How to select Hewland Gearing?

[Jay Laifman]

Edit: I found tons of threads with people talking about gear selection for various courses. That helped a lot. So I should probably withdraw these questions! I'll see if I can delete the thread.



We need to select the gearing for our Elva Mk6 with Alfa engine. I've now looked at quite a few different gearings that various people have on their cars. I understand that there are differences between short courses and long courses, and differences between wide torque bands and short bands, and torque curves that drop off at the top quickly. But I'm not fully understanding some of the differences.

Here are my main questions - we're trying to get to an initial gear set that we'll change as we get experience with the car on the track. But before that, we still need something to start with!

(1) I've seen gear selections that have different spacing between them. For example, measuring from speed at redline to redline, I've seen gaps from 17 mph to 28 mph. I'm assuming this is based on how long of a power band a car has. Are there any rules of thumb for determining what bands will work with a given engine?

(2) I've seen some that have equal spacing all the way up, and some that get narrower as they go up in gears, and some that have 3rd pulled down. Am I understanding this right that absent a big drop off in torque curve, it is almost always better to go up to redline than to shift up? And if that's true, does that mean that absent any other factor that equal gaps would be correct? So what factors would cause someone to have narrower gaps as they go up? Just to have the right rpm for turns at a particular track configuration? Like having to shift mid corner.

(3) I have also seen that some people push up 1st. This means the car will have slower rate of acceleration, but it will allow the car to stay in 1st longer for higher acceleration before having to shift into second - which also allows to push up the ultimate top speed, as each gear goes up. Are there any rules of thumb for setting 1st? Obviously it doesn't need to be as low as a non-race car, but you don't want it so high that it doesn't accelerate fast enough. How do we set that?

Thanks for any thoughts on any of these.

[TimH]

Rough rule: you'll want to shift up before HP begins to roll off. Your engine builder may also recommend a particular redline for safety. Then choose your top gear so you'll be near redline at your maximum speed (typically at the end of the longest straight) at a given track. You'll choose first gear to give reasonable HP in the slowest corner. Middle gears then can be evenly spaced, or with slightly diminishing intervals up through the sequence - opinions differ (I think the reasoning behind narrowing the gaps is that as speed builds there is more aero drag and you need to stay higher up in the power band). Then for a particular track you might go a little taller or shorter on a middle gear to avoid having to shift in an inconvenient place.

[Daryl DeArman]

To piggyback on what TimH said. Ideally, you don't want your rpms to drop below the rpm where your motor delivers its peak torque. You don't want your rpms to go beyond where you make your peak HP. Given that we only have 4 gears to select and the speed delta between slowest corner and fastest straight speed might be too much to accomplish both, some compromises are made. Compromises made in lower gears (higher numerical ratio) will have less effect on acceleration.

In practice you may find that splitting the gears to accomplish the best acceleration from 1st to 4th puts a shift in a spot(s) on the track where you don't want to shift, so changing a ratio away from ideal will hurt acceleration, but could help lap times and minimize mistakes.

[Jay Laifman]

For what it's worth, in playing with various gearing calculators, it became pretty clear that it is always better to go to redline, even after power drop off, unless there is an unusual spike down in power.

I did come across this chart: https://www.apexspeed.com/resources/...ne_gearing.htm

It is for FC. But it really helps show the variations. And I believe this is for three different engines, 1100 cc, VW, and FF, yet all same baselines. It also shows how there are no rules of thumb! I hear that's for people with all thumbs anyway.

It's too bad this chart doesn't have Buttonwillow or Laguna Seca. But I have some other sources on those now.

[Jim Edmonds]

Good chart. The Phoenix data looks good for the East Track and West Track. There is probably a transposition in the first gear. S/B 16/34, not 16/43.

The Main Track takes taller gears.

Have fun today

Jim

[Daryl DeArman]

it became pretty clear that it is always better to go to redline, even after power drop off, unless there is an unusual spike down in power.

It also shows how there are no rules of thumb!

You should always avoid speaking in absolutes

[Jay Laifman]

I put the chart in a spreadsheet and calculated the speed at redline (assuming 7k, 22" tires, 10/31 R&P) and then the spread of mph between redlines for the gears. It further shows how little consistency there is in that as well, and how track specific it really is.

