Regal Sulkies: On Going Faster

This sulky was fitted with sensing devices and apparatus so that it automatically compensated for the driver's weight and seating position by shifting the wheel axles backwards or forwards under its own electric power. The shafts were HOLLOW, and made from laminated maple. The whole bike weighed less than 17 Kgs, or about 60% of the weight of the standard racing cart of the day.

Pickard published his data and test results in a fascinating article in the February 1971 edition of HOOF BEATS. Pickard's study was of such excellence that we can still learn much from it today.

Typical of Pickard's thoroughness, the following table demonstrates the relationship between wheel load, tyre inflation and rolling resistance, and was derived from an actual test on a 24 X 2.25" wheel/tyre combination operating on a nylon belted variable speed treadmill:


Axle wt in lbs	resistance @ 30 psi*	resistance @ 20 psi	resistance @ 10 psi
72	1.50	2.00	2.25
110	2.00	2.25	2.75
148	2.75	3.25	3.75
186	3.50	4.00	4.25
224	4.25	4.75	5.25

Note that the rolling resistance increases proportionally with weight, as the following graph shows ( note that this is for a tyre operating on a fairly hard surface, i.e; a nylon belt, and that the figures on the vertical Y scale are actual resistance x 100). The formula for rolling resistance is

R=(W * f)/r, where R= Rolling resistance in pounds, W= Total load on wheel in pounds, f = coefficient of rolling friction in inches, and r = radius of wheel in inches.

The "wavy" nature of the 20# and 10# lines indicate that Pickard "rounded off" his results to the nearest quarter pound. Pickard noticed that an increase in tyre width (from 1.5" to 2.25") resulted in a "slight reduction" in rolling friction. This agrees with the recent Swedish paper on the subject (reviewed in the October and November issues of "Harness Racer"), and with common sense.

Pickard further noted that,"Speed does not appear to affect the above values (rolling friction) appreciably. Once the bike passes the "break-away torque" and when moving from a 2:10 rate to a 1:55 rate (11% increase in speed), no measurable difference was detected."

The alert reader will have noticed that there is no variable associated with velocity in the formula for rolling resistance, and that we should therefore not be surprised when Pickard's results confirm that no allowance for speed is needed. Especially when Pickard derived his results from a wheel operating on a treadmill; ie; not moving forward. In such circumstances, it is little wonder that the dramatic increase in aerodynamic drag - which increases as the square of the airspeed (ie; the velocity to the second power) - went undiscovered.

1. WEIGHT. There is no disputing that rolling friction increases in proportion to the increase in load on the axle. A sulky with axle loads of 100 pounds (45.5 Kgs) will require twice as much effort to pull as one with axle loads of 50 pounds (22.75 Kgs). Pickard stated, " Disregarding the weight of the bike, it takes twice as much energy to pull a 200 lb man as it does to pull a 100 lb man."

This is an interesting contradiction of a much later article published in Australia in 1981 by another American, Bruce Kennedy, under the title "How Significant is a Driver's Weight", in which Kennedy cited an experiment by John J. Jackiewicz, that allegedly showed that driver weight made no noticeable difference in the effort required to pull a sulky. Kennedy reported that pulling a sulky at a 2:00 mile rate with an 85 lb driver weight, required a pull of 4.75 to 5 lbs from the horse. That would indicate a force of (say) 4.88 lbs = 0.39 horsepower.

On a hard dry track, using his ultra-sophisticated bike with a 110 lb driver, neutral balance and 28" wheels, Pickard achieved a power requirement of only 0.32 horsepower, or, to put it another way, a pull of 3.98 lbs. Had he had a driver weight of 85 lbs, as had John J Jackiewicz, who's tests Kennedy was quoting, the pull would have been reduced by the same percentage as the total weight, -17% - which would give 0.26 HP = 33% LESS EFFORT!

Pickard's bike demonstrated that the enormous sophistication that he brought to sulky design achieved precisely the reduction in power requirements that he had sought! He had made a sulky that, with identical driver weight to a "modified" sulky, required only 66% as much effort from the horse! That sort of achievement should have won him nearly every race in which the cart was used. History records that it didn't. There was no clearly discernable advantage to using the Pickard sulky that was reflected in race results.

