree of deceleration is achieved.

The only other situation I can think of is the use of engine braking, combined with the actual brakes during a down-change. If that is the case then, sorry but its plain wrong and theory does not support it at all.

I posted the following over in the other thread on downshifting to slow down, which is very relevant to this discussion

My thoughts match Duke's on this one.

The following may be of interest, its a post of mine from the GT4 & Brakes thread, which looked at how GT4 braking matches with real world braking. The following post covered engine/compression braking and may be of interest.

Compression Braking

Compression braking or as it is commonly know Engine braking is an often used and just as often misunderstood term. I hope in this post to clarify exactly what it is and how it functions as well as discuss its use, miss-use and disadvantages for track use.

In a future post I will also be looking at how Compression braking has been implemented in GT4.


How does Compression Braking work?

In the simplest terms Compression Braking is using the engine to slow the car. It occurs because unless the clutch is engaged and/or the brakes are applied the speed of the driven wheels is determined by the engine; remove your foot from the throttle and the engine will slow and with it the driven wheels. However if you remove your foot from the throttle and engage the clutch, the engine and driven wheels are not connected and the only thing slowing the car will be friction from the tyres and the resistance of the air, as a result the car will take much longer to stop.

This is all fairly simple stuff and it is possible to judge approximately what speed a car should be doing at any engine speed in a particular gear, using the ‘MPH per 1,000 rpm’ figure, which is derived from the cars gear ratios, axle ratio and tyre diameter.

If for example we take the Ford GT, with the following MPH per 1,000 rpm figures as an example.

1 = 9.8
2 = 15
3 = 20.1
4 = 27.2
5 = 33.2
6 = 40.6

Now using these figures we can examine Compression braking both in gear and through the gears.

If the car is in 3rd gear at 5,000rpm the car speed would be approx 100.5 mph, if you lifted off the throttle completely and relied on the drop in engine speed to slow the car the speed would drop according to the engine speed (if you remain in 3rd gear) as follows.

5,000 rpm = 100.5 mph
4,000 rpm = 80.4 mph
3,000 rpm = 60.3 mph
etc

This clearly shows the relationship between allowing engine speed to drop and it slowing the car; the amount of time this would take depends on a number of factors, such as engine compression ratio (generally the high this is the quicker the drop in engine speed) and flywheel and driveshaft weight. Using this method to slow the car down will always be far, far slower that using the brakes.


Compression braking when changing down through the gears can have a far more dramatic effect as illustrated below (using the Ford GT figures).

The car is travelling at approx 100 mph in 3rd gear (approx 5,000 rpm) when the driver begins to brake slowing the car to 50mph, the speed required to take our imaginary corner.

At this point the gear is changed to 2nd with the revs at approx 2,000 rpm, when the clutch is disengaged the engine speed dictates a speed of 30mph (2 * 15mph per 1,000 rpm).

The car is travelling at 50mph, but the engine speed and gearing dictate that it should be travelling at 30mph. This 20mph difference in the speed the car is travelling at compared with the speed the engine wants to travel at is going to be transferred to the driven wheels and slow the car in a dramatic and un-controlled manner. In turn the wheels will also try to resist this rapid deceleration and speed up the engine.

If however the driver had ‘blipped’ the throttle to approx 3,300 rpm before the clutch is disengaged then the engine speed would match the car speed and no additional braking would be experienced, keeping the car stable.

Compression braking on the road
Many driver use compression braking on the road when changing gear to help slow the car and with older cars it can help if the brakes are poor or fading. With modern cars it is generally considered unnecessary as the braking system is more that sufficient.

Many people would argue that using compression braking saves on wear and tear on the brakes, however just as many people would say that what you save in pads and discs you lose in wear and tear to the engine and drive-train. Its hard to say who is 100% right on this one as it would vary from car to car and driver to driver.

On a personal level I don’t using compression braking when changing down as I consider the sudden braking force it generates too sharp and prefer the smoother experience you get with heel and toe downshift that match engine speed to road speed.

One useful aspect of in-gear compression braking is often used in off road driving when descending a steep slope, here the rev limiter and knowledge of mph per 1,000 rpm can be used to control your descent speed without touching any of the pedals.


Compression braking on the track
I would be surprised if many professional drivers use compression braking (even on endurance events) and I am yet to come across a race school that recommends its use. While it does help conserve the pads, it does no good to a race tuned engine; they are not designed to slow the car, but to power it.

With Compression braking it is very difficult to judge how much additional deceleration you will get, as a result if you are already at the threshold of braking (or near it) you could well overload the tyres grip level and lose control, flat spots on the tyres will result which will hammer your lap times.

Changing the pads on an endurance race spec car is relatively quick and easy, and a task regularly carried out in the pits during endurance racing. It is also far less time consuming to have to change the pads, than it is replace an engine component (or retire through engine failure) or lose time over a number of laps because you've just flat spotted a tyre.

