Automotive History: Chrysler’s CAP, The First Effective Exhaust Emission Control (With ’63 Dodge and Dart Road Tests)

By the late 1950s, it was grossly and painfully apparent that “traffic smog” was a real and really bad thing. Dutch-born scientist Dr. Arie Jan Haagen-Smit figured out unburnt hydrocarbons and nitrogen oxides—noxious in their own right—react with sunlight to form a toxic stew of ozone; peroxyacetyl nitrate; nitrogen dioxide, and aldehydes.

And automobiles, more and ever more of them as the postwar boom boomed along, were the main source of atmospheric unburnt hydrocarbons and nitrogen oxides. Add sunshine and unfavourable valley- or bowl-shaped topography, and the result was poisonous air so thick it could almost be shovelled. Los Angeles was the American poster child…

Los Angeles Civic Center one smoggy day in 1948. Buildings just one block away are scarcely visible.
Los Angeles Times Photographic Archive, Department of Special Collections, Charles E. Young Research Library, UCLA

…but far from the only affected locale; Denver—300+ sunny days a year; bowled in by the Rocky Mountains, and up at high altitude so rich-running cars chuffed out enormous amounts of unburned hydrocarbons—had its persistent “brown cloud”:

That’s downtown Denver on 15 December 1962…somewhere under that cloud of smog.
Charles Grover image

Smog was ugly and smelly and eye-burning and far, far worse. Car reviewers of the day added colour to their stories by resignedly describing the need to hurry and get all the brake testing (or whatever) done before noon when breathing outside became too painful. People got really sick. People died. Crops withered. Its severity cannot be properly understood by first-worlders who didn’t live through the worst of it back then, and who’ve never been someplace like Beijing or Delhi or Mexico City.

Los Angeles pedestrians in 1958 bat at their smarting, watering eyes and take refuge behind gas masks. They’re under chemical attack not by some belligerent foreign military, but by all those automobiles.
Herald-Examiner Collection/Los Angeles Public Library

A first whack at the lowest-hanging fruit was ducting engine blowby into the intake tract rather than out to the atmosphere; PCV (positive crankcase ventilation) became mandatory on 1961 cars in California, 1962 cars in New York, and by 1964 it was standard equipment—more or less; hold that thought—throughout North America. That helped, because those crankcase fumes and vapours were very high in unburnt hydrocarbons. And despite a bit of initial turbulence—some early PCV valve designs made a maintenance hassle of themselves, especially at the hands of the uneducated or skeptical—positive crankcase ventilation made life better. The wretched odour of crankcase fumes was much reduced for those behind (and inside) the car; the engine oil stayed a lot cleaner, and here’s Gus Wilson to describe a less obvious advantage; click the pic for the episode as a PDF:

Subsequent whacks were going to be a lot trickier, because tackling tailpipe emissions meant automakers suddenly had to figure out how to make engines run as well as buyers had come to expect, but with much sweeter breath, and without costing significantly more. The technology, technique, and knowledge of engine emissions and combustion dynamics were primitive, and for years the standard fix for any driveability faults had been to fatten up the fuel mixture. Balky cold starts? Add more fuel; richen the choke. Stumble on acceleration? Add more fuel, bigger accelerator pump shot. Surge at cruise? Add more fuel, put in a bigger jet. Hesitation on flooring the accelerator to pass another car? Add more fuel, loosen up the power valve. This was easy, cheap, and effective, but the resultant exhaust was filthy (and fuel mileage was stinkin’ to match). Once emissions started mattering, that method was right out.

As they bumped and bottomed along the rocky learning curve, under tight deadlines and budget constraints, automakers flailed around with a variety of tweaks and twiddles, gadgets and widgets, thoughts and prayers—mostly to no viable, commercially-acceptable effect. Chrysler’s engineers wanted to avoid afterburners, air pumps, and other costly, problematic hang-on devices other automakers were kludging around with; instead to improve combustion so less unburnt fuel would wind up in the exhaust.

Groping around for the beginnings of an understanding of the factors at work, they found that tailpipe emissions were especially dirty at idle and while decelerating. The common factor between those two conditions: the throttle was damn near completely closed.

Idle cleanup was relatively easy. Curb idle speeds at that time were very slow—500 rpm was pretty normal—and so at closed throttle there was very little air admitted to the engine.

A wider throttle opening admitted more air, but raised the idle speed too high. Retarding the basic ignition timing by as much as 10 degrees slowed the idle back down despite the wider throttle opening; now they had their market-acceptable idle speeds with much cleaner exhaust.

But retarding the spark and just leaving it like that would do a real job on driveability, performance, and economy. Ripping off all those extra-fuel band-aids was an obvious and necessary first step to clean exhaust, but then how to provide acceptable driveability? Engines run poorly with retarded timing and lean carburetor jettings and settings; they have no power and they guzzle gasoline. That was where pretty much the whole rest of the auto industry was stuck.