Track	1st Gear Speed at 7k redline	2nd	3rd	4th	Gap b/w 1st and 2nd @ Redline	2nd and 3rd	3rd and 4th
Atlanta Motor Speedway	19/31 91	22/30 108	24/28 127	23/24 141	17	19	14
Blackhawk Farms Raceway	18/32 83	20/31 95	22/29 112	21/25 124	12	17	12
Brainard International Raceway	18/34 78	20/30 99	23/28 121	23/24 141	21	22	20
Carolina Motorsports Park	17/29 86	17/25 100	22/29 112	24/27 131	14	12	19
Charlotte	19/31 Same as Atlanta	22/30	24/28	23/24	17	19	14
Corpus Christi (2.8 mi. Course C)	19/33 85	20/31 95	21/27 115	24/26 136	10	20	21
Daytona International Speedway	20/30 99	22/30 108	24/28 127	23/24 141	9	19	14
Des Moines	17/34 74	19/33 85	21/30 103	23/29 117	11	18	14
Firebird International Raceway	19/33 85	21/31 103	22/29 112	24/27 131	18	9	19
Gateway International Raceway	18/32 83	20/30 98	22/28 116	24/26 136	15	18	20
Gingerman Raceway	17/33 76	19/31 91	21/30 103	22/28 116	15	12	13
Grattan Raceway	17/34 74	19/32 88	22/30 108	24/26 136	14	20	28
Hallett Motor Racing Circuit	16/35 68	18/33 80	20/30 99	23/28 121	12	19	22
Heartland Park Topeka	17/34 74	20/31 95	22/29 112	21/24 129	21	17	17
Indianapolis Raceway Park (Oval)	19/31 91	21/31 103	24/28 127	23/27 126	12	24	-1
Indianapolis Raceway Park (Road Course)	18/32 83	20/30 98	23/29 117	25/27 137	15	19	20
Lakeland	18/32 83	20/30 98	22/29 112	23/29 117	15	14	5
Lime Rock	20/30 99	21/29 107	23/29 117	24/26 136	8	10	19
Memphis Motorsports Park	17/33 76	20/33 90	22/29 112	24/26 136	14	22	24
Mid Ohio	18/33 81	20/31 95	21/29 107	21/24 129	14	12	22
Moroso	17/33 76	19/31 91	24/28 127	24/26 136	15	36	9
Mosport	18/34 78	20/30 98	22/28 116	24/27 131	20	18	15
Nelson Ledges	18/33 81	20/30 98	23/29 117	24/26 136	17	19	19
New Hampshire International Speedway	17/33 76	20/31 95	23/29 112	24/26 136	19	17	24
Phoenix (Road Course)	16/34 70	19/31 91	21/29 107	21/24 129	21	16	22
Portland International Raceway	19/31 91	21/29 107	24/28 127	24/26 136	16	20	9
Road America	18/32 83	20/30 98	23/28 121	23/24 141	15	23	20
Road Atlanta	19/32 88	21/31 103	23/28 121	23/24 141	15	18	20
Roebling Road	19/32 88	21/30 103.5	23/28 121	24/27 131	15.5	17.5	10
Sears Point	17/34 74	20/31 95	22/29 112	24/27 131	21	17	19
Sebring (Long Course)	18/34 78	20/30 98	24/28 127	24/26 136	20	29	9
Sebring (Short Course)	17/33 76	19/31 91	22/29 112	24/27 131	15	21	19
Shanonville	16/34 69	19/31 91	21/29 107	23/28 121	22	16	14
St. Petersburg	17/33 76	19/31 91	22/29 112	24/27 131	15	21	19
St. Jovite	17/35 72	19/31 91	22/30 108	24/27 131	19	17	23
Summit Point	19/33 85	20/30 98	23/29 112	24/26 136	13	14	24
Texas World Speedway (1.8 "school" config)	19/32 88	22/30 108	23/28 121	23/24 141	20	13	20
Texas World Speedway (2.9 config.)	18/33 81	20/30 98	23/29 112	24/26 136	17	14	24
Trois Rivers	17/33 76	19/32 88	21/30 103	23/28 121	12	15	18
Virginia International Raceway	18/33 81	20/30 98	22/29 112	24/26 136	17	14	24
Waterford Hills	17/34 74	19/31 91	21/31 100	22/29 112	17	9	12
Watkins Glen (Long Course)	20/30 99	21/30 103	23/28 121	24/26 136	4	18	15
Watkins Glen (Short Course)	20/32 92	21/30 103	23/28 121	24/26 136	11	18	15
Willow Springs	19/32 88	21/30 103	22/28 116	24/26 136	15	13	20

