TurboPirate

Hello everyone, my nephew got a 1990 Honda Civic HB with D16 SOHC engine wish it was made Turbo already, the first thing I notice when he brought it home was the black smoke and way to rich idle, saw melted plastic and found the factory MAP sensor melted, I installed a 3 Bar GM MAP sensor now idle is 99% perfect, trying to step on the gas pedal and instead of taking off it simply dies like it ran out of gas, checked the TPS and it was dead, bought a brand new Japan made one installed it and the problem was there, I took voltage readings at the TPS connector and had 2.5Volts at idle and went to 0 once I started to step on the gas, checked the cables and previous owner inverted the 5 Volts with the Ground, when this was done the plug got busted I knew I needed to replace it, went the local junkyard and got the plug installed the wires just like I found on the net and now I have 0.5volts at idle and almost 5Volts at WOT, the oxygen sensor was no where to be found so I installed a new one also and found the cable cut inside the wire's plastic cover, connected the O2 sensor and got better but the problem I have now seems to be the TPS, if I just try to cruise or the civic parked with hood open and try to start to accelerate a bit almost at 2Volts the engine starts to go up and down real quick revs up and down, checked the ECU for codes and found the codes for TPS, Map and idle air control, since I installed the TPS and the MAP sensor I reset the ECU started the engine again and now the only code that shows Up is the Idle Air control, I doubt the idle air control is the cause of the up and down revs but to be sure a friend lend me the idle air sensor and the problem continues after resetting the ECU riding the civic for a few minutes codes don't show again Check engine light is off, I have checked every thing and can't fix this problem, if you drive the civic stepping hard on the gas the problem doesn't show it shows only when you try to drive normal every day driving, then the problem appears and its really annoying, Thanks in advance Turbo

4drEF

Did you get a new tune when you installed the GM map sensor?
If not, try another stock MAP and see if it clear up.

TurboPirate

Thanks for the response, I did try with another factory MAP sensor and code 3 wish it's MAP Sensor didn't show up, I used the GM 3 Bar because it is the best MAP sensor for turbo applications, the original 1 Bar Map sensor cannot handle the Boost, the Code I am getting with the Original Idle Air Valve is the 14, but code hasn't show up with a borrowed Idle Air Valve, right now I got no codes but have the revs up and down problem..............the MAP sensor voltage readings are, almost 1 Volt at idle and 4 Volts at WOT (no vacuum)
The TPS is set at 0.5Volts at Idle and almost 5 Volts WOT.
The intermittent problem sounds like when it has a bad coolant temp sensor at cold start ups that until it warms up you can hear and see on the tach revs go from 1,000 and 1,500 very fast.......I get that problem but when I step on the gas a bit just to take off, I used a voltmeter installed it inside the civic and when I start pressing the gas between 1.5 and 2 volts I get the revs up and down thing, in any test I get the same problem, at neutral or with the 1st speed selected, 5 speed manual tranny I forgot to mention.

the code 14 is the one that will come back when I return the borrowed Idle air Valve I am using, but the rev problem is there with no codes showing up...........a friend of my nephew said he had the same problem with a civic and it turned out to be the ignition, a bad ignition, if this is true I can't get a borrow ignition for the test so I'll have to replace all the internal components 1 by 1
Turbo

TurboPirate

I just found out this info, it might be useless but I prefer to share it now, it seems this Civic is Canada Made and had a D15 with 2 Injectors with an auto tranny and cruise control, was converted to D16 with 4 Injectors and manual 5 speed tranny.........Turbo

wantboost

iTrader: (2)

you CANNOT use a gm 3bar on a STOCK ECU! This is why the car runs like ****. The stock ecu has a voltage scalar table for map voltage vs. load.... a map sensor that has a larger pressure reading ability also means a different voltage scalar table.... this is why it might idle but doesn't do anything else.... the stock ecu has no idea how much load is actually on the motor and cannot fuel the engine properly

you must use a stock map sensor until you get some sort of TUNEABLE engine management system that will allow you to PROPERLY use an aftermarket map sensor

before you start ******* about and destroying the car, read the FAQs stickied at the top of this forum, MULTIPLE TIMES...

the car and your wallet will thank you, not to mention you avoiding asking the dumbest questions on here lol

wantboost

iTrader: (2)

This is so not true... it's a plastic/polymer bodied sensor that can have issues such as the case cracking under high boost pressures, the remote mounting using a length of hose delays map readings slightly...

it has a pressure limit, 1 bar of the sensor is devoted to reading vacuum, so that leaves 2bar or 29.4psi of boost pressure that the sensor can read. dont know about you but the average honda setup easily has boost pressures that require a larger sensor, like a 3.5 or 4bar map sensor.