Meanwhile, exhaust is dirty under deceleration because at road speeds with the driver’s foot off the accelerator, the vehicle motorises the engine and creates very deep manifold vacuum, but the nearly-closed throttle limits the amount of mixture drawn into the cylinder. The density of the mixture is thus low, and so combustion is uneven and incomplete. Camshaft overlap aggravates the problem; the exhaust valve opens before the piston reaches the bottom of the power stroke, so some exhaust is pulled into the cylinder, which dilutes the cylinder charge and further interferes with combustion.And with the retarded basic ignition timing there isn’t enough time to burn this raggedy mixture. Net result: a great deal of unburnt gasoline flies out the tailpipe.

The wider throttle opening at idle helped some, but not enough. The core problem was that under deceleration when a lot of spark advance was needed to give a complete burn in the cylinders, there wasn’t any advance provided. The engine wasn’t typically spinning fast enough for much centrifugal advance, and because the vacuum advance was tapped in above the throttle plate, there was no vacuum there when the throttle was mostly or fully closed. But if they hooked up the distributor to manifold rather than ported vacuum (below rather than above the throttle plate), it caused other emissions and driveability problems: two steps forward, two steps back.

Advanced timing under deceleration would start the fire earlier, giving more burn time, and so what was needed was to somehow provide retarded timing at idle but advanced timing under deceleration—two closed-throttle conditions differing only in engine speed, driven in turn by whether the car was moving or still.

Chrysler’s emissions boffins leveraged the strong manifold vacuum that was sucking unaerated fuel into the intake manifold under deceleration. Their solution was to send either ported (above throttle) or manifold (below throttle) vacuum to the distributor, depending on driving conditions. This was achieved with what was called a distributor vacuum control valve:

The control valve has three vacuum nipples: one to manifold vacuum, one to the vacuum advance port on the carburetor, and one to the distributor vacuum advance. The spring has a tension adjuster screw running through the centre of its keeper cap.

This is a pneumatic logic gate—really, the first engine control module. It advances the timing under closed-throttle, high-vacuum situations: deceleration or coasting with foot off accelerator. When manifold vacuum (below the throttle) is equal or less than ported vacuum (above the throttle), the distributor receives an ordinary ported vacuum signal as though the valve weren’t there. When the throttle is closed and there’s no ported vacuum, the strong manifold vacuum under the closed throttle at higher-than-idle engine speed overcomes the control valve’s preload spring, opening the valve and sending manifold vacuum to the distributor, which advances the spark.

This handily resolved the timing conflict, and so Chrysler were first with an emission control system that met all the targets: it radically cut emissions without spoiling driveability, performance, or fuel economy.

The system comprised this control valve, a carburetor with leaner and more precise calibration, and a distributor with a modified centrifugal advance curve to provide normal levels of spark advance for acceleration and cruising despite the retarded basic timing.

It was called CAP, the Clean Air Package, and was first launched on a trial basis in California in 1963. That trial was a grand success; equipped cars ran well, with good driveability and fuel economy and much cleaner exhaust. Here are two tests of the time (again, click the pic for the article as a PDF); a Popular Mechanics article from November of 1963…

…and one from Car Life, September ’63:

These 1963 installs show the valve with spring exposed; a snap-on plastic cover was added for volume production to attenuate valve rattle and discourage errant screwdrivers. The distributor vacuum control valve wasn’t a service nightmare, but it did add some check-and-adjust points at tune-up time. With insufficient spring tension it would give vacuum advance action at idle, which would interfere with setting the idle mixture and speed for a smooth, stable, clean idle. With excessive spring tension it was as though the valve weren’t present. It didn’t tend to go out of adjustment; as with many such things, it worked fine and best when left alone.

Starting in 1966, CAP became standard equipment on all Chrysler Corporation vehicles sold in California, and Chrysler even offered “UCCAP” (Used Car Clean Air Packages) at dealer parts counters, for retrofit to any 1955 or newer Chrysler vehicle.

All in all, CAP gave a much better result than other automakers could come up with for awhile, as these Popular Mechanics “Detroit Listening Post” excerpts show:

September 1964. Will Ford switch to new low-maintenance PCV valves?
(And, er, bus exhaust with a lovely floral fragrance…?
God only knows what hideously toxic chemicals that would’ve taken…)


June 1965: No, instead Ford just stopped putting PCV on non-California cars.
Chrysler’s CAP is out in front. (V6? What V6?)


September 1967: other automakers get hip.

The distributor vacuum control valve was used on some CAP-equipped engines from 1963—those prototype CAP cars—through 1969. CAP became standard equipment on all US-market cars for ’68, and in ’69 Chrysler renamed it CAS, for Cleaner Air System. A few years after Chrysler, Ford and AMC began using more or less similar valves. Other emission control strategies supplanted the valve by 1970.

This seminal work on the original CAP poured a foundation for Chrysler’s later Electronic Lean Burn system, which also controlled ignition timing in unusual ways to allow for nonstandard carburetion. As emissions regulations grew tighter, Chrysler did eventually use such add-ons as the air pump and catalytic converter, just like everyone else, at least on some models. But for awhile there at the start, Chrysler were the only automaker putting out cars that ran well and put out exhaust clean enough to find favour with California’s air guardians.

Ironically, Dr. Haagen-Smit died aged 77 of lung cancer. While he was busily working to clean up everyone else’s air, he was just as busily toxifying his own as a heavy cigarette smoker.