[BorkRacing]

Let's define redline. The highest sustainable RPM at which the motor will run for a race/season. It is not peak torque or HP. By this definition Torque and therefore HP drops as you near redline. For example a FF motor has peak torque at 6200 RPM, redline is 7000. Shift at 6200, go to 7000 if in a draft (less aero drag)

Aero drag continues to climb with speed.

The force (torque) required to increase your hypothetical speed from peak HP to redline increases dramatically due to the increasing aero drag. Remember that when you double your speed you quadrouple your aero drag. Your HP/Torque is decreasing so your rate of acceleration is also decreasing. The bottom line is:

Your hypothetical speed from your gears at max HP is more true to the real world results than the speed calculated for redline.

[Robert J. Alder]

To piggyback on what TimH said. Ideally, you don't want your rpms to drop below the rpm where your motor delivers its peak torque. You don't want your rpms to go beyond where you make your peak HP. Given that we only have 4 gears to select and the speed delta between slowest corner and fastest straight speed might be too much to accomplish both, some compromises are made. Compromises made in lower gears (higher numerical ratio) will have less effect on acceleration.

In practice you may find that splitting the gears to accomplish the best acceleration from 1st to 4th puts a shift in a spot(s) on the track where you don't want to shift, so changing a ratio away from ideal will hurt acceleration, but could help lap times and minimize mistakes.

Nicely put.

[BorkRacing]

To piggyback on what TimH said. Ideally, you don't want your rpms to drop below the rpm where your motor delivers its peak torque. You don't want your rpms to go beyond where you make your peak HP. Given that we only have 4 gears to select and the speed delta between slowest corner and fastest straight speed might be too much to accomplish both, some compromises are made. Compromises made in lower gears (higher numerical ratio) will have less effect on acceleration.

In practice you may find that splitting the gears to accomplish the best acceleration from 1st to 4th puts a shift in a spot(s) on the track where you don't want to shift, so changing a ratio away from ideal will hurt acceleration, but could help lap times and minimize mistakes.

OMG I agree with Daryl.
Say goodbye to your loved ones. The world ends today! LOL

[Jay Laifman]

Ok, just to throw another factor in. I am the racing newby here (sort of). So happy to be shown where this is wrong. But I do want to contribute to the discussion.

Yes, I agree that hp and torque drop off at peak. Yes, I agree that it takes more and more power to extract more acceleration. However, that doesn't mean at the very second hp/torque starts to drop, it is time to shift. Gearing still comes into play. Upshifting the higher gear immediately increases the force needed to extract more acceleration - just like wind resistance, and the declining power. The cross over point is therefore somewhere past peak hp/tq.

I have seen multiple comments online that in fact, because of the gearing, there is actually more force through the wheels onto the ground in the lower gear to redline, no matter what - unless there is some excessive drop off in power for some reason. There is even an equation that proves this: Force = Tq*Gearing*Ring/Pinion/Diam*24

I actually have a spreadsheet for gearing that has this equation tied to the torque curve. I did put in all sorts of torque curves and even artificially adjusted them - putting in bigger and bigger hp peaks, etc., and nothing ever resulted in the next gear up being a better choice - unless I did something really odd with the top of the torque curve, adding an excessive drop or rev limiter.

