I've heard of some people widening the spark plug gap beyond the recommended gap, to compensate for altitude.

Why? Because recommended spark gap specs are related to air density at sea level. The higher altitude = lower air density. I live in utah, at approximately 4,500 feet above sea level, and the air density here is about 85% of sea level. This translates into lower compression, and the theory is that you can compensate for the lower air density by increasing the spark gap a bit. The question being, how do you determine how much?

Anyone else tried this?

 

for a racecar, I would consider it, but it would take a CONSIDERABLE amount of testing to see what a car liked.

For a DD, I honestly wouldn't bother

 

You have to understand the people that are at that level of racing are mixing very precision fuel and they will be making adjustments for when the sun goes down or if a storm front passes. Trying these types of fine adjustment on a car that is burning gas bought from a gas station is like buying high end range finding 400 yd scope and mounting it on a 10-4 shot gun that is loaded with bird shot.

When you open the gap on a plug you increase the voltage needed for it fire, in effect you make a hotter spark. You decrease the gap you reduce the voltage making a cooler spark. The recommended way to do this is to just buy hotter or colder plugs. The only time you would a just the gap is if you believe that one plug is too hot and another one is too cold and there is nothing in between.

I am not even sure the testing process would work on a modern car. When we dyno tuned our race car the goal was to run the hottest plug we could that did not cause pre-ignition (spark knock) with the CAM2 racing fuel. We never went as far as adjusting the gap. Your car is just going to retard the spark timing the second it feels any pre-ignition long before you can hear it. Even after we figured out the best plug we would back up one or two full heat ranges because we knew dryer air or slightly different fuel could cause a problem. Plug where pulled after every race, heads where pulled every 4 races any sign of fouling we went hotter and sign of pre-ignition we would cool them down.

Edit: Now that I think about it figuring out the best heat range on a newer car would be easier because you could use a scan tool and watch for Knock sensor activity. Sure seem like a lot of work, unless you have one of those 4th gens that are easy to remove spark plugs. I would expect you would need 3 or 4 sets of plugs just to start testing.