Reving past red line

[Bruin]

I run a Honda '82 XL 500 four valve, dual exhaust thumper. It is stock other than a 2mm larger carb, velocity stack and straight pipes. The factory says it generates 32hp at 6500rpm. On the salt I can wind it up to 8,000rpm in 1,2,3 gear and then it drops to 7,600 in 4th and about 6,800 in 5th. My question is, "Should I shift at 6,500rpm or is there a benefit taking it to 8,000rpm?" I wonder if the power peak drops off after 6,500 and I'm fooling myself.

[fordboy628]

I run a Honda '82 XL 500 four valve, dual exhaust thumper. It is stock other than a 2mm larger carb, velocity stack and straight pipes. The factory says it generates 32hp at 6500rpm. On the salt I can wind it up to 8,000rpm in 1,2,3 gear and then it drops to 7,600 in 4th and about 6,800 in 5th. My question is, "Should I shift at 6,500rpm or is there a benefit taking it to 8,000rpm?" I wonder if the power peak drops off after 6,500 and I'm fooling myself.

It's hard to give an intelligent answer without knowing the shape of the bhp curve, but in general, the fastest acceleration uses the largest amount of area under the bhp curve, IF, the engine is reliable at the increased revs above peak bhp.   Knowing the rpm drops through the gear ratios is also a requisite, because a gearbox with progressive "drops" will normally require differing shift points to utilize the power band properly, Vs a gearbox with straight drops.

These two samples REQUIRE different gear ratios:


Notice the SIGNIFICANT difference in area under the bhp curve in these 2 samples.
One was "thrown together", the other was "engineered".     You pick how you want to do it . . . . . .
And, BTW, the "engineered" version was about 15 seconds per lap faster @ Road America.

Sumner's site has some free gear ratio and rpm drop software, I think in spreadsheet format, that you might find useful.

Alternatively, you might want to invest in one of these pieces of software:

1/   Bonneville Pro   or
2/   Performance Trends
      A/   Transmission Gear Calculator reg or Pro
      B/   Land Speed Record DRA Pro

Several guys on the site use these and can offer advice.   I use Bonneville Pro and Transmission Gear Calculator Pro and I am satisfied with both.   They do different things though, so be aware.

When you want the best results, it is always better to take the time to calculate things out.   It avoids having to try out every possible permutation, with the attendant cost and time penalties.

AND, just for the record, calculations and software might NOT find you the best combination, BUT, they will keep you away from the dumb, time/money waster choices . . . . .

The process used is:
A/  Calculate
B/  Test and verify

Just ask Woody, Rob or the midget . . . .

 
Fordboy

[fordboy628]

Here is a better sample graph:


Although there is not a massive difference between the red and black permutations, the black version with the short trumpets and proper gearing was faster.    About 1.2 seconds/lap    We are not even going to talk about the green version.

This should be enough to get you thinking.

 
Fordboy

[wobblywalrus]

There is an rpm drop when shifting from fourth to fifth gear due to the different ratios.  This can be called a numerical drop.  Plus, there is an additional drop due to the time it takes to make the shift and plus operator error.  A nice and tight linkage that is easy to use is an added 500 rpm, a sloppy or hard to use linkage is an additional 1000 rpm, is what I use.  Added up, this gives a range in RPM.  Usually I try to set the gearing so this rpm range bracket is set evenly between, or evenly overlapping the gap between peak torque and peak HP.  Data needed are trans ratios and dyno curves.  As best as I can tell, this is what Mark is recommending or real close to it.

Make sure that checking the valve spring tension is part of your annual teardown.  Hondas are notorious for annealing their springs.  The tension is lost, the valves float, a valve head breaks off or retainer breaks, and kaboom.  A lesson it took two times for me to learn.     

[SPARKY]

shift chart --- well I flunked this upload   lol

This is just graphing the shift points using eng speed and vech. speed  if you have a graph using the same scale to plot your HP and Torque 

notice that I keep changing the shift point to keep the eng in its sweet spot ---my eng really get going at about 5750 and peak HP is just above 7200 and peak torque is just about 6500

[hotrod]

It can be an advantage to shift at different rpms as noted above. The key is figuring out the actual power delivered to the wheel before and after the shift points.

