The Mark Ortiz Automotive
Reproduction for sale subject to restrictions. Please inquire for details.
Mark Ortiz Automotive is a chassis consulting service primarily serving oval track and road racers. This newsletter is a free service intended to benefit racers and enthusiasts by offering useful insights into chassis engineering and answers to questions. Readers may mail questions to: 155 Wankel Dr., Kannapolis, NC 28083-8200; submit questions by phone at 704-933-8876; or submit questions by e-mail to: firstname.lastname@example.org. Readers are invited to subscribe to this newsletter by e-mail. Just e-mail me and request to be added to the list.
HYSTERESIS EFFECT WITH BRAKE PROPORTIONING VALVES
I have been an enthusiastic reader of your newsletters when I could get them since 2000. Another subject I am excited about is brakes. I captured the data pictured below in support of my typical proportioning valve discussion. It starts with something like MO in the October 2016 newsletter:>a proportioning valve in the rear brake line [snip]. This allows the car to have more rear brake on gentle application than on hard application. This is particularly helpful when it rains But it ends with something like ES in an August 10th Electron Speed Facebook post:>As the brake pedal is released during trail braking at corner entry, the brakes bias more toward the rear. As a school of hard knocks kind of guy with a few years under his belt, I have an observation that you may appreciate. I suspect that this is why you typically see the inner rear wheel lock on corner entry on a traditional tin top (proportioning valve). And you typically see the inner front wheel lock on corner entry on a traditional formula car (bias bar). Upon review, I notice at least 2 typos in the picture below. I hope that as a technical writing reviewer these are not the focus of our upcoming conversation. I look forward to the technical aspects of this discussion.
This appears to be a data acquisition plot from a real car, set up for 50% front brake or at least identical front and rear line pressure, when the proportioning valve is below the knee point, a 400 psi knee point, and about 4/7 or 57% pressure reduction beyond the knee point. The data appears to be taken with the car stationary and the brakes applied and released more gradually than would actually be seen when racing, to show the action of the proportioning valve more clearly. If the car were actually racing, we would expect much more abrupt brake application and also a brake release lasting a lot less than nine seconds unless the car is on a large oval track.
Anyway, the hysteresis effect – higher rear brake pressure during release than during than during application – is real, and is widely recognized. I will also send with this newsletter a .pdf file from AP Racing (available for free download at their website) that shows an idealized graph of input and output pressure of one of their proportioning valves. The AP graph shows a similar effect to the traces above. For a given input pressure, the output pressure is higher during brake release than during brake application.
The AP graph shows no drop in output pressure as the input pressure first decreases, but it shows the output pressure starting to drop before it is completely equal to input pressure. The trace above actually does show a slight drop in output pressure when input pressure first begins to decrease, but then output pressure holds steady while input pressure drops from 1500 to 1000 psi. 1000 psi is where input and output pressures are equal during release. They are only equal up to 400 psi during application.
At around 16.5 seconds, which is during release, the input pressure is around 1300 psi and output pressure is 1000 psi. During application, output pressure doesn’t get that high until input pressure reaches 1800 psi. 1300 psi input pressure on application produces only 800 psi output.
AP notes that the hysteresis effect is quite variable. It is affected by where the knee point is set. An earlier knee point results in more hysteresis. It is also affected by fluid volume required to operate the brakes. AP doesn’t say, but I would expect that larger cylinders, more fluid displacement, and lower pressures would increase the effect (and would certainly require a lower knee point pressure for a given application, which alone would increase the effect).
Does this mean we shouldn’t use proportioning valves in race cars at all? Or, if we do use them, what are the effects on car behavior and what allowances for those effects are appropriate?
Even without hysteresis, just having more rear brake on light application than we would have with only a balance beam makes the car freer (adds oversteer) when trailbraking. The hysteresis further exaggerates this. But is that bad? For trailbraking into a high speed sweeper, maybe. On the other hand, for a situation where we want to toss the car with the brakes, to use the brakes to make the car rotate, as in autocross or rallying, it may be an advantage. In most cases it will require an earlier knee point than would be needed if hysteresis were absent, if the car has only a proportioning valve, or a later knee point but more front brake on the balance beam if the car has both a balance beam and a proportioning valve.
Can we tell if a car has a proportioning valve based on whether the inside rear or inside front wheel locks first in trailbraking? Not necessarily. That will depend on the lateral load transfer distribution as well as the brake force distribution, and how the brake bias is overall. However, it is true that a car with a proportioning valve will have more tendency to lock the inside rear, and less tendency to lock the inside front, than the same car with only a balance beam, other things being equal.
Would it be possible to eliminate hysteresis? Perhaps. I don’t know if this has been tried, but it would be possible to create a knee in the rear brake pressure curve with a return spring at the rear master cylinder, arranged to only operate once the piston has moved to a particular displacement. This spring could be either inside the cylinder or outside, perhaps coaxial with the push rod. This would have no hysteresis. It would create a knee point that would depend on master cylinder piston travel rather than line pressure. I am not sure whether this would create unwanted changes in the knee point in actual use. Probably an external spring would be better than one inside the cylinder, to allow for knee point adjustment.
In any case, it is definitely possible to race successfully with a proportioning valve, but it is necessary to understand the device’s properties and effects, including hysteresis, and tune around them.