The only reasonable explanation for this lack of competitive success being that, from the HORSE'S point of view, he wasn't feeling any advantage from the Pickard cart. But how could he not? Surely any of us could tell the difference between having to exert a pull of 3.9 pounds compared to 4.9 pounds?

The answer is to be found in an analysis of Pickard's opposition. What does a "modified" sulky do that the Pickard sulky does not? One thing the "modified" carts do is require more horsepower for the same speed, as the following graph illustrates:

The lower horsepower requirement of Pickard's bike compared to a current U.S. "modified" sulky is shown in Fig 3. above. The Pickard cart uses between 0.32 hp and 0.49 hp as the driver weight increases from 110 pounds to 220 pounds. By comparison, the modified cart uses between 0.37 and 0.59 hp over the same range of driver weights.

WHEEL DIAMETER. Pickard stated: "By increasing the size (or radius) of the wheel, it decreases the amount of energy required by the horse for any given weight of driver. For example: A 28" wheel is 7% more efficient than a 26" wheel, and if it could be made a 30" wheel, it would be 14% more efficient over a 26" wheel, thus a significant reduction."

This is yet another confirmation of Billy Hughes' remark: "The only thing we learn from history is that we don't learn from history!" Had Pickard studied the history of wheel sizes used last century, he would have been unlikely to repeat their mistakes, ie; wheels up to eight feet in diameter that produced NO increase in speed!

BALANCE. Pickard stated: " Balance weight is probably the most misunderstood of all the factors affecting the efficiency of the racing process. It is caused by the leverage effect in relation to the driver's position, the angle of the shafts and the location of the load-bearing axles. A negative balance weight actually takes weight off the horse but adds it to the axles. A positive balance weight adds weight to the horse but reduces it from the axles. A completely balanced bike does neither - each weighs as they would individually."

In 1968, in Germany, a non-engineer - Dr Walter Weber - had constructed a "crude and heavy" sulky with enormous negative balance. It was Weber's cart that was to revolutionise the world's sulkies. The later single-shafted and "modified" sulkies of Joe King all followed Weber in employing huge amounts of negative balance.

1968 - The Weber Sulky. Bill Haughton in the seat, Delvin Miller holding the shaft

They all required more effort to pull than Pickard's beautifully crafted bike, and they all went faster!

Do horses actually "enjoy" working harder, and go faster pulling a heavy load than a lighter one?

Pickard's fundamental error was in not allowing for the variation in effort required from the horse to move HIMSELF as the balance of the bike was changed. He assumed the horse ran best with no load. But the horse never runs with "no load". He must always move himself, and he weighs vastly more than the cart and driver.

Let us look for a moment at the horse's side of the question. From oxygen consumption data, it has been shown (Evans et al) that a fit racing standardbred develops about 42 horsepower over the two minutes or so of a sprint race, better than 132 times as much energy as that required to pull the sulky and driver.

If you're wondering how one horse can produce forty two horsepower, just remember that the definition of horsepower was based on the average power put out by an average horse during an entire day ie; 33,000 foot-pounds per minute, every minute, all day long. We are interested in the power we can coax out of a highly conditioned champion in two minutes.

Suppose now that we negatively balance the bike so that we remove 50 lbs from the horse and add it to the bike. The INCREASED effort required to pull the 34% heavier bike will mean a jump from 0.32 to 0.40 HP, an increase of 0.08 HP The REDUCED effort required from the horse to pull himself will be: (50/900) * 3.90 = 0.22 HP. The NET advantage will be a reduction in output required of: 0.22 - 0.08 = 0.14 HP = 3%.

In practice, it is very difficult to design a practical racing bike that will achieve 50 pounds of uplift with a 110 pound driver. A more achievable figure is about 24.50 pounds. The TOTAL horspower required for a 110 pound driver on a 37 pound modified cart, from the 900 pound horse is 4.17 hp. By comparison, the TOTAL horsepower required for a 110 pound driver on the 37 pound Pickard cart, neutrally balanced, from a 900 pound horse is 4.22 hp, which is 0.05 hp (1.2% more).