I have watched numerous endurance races, and have session reviews of Le Mans dating back to the 1960's, also a documentary on the Morgan team at Le Mans and Bathurst. I can't remember ever hearing a driver talk of using compression braking out of choice. Now if the brakes have failed for some reason, you may not have a choice; but that’s a different thing.

You should also remember that compression braking will only directly effect the driven wheels, shifting the brake bias in that direction. With the majority of race cars being rear wheel drive, this would mean a major brake bias to the rear wheels, if the car is anything other than straight you may now be in a situation of just trying to control the car rather than brake and setup the car for the corner entrance. Even if the car is in a straight line, if the car is running a high compression engine (which increases the effect of engine braking) it can be enough to lock the rear wheels and get the back of the car twitching. Neither of these scenarios is worth conserving a little bit of brake pad material. You would get more of a saving on the brakes through good brake control and technique than you ever will through engine braking.

A quick quote from Danny Sulivan (of the Skip Barber racing school) illustrates this point
"To put it into perspective, at Laguna Seca, which is hard on brakes, Rick Mears and I were team mates at Penske and Rick finished the race with only 70 thousandths of an inch of brake pad material left. I only used 70 thousandths of the pad in winning the race. People brake differently but can still run the same lap time, especially in a race"


The following is an extract from the Russ Bentley books "Speed Secrets - Profressional Race Driving Techniques"

"Again, the reason for downshifting is not to slow the car. I can't emphasise this enough. That's what brakes are for. Too many drivers try to use the engine compresion braking effect to slow the car. All they really achieve is upsetting the balance of the car and hindering braking effectiveness (if the brakes are right at the limit before locking up and you then engine braking to the rear wheels, you will probably lock up the rear brakes), and more wear and tear on the engine. Brake first, then downshift."


This is from Skip Barber's "Going Faster"

"What downshifting is really for.
We ask this basic question of every racing school class. The most frequent (and incorrect) answer is, "to help slow the car down." In a racecar with good, durable brakes (the majority of modern racecars), downshifting to help slow the car down is unnecessary. The brakes slow the car down. You downshift to get the car in the proper gear to exit the corner."

Now Russ Bentley has raced Indy cars, World Sports cars (including endurance) and is now a race instructor, Skip Barber should need no introduction, but the book I refer to has been written with the assistance of ten instructors from the Skip Barber Racing School.

Brakes slow the car, not the engine; unless your brakes are shot in which case you do not have a lot of choice, but unless you're Moss or Fangio you're not going to win like this.

Problems with Compression braking on the track

1. Its less effective than normal braking.

As Compression braking effects only the driven wheels it will have a major effect on the brake bias of the car; incorrect brake bias (or brake balance) can increase braking distances significantly. Even with 4WD cars the effect of engine braking is limited to the front/rear split of power distribution and will normally affect the brake bias negatively.

It is also not possible to accurately predict the level of Compression braking or modulate it once it has been applied, making accurate and controlled braking almost impossible.


2. It does not give ‘more’ braking force if you are already at the limit

This one is a common myth of Compression braking, that it will allow you to get more braking force for free. While in the distant past braking systems were not powerful enough to exceed the grip limit of tyres (and this is a long way in the past), modern braking systems are more than capable of exceeding the grip levels (measured in straight line braking as the slip percentage) of the tyres.

If you are already at the braking ‘threshold’ adding more braking force through compression braking is just going to exceed the slip percentage and lock the tyres.

Using the Ford GT example from above and assuming that the car was already braking at the tyres threshold (assume max 10% slip percentage for this example). If when the clutch is released (with the car at 50mph) the tyres are already using the full 10% slip in braking, the additional braking caused by compression braking (20mph vs. 50mph) will exceed the 10% max by a large margin, the rear wheels will lock, the tyres will flat spot, braking distance will actually increase and unless the car is totally straight the major rear bias will possibly cause a loss of control.


When to use Compression braking?
Obviously with road driving compression braking and its use is very much a case of personal preference; however on the track most instructors and drivers share the belief that the disadvantages more than outweigh the advantages (and some would argue that it has no advantages).

However should you find yourself in an older car (historic racing), suffering from brake fade or even brake failure then you have little choice but to use what ever method you can to slow the car.

As Skip Barber’s book says:

”In this case you certainly do use the downshift to slow the car down – but it’s a last resort”

From a personal point of view I always heel and toe downshift, all of them, I have an issue with uncontrolled and unbalanced braking, even braking from high speed and at the threshold of the tyres should be smooth and controlled. I agree with Duke that brake components are far cheaper and easier to replace that driveline components, and I for one would much rather be in control of the braking process. Using the brakes allows you to modulate the level of force being used, engine braking does not allow this and if the conditions are poor its more than possible to lock-up the driven wheels. Not too bad in a front wheel drive car, but a potential accident in a rear wheel drive car.