For example when I was drag racing with my gearing (stock gears no choice) I had more power at the wheel in low at red line than at any rpm in second gear.
So to get best times I shifted at the highest rpm in low gear I could without triggering fuel cut due to the built in rev limit.

As noted take your dyno plot and your gear ratios and plot out the power delivered to the transmission output in every gear through out the rpm range, also include consideration of the rpm drop mentioned above. You want to shift at the point where the power delivered to the output after the shift is as close as possible to the power delivered in the prior gear just before the shift. (this will also minimze power shock to the wheels during the shift from a large mismatch).

Your objective is to get the maximum area under those curves that you can.

In my drag racing situation, I  launched at 6200, (which dropped to about 5800 after I stopped spinning the wheels out of the hole, shifted at 6250 low - second, 6000 second-third, 5800 rpm third-forth
In my case if I held it in third through the traps it reved out and slowed down before the lights, if I waited until 6200 rpm (the nominal power peak rpm) I was shifting just prior to the lights and got no benefit from the shift because I was coasting for the 1/3 second it took to shift.

Every engine, transmission will come out different when you do the plots. Don't be a slave to the peak power rpm, pay attention to band of rpm on both sides of the power peak, that the engine can effectively pull in, and get the best overall average power out.

That means that at the shift point you will normally be a little past peak power, and during the shift, engine rpm will drop a little below peak power, then pull through the peak power rpm before the next shift.

Out at bonneville you have two other issues to pay attention to also. Due to the very low final drive ratios some use, and air drag, if you let the rpm drop too much during the shift the engine will not be able to pull back into the power band in the next gear and either bog or just hang at the rpm after the shift.

Good luck!

[fordboy628]

There is an rpm drop when shifting from fourth to fifth gear due to the different ratios.  This can be called a numerical drop.  Plus, there is an additional drop due to the time it takes to make the shift and plus operator error.  A nice and tight linkage that is easy to use is an added 500 rpm, a sloppy or hard to use linkage is an additional 1000 rpm, is what I use.  Added up, this gives a range in RPM.  Usually I try to set the gearing so this rpm range bracket is set evenly between, or evenly overlapping the gap between peak torque and peak HP.  Data needed are trans ratios and dyno curves.  As best as I can tell, this is what Mark is recommending or real close to it.

Make sure that checking the valve spring tension is part of your annual teardown.  Hondas are notorious for annealing their springs.  The tension is lost, the valves float, a valve head breaks off or retainer breaks, and kaboom.  A lesson it took two times for me to learn.     

Yes, this is exactly what I am recommending.

shift chart

And yes, this is the type of gear selection chart.    Either by hand or by computer.   I can do it both ways, but the laptop saves wear and tear on my brain and my sliderule.   

 
Fordboy

[fordboy628]

It can be an advantage to shift at different rpms as noted above. The key is figuring out the actual power delivered to the wheel before and after the shift points.

For example when I was drag racing with my gearing (stock gears no choice) I had more power at the wheel in low at red line than at any rpm in second gear.
So to get best times I shifted at the highest rpm in low gear I could without triggering fuel cut due to the built in rev limit.

As noted take your dyno plot and your gear ratios and plot out the power delivered to the transmission output in every gear through out the rpm range, also include consideration of the rpm drop mentioned above. You want to shift at the point where the power delivered to the output after the shift is as close as possible to the power delivered in the prior gear just before the shift. (this will also minimze power shock to the wheels during the shift from a large mismatch).

Your objective is to get the maximum area under those curves that you can.

In my drag racing situation, I  launched at 6200, (which dropped to about 5800 after I stopped spinning the wheels out of the hole, shifted at 6250 low - second, 6000 second-third, 5800 rpm third-forth
In my case if I held it in third through the traps it reved out and slowed down before the lights, if I waited until 6200 rpm (the nominal power peak rpm) I was shifting just prior to the lights and got no benefit from the shift because I was coasting for the 1/3 second it took to shift.