Here's an example:
2nd Gear: 15:35 (2.3)
3rd Gear: 16:29 (1.8)
R&P: 10:31
Tire: 22
Force = Tq * gearing * ring/pinion/diam *24
2nd gear Force = Tq * 2.3 * 10/31/22 * 24
3rd gear Force = Tq * 1.8 * 10/31/22 * 24

Using these equations, if I put in 70 lbs in 2nd gear, and keep increasing power in 3rd gear, I have to get to 90 lbs in 3rd gear before it exceeds the force in 2nd. That means that 2nd has to drop off to 70, and my shift into 3rd has to get me back to 90 lbs, before I get more force for upshifting. A 20 lb spread in a torque curve has got to be pretty big!

Let me know what I'm missing.

I do agree that it is really the track that dictates the gear - that is, if you need to have another gear before the end of a straight, or if you'd run out of straight before you'd run out of gear, or a different gear to pull out of a corner, or how soon the first corner is from the start line, etc.

[TimH]

Let me know what I'm missing.

Just one thing that Robby pointed out. Nearly every motor starts rolling off the torque surprisingly early. LONG before redline. There comes a point (about 6200-6500 for FF motors as he said) when the torque drops so quickly that HP is dropping, too. By then you'll often gain more torque at the driven wheels by upshifting back into the power band. Done right, you'll feel the surge each time.

[problemchild]

The proper answer to this question:

Ask knowledgeable experienced racers that are successful (and faster than you) with similar cars what ratios they use.
Use those ratios until you find a better advisor or identify a problem that can be solved by some of this engineering mumble-jumble .... or just plain common sense.

[alangbaker]

Rough rule: you'll want to shift up before HP begins to roll off.

To piggyback on what TimH said. Ideally, you don't want your rpms to drop below the rpm where your motor delivers its peak torque. You don't want your rpms to go beyond where you make your peak HP.

I'm sorry, but those are both incorrect.

The correct way to select shift points is to maximize the area beneath the power curve.

Power combines the effects of torque and RPM to let you know what torque you will get at the rear wheels.

If you automatically shift up at the HP peak, you will have less area under the curve from lowest to highest RPM used than if you shifted a little later. Similarly, if you tried to shift a Kent such that you never went below peak torque, your RPM could never drop below about 6000-6100.

There is a spreadsheet I have that lets you input your dyno data, your gear sets, your final drive and your tire diameter and it lets you see precisely where the torque AT THE REAR WHEELS goes from being more in the lower gear to more in the next higher gear. Let's take a single example:

With my 17:33 first and 17:28 second gears, the torque at the rear wheels lines cross at approximately 6420 RPM in first which then drops to 5500RPM (actually 5497) in second. At the first data point after that—6500 RPM in first, the HP has dropped to 103.5 where at the RPM you'd be at in second (5565) the engine is producing more than the 104.6HP at the 5500 data point.

The only correction applicable is for frictional losses. The actual answer is area under the HP curve after taking into account the power lost to mechanical and aerodynamic drag, and it is aero HP that rise as the cube of your speed (double your speed, and you move twice as much air, twice as fast, while covering twice the distance—i.e. 8 times the power required).

This is why the ratios get closer together through the gears.

[Jay Laifman]

I wonder if your spreadsheet uses the equation I had above for those calculations: Force = Tq * gearing * ring/pinion/diam *24

[Daryl DeArman]

delete

[Daryl DeArman]

I'm sorry, but those are both incorrect.

The correct way to select shift points is to maximize the area beneath the power curve.

If you automatically shift up at the HP peak, you will have less area under the curve from lowest to highest RPM used than if you shifted a little later.

That would be completely dependent on the shape of said curve and the splits between gears.

[Daryl DeArman]

Similarly, if you tried to shift a Kent such that you never went below peak torque, your RPM could never drop below about 6000-6100.

No Kent dyno sheet that I've ever seen makes it's peak torque anywhere remotely close to 6000-6100rpm. You are talking about peak HP rpm or you're off by a couple thousand RPMS. Or, I completely missed your point.

[alangbaker]

That would be completely dependent on the shape of said curve and the splits between gears.

No. It absolutely would not.

This is mathematics, and it is undeniable.

[alangbaker]

No Kent dyno sheet that I've ever seen makes it's peak torque anywhere remotely close to 6000-6100rpm. You are talking about peak HP rpm or you're off by a couple thousand RPMS. Or, I completely missed your point.