Once again nothing in this scenario would effect the load transfer, other than the three factors already outlined, in fact the additional deceleration force that would result from engine braking would increase the load transfer to the front, not reduce it. Its this increased load transfer to the front, which will unload the rears and with the additional engine braking to the rear that a RWD car has, will actually increase the chance of rear wheel lock-up.

The quickest way to stop a car, with fully functioning brakes, in any given situation is to make maximum effective use of the brakes and the tyres. As has already been mentioned by M, tyres are what stops your car, its the single biggest factor involved. Load transfer, brake bias, etc are simply factors that determine how much of the tyres ultimate grip you can use, and on a modern road car you always have the ability to exceed the tyres grip level.

Compression/engine braking does not in anyway give you more braking ability than your tyres can ultimately offer, and automotive engineers know this, its why brake balance is always given a significant bias to the front of a car is. The place than under deceleration will have the great load and the greater grip. Utilising the maximum potential of the tyres is all about balancing the front and rear brake force and the front tyres will always be able to offer more grip under braking than the rears.

You are quite right that compression braking can help if you are suffering from brake fade and can allow you to lower the strain on the brakes to help keep it at bay. However it is neither as effective as the cars braking system, nor can it be modulated in anyway, brakes can. The single biggest reason that uprated brakes are fitted to cars is to alleviate the braking and ensure that repeated stops do not result in an early onset of brake fade. What it will not do is actually stop the car any quicker; providing the current braking system is capable of exceeding the tyres grip limits and for any remotely modern braking system in good order that I can assure you will be the case (unless the manufacturer serious messed up on the design).



Regards

Scaff

 

[Perfect Balance]

Thanks for that post Scaff. Let me explain my idea a little better. In my friends Miata, the front wheels have a lot more braking power than his rear, like most cars. His rear rotors need to be changed though, so when the front wheels are at their threshold, the back ones are still rolling. By down shifting, you can use compression braking along with the back brakes to bring the rear tires to their threshold without locking the front. A normal car shouldn't do that, but since his rear rotors are worn, there is significantly less braking force there, not even enough to lock the wheels with the handbrake. The car has very neutral handling, so even when you release the brakes, it doesn't react very different than while braking, just the speed is higher, making the car easy to control even though the rear tires are almost sliding.

So I guess you are right, or Skip Barber Racing school, but when the brakes are messed up, It's one thing you can do to solve the problem.

Edit:This isn't exactly the same as my idea before, but I found a fault with that one, and I fixed it in this example.

 

[wfooshee]

I still don't see any reason to do it. As long as you get the revs right, the gear will go in smoothly, whether you double clutched or not. The transmission is always moving because the wheels are rolling, so all you have to do is line up the engine speed with it. If done right, it will be very smooth, we tried it on my friends Miata today a couple times. If done right, you don't feel the shifts at all. Sure, if you don't have synchronizers, then I see why, but I'm sure all the people who do it on here have synchronizers in their cars. I'm either overlooking something here, or there truly is no point. The synchronizers barely have enough time to slow down on a downshift anyways, and unless you have a really wide gear ratio, it should only be a difference of 1,500 rpms or so.(which isn't a lot) Even less since you are slowing down between the downshifts too, letting the engine drop a couple hundred rpm's in between.

EDIT: Is it so desirable to be a little smoother(from almost perfectly smooth) to go to the length of learning a complete new driving technique?

What you're looking at is my Probe with the end cover off of the transmission, as I prepared to replace 5th gear for a worn syncro and shift fork. Bear with me here because I'm going to show you exactly why double-clutch downshifts are better for your parts. To the left is the input shaft gear for 5th, and to the right, the end of the stack, is reverse. In between is the shift mechanism, and at the bottom you see the shift fork, with a worn section cut out of it. (It was worn by pressure from fifth gear not resting in place, it wanted to pop out, so the selector actually lathed the fork down over time.)






These views are from the end of the gearbox, showing the position of the output shaft down from and beside the input shaft. From the end, you're looking at reverse.


A modern manual street transmission is synchronized constant mesh transmission. That means the actual gears are ALWAYS meshing, whether they're shifted in or not. Each ratio has a gear on the input shaft which freewheels (not splined to the shaft) and a gear on the output shaft which is fixed to the shaft with splines. The output shaft NEVER releases from the drive wheels, and ALL of its gears are always spinning at the output shaft's RPM. The input shaft is connected to the engine by the clutch. If the clutch pedal is down and the box is in neutral, the input shaft floats free and may even stop completely. (Not realistically, there is enough friction and viscosity to drag it along a little bit.) If the clutch pedal is released, the input shaft is spinning at engine RPM.

