Rough Idle
 

I have had a rough idle since I finished my hf vx swap. I have replaced all ignition components and meticulously checked for vacuum leaks. When I pump the brakes the idle jumps from 700 to about 1500. Is that normal or could this be the leak I've been looking for. Can somebody that has experience with this help me fix this idle problem. When I shift into neutral while coasting the revs hang at about 1000rpm and then drops down and runs rough.

How about temporarily plugging the brake booster vacuum hose and seeing if the problem goes away? (Drive slowly during this test and be prepared for a hard brake pedal)

Do your power brakes still have vacuum assist for at least a minute after you turn off the car? If not, the brake booster may have a leak.

Yea do what erik said but i wouldnt drive the car with the booster vac not hooked up (ive tried, takes ALOT of force)

yes u should have some vacum left to be able to pump the pedal 2-3 more times after the engine is off then it should get hard as a rock.

does this car have an EGR valve? id temporarily disable it (unplug the vacum hose and jam a golf tee in the hose) im almost betting if it does have an egr valve its sticking, it can cause rough idle, hesitation, stumbiling, etc.

From what you said, even if the booster does have a leak it isn't the source of your rough idle. If it was leaking enough to cause a rough idle it wouldn't increase idle speed when you press the brakes. You might very well have a leak, but it isn't your problem. The only leaks that normally cause a problem are leaks in the intake pipe after the mass air flow sensor making the ECU think less air is entering the engine causing a low, lean idle. Your typical vacuum leak causes an idle increase, to a point.

Have you checked your ignition timing?

Does it idle rough when cold, hot, or both? If it does it while the engine is hot, what does the exhaust smell like?

I've stopped a 6 ton truck before when the power brakes failed. (wasn't easy, but I did it.) Surely a little civic would be nothing like that.

It is rough when warm. The exhaust doesn't smell like anything abnormal. Isn't the brake booster after the map sensor in the vacuum diagram. I was going by where it attached to the manifold. Why would a vacuum leak in the booter be any different than anywhere else behind the map sensor. I checked my timing and it seems to retard at regular intervals with the service connector jumped. I thought that it shouldn't be retarding so maybe the #1 plug wasn't firing as well as it should. I did a full tune up. Does anybody elses rpm's surge when they pump there brakes while the car idles?

MAP sensor is even better because a vacuum leak is registered by the MAP sensor so your mixture always stays good.

The ignition system showing a periodic retard when the engine is idling with fixed timing means the ignition system is firing late. Did you replace your ignition coil when you did the change? Connect the timing light to the high tension wire from the coil to the distributor and watch if it retards a few times in a row or not.

It's entirely possible for a Civic to be more difficult to stop, depending on system design. It's a matter of what kind of leverage the hydraulic system gives you, the swept area of the brake pads/shoes and the leverage that swept area has over the wheels.

Yes. The brake booster runs on engine vacuum, so it's normal for some air to be released into the intake during normal operation.

Doesn't exist. The coil is built into the distributor housing, so the primary lead is integrated into the distributor cap.

I looked at the timing again and I don't think it's retarding. I think every few go rounds it isn't firing. Can somebody with a civic pump his or her brakes with the idling and tell me how much it surges.

Base idle is about 670 RPM. Pressing the pedal once and holding it there results in the engine speed revving up to about 1050, then settling to about 800. Rapidly pumping the pedal brings it up to about 1200 RPM.
As stated, Hondas use a speed/density system with a Manifold Absolute Pressure sensor to judge air flow. Unlike Mass Air Flow systems, vacuum leaks will not cause misfires due to leaning of the air/fuel mix. In a MAP system, the vacuum leak shows up as an increase in manifold pressure, which the ECU reads no differently than opening the idle set screw, or the fast idle thermo-valve being open. Try it for yourself: Start up the engine, and pull the brake booster vacuum line off it's intake manifold nipple (you'll have to move the spring clamp up the hose a couple inches and maybe un-stick the hose first). That will give you a roughly 1/4" diameter vacuum leak... The engine speed will rev up to about 3k RPM or maybe start reving up and down (alternating between idle and decelleration fuel-cut modes). That's it.