I beg your pardon!

I referenced the HP peak by mistake from my own dyno numbers. Peak torque was 4300.

It doesn't change the fact that what one is trying to do is maximize the area under the horsepower curve... ...after taking into consideration HP lost to friction/drag. That is the only mathematical basis for shift points.

[alangbaker]

I wonder if your spreadsheet uses the equation I had above for those calculations: Force = Tq * gearing * ring/pinion/diam *24

Nope. It stops at rear wheel torque which is sufficient because the rear wheel diameter is a constant.

[mark elder]

I have never quite understood why so many people use torque in any of these sort of calculations. I would have thought that we are only interested in power (and for gearing, power under the curve between the change points; third gear might be 5500 and 6700). We derive power as work done in a given amount of time. We use torque to get the forces and use rpm to get time (and we multiply them together to get power). Time is just as important in this equation. I am not that interested in torque at the wheels, I want power at the wheels over the rpm determined by my gearing.

Just my pennies worth.

[BorkRacing]

I have never quite understood why so many people use torque in any of these sort of calculations. I would have thought that we are only interested in power (and for gearing, power under the curve between the change points; third gear might be 5500 and 6700). We derive power as work done in a given amount of time. We use torque to get the forces and use rpm to get time (and we multiply them together to get power). Time is just as important in this equation. I am not that interested in torque at the wheels, I want power at the wheels over the rpm determined by my gearing.

Just my pennies worth.

HP is a derived number from torque based on RPM. The higher you spin the motor the more HP you have.
Once you are beyond the peak of the torque curve the HP numbers begin to become useless except for bragging rights.

[Jay Laifman]

Here are two pages supporting why it is best to shift at redline (or as high above or below that as you feel you can, based on how long you want your engine to last!).

http://www.welltall.com/ymc/discovery/car/shiftpt.html

http://glennmessersmith.com/shiftpt.html

[BorkRacing]

Here are two pages supporting why it is best to shift at redline (or as high above or below that as you feel you can, based on how long you want your engine to last!).

http://www.welltall.com/ymc/discovery/car/shiftpt.html

http://glennmessersmith.com/shiftpt.html

You are missing the important information here or have never seen a FF or FC torque graph.

From your source: Optimal shift point is defined as the flywheel rpm for which the difference in transmission output torques between two gears is near zero for the same output shaft speed (and same road speed). If the difference never approaches 0, the optimal shift point is at redline.

This utility does this calculation for every 50 rpm step within the range of data you provide, using straight line interpolation between data points. It returns the rpm in each gear which yields a torque drop to the next gear as close to zero as possible. In a output torque vs road speed graph, this would be where the torque curve lines for two adjacent gears intersect. In a car with a flat-ish torque curve and optimal street gearing, the gear-torque curves do not intersect. This means the optimal shift points will be redline in each gear.


We do not have flat torque curves and when you graph the speed vs torque for each gear you will see that they do intersect below redline.

Follow Problemchild aka Greg Rice's advice. Use what other people are using until you beat them consistently.

[Daryl DeArman]

No. It absolutely would not.

This is mathematics, and it is undeniable.

I'm going to let you believe you are correct.

For the other folks who are open to learning, yes we are trying to maximize the area under the curve. Whether or not maximizing the area under the curve dictates going beyond the peak depends on the shape of the curve. A plot that nosedives faster than it climbed will have less area under the curve after the peak than before. The farther we go beyond that peak the more area we are losing. A curve that falls off slower than it climbed will give more area under the curve if we go beyond the peak, and the farther beyond peak we go, the better. A curve that is symmetrical will yield the same area on each side of the peak, but the most area the closer we stay to peak. 500 rpms each side of peak (1000rpm spread) will net more area than either 1000 before to peak, or peak to 1000 after. The narrower our gear ratios the more we can maximize the area under the curve. Staying 100 rpms each side of peak will produce more area under the curve (average).

Hopefully I explained it in a way that clearly illustrates how the shape of the curve and the ratios we are dealing with have an effect on the results.

[Jay Laifman]

Thanks!