Every engine, transmission will come out different when you do the plots. Don't be a slave to the peak power rpm, pay attention to band of rpm on both sides of the power peak, that the engine can effectively pull in, and get the best overall average power out.

That means that at the shift point you will normally be a little past peak power, and during the shift, engine rpm will drop a little below peak power, then pull through the peak power rpm before the next shift.

Out at bonneville you have two other issues to pay attention to also. Due to the very low final drive ratios some use, and air drag, if you let the rpm drop too much during the shift the engine will not be able to pull back into the power band in the next gear and either bog or just hang at the rpm after the shift.

Good luck!

Yes.

And also important to note, as Hotrod suggested, when the drag hp = the bhp available, you are done.   In 2013, the Milwaukee Midget had a "small" dip in the bhp curve, at about 7700/7800 rpm.   Once into 4th gear, the engine hit the "dip" @ around 118 mph, and that was "all she wrote".   The "Grenade" just could not pull up out of the dip in top gear, although it pulled easily to 8500 in the lower gears.

 
Fordboy

[RichFox]

All this science is nice. But i would think direct observation would be helpful. Try shifting sooner and see what happens.

[SPARKY]

Hot Rod 

I run my dyno results for each 100 RPM   through my Dream Charts FDR final drive ratios and correct for tire size

In my low gear 1.76 (FDR 4.10)  I have over       2700 ft. lbs. of TE
In my hi gear     .78 (FDR 1.66)   only just under 1200  that fall off pretty fast at the speeds that I am trying to run

This way I can predetermine what combo maxes out best

[hotrod]

Sparky

That is exactly what I did for the drags, plotted out the final power delivered to the track for each gear and rpm then looked at the cross over points between gears and started experimenting to see how much rpm drop I had on the shifts. (that forced me to learn how to do no-lift shifts to set my record) If I took too long to make the shift the turbo dropped off boost and it fell on its face. Lots of little variables to sort out, just solve them one at a time, but the shift chart and graphing things out is big help to get a handle on what will likely work rather than experimenting for 50 runs to learn the same thing.

[Interested Observer]

Bruin,
As you may surmise from most of the foregoing, to make any rational judgements about gearing and shift points, knowledge of the power curve is essential.  Just the manufacturer’s quoted maximum values are nice, but really inadequate.  The best thing, of course, is to get the curve for your particular engine, as it sits, from a reputable dynamometer.  Guessing that you may not be up to that at this point, you may be able to find a curve from one of the motorcycle magazines that generally do at least something of a dyno test, generally when the bike is newly released, assuming you are still in a pretty much stock configuration.  That might give you a better idea of how the power is spread around.

To really get serious about predicting top speed and appropriate gearing for that, you need to measure or closely estimate the aerodynamic drag of the bike.  If you know the hp curve and find the terminal speed at some gear and rpm at WOT, a reasonable guess could be made.  Or, perhaps do a coast-down test.

Also, at Bonneville, fast shifts are not quite as important as they are at the drags.  The finish line is generally quite a ways further up the track.  You don’t want to be too laggard about it and lose speed, but a bike will shift pretty quickly anyway.

[Bruin]

Thanks a lot for all the info and guidance. You gave me a lot to think about. I am dyno shy for two reasons, I'm on a penny budget and the one time I hooked up my 250 cc 1970 Triumph to a sponsored dyno, the tech kept telling me to "lean it out".... until my engine seized... one week before Bonnie. But I get the drift about doing some homework and will try to google some pre-existing power curves on my engine as a rough guide. Concurrently I am also trying to reduce my wind profile.
Thanks again.

[SPARKY]

   have fun---you are on the right track   

[55chevr]

My experience on the big white dyno is that once you reach the highest point of the torque curve in high gear you are all in.  The engine  wont pull past that peak.