Started right up after winter storage. Constant smart battery charging always showing 100%, 12.7 v. But when brake pedal was released, engine stopped. Craftsman schematic shows mechanical grounding switches used to kill magneto. The seat and PTO switch were removed and buzzed out ok including a stress test (haha) of the PTO contacts (in vise , wiggle blades). Again no fault found with buzz out. Prepared for deeper look with a battery charge....Smart charger screen said to connect clamps...they were already in place. So apparently my smart battery charger (Viking and EverStart) will not recognize a battery with only 2 v. on the terminals. Checked with VOM...sure enough , only 2.1 v. A hydrometer test showed low electrolyte reading. Apparently the cranking to start (load test) had revealed a failed battery. The PTO switch terminals were lightly coated with bulb grease and repeatedly inserted and withdrawn from the mating female (dash) connector in an attempt to improve contacts. Then it and the seat switch installed. A used battery was installed, itself only fair electrolyte. EUREKA ! The engine booms to life and the brake pedal problem is gone. The failed battery must have had a surface charge. But is the magneto kill associated with voltage? The schematic of the PTO switch shows contact A should not touch contact D. If it does it looks like grounding is available through D to the magneto, stopping engine. Did the bulb grease exercise temporarily help? A buzz out can not simulate an operating current. There are many PTO switch sales on ebay, whether changed out by guess or by golly. I may have to replace it. Your thoughts ?