I can't diagnose the erratic idle issue. 94 Civic DX started that rapid idle cycle around 1500-1650 RPM, once per second making car undrivable. Erratic idle started on drive home, magically disappeared 3 miles later after stop/restart and came back 50 miles later for good. Triigered a code 14, but before you recommend obvious IAC let me go through what I've tried.

- 94 DX with 5 spd Manual Tans
- Engine is 1500 D15B7 but chassis wiring color code appears to be D15B8; for example MAP connector is Y/R-G-W and MAP location is on throttle body.
- Replaced IAC with salvage used part, then new part from discount AP.
- IAC voltage reads in spec. mechanically sound, clicks wide open when connected.
- IAC air is rushing through TB bypass port full open
- Engine still idles erratically with IAC disconnected but slightly longer cycle.
- Replaced TPS
- All 4 plugs are black, replaced plugs, replaced fuel filter
- Checked fuel injectors - passed spec
- Checked O2 sensor - in spec.
- Checked TW temp sensors - in spec.
- Cleaned warm idle air bypass valve (bottom throttle body)
- Replaced MAP with salvage MAP, then new MAP from discount A/P
- Voltage at MAP connection D17-D21 at ECU pins are 2.47 V (1/2 voltage spec).
- Replaced Engine Control Unit (ECU) New from discount A/P - No change.
Nothing I've tried gets rid of poor erratic Idle.
I'm out of ideas. Help please?!!?
Thanks

 

With IACV disconnected, there should be no air flow through the IACV bypass port. The engine should slow down to almost stall. So something else is letting too much air into the intake. Often it is the FITV, (Fast Idle Thermal Valve) which has bypass port in the bottom of the throttle body. That valve is actuated by the coolant temperature, it should be fully closed above 60F. In other words it is designed to open only for cold starts in cold weather.

 

Thanks for the response. Here's a couple more details which may help.
> The car surges with or without the IAC connected.
> The only difference is surge is every second with IAC connected, and every four seconds with IAC disconnected. All valves are staying 50% open when disconnected, including the brand new one.
> I've tried two used IACs and one new IAC with exactly the same surging results.
> Majority of the air is surging through IAC, some through FITV.
> I've tested by closing off the IAC port with rubber hose and the car surges heavily 15-30 MPH range although I can get the idle to smooth out after tweaking idle adjustment.
> the FITV was removed and clean. It is passing some air but not nearly that of IAC.
> Tested the FITV by choking off with hose, and still get same massive surge, undrivable car.
> What's got me stumped is that I can disconnect IAC, close off IAC and FITV and still get surging. I can get the idle to smooth out by applying vacuum to offline MAP. Car surges as soon as I open the throttle.

> Could this be related to bad ground or sensors inside the distributor? Anyone have this cause surging?

 

As the car doesn't drive properly either, there is a different, perhaps more fundamental problem. The IACV and FITV only affect idling.

Start by checking the spark timing. Also test the MAP and TPS sensors with a voltmeter. MAP output voltage should be about 3.5 volts engine off and drop with engine running or vacuum applied. Remote MAP sensors need to be connected via a dedicated hose to the port on the top of the throttle body.

 

Reset spark timing 3500 miles ago when I replaced the leaking o-ring on distributor and was very careful to make sure it was right since that was right before I drove on 3000 mile trip and car ran flawlessly @ 43MPG. I used medium lock-tight on 3 bolts to make sure nothing moved. I'll verify nothing has changed.
TPS checks OK @ 4.5 V to 0.5 Volts exactly in spec. It too was replaced early in diagnostics.
MAP - Heres where it gets weird. the MAP is located on top of TB, not on firewall as it should be for D15B7 1500cc v-tec engine in the car. The wiring in car is for the D15B8 chassis and is reading 5V to Y/R-Grnd, 5V to Y/R-G/W, and 2.47V to W-G/W (should be 5V). Trace it all the way back to the D connector at the ECU and terminal D17-D21 and it is 2.47V. Opened up ECU and nothing looks burned out on circuit board path. Anyway, replaced the ECU with $273 Advance Auto ECU and same old crap...no change in surging; returned computer. I've also substituted another OEM used MAP and new $80 aftermarket MAP with no change. I can disconnect the harness to the installed ECU, plug into the spare ECU and apply vacuum with IACU disconnected and everything smooths out at idle with surging magically disappearing, although I cannot drive the car without interactive MAP feedback to ECU.
I'm at a loss to determine the one electrical thing that went bad, especially since It was intermittent for the final 55 miles before the big downfall.
Its become a expensive guessing game of swapping out perfectly good parts

 

Sometime during the 5th generation, Honda switched to TB mounted MAPs. My '95 DX had one it was definitely stock.

The voltage at the output wire (I think is white) must vary with manifold pressure. The other two wires are constant 5V and ground. If the white wire were crossed to another wire in the harness, MAP readings will be disturbed. Unplug MAP and ECU and confirm infinite resistance from the wire to anything else.

Make sure the vacuum passage from the sensor through the TB and into the manifold is open and not leaky. Using the wrong gasket or a bunch of silicone sealant between the TB and the manifold could block it. There needs to be an o-ring under the MAP sensor, around the nipple that mates to the passage.

 

Thanks for all the great pointers. I've been following along the same line of troubleshooting you have hit on.

I've checked the MAP according to the Civic OEM shop manual and here's what I've found. According to the book, ther MAP has 3 connections, Y/R, G/W, White:
MAP disconnected Y/R-Grnd = 5 Volts (Yep, 4.97V), MAP Y/R-G/W = 5V (Yep, 4.97v), G/W-White 5V (Nope - 2.47V). Same G/W-White with MAP connected = 3V (Nope - 2.47v).

So I traced all the way back to the ECU for multipin connector D17-D21 and measured 2.47 with MAP connected, and 2.47V with harness multipin connector removed. Also measured 5V from D19-D21 per the book. Just for diagnostics, I clipped the harness wire at D17 (to the MAP white connection) and jumpered it to D19 to put 5V on the white MAP connector. The engine will idle smooth about 1000 RPM only with the IACU disconnected, but surges rapidly and significantly once I crack the Throttle or with the IACU connected. Disconnected IACU is still allowing a moderate amount of air to pass through bypass port, about equal that of the warm idle control unit port on bottom of throttle body. The car is still very much undrivable, bucks, kicks and surges with rich gas smelling exhaust.

The only other thing I can think of that I have not tried is to replace the TDC/Crank/Cyl sensors which is the entire distributor. Bad head or plugged catalytic converter? Before I rip the intake or exhaust off, I want to rule out electrical gremlins. Thoughts, suggestions, advice, similar experience anyone? Help.

I'm getting a little closer to sending it over the cliff and may start selling tickets to the big event. first come, first serve......takers?

 

 

It sounds like the white wire lacks continuity from the ECU to the MAP itself. Maybe the plug on the MAP end is not making contact with the pin in the sensor. Check resistance from the ECU plug (unplugged) through the MAP sensor (plugged in) to the green/yellow wire. Compare to what you read at the sensor pins directly with the plug off.

This isn't real complicated. The MAP puts out a voltage proportional to manifold pressure. The ECU receives that voltage to act on it. It's not getting there.

 

open you raidaitor cap let it open up for air to get out and let it idle till it bubbles put the cap back on there could be air trapped in it, also make sure your dist is in time i tried making one hundred percent mine was to but it never goes that way. try to change the oil.
after i did that to my 5 speed accord it fixed it
and i did everything youve already done too