Non-Prevalent Order Random Sprocket BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The invention pertains to the field of sprockets. More particularly, the invention pertains to sprocket and chain systems having reduced engagement noise.

DESCRIPTION OF RELATED ART

Figure 1 shows an example of a chain drive system for a typical engine - for example, this arrangement might be used in an inline four-cylinder diesel engine. There are five sprockets - sprocket 30 is on the crankshaft, sprocket 31 is an idler which is driven by the crankshaft sprocket 30 and drives the exhaust camshaft sprocket 33 and intake camshaft sprocket 34. Sprocket 32 drives a fuel pump. There are two chains - fuel pump chain 10, made up of strands 1-3, is driven by the crankshaft sprocket 30 and drives idler sprocket 31 and fuel pump sprocket 32, and cam timing chain 20, made up of strands 4-6, is driven by idler sprocket 31 and drives intake camshaft sprocket 34 and exhaust sprocket 33.

A chain or toothed belt drive is subjected to oscillating excitations. For example, a chain or toothed belt drive can be used between an engine crank shaft and cam shaft. The oscillating excitation could be the torsional vibrations of the crank shaft and/or fluctuating torque loads from the valve train and/or a fuel pump. Random sprockets are sometimes used on chain drives to reduce chain engagement noise. Traditional random sprockets use a fixed "random" pattern of pitch radius variation to vary the timing of engagement. This breaks up the engagement noise so it is not an objectionable pure tone. With a nonrandom sprocket, engagement noise is all concentrated at engagement frequency or engagement order. With a random sprocket the engagement noise is spread out over many orders - predominantly at low orders and orders around engagement order. While it reduces objectionable noise, the radial variation of the random sprocket creates oscillating tensions in the chain drive - mostly at low orders (several times per sprocket revolution).

Most chain drives have one or more torsional resonance frequencies. If the tension fluctuations caused by the random sprocket are at an order that has a frequency at, or near, a chain drive torsional resonance frequency, the tension fluctuations will be amplified - possibly causing substantial variation in the chain tensions.

If the chain drive has no external causes of tension variation (such as crank TVs, cam torque, fuel pump torque), the oscillating tensions caused by the random sprocket will add to the mean tension to increase the overall maximum tension. In this case a random sprocket will always increase maximum chain tensions. If the maximum tension remains within the acceptable range for the chain, this may not be a problem.

If the chain drive has external causes of tension variation, it is likely that the random sprocket will create tension variations at orders where tension variations already exist from the external sources. In many cases this will create tensions that will add and increase the overall maximum tensions. Using existing random sprocket methods almost always results in an increase in the overall maximum chain tension. In many cases the tension is increased beyond the acceptable tension level for the chain. For this reason, it is not possible to use traditional random sprockets on many chain drives with external sources of tension variation such as engine cam or balance shaft drives.

Tension reducing random sprockets were developed by defining the random sprocket radial variation pattern to create tensions at one or two specific orders. The tensions created by the tension reducing random sprocket are phased to be opposite tensions caused by external sources of tension variation. This results in a cancellation of tensions and a reduction in overall maximum tension levels. Because tension reducing random sprockets use uniform repeating patterns, and amplitudes must be selected to generate the correct tension levels, tension reducing random sprockets often do not have much effect on engagement noise.

A tension reducing sprocket uses orders prominent in the chain tensions to create tension fluctuations that are phased opposite the tensions caused by the oscillating excitations so that the tensions at those orders cancel or partially cancel at the chain or belt drive resonance (when the overall maximum tensions are dominated by one of these orders) The problem is how to design a random sprocket for sufficient noise reduction while limiting the increase in overall chain tensions.

SUMMARY OF THE INVENTION

The invention discloses a random sprocket or pulley with varying pitch radius intended to reduce the engagement noise of the drive. The random sprocket or pulley uses a radial variation pattern made of orders that are not already prominent in the chain tension, which are caused by the oscillating excitations.

The chain or toothed belt drive with oscillating excitations and torsional resonance creates oscillating tensions at specific orders. These orders should be avoided in the random sprocket radial variation pattern to minimize the increase in chain tension caused by the random pattern.

Using a random sprocket with radial variation made up of orders not prominent in the chain tension reduces tensions caused by the random sprocket exciting the drive resonance at the same time as tensions caused by the oscillating excitations, minimizing the increase in chain tension caused by the random sprocket.