TurboPirate

So far the 3 Bar GM sensor has done the job, I simply use a different port from the intake not the port Honda says its the one for MAP, because every time I use the factory Map sensor port the GM MAP sensor gives me 4 Volts at idle when I use a different Port at the intake then I get .33 to almost 1Volt at idle, I have used the GM 3 BAR MAP sensor on almost anything that was brought to me, Toyota Mr.2 Super Charged or turbo, 2.2L 2.5L Chrysler Dodge Daytona Turbo, 77 Chevy Nova EFI converted VW Corrado, and the list goes on and on and the GM Map never failed me, thanks for the advise and info tho.
To bad I got no signature but my signature all over the net is

There are no stupid questions only stupids that don't ask. Author: Anonymous.

wantboost

iTrader: (2)

But a Honda ecu doesn't know how to read the voltage data from the gm map

Try driving the car... It will run like ***.

The voltage the sensor puts out doesn't matter if the ecu has a different map voltage table

I don't know how to explain this any better

Stock ecu has a unique map voltage lookup table that corresponds to the stock map sensors voltage vs load scale

The gm 3bar has a different load vs voltage scale than the stock map sensor and the ecu can't adjust its voltage table to match it. You have to have a chipped ecu and enter the scalars for the 3bar so it scales the map to fit so the car can run properly

It might idle but you won't be able to drive it, especially if you throw a turbo kit on with a stock ecu and fuel system. It'll throw a cel the second the ecu sees boost and put it into limp mode. It'll be even worse with the 3bar. The car probably won't go anywhere above 1/4-1/2 throttle

TurboPirate

I hear you, maybe the ecu is not that stock because after I fix a few details I tried the Honda MAP (borrowed) used the small port at the throttle body where Honda says is the Port for the MAP connect the battery start the car ran a 5 minutes test did the same with the GM MAP sensor using different Port connect the battery ran another 5 minutes car runs the same, trying to cruise in the Mall's parking where you barely press the gas pedal is where I get the up and down revs but any driving in the highway car runs great, maybe I can post some photos..........maybe the voltage readings are affected by the port where the MAP gets connected, mine gave me a few problems until I found the MAP's port at throttle had very little flow I used another near the brake booster port and got the MAP and car working great,,,,,,,,,,,,,,,like I said let me see if I can upload photos of what my nephew bought and got me involved to help him fix, Thanks in advance Turbo

wantboost

iTrader: (2)

port should not affect readings.

the easiest way to explain it is this

at 5 volts (4.99) the stock map sensor reads 10psi(positive pressure)
at 5 volts(4.99) the GM 3bar sees almost 30psi(positive pressure)

at 0 volts both read maximum vacuum. so you can see how from 0-5VDC how each sensor at idle or under load will make the ecu think it's under more load than it actually is based off of the voltage breakpoints in the stock ecu

if it runs like crap with a stock map and throws codes but runs fine with the 3bar then you could have a chipped ecu

do me a favor, take a picture of the sticker/barcode on the ecu case then take the top off and snap some pictures of it. that way I can determine if the ecu is stock or chipped

wantboost

iTrader: (2)

take the case off and take pictures of the circuit board.

if the connector is on your left side while looking inside the ecu then focus on the bottom right corner of the PCB. that's where the ecu would be socketed/chipped