For what it's worth, we're not FF or FC. We are restoring a 1962 Elva Mk6 that uses a 1300cc Alfa Romeo engine. We got it with a Hewland that has completely the wrong gearing - so we have to pick new 1-4 gears. We will be going historic racing with it. It is me and my two college sons, and we're doing this as a team project - like the Prisoner said about his Lotus, they want "to know every nut and bolt on the car." They are accomplished kart racers, with huge understandings of chassis dynamics.

Learning gearing is of course all part of it. We want to understand the whys behind it - and not just throw in what someone says to put in. We've learned a ton with this thread. On the plus side, we won't be racing against FF where there is so much similarity, that picking the right gears for the right track are critical. We probably have a bit less sensitivity to exact gear choice. But, we still want to be competitive, and not disadvantaged by the gears.

So we have to decide if we select a gear set we think is optimum for one of the tracks we will ultimately race at, or if we try to pick a more generic, middle of the road set, that will allow us to start off with. On the other hand, I realize that's kind of like a crossover bike - it has road features and off-road features, but it's not really good at anything.

We are in So Cal, and figure our initial circle of tracks will be Buttonwillow, Willow Springs, Laguna Seca and Sears Point. And from the charts, they each have very different gearing for FF/FC, making it all the harder to figure out an initial set.

[BorkRacing]

Jay,
Not being rude but you just don't seem to get it. You read into things to justify your own ideas. Follow advice, you will be much happier and have less expense rebuilding.

For example in the real world shifting at redline means many times the shift won't be done till 200-400 RPM past your intended spot. How many engines do you plan on using a year? Alfa motors are not happy over-reving.

A Hewland has easily changeable gears, you don't pick one set that will work for most tracks. You have an inventory and change to suit the track.

You're not the first person to run an Elva with a Hewland. Much can be learned from doing what others do and then figuring out why.
There is no need to reinvent the wheel, pun intended.

Robby

[Daryl DeArman]

Thanks!

For what it's worth, we're not FF or FC. We are restoring a 1962 Elva Mk6 that uses a 1300cc Alfa Romeo engine. We got it with a Hewland that has completely the wrong gearing - so we have to pick new 1-4 gears. We will be going historic racing with it. It is me and my two college sons, and we're doing this as a team project - like the Prisoner said about his Lotus, they want "to know every nut and bolt on the car." They are accomplished kart racers, with huge understandings of chassis dynamics.

Learning gearing is of course all part of it. We want to understand the whys behind it - and not just throw in what someone says to put in. We've learned a ton with this thread. On the plus side, we won't be racing against FF where there is so much similarity, that picking the right gears for the right track are critical. We probably have a bit less sensitivity to exact gear choice. But, we still want to be competitive, and not disadvantaged by the gears.

So we have to decide if we select a gear set we think is optimum for one of the tracks we will ultimately race at, or if we try to pick a more generic, middle of the road set, that will allow us to start off with. On the other hand, I realize that's kind of like a crossover bike - it has road features and off-road features, but it's not really good at anything.

We are in So Cal, and figure our initial circle of tracks will be Buttonwillow, Willow Springs, Laguna Seca and Sears Point. And from the charts, they each have very different gearing for FF/FC, making it all the harder to figure out an initial set.

Jay,

Gears aren't expensive and are pretty darn easy to change. Are you familiar with Santo's? Reach out to Anthony (if you haven't already) for some suggestions on gearing for those tracks with that engine and chassis.

Have you thought about doing a couple events with the AROSC before VARA or SVRA? You can get a ton of seat time to sort things out and get the gearing close for a lot lower entry fee and stress-free environment.

Whose car did you acquire?

[Jay Laifman]

Thanks Robby. That's the point - I'm trying to get advice here to learn - not just take a set of gears that someone else thinks is good. I do not see that I'm trying to justify my ideas. Yes, I've added information that I've used trying to understand gearing - but each time saying I don't understand and asked for further input. Yes, I've seen that there is a disagreement about whether or not it's better to shift at redline, and I've found links online that seem to support that view. I provided it here to help understand. And note that I followed that with "or as high above or below that as you feel you can, based on how long you want your engine to last!" - which was your recent point.