BRIEF DESCRIPTION OF THE DRAWING

1 shows a prior art two stage chain drive for an 1-4 Diesel application.

. 2a-2c show graphs of excitations from crank torsional vibrations in the drive of figure 1.

. 3a-3c show graphs of excitations from exhaust cam torques in the drive of figure 1 . 4a-4c show graphs of excitations from intake cam torques in the drive of figure 1. . 5a-5c show graphs of excitations from fuel pump torques in the drive of figure 1. Fig. 6a shows engine speed vs strand tension for the fuel pump chain (strands 1-3) in figure 1.

Fig. 6b shows engine speed vs strand tension for the cam timing chain (strands 4-6) in figure 1.

Figs. 7a-7c show strand tensions in strand 1 of figure 1. Figs. 8a-8c show strand tensions in strand 2 of figure 1. Figs. 9a-9c show strand tensions in strand 3 of figure 1. Figs. 10a- 10c show strand tensions in strand 4 of figure 1. Figs. 1 la-1 lc show strand tensions in strand 5 of figure 1. Figs. 12a- 12c show strand tensions in strand 6 of figure 1.

Figs. 13a-13f show summary graphs for a baseline system as shown in figure 1, with

"straight" sprockets.

Figs. 14a-14f show graphs for a system as shown in figure 1, with traditional random

sprockets and no external excitations.

Figs. 15a-15f show graphs for a system as shown in figure 1, with traditional random

sprockets and with external excitations.

Figs. 16a-16f show graphs for a system as shown in figure 1, with the non-prevalent order (NPO) random sprockets of the invention, and no external excitations.

Figs. 17a-17f show graphs for a system as shown in figure 1, with the non-prevalent order (NPO) random sprockets of the invention, and with external excitations.

Figs. 18a- 18b show graphs of tensions for a system as shown in figure 1, with a traditional random sprocket mounted at differing orientations.

Figs. 18c-18d show graphs of tensions for a system as shown in figure 1, with an NPO random sprocket mounted at differing orientations. Figs. 19a-19f show graphs for a system as shown in figure 1, with tension reducing random sprockets and no external excitations.

Figs. 20a-20f show graphs for a system as shown in figure 1, with tension reducing

random sprockets and with external excitations.

Figs. 21a-21f show graphs for a system as shown in figure 1, with a combination of NPO sprockets and tension reducing random sprockets and no external excitations.

Figs. 22a-22f show graphs for a system as shown in figure 1, with a combination of NPO sprockets and tension reducing random sprockets and with external excitations.

Fig. 23 is a graph of chain load vs engine RPM for a system as shown in figure 1, showing straight sprockets, traditional random sprockets, tension reducing random sprockets, NPO random sprockets and a combination of NPO random and tenstion reducing random sprockets.

Fig. 24 is a graph of sprocket radial variation vs root number for straight sprockets,

traditional random sprockets, tension reducing random sprockets, NPO random sprockets and a combination of NPO random and tenstion reducing random sprockets.

Fig. 25 is a flowchart of a method of reducing tensions using NPO random sprockets.

Fig. 26 shows the legend for Figures 2a, 5a, 7a, 8a, 9a, 10a, 11a, and 12a.

Fig. 27 shows the legend for Figures 2b, 3b, 4b, 5b, 7b, 8b, 9b, 10b, 1 lb, 12b, 13c, 13d, 14c, 14d, 15c, 15d, 16c, 16d, 17c, 17d, 19c, 19d, 20c, 20d, 21c, 21d, 22c and 22d.

Fig. 28 shows the legend for Figures 2c, 3c, 4c, 5c, 7c, 8c, 9c, 10c, 11c, 12c, 13e, 13f, 14e, 14f, 15e, 15f, 16e, 16f, 17e, 17f, 19e, 19f, 20e, 20f, 21e, 21f, 22e, and 22f.

Fig. 29 shows the legend for Figures 6a, 13a, 14a, 15a, 16a, 17a, 19a, 20a, 21a, and 22a.

Fig. 30 shows the legend for Figures 6b, 13b, 14b, 15b, 16b, 17b, 19b, 20b, 21b, and 22b.