TurboPirate

I'll take those photos tomorrow, let me say what I found today, the ECU says P4, Changed the ignition Cap car got better, changed the coil and finally got rid of the up and down rev when trying to cruise, another thing I found went to high way took advantage no cars around stepped on the gas WOT and tach went all the way to #8 and no cut off, it seems this ecu is not virgin lol I recall the Civics getting cut off at 6,200 RPM so this one went to 8,000 and nothing????
OK Wantboost, I also believe the port should not be a problem but since I don't know what was done to this car before my nephew got his hands on it I checked with Boost Gauge installed inside the Civic and a Vacuum Gauge I have for diagnostics and connected them at the Port in the throttle body where the factory MAP is connected and the readings at both gauges is almost 10, 8-9.1 and needle sweeping very fast, then moved both to a Port near the Brake Booster and the readings are 20-19.8 needle moves very slowly, maybe previous owner or the mechanic did something to the 2 ports that are at the top of the throttle........here photos when I started fixing this Civic, first was fixing the low, high beam, parking lights and signal lights of the front passenger side, wires were burned previous owner said it was not an electrical problem that was fire caused by a sprayed gasoline from the gas filter, all lights working properly........Thanks in advance Turbo

http://i455.photobucket.com/albums/q...ps1722750f.jpg
http://i455.photobucket.com/albums/q...ps1f644f44.jpg
http://i455.photobucket.com/albums/q...psa692c9d8.jpg

KRITIK

after all of this have u ever checked for a vacuum leak, or check to see if the coolant line to the manifold is on

TurboPirate

Hello, Glad you asked, Before registering at this forum I did notice some coolant lines eliminated, example and bare with me here trying to translate as best possible, the throttle body has 2 lines where the coolant travels both of those metal lines have no hose connected to them, the Idle Air Valve has a small house in a U shape it seems the Idle Air Valve had some coolant traveling and with this U shape no coolant gets to that sensor, I did fix a small coolant hose connected near the #4 piston runner at the intake manifold, since the Civic has no up and down revs my nephew is using it I asked him to bring it home to take the photos you guys solicited, inside the ECU and I will really like to take photos of all the coolant hoses I mention are no longer connected and the ones connected, I really hope he brings it home today before night, yes I did check vacuum leaks the only one I found was the melted factory MAP sensor, at the photo where the GM MAP is installed below you can see the black melted plastic.
thanks to you all for your time Turbo

wantboost

iTrader: (2)

the ones on the throttle body are useless. lots of people delete/loop them as they are only to warm up the throttle body (and thus the air passing through it) to help warm up the car on cold days

TheShodan

iTrader: (14)

or use them to cool DBB turbos.

TurboPirate

And the ones at idle air valve are needed? I can see the logic of delete those but the idle? Thanks turbo

lsturbo33

You can delete the ones for the idle air control valve. It cleans up the bay. Also, I wouldn't be revving that single cam to 8,000rpm.

wantboost

iTrader: (2)

that is valve float city, either that or having a pump fail

TurboPirate

Thanks for the advise but I rev the engine because I wanted to see where the ecu start the rev limiting and it doesn't. Fuel pump gives me a constant 35psi at idle and 70 at no vacum or boost found someone that can dump the info in the ecu's chip and tell me what was written, as soon as I get this info will share with you for opinions, thanks turbo

TheShodan

iTrader: (14)

C'mon.. that's not true either. The GM 3Bar never had those kinds of issues on what we used in the TAs or SS Camaros from years ago. those things are MUCH more robust than anything that Hondata or Omni ever put out.

GM 3 Bar map sensors have been used for 20 years with great success. However, they cannot be used as a direct replacement for the OEM sensor on a stock ECU. The computer like Hondata system would have to be able to read the correct voltage in order for it to work properly, that's all.

The only reason why the Hondata and Omni 4 bar maps were used for for those going over 29psi of boost pressure AND the fact that it fits in the stock OEM MAP location. Nothing more, nothing less.

TurboPirate

Good info, but nephew exchanged d civic for a Honda del Sol, the stock ecu of the civic was not stock, it was chip for turbo and for reading gm map sensor, this I found out on Saturday before nephew exchanged the civic we went to an auto show and the person who made the civic turbo came to us and told what he did to it and when he sold it he told the guy to install a 3 bar gm sensor but instead he just started to eliminate sensors.......