I will even point out that I said early on that I was seeing that the shifting at redline was best - unless there was a big drop off in power after peak. That's exactly your point. So I do not see what you are complaining about.

Daryl, we bought 60/23 which has a long history. The most recent owner before us was John Feng (great guy). It was raced in the UK with a 1220cc Coventry Climax (not the standard 1100cc), then when it came to the US, it got the Alfa 1300. With the Alfa, it won the CenDiv Regional Championship one year. Right now it's 100% in pieces. We're slowly going through everything to make sure they are in great condition and to spec. We are restoring it to be very period correct. We expect it will take at least another year before we get it race ready.

[BorkRacing]

Thanks Robby. That's the point - I'm trying to get advice here to learn - not just take a set of gears that someone else thinks is good. I do not see that I'm trying to justify my ideas. Yes, I've added information that I've used trying to understand gearing - but each time saying I don't understand and asked for further input. Yes, I've seen that there is a disagreement about whether or not it's better to shift at redline, and I've found links online that seem to support that view. I provided it here to help understand. And note that I followed that with "or as high above or below that as you feel you can, based on how long you want your engine to last!" - which was your recent point.

I will even point out that I said early on that I was seeing that the shifting at redline was best - unless there was a big drop off in power after peak. That's exactly your point. So I do not see what you are complaining about.

Daryl, we bought 60/23 which has a long history. The most recent owner before us was John Feng (great guy). It was raced in the UK with a 1220cc Coventry Climax (not the standard 1100cc), then when it came to the US, it got the Alfa 1300. With the Alfa, it won the CenDiv Regional Championship one year. Right now it's 100% in pieces. We're slowly going through everything to make sure they are in great condition and to spec. We are restoring it to be very period correct. We expect it will take at least another year before we get it race ready.

Jay,
You seem intent on shifting at redline.
We all tried to help and it seems you will do as you wish.
Best of luck in your restoration.


Robby

[Daryl DeArman]

Daryl, we bought 60/23 which has a long history. The most recent owner before us was John Feng (great guy). It was raced in the UK with a 1220cc Coventry Climax (not the standard 1100cc), then when it came to the US, it got the Alfa 1300. With the Alfa, it won the CenDiv Regional Championship one year. Right now it's 100% in pieces. We're slowly going through everything to make sure they are in great condition and to spec. We are restoring it to be very period correct. We expect it will take at least another year before we get it race ready.

Reach out to Anthony Rimicci at Santo's in Northridge. They've been racing all things Alfa for decades and are familiar with all the tracks you intend to run. They can give you their starting point/suggestions on gearing. You can plug in the numbers and see what you believe would be the best shift points for those gears.

Once you guys get on track you'll find that you sometimes need to shift a bit early because shifting at a pre-determined RPM may dictate a shift right at an apex. Sometimes you might need to hold a gear a bit longer to shift after a crest instead of on-top of a crest. Or upshifting 100 feet before a braking point, just to shift back down again. Alternatively, those nuances probably explain why a particular gear might be a tooth taller/shorter than what the math says should be optimum for acceleration.

[Jay Laifman]

Jay,
You seem intent on shifting at redline.
We all tried to help and it seems you will do as you wish.
Best of luck in your restoration.


Robby

I must be having a terrible time communicating because I think we are agreeing on everything. And I hate to have a misunderstanding like this!

I agree that we shouldn't shift at redline if we want the engine to last longer. I said so above more than once.
I agree that shifting at redline is not correct if the power drops off quickly before redline. I said so above more than once.
I also asked for people to tell me where anything I said was wrong.

I will admit that I think that shifting at peak hp or torque was not always correct. But that is certainly not insisting on shifting at redline.

Maybe you are thinking of things that other people posted above - because others have pushed these points harder than I did.

My deepest apologies if I was not clear.

[problemchild]

I must be having a terrible time communicating

Apexspeed is an incredible resource.
It is, however, on the internet. No level of expertise or knowledge is required to post ..... usually just an opinion.
Being successful in racing, is totally about the people you choose to guide you, and what guidance you choose to embrace. Now ...... that is undeniable!