ALTERNATIVELY, if we hold the output constant and reduce our time (increase velocity) by 1.2%, we see a reduction in time from 2:00 to 1:58.6. In fact we DO achieve just such an improvement with negative balance, a fact that was born out in the field when the Macau Trotting Club switched from traditional Australian carts to a "modified" type in 1983: "Generally, horses improved their personal best from 1:5 to two seconds to the mile. The track record at Macau was quickly reduced to around 1:58." (Trotguide. November 1st, 1984)

Macau experienced a more than the 1.4 second improvement predicted, because they were changing from the long, low-uplift, Australian carts, to a short, high-uplift U.S. style carts, thus gaining an aerodynamic advantage as well as the lower horsepower advantage. The additional speed gained from moving the driver and rear of the bike closer to the horse accounts for the 0.1 to 0.6 second difference.

If we graph the combined or total output required at various driver weights for a 37 pound Pickard cart compared to a 47 pound "modified" cart, we have the following (FIg. 4.):

In the above graph, the total energy requirement for the Pickard cart increases from 4.22 to 4.39 hp (the vertical scale showing only the fractions of horsepower in excess of 4 horsepower), while the "modified" cart's increases from 4.17 to 4.27 hp. It is important to note that the increase in power requirement is LESS for the "modified" cart as the driver weight increases. Thus heavier drivers suffer less by comparison to lighter drivers in a "modified" sulky. In either cart, however, the lighter driver is at a distinct advantage.

When Pickard raced his cart against other types he found no discernable advantage. He summed up the experience: "It (the sulky) has been raced a number of times, but no one miracle idea can be offered for instant success. If it were that simple, it would have been discovered years before."

"Today, with the high degree of competition of evenly matched horses, one looks for inches. . . . . Just imagine the benefits to be gained by improving your horse's time by one-fifth of a second or putting him up the track 8 feet farther at the finish line." (emphasis mine)

Contrast those modest aims with the introduction in Australia of the offset sulky. Three horses use it once each for three life-time best times. The world record for a 2 YO gelding on a half mile track is broken by an extraordinary 3.4 seconds (ie; we "put him up" 150 feet)!!

However, the general tenor of Pickard's work is thoroughly commendable. He approached his analysis professionally, in full accord with the scientific method. His final comment; " In fact, it is my belief that if a continued scientific approach is taken to the subject (sulky design), even more significant improvements can be made." - sums up the calibre of the man. It is only with the wisdom of hindsight that we can look back and say that Pickard had failed to be aware that Weber's and King's carts had already rendered his fundamental assumption obsolete.

Today most of the world is unaware that the current "modified" cart is obsolete. The "modified" cart is now sixteen years old. It lacks structural redundancy, it has a short service life, and the records set in it are starting to fall to the newer carts. It will only cease to be widely used when enough people become aware of its inferior performance.

Summing up; what have we learnt about setting ourselves up for maximum speed, based on a century of accumulated knowledge?

WEIGHT. It is vitally important to realise that nothing I or Pickard have written disputes the fact that, ALL OTHER THINGS BEING EQUAL, it is best to have as light a sulky/driver combination as possible.

Jackiewicz's experiment was simply too crude to demonstrate the influence of weight. Both Jackiewicz and Pickard had some interesting comments about the standard of wheel building in the U.S.A.:

Jackiewicz:"Measurements of the force required to pull an empty sulky could not be made because it was impossible to keep the sulky from bouncing. The addition of a 50 pound weight limited the bouncing somewhat. It required 85 pounds to stabilise the sulky." (as quoted by B. Kennedy. Emphasis mine)

Pickard:"At all speeds from approximately a 2.57 rate (29.8 ft/sec) to a 1:48 rate (48.7 ft/sec) and especially at the higher speeds, the wheels used in the test vibrated excessively. In fact, with no weight on the bike (unloaded), the oscillations were so excessive that the forward pull values could not be read. Only after two trips to the local bike shop and bringing the outer circumference to a diameter trueness of under 1/8" was the wheel reasonably quiet. The wheel was also found to "wobble from center" and was "out of balance". ( Emphasis mine). It is worth remembering that these tests were conducted ten years apart, and they both appeared to have similar wheel trouble.