The only things splined to the input shaft are the baulk rings. The "sleeve," if you will, that is carried by the shift fork. Inside the sleeve is a set of grooves that match up to the dog teeth you see on the brass-colored ring and the actual gear behind it. The brass-colored ring floats freely on the shaft, and has a cone-shaped surface inside that matches the surface on the end of the actual gear. Now, to engage a ratio. When you move the shift lever, a mechanical link slides one of the shift forks towards the gear you're selecting. Let's use 5th, since that's in my picture. Those grooves in the sleeve are spinning at the same speed as the input shaft. When you shift, the input shaft is released from the engine, the sleeve slides toward the brass-colored teeth. It can't slide its grooves over the teeth until its speed is matched to the gear (hence "baulk ring"), which is being driven by its output shaft partner. That's what the cone-shaped ring does: it provides a friction surface between the gear and the selector sleeve. When the speeds match, the sleeve's grooves slide over the syncro ring all the way onto the dog teeth of the gear. Now the input shaft is tied to the gear, and the ratio is selected. You release the clutch and the shift is complete. If the sleeve slides on before the speeds match, you get the classic crunch sound. It's not from gears not meshing, it's from the dog teeth being at a different speed from the selector ring.

Now. You're cruising along in top gear and need to downshift. All those selector rings are spinning at engine RPM (they're splined to the input shaft.) You depress the clutch, blip the throttle, move the shifter, etc. The selector rings begin this process at a slower RPM than they need to be, because the input shaft was turning at the engine's top gear RPM, not it's 4th or 3rd gear RPM (much higher). They may even be coasting down a bit during the de-clutched period. That means the syncro has to do just that much extra work to match the dog teeth to the selector ring. If you release the clutch briefly in neutral and blip the throttle, you spin up the selector rings. Remember, they're the only parts splined to the input shaft. Now clutch pedal down, select the gear, another blip, and release the clutch. The syncro has less work because you've spun them up between gears. The first blip, in neutral, spins up the selector rings, the second blip, in gear but de-clutched, matches the engine revs to the input shaft of the tranny.

That's why double-declutch downshifting is both smoother and better on the equipment. It's also why double-declutching makes no sense on upshifting. You actually want the baulk rings to slow down during the shift.

Some gearboxes have stouter synchronizers than others. Obviously mine had a bit of a problem in 5th, but not from downshifting. I had a problem where a shop used the wrong part in a CV joint replacement, and the axle slid partway out of the box, allowing transmission oil to spill onto I-10, in turn causing the gearbox to get quite warm. Normal driving, you won't really notice much difference, if any, using single- or double-clutching. But double is better on the internal parts.

Thanks for that post Scaff. Let me explain my idea a little better. In my friends Miata, the front wheels have a lot more braking power than his rear, like most cars. His rear rotors need to be changed though, so when the front wheels are at their threshold, the back ones are still rolling. By down shifting, you can use compression braking along with the back brakes to bring the rear tires to their threshold without locking the front. A normal car shouldn't do that, but since his rear rotors are worn, there is significantly less braking force there, not even enough to lock the wheels with the handbrake. The car has very neutral handling, so even when you release the brakes, it doesn't react very different than while braking, just the speed is higher, making the car easy to control even though the rear tires are almost sliding.

So I guess you are right, or Skip Barber Racing school, but when the brakes are messed up, It's one thing you can do to solve the problem.

Edit:This isn't exactly the same as my idea before, but I found a fault with that one, and I fixed it in this example.

A lot of cars with rear disc brakes won't lock the wheel with the handbrake. The brake mechanism is different with discs than drums. There's either a small clamp that feebly squeezes the pads, or a little-bitty drum brake in the hub.

You generally want the fronts to lock up before the rears. Going straight with no steering, then hitting something straight on (where belts and airbags work best) is better than spinning out of control and hitting something sideways or backwards.

 

[Perfect Balance]

A lot of cars with rear disc brakes won't lock the wheel with the handbrake. The brake mechanism is different with discs than drums. There's either a small clamp that feebly squeezes the pads, or a little-bitty drum brake in the hub.

You generally want the fronts to lock up before the rears. Going straight with no steering, then hitting something straight on (where belts and airbags work best) is better than spinning out of control and hitting something sideways or backwards.

I'm not talking about in general, I'm talking about this specific car. It can lock the wheels using the handbrake normally. And yes, I know you want to front to lock up first, but by downshifting you can provide some extra braking force to the rear.

 

[Keef]

You can. But it isn't advised, because it reduces the grip available to the rear tires that could be used to hold a line in a corner, or even going in a straight line. If you get the rear tires closer to their braking threshold, there could be negative side effects in the form of oversteer.