And many people would rather do it their own way, than be successful.

Enjoy your project!

[alangbaker]

I'm going to let you believe you are correct.

For the other folks who are open to learning, yes we are trying to maximize the area under the curve. Whether or not maximizing the area under the curve dictates going beyond the peak depends on the shape of the curve. A plot that nosedives faster than it climbed will have less area under the curve after the peak than before. The farther we go beyond that peak the more area we are losing. A curve that falls off slower than it climbed will give m
ore area under the curve if we go beyond the peak, and the farther beyond peak we go, the better. A curve that is symmetrical will yield the same area on each side of the peak, but the most area the closer we stay to peak. 500 rpms each side of peak (1000rpm spread) will net more area than either 1000 before to peak, or peak to 1000 after. The narrower our gear ratios the more we can maximize the area under the curve. Staying 100 rpms each side of peak will produce more area under the curve (average).

Hopefully I explained it in a way that clearly illustrates how the shape of the curve and the ratios we are dealing with have an effect on the results.

Daryl...

In my first post in this thread, I said:

"The correct way to select shift points is to maximize the area beneath the power curve."

So you're not correcting me when you say that. I already understand it.

The point is, that it is the area under the POWER curve that must be maximized. If you want to use the maximum area of a torque curve, it must be torque AT THE WHEELS (actually, at any stage after the last changeable gear ratio; torque at the pinion shaft would do as well).

Let me offer a very simple example:

If you have an idealized engine—say an electric motor—that has a completely flat torque curve, you could achieve the same area under the curve by shifting gears anywhere, right?

But 100ft-lbs engine torque @ 1000rpm doesn't deliver the same torque at the rear wheels as 100ft-lbs @ 2000 running through twice the reduction ratio, does it?

[alangbaker]

Here are two pages supporting why it is best to shift at redline (or as high above or below that as you feel you can, based on how long you want your engine to last!).

http://www.welltall.com/ymc/discovery/car/shiftpt.html

http://glennmessersmith.com/shiftpt.html

Jay, neither of those references actually says that. Sorry.

From welltall.com:

"As we can see there are intersections between the transmission torque curves, this confirms the existence of optimal shift points before red line."

From your other source, glenmessersmith.com, the actual example data they give you shows that:

1. Redline is 7000rpm
2. The calculated shift points are: 6850, 6700, 6300, 5900, 5950.

So...

[alangbaker]

HP is a derived number from torque based on RPM. The higher you spin the motor the more HP you have.
Once you are beyond the peak of the torque curve the HP numbers begin to become useless except for bragging rights.

Sorry, but that is completely and utterly wrong.

A simple thought example.

Imagine an engine with a very low torque peak at (say) 2000rpm, but where that torque only falls very slightly to the (say) 10,000rpm redline, say from 100ft-lbs to 90ft-lbs.

Now, if you can choose between being at 10,000rpm with 5:1 reduction compared to 2000rpm at 1:1 relative reduction (for the same output shaft speed:

Which will give you more output torque? The math is easy.

5:1 speed reduction means 5 * 90 = 450ft-lb torque at the output shaft.

1:1 means you get 100ft-lbs at the output shaft.

Still think HP beyond peak torque is just for bragging rights?

The power figure combines both the torque you're getting AND the amount of reduction you can apply to that torque before it gets to the wheels.

The power curve rules.

[WRD]

I think most everything is covered above, but we do have a nice spreadsheet available on our website to help you work it all out!!

http://www.willrace.com/technical-information.html

[Steve Demeter]

Finally a simple answer to a simple question.

The OP simply asked about gearing, not a many post debate on HP and where to shift.

He is smart enough to figure out how high RPM he wants to shift at based on many things (engine builder advice, how long he wants motor to last etc.)

Tony Kester used to travel around giving a driver's seminar and he said to gear a tad tall. It makes you drive the corners harder cause it sounds like you are going too slow. YMMV

[TimH]

Tony Kester used to travel around giving a driver's seminar and he said to gear a tad tall. It makes you drive the corners harder cause it sounds like you are going too slow. YMMV

..and you never know when you might pick up a good tow...