Pickard provided his usual accurate data with respect to driver weight. There is a linear relationship between driver weight and uplift (negative balance) such that a 110 pound driver with the sulky hitched to a 14-hand horse achieved a maximum of 9 pounds of lift, while a 210 pound driver in a sulky hitched to a 17-hand horse achieved a 48 pound lift.

The reason for the higher negative balance achievable on a bigger horse is that; the higher the shafts are raised, the more the wheels move away from the driver's centre of gravity and towards the horse. Thus the ratio that determines uplift is changed to enable more uplift to be achieved with no other changes to the sulky's configuration. This is why I always advise purchasers of steel shafted sulkies to hitch the shafts as high as possible on the saddle.

Thus, for a given sulky configuration, the greater uplift achieved by the heavier driver will tend to offset the increased rolling resistance resulting from his extra weight, but, as the graph (Fig. 4.) shows, the greater uplift NEVER entirely offsets the weight disadvantage of the heavier driver.

BALANCE. Negative balance is the only way to go. We want to achieve a negative load (ie; lifting the horse via the girth strap) on the horse of between 20 and 50 pounds (9 to 23 Kgs) depending on driver weight and sulky configuration.

WHEELS. As light as possible, and fitted with lightweight (white) discs that extend all the way to the rim on both sides of the wheel with NO valve access holes. Tyres should be lightweight, as near to 4.5 cms wide as possible, inflated to the recommended pressure. The wheels MUST be checked for skew; ie; no toe-in or toe-out is permissable.

HARNESS. As light as possible, short traces, wide elastic girth, with the sulky hitched as high as possible on the saddle.

SULKY CONFIGURATION. On Australian and North American tracks of 1000 metres or less, an asymmetric (offset) sulky may be built smaller and lighter for its length than any other current configuration. A smaller sulky weighs less, and produces less aerodynamic drag, which means greater speed. Additionally, the horse is able to cover less distance per lap (because he runs closer to the running rail) than in a symmetrical (conventional "modified" sulky). Less distance inevitably results in faster times.

TRACK. Choose a track that has fast average times. They are the best pointer to track potential speed. Differences of more than 3.0 seconds in mile rate are commonplace between tracks of the same circumference. It is a waste of time (literally) trying to do well on a track limited by its geometry to slow times. Track surface condition is also critical. It is well known that a "heavy" track produces slow times. Pickard found that the effort required to pull any sort of bike on a heavy track increased by two to three times!.

Don't forget, when judging average times, to take into account the standard of horses racing and the race distance. Certain learned gentlemen have put it to me that Mooney Valley looks a lot better than it is, due solely to the high standard of the fields racing there.

I am indebted to Bede Ireland and John Peck for the following table of sub-two-minute miles recorded at the major tracks over a seven year period. Note that in this period, Albion Park and Gloucester Park were rebuilt to a design by Bede Ireland, a fact that shows up in a sharp increase in speed as soon as the new tracks were opened:

Fig.5
TRACK	1980	1981	1982	1983	1984	1985	1986
Albion Park	5	5	5	5	6	223	328
Gloucester Park	20	25	34	48	92	143	255
Moonee Valley	19	25	41	69	113	147	173
Harold Park	24	25	34	45	59	72	81
Richmond Raceway	2	2	3	4	6	19	21
Globe Derby Park	2	2	3	3	8	9	11
Hobart	0	2	2	2	2	6	11
Townsville	0	0	0	0	0	0	3
TOTAL	72	86	122	176	278	619	883

If we graph these results, and restrict ourselves to (a) the average for all tracks, (b) the Albion Park results, and (c) the Moonee Valley results, we find:

Please note that:
1. Albion Park showed a prodigious increase in sub-two minute miles after the opening of the new Ireland-designed track in 1984. Fig.5 shows a similar result after the opening of the remodelled Gloucester Park track (again by Bede Ireland).

2. The number of sub-two minute mile recorded at Moonee Valley follows the National Average closely, as you would expect, since the track has not been re-built in the period under review.

There is the additional risk (on tracks of poor geometry) of injury to the horse in attempting fast times on inefficient tracks.

WIND. None is optimum. Any wind is a disadvantage and must result in a time below the horse's potential best. Optimal conditions are a warm, still, humid day. Humid air is less dense than dry air and so offers less resistance to the horse.