Page 1 of 1

Some clearifcation need on advance under load/vacuum

Posted: Mon Jan 25, 2010 1:40 pm
by Desp
I'm a little confused as I've read it many different ways and I don't have a solid understanding of this yet, so I'll explain my situation as it stands and hopefully someone can clear this up for me.

I have an old VW air cooled engine, running MAP to the mjlj. At idle my vacuum reading is 97 kpa, acceleration increases vacuum, therefore the number drops. Not sure if this is backwards from other motors, but it seems to be. I have the vacuum line connected to the vacuum advance on the carb where the old dizzy was connected to. It started out confusing as at idle/no load the MJLJ mapping starts out in the bottom left of the program, instead of top left. Under acceleration the vacuum drops to around 85 kpa.

Now I've seen people say it both ways and I'm a bit confused. Under load do I want to advance the ignition or retard it? At around 1500 RPM no load I'm running at 15 deg, do I want to increase or decrease the advance, people have said pinging will occur under load so that makes believe it needs to be retarded, but with more full mixture it needs longer to burn, so it needs to be advanced right? Right now I'm retarding it, and under load at low RPM i have no power, which makes me think it's burning after TDC thus = no power.

Posted: Mon Jan 25, 2010 3:06 pm
by DannyP
Under load, less advance. Or rather, max advance without detonation or pinging. The vacuum or MAP sensor merely allows you to add advance at lighter loads which lets you run more economically, i.e. save gas. You should not run more advance at low vacuum/high engine load conditions, unless you like burned pistons, bent rods, and cracked blocks!

Posted: Mon Jan 25, 2010 5:34 pm
by Quagmire
So your vacuum increases (kPa drops) under acceleration? Sounds odd. Really it should be the other way around- what side of the throttle butterfly is the map takeoff? Filter side or manifold side? Should be manifold side.

Really you should see your vacuum highest (kPa lowest) when you are at idle or over-running with your foot off the throttle. Thats becuase the throttle buttefly is closed but the engine is still running revs and so trying to draw air past the closed butterfly. This causes the drop in manifold pressure.

Conversely you should see no or very llittle vaccum when accelerating or at wide open throttle. This is because in this position the engine can draw as much air as it likes as the buttefly is not restricting and so you effectively get a reading approaching atmospheric pressue (~100kPa)

For reference mine idles at 600rpm drawing 35kPa (roughly) and if i give it lots of beans i can get to almost 100kPa :shock: . Not sure how much these readings vary across different engines, or if all engines operate in the same areas...

For more info on how advance works, read this thread:

http://www.autosportlabs.org/viewtopic.php?t=2561

As Danny P says- Less advance under load is the way to go.

Posted: Mon Jan 25, 2010 8:41 pm
by Desp
Thanks for the responses. I think the vacuum issue is what caused a lot of confusion for me, since I'm basically having to write the map backwards. I'll try and tap the manifold with a barb to attach the vacuum line to and see if I can just plug the carb.

I have a map from a centrifugal dizzy, with that type it takes into account its always under load so whatever range it goes through its safe to make that the maximum advance under load, right? Because it has no way of knowing load since it’s based entirely off of rotational speed.

Posted: Tue Jan 26, 2010 1:24 am
by DannyP
FYI, sometimes the carb vac port is over or under the throttle plate, depending on throttle position. These ports are usually designed to work with a specific distributor. ALWAYS plumb the MAP sensor directly to the intake manifold. Or, in the case of multiple carbs, either tap each runner to a common vac chamber, or just use a TPS like me!

Desp, you are right about the centrifugal map, that was how I did my first few maps until I felt comfortable with the load sensing.