My 230SL almost always stalled at the end of warmup (at or after WRD termination). It still occurred after months of effort to determine possible reasons and to readjust everything adjustable.
I discovered by multiple Split Linkage Tests that Lean Dip on my car was neither lean nor rich but was stoichiometric. I used a table of %CO values with their corresponding Air Fuel Ratios (provided by some Gunson Exhaust Analyzer suppliers). I used it to "translate" the typical base curve of the sl113.org Technical Manual's Fuel Injection Pump interactive from %CO to Air Fuel Ratio vs Time. The Fuel Injection Pump interactive was written by Mrfatman and is excellent way to learn how the Fuel Injection Pump behaves. The lowest %CO on the curve (leanest point on the Lean Dip curve) translates to 14.17:1 A/F whereas stoichiometry is 14.7:1 A/F, which is leaner than the leanest Lean Dip point. Lean Dip is a an appropriate description for an easily recognized portion of the %CO vs time-since-start curve of a Pagoga cold start and warmup. Although none of the curve is actually leaner than stoichiometry, that portion of the curve "dips" into a portion of the graph that is "leaner" than the other parts of the curve. It also corresponds with a part of the engine curve that tends to be behaviorally problematic for the engine warmup.
On my 230SL, enleanment and enrichment via Split Linkage manipulation as idle began to deteriorate toward Lean Dip stall inevitably caused instant stalls. I could not get the rpm to respond to fuel or air except to stall.
I would describe the mixture condition during the Lean Dip idle deterioration as "fragile", and the rpm drop inevitable, making it vulnerable to even slight air-fuel ratio changes which caused instant stalls.
SUSPICION: My warmup stalls were not caused by Lean Dip, except perhaps a lean dip in rpm caused by cold or not warm enough engine induction system hardware leaning out the mixture through condensation. My problem appeared to be caused by Combustibles Starvation at the end of WRD function.
When I searched for a reason to justify that suspicion. Given that the "base fuel and air calibration" after WRD shutoff would have been sufficient to maintain a reasonably rich idle if the engine were warm enough, I could only imagine the following HYPOTHESIS: At the end of WRD function, the engine (intake manifold, fuel injection pump, fuel injection lines, injectors,are still not hot enough to vaporize "enough" of the provided idle fuel and air mixture to maintain good combustion. Only the spark plugs, combustion chamber, and perhaps the exhaust valves are at a warm enough temperature. But they are perhaps not in contact with the mixture long enough to add much to the vaporization. It therefore might tend to stall.
Since I know of no way to warm up the engine faster (without driving away), I conclude that I need to do something that Mercedes probably did not anticipate when these engines and fuel injection systems were new - 60 years ago. That is to stretch out the WRD operation time to keep it delivering warmup fuel and air until the engine has been heated up and completely "weaned" from the warmup fuel and air volume and mixture ratio. "THAT", I know how to do.
The entire operation of the WRD is driven by the heat provided by the engine coolant as the engine heats up and the heating coolant's effect on the WRD wax filled thermostat, as it expands and controls the WRD air and fuel mixture and quantity. Consequently, it is simply necessary to determine how much to slow down the WRD by reducing the warming engine coolant flow to the WRD as the engine warms up.
As a first test of the hypothesis, I reduced the flow to the WRD by significantly pinching the coolant supply hose feeding the WRD, using a pair of needle-nose vice grips.
I have always been lucky. My luck continued. During the first test of the hypothesis there was no Lean Dip and consequently, no Lean Dip stall.
I bought a brass 1/2 inch PEM ball valve at a plumbing store ($ 10.00 Canadian), installed it in the top hose of the WRD and ran two tests on consecutive days with different valve restriction settings to bracket my best-guess vice-grip restriction setting.
The first test, quite restrictive, slowed down the WRD but not enough - the Lean Dip stall occurred anyway, but later and at a higher engine temperature than usual. The second test, even more restrictive, effectively turned off the WRD and the engine would not keep running at all. The WRD did not become warm to the touch.
The third test, a best guess between the previous two settings worked perfectly. The engine ran, did not falter and did not even think about stalling. Everything was as it should be.
I would like to eventually replace the valve with a fixed restrictor inside the hose so that it looks original. But I will test a bit more first. Based on the angle of the valve knob, and internet ball valve flow curves, it is flow-reduced to about 13% of the fully open flow, or hose-only flow. I have not timed it yet but the time to "stable engine thermostat temperature" (78-79 deg C) is about 7-8 minutes.
Is this a definitive solution? For my car, I will let you know in a few days. I welcome discussion, suggestions and questions.
Tom Kizer
