Page 4 of 5 FirstFirst ... 2345 LastLast
Results 31 to 40 of 43

Thread: 2000+rpm hesitation... fuel pump?

  1. #31
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    Quote Originally Posted by umran9 View Post
    Don't the Bosch Coils need to be machined etc so they can fit the Nsx?
    as far as i can tell, the 'machining' is tweaking the rubber to fit in the cylinder sleeve..... the first pic on 'Operators' thread shows the difference, and he only mentions having a 'rubber grinder'... to anyone this side of the atlantic that equates to a belt sander



    original thread text....
    These coils dont need adapters. I have macined rubber seals to fit OEM covers . Perfect fit . Thats why we dont sell harnes without coils > Coils are special mede for NSX
    ie. rubber tweaked....

    tho looking at it, they are thin seal rings, but we'll see once i get these to hand... i'm also lead to believe that the coil packs clip onto the hex of the sparkplug, giving a better/secondary earth path, so the top just has to make a snug seal


    and it's annoying that the pic i chose previously doesn't show up in the thread, but shows up on mine because i looked at it on their site!!!! grrrr
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  2. Default

    Good Stuff il be keeping an eye on this thread...

  3. #33
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    receive this today from Lithuania!

    WP_20180117_18_23_43_Pro by jon sutherland, on Flickr

    WP_20180117_18_24_02_Pro by jon sutherland, on Flickr

    the one on the right (slightly lighter) is my possibly defective original, the one on the left is the second hand, slightly used 1998 honda Legend one!

    it's on, and the car fires up... a good sign!!! ... I shall take it for a longer test run when i get a break from the major overhaul of the hotrod!... to be continued
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  4. #34
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    early indications are positive, the engine smoother with the replacement ignitor.....

    it'll get another long run out next weekend.....
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  5. #35
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    so... the weekend is here, i get a break from re-engineering the hotrod to do some NSX investigation.....

    the lithuanian ignitor was working ok, then came a stutter... so in went the USA ignitor..... that didn't cure it either, so the lithuanian one was put back in as it seems the best of the 3....

    next port of call was to replace the coilpacks, as i have a set that was a deal on ebay a while back (£92 the set of 6!!).... being aftermarket they need a little fettle to fit with the coil covers, so a quick tweak and in they went...

    i also rigged up the fuel pressure gauge where i could see it, and with Sarah doing the fuel pressure monitoring, off we went, the same route that had shown the problems previously.....

    i'd narrowed it down to a 2000-2500prm misfire, on an incline (so increased load) but as soon as you got over 3000rpm it seemed fine pulling well and no misfire!!!!

    we followed the same route, car already at working temp,Sarah calling out the fuel pressures......

    36psi no load/overrun
    42psi light load
    50psi when the pump gets full power with the resistor shorted out

    the pressures were stable and predictable throughout...

    fingers crossed, the problem didn't re-occur during this drive, so maybe we've got there, but i'll keep an eye on it!
    Last edited by britlude; 28-01-2018 at 03:39 PM.
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  6. #36
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    well, at the weekend we took the car down to canterbury, about 38 miles