Fig. 31 shows the legend for Figures 3a and 4a. DETAILED DESCRIPTION OF THE INVENTION

This invention addresses the problem of increasing chain tensions by creating the random pattern using only orders that are not prevalent in the chain tensions. By doing this the interaction between tensions caused by the random sprocket and tensions caused by external sources is minimized. In addition, the tensions caused by the random sprocket will not excite resonances at the same time as tensions caused by external sources.

Furthermore, the orientation of the radial variation on the random sprocket is no longer very important (because it does not add or cancel orders caused by external sources). It is possible to combine this method with a tension reducing random sprocket. This can be done by first defining the orders, amplitudes and phasing for a tension reducing random sprocket. Then other orders not already prominent in the chain tension can be added to further improve engagement noise reduction with minimal increase in overall maximum tensions.

The chain or belt drive could have many configurations and a variety of oscillating excitations. What is important is that the drive is subject to oscillating excitations and the drive has a torsional resonance that can be excited by these excitations. There are many possible radial variation patterns that could be used. What is important is that the pattern does not contain orders that are prominent in the chain tensions. Or, if it does contain orders that are prominent in the chain tensions, they are selected with proper amplitude and orientation to cancel tensions caused by the oscillating excitations.

This concept could be combined with the radial variation of a tension reducing sprocket to reduce both noise and maximum overall chain tensions.

It should be noted that the legends for Figures 2a, 5a, 7a, 8a, 9a, 10a, 11a, and 12a is found in Figure 26. The legend for Figures 2b, 3b, 4b, 5b, 7b, 8b, 9b, 10b, 1 lb, 12b, 13c, 13d, 14c, 14d, 15c, 15d, 16c, 16d, 17c, 17d, 19c, 19d, 20c, 20d, 21c, 21d, 22c and

22d is found in Figure 27. The legend for Figures 2c, 3c, 4c, 5c, 7c, 8c, 9c, 10c, 11c, 12c, 13e, 13f, 14e, 14f, 15e, 15f, 16e, 16f, 17e, 17f, 19e, 19f, 20e, 20f, 21e, 21f, 22e, and 22f is found in Figure 28. The legend for Figures 6a, 13a, 14a, 15a, 16a, 17a, 19a, 20a, 21a, and 22a is found in Figure 29 and the legend for Figures 6b, 13b, 14b, 15b, 16b, 17b, 19b, 20b, 21b, and 22b is found in Figure 30. The legend for Figures 3a and 4a is found in Figures 31.

The chain drive arrangement in figure 1 has high oscillating excitations from crank torsional vibrations. This can be seen in the graphs in figures 2a-2c. Similarly, figures 3a- 3 c and 4a-4c show graphs of cam torques in the exhaust camshaft and intake camshaft, repsectively, and figures 5a-5c show fuel pump torques.

Figures 2a, 3 a, 4a and 5 a show time traces of the oscillating excitation for one engine cycle at different speeds. These time traces repeat every engine cycle (2 crank revolutions). Because this is a four cylinder engine, the oscillating excitations have a pattern that repeats four times in the engine cycle. This creates a dominant fourth engine cycle order (the excitations increase and decrease four times per engine cycle). Because the time traces are not sinusoidal, there are harmonics - or orders - that are multiples of the fourth engine cycle order (8, 12, 16, ... ).

It will be understood that the word "engine cycle order" or just "order" as used herein to apply to a four-cycle engine will refer to the "engine cycle" order. The graphs also note a "crank sprocket order" in parenthesis. The engine cycle order is twice the crank sprocket order for a four-cycle engine, because the crank rotates twice per engine cycle.

For this application the time traces can be represented by the order amplitudes and phases using the equation:

X=A ₄ x sin(4 x 0 + ) + Ag x sin(8 x Θ + φ ₈ ) + A ₁₂ x sin(12 x Θ + φ ₁₂ ) + ...

Where: A _n = amplitude for order 4, 8, 12, .. .

φ _η =phase for order 4, 8, 12, ...

Θ = engine cycle or cam angle (0-360 over one engine cycle)

The amplitude and phase of each order varies with engine speed. The variation of amplitude with speed for orders 4, 8, 12, 16 and 20 is shown in figures 2b, 3b, 4b and 5b, and the same plot is shown for orders 6, 10, 14, 18 and 22 in figures 2c, 3c, 4c and 5c.

Figures 6a and 6b shows the maximum and minimum tensions in each strand of the fuel pump chain 10 and cam timing chain 20 vs engine speed (calculated using a chain drive dynamic simulation computer program). These tensions are the result of the external oscillating excitations. The maximum tensions are very close to the endurance limits of the chains (for example, 2800 Newtons for the primary chain and 2300 Newtons for the secondary chain).

The time traces and order amplitudes for each strand tension are shown in Figures 7a-12c, in which figures 7a, 8a, 9a, 10a, 11a and 12a show strand tension vs crank angle at various engine speeds for strands 1-6, respectively; figures 7b, 8b, 9b, 10b, l ib and 12b show strand tensions vs engine speed for orders 4, 8, 12, 16 and 20; and figures 7c, 8c, 9c, 10c, 11c and 12c show strand tensions vs engine speed for orders 6, 10, 14, 18 and 22.

Note from figures 7a- 12c that there is a major peak in tension at around 3500-4000 RPM engine speed due to the first torsional resonance of the chain drive. This is most noticeable in the 8th engine cycle order. The amplification of the 8th engine cycle order causes the highest chain tensions.

Figures 13a-13f show summary views of a baseline system as shown in figure 1, in which the sprockets 30-34 are "straight" sprockets - that is, sprockets without

randomization or other modifications for noise or torque or tension adjustment.

Figures 13a and 13b correspond to figures 6a and 6b, respectively, showing the maximum and minimum strand tensions in the fuel pump chain 10 (strands 1-3) and cam timing chain 20 (strands 4-6). Figures 13c-13f show the order content of the strands with the highest tension for the fuel pump chain 10 (strand 3) and cam timing chain 20 (strand 6). Figure 13c corresponds to figure 9b, figure 13d corresponds to figure 12b, figure 13e corresponds to figure 9c, and figure 13f corresponds to figure 12c.

The engagement noise for this drive arrangement is not acceptable. The worst noise comes from near the fuel pump sprocket. A random fuel pump sprocket could be used to reduce engagement noise levels to address this issue, as is known to the prior art.

Figures 14a-14f show the chain tensions created by a traditional random sprocket when no external excitations are present - that is, without considering any torque variations introduced by the operation of the crankshaft, camshaft(s) or fuel pump. Figures 14c-14f show that the traditional random sprocket excites tensions at many orders. Some of the more prominent orders include 4th, 10th and 12th engine cycle orders.

However, it will be understood that external excitations will be present in any real- world engine, created by the impulses of the pistons on the crankshaft, or the valves on the camshaft(s) or the operation of the fuel pump. When combined with the external excitations, it is likely that the 4th and/or 12th order contributions will add together, thereby increasing chain tensions. This is shown in figures 15a-15f. Relative to the baseline shown in figures 14a-14f, there is a large increase in strand 1, 2, 4, 5 and 6 tensions and a smaller increase in the strand 3 tensions. This kind of increase in chain tensions is not acceptable.

Figures 16a-16f show the chain tensions created using a random sprocket with pitch radius variation based only on orders not prominent in the baseline system

(herinafter referred to as "Non-Prevalent Order" or NPO random sprockets), as taught by the present application. As with figures 14a-14f, above, figures 16a-16f do not take external excitations into account. In this case, the orders that can be used for the NPO random sprocket pattern are 1 ^st , 2 ^nd , 3 ^rd , 5 ^th , 6 ^th , 7 ^th , 9 ^th , 10 ^th , 11 ^th , 13 ^th , ... engine cycle orders.

In the engine of the example, the fuel pump sprocket rotates twice per engine cycle - that is, at the same speed as the crankshaft. This means that fuel pump sprocket orders are two times engine cycle orders - 1 st fuel pump sprocket order (a sine wave that repeats once per fuel pump rotation) equals 2nd engine cycle order, and so on. This means the fuel pump sprocket can only be used to create 2 ^nd , 4 ^th , 6 ^th , 8 ^th , 10 ^th , 12 ^th ... engine cycle orders.

Eliminating the orders that are already prevalent in the baseline tensions leaves 2 ^nd , 6 ^th , 10 ^th , 11 ^th , 14 ^th , 18 ^th ... engine cycle orders - or 1 ^st , 3 ^rd , 5 ^th , 7 ^th , 9 ^th ... fuel pump sprocket orders. The order plots in figures 16c-16f show that, for this example, the NPO random sprocket generates tensions mostly at 6 ^th , 10 ^th , 14 ^th and 18 ^th engine cycle orders. The tensions generated at 4 ^th , 8 ^th , 12 ^th , 16 ^th ... engine cycle orders are very small and will not add significantly when combined with the external excitations.

Figures 17a- 17 shows graphs of the engine with an NPO random sprocket, but taking external excitations into account as in figures 15a-15f. Comparing the order content from the baseline (Figures 13a-13f), figures 17a-17f show only small changes in the dominant orders, and the addition of new orders from the PO random sprocket. The maximum tensions in strands 1, 2, 4, 5, 6 only increase slightly. The tensions in strand 3 increase a little more due to the 10 ^th engine cycle order contribution of the NPO random sprocket. This could be improved by reducing the 5 ^th sprocket order amplitude used to create the NPO sprocket.

Figures 18a-18d shows how the tensions change with orientation of the traditional random sprockets (figures 18a- 18b) and the NPO random sprockets of the present application (figures 18c-18d). As used in these figures, the term "orientation" means the angular position of the sprocket at the start of the engine cycle. The sprocket orientation can vary in one tooth increments.

With a traditional random sprocket there can be a large variation in maximum tensions due to sprocket orientation, as shown in figures 18a- 18b. Also, selecting an orientation that reduces tensions at some conditions is likely to increase tensions at other conditions. Therefore, when manufacturing or servicing an engine with traditional random sprockets, the manufacturer or mechanic needs to be very careful how the sprocket is assembled onto the shaft - taking a sprocket off and reinstalling it at a slight variation in orientation can result in major changes in chain tension. The tension variation with orientation for the NPO random sprocket is much smaller, providing an advantage for manufacturing and servicing of engines using the NPO sprockets.

Figures 19a-19f show an example of the tensions generated by a tension reducing random sprocket of the prior art, without considering external excitations. In this example the tension reducing random sprocket is designed to generate tensions at 4 ^th and 8 ^th engine cycle orders (2 ^nd and 4 ^th fuel pump sprocket orders). Under the conditions considered, the 8 ^th engine cycle order excites the chain drive torsional resonance in a range between 3000- 4000 RPM.

Sprocket orientation is very important for tension reducing random sprockets. The tensions generated by the tension reducing random sprocket must be timed to have the opposite phase relative to the baseline tensions - when the baseline tensions are high, the tensions from the tension reducing random sprocket must be low, and vice versa. Figures 20a-20f show the effect of the tension reducing random sprocket when combined with the external excitations. The maximum tensions in strands 1, 2, 4, 5, and 6 are significantly reduced from those using the traditional random sprocket shown in figures 15a-15f, while the strand 3 maximum tension is reduced a smaller amount.

The PO random sprocket pattern made up of odd fuel pump sprocket orders of the present application can be combined with the 2 ^nd and 4 ^th fuel pump sprocket orders of the tension reducing random sprocket of the prior art to combine the benefits of each. The tensions generated by just the sprocket are shown in Figures 21a-21f and the result when combined with external excitations is shown in Figures 22a-22f.

As can be seen in figure 23, the noise improvement of the new random sprocket pattern is gained with almost no increase in chain tensions. Note that because this design includes the tension reducing orders, the combined NPO random and tension reducing random sprocket must be oriented correctly relative to the engine cycle.

Figure 23 shows engagement order chain force on the fuel pump sprocket. This is a measure of the system noise characteristics. In this example, suppose the noise of concern occurs when the engine is operated in the 1500 to 2500 RPM range.

The traditional random sprocket (trace with circles) does not show an improvement below 1700 RPM, but does show a significant improvement above 1700 RPM. The acceptability of this will depend largely on what frequencies are amplified and attenuated in the noise path from the chain to the listener. The tension reducing random sprocket (trace with diamonds) shows some improvement in certain speed ranges.

The NPO random sprocket of the present application (trace with squares) shows a significant improvement, starting around 1400 RPM, and between about 1400 RPM and and about 2200 RPM it is the least noisy. Combining the NPO random and the tension reducing random sprocket (trace with triangles) gives noise results similar to the NPO random sprocket.

Table 1 shows the orders, amplitudes and phases used to construct the radial variation patterns for the example NPO random and tension reducing random sprockets. The traditional random sprocket is not included because the pattern is not generated based on order content. The radial variation is calculated using

AR = AR _P + Ani x sin(nl x Θ + φ _η ι) + A„2 x sin(n2 x Θ + φ^) + A _n 3 x sin(n3 x Θ + φη3) + ...

Where: A _n = = amplitude for order 1, 2, 3, ...

φη = phase for order 1, 2, 3, .. .

Θ = sprocket angle

n* = ^: sprocket order for order 1, 2, 3,

AR _P = pitch radius mean shift

A pitch radius mean shift is included to maintain a constant pitch length between all seated pin positions. For the NPO random sprocket orders, the phase, like the sprocket orientation, has only a minor impact on the resulting chain tensions.

Sprocket Order 3 5 7 9 11

NPO Random Amplitude (mm) 0.2 0.2 0.2 0.2 0.2

Phase (deg) 0 60 0 60 0

Sprocket Order 2 4

Tension Reducing Random Amplitude (mm) 0.25 0.3

Phase (deg) 0 15

Sprocket Order 2 4 3 5 7 9 11

NPO Random + Tension Reducing Random Amplitude (mm) 0.25 0.3 0.15 0.15 0.15 0.15 0.15

Phase (deg) 0 15 30 45 60 75 90

Table 1

Table 2 shows a table of radial variation and angular variation for a traditional random sprocket for each sprocket root around the sprocket. Angular variation must be included to keep the pitch length between seated pin centers constant. The far right column contains numbers assigned to each pitch radius variation. One is the highest radial

variation, two is the next highest radial variation and so on. In this case there are only three different radial variations. A typical traditional random sprocket only uses a few fixed radial variation values. The numbers in the far right column are referred to as the sprocket pattern. In a traditional random sprocket, repetition in the pattern is avoided.

Traditional Random

Root Radial Angular

Number Variation Variation Pattern

(m) (deg)

1 -0.00003724 0.00000000 2

2 0.00056276 -0.13710377 1

3 0.00056276 -0.36622326 1

4 0.00056276 -0.59534276 1

5 -0.00003724 -0.73244652 2

6 -0.00063724 -0.62141636 3

7 -0.00003724 -0.51038620 2

8 -0.00003724 -0.49497202 2

9 -0.00063724 -0.38394186 3

10 -0.00063724 -0.11573631 3

11 -0.00003724 -0.00470615 2

12 0.00056276 -0.14180991 1

13 0.00056276 -0.37092941 1

14 -0.00003724 -0.50803318 2

15 -0.00003724 -0.49261900 2

16 -0.00063724 -0.38158884 3

17 -0.00003724 -0.27055868 2

18 0.00056276 -0.40766244 1

19 0.00056276 -0.63678194 1

20 -0.00003724 -0.77388571 2

21 -0.00063724 -0.66285554 3

22 -0.00063724 -0.39464999 3

23 -0.00063724 -0.12644444 3

24 -0.00003724 -0.01541428 2 Table 2

Table 3 shows the radial variation and pattern for the example NPO random sprocket. The sprocket pattern contains many root radii and there are no regular repeating patterns. This will be true for most NPO random sprocket cases. However, depending on the orders being used and how they are phased, it is possible to have an NPO random sprocket with a regular repeating pattern.

N PO Random

Root Radial Angular

Number Variation Variation Pattern

(m) (deg)

1 0.00045216 0.00000000 1

2 -0.00014849 -0.09191165 14

3 -0.00022102 -0.01556084 16

4 -0.00027698 0.08784889 17

5 0.00035263 0.03982474 4

6 -0.00052479 0.01229768 20

7 -0.00020601 0.15653050 15

8 0.00033360 0.10642216 5

9 -0.00014650 0.04903780 14

10 0.00042817 -0.03587976 3

11 -0.00004409 -0.13300690 13

12 -0.00027728 -0.07073505 17

13 -0.00054637 0.09551772 21

14 0.00006078 0.16622696 10

15 0.00012561 0.12745052 9

16 0.00018183 0.06390381 7

17 -0.00044885 0.08664837 19

18 0.00043023 0.02718861 2

19 0.00013455 -0.09575846 8

20 -0.00043694 -0.05980716 18

21 0.00005215 0.00050452 11

22 -0.00052388 0.07135210 20 23 -0.00004314 0.17045394 12

24 0.00018644 0.13658762 6

Table 3

Table 4 shows the radial variation and pattern for the example tension reducing random sprocket. The tension reducing sprocket contains a pattern that repeats twice. Note that tension reducing random sprockets often contain patterns that substantially repeat - but this is not necessary.

Tension Reducing Random

Table 4

Table 5 shows the radial variation and pattern for the example NPO random + tension reducing random sprocket. Like the NPO random sprocket, there are few, if any, radial variations that repeat. There is no repetition in the pattern. Pattern repetition in a combined NPO random and tension reducing random sprocket is highly unlikely.

NPO Random + Tension Reducing Random

Table 5

Figure 24 contains a plot comparing the radial variation versus sprocket root number for the example sprockets. This shows the repetition in the tension reducing random sprocket and the lack of repetition in the other sprockets. It also shows the few discrete radial variations used in a traditional random sprocket.

The engine cycle orders that can be used for a new random sprocket will vary depending on the engine configuration. The sprocket orders that can be used for an NPO random sprocket will depend on the speed ratio between the crank and the new random sprocket. Table 6 shows examples of engine cycle orders which can be used for an NPO random sprocket:

Table 6

These orders may change if the engine has a fuel pump that does not have torques that are dominated by firing order and/or its harmonics. In some cases there may be additional orders that should not be used due to orders in the crank torsional vibrations that are not firing order or its harmonics.

If the NPO random sprocket rotates at crank speed, the sprocket orders that can be used will be the above orders times two. If the NPO random sprocket rotates at cam speed (half crank speed), the sprocket orders will be the same as the orders above. Likewise, if other speed ratios are used, the sprocket orders that can be used will be twice the orders above times the ratio of the NPO random sprocket speed over the crank sprocket speed. Fig. 25 is a flowchart of a method of reducing engagement noise without increasing maximum chain tensions using NPO random sprockets.

In a first step, external excitations within the chain drive are determined to determine engine cycle orders in which tensions occur (step 200). Alternatively, excitations can be applied to the chain drive in order to determine the tensions and then determine the prevalent engine cycle order(s).

Engine cycle orders at which tensions are prevalent within the chain drive are converted into sprocket orders (step 202). For example, sprocket orders are half as much as engine cycle orders. So, a fourth engine cycle order is a second crank sprocket order.

The orders which are non-prevalent are determined (step 204). These are the orders which were not discovered in steps 200-202.

The non-prevalent orders are incorporated into the radial variation pattern of a sprocket (step 206) to create a non-prevalent order (NPO) sprocket. The amplitude of the non-prevalent orders determines the tensions that will be introduced into the chain drive and corresponds to the radial variation pattern of the sprocket. In a preferred embodiment, non-prevalent orders chosen are less than half of the number of teeth of the sprocket and first and second orders are typically avoided, since orders above half the number of sprocket teeth are aliased and appear as lower orders. Low orders, like the first and the second orders, typically have little impact on engagement noise because they do not cause enough change in engagement timing and is why there is not much reduction in engagement noise with tensioner reducing random sprockets.

The non-prevalent order (NPO) sprocket is installed in the chain drive of the internal combustion engine (step 208) and the method ends.

It should be noted that that while the chain drive system described above refer to camshafts, fuel pumps, etc., the NPO sprocket of the invention is equally useful with other engine accessories and components driven by a chain, such as balance shafts or water pumps, etc.

It should also be noted that while the description above is in the context of internal combustion engines, the NPO sprocket of the invention can also be used with other chain applications such as transmissions, transfer cases, hybrid drives, and so on. It will be understood that the PO concept can be applied to any chain-driven system that has excitations at specific orders or interactions with other systems at specific orders (creating orders than need to be avoided in the random pattern).

It should also be noted that the term "sprocket" includes the pulleys used in toothed belt driven systems.

Accordingly, it is to be understood that the embodiments of the invention herein described are merely illustrative of the application of the principles of the invention. Reference herein to details of the illustrated embodiments is not intended to limit the scope of the claims, which themselves recite those features regarded as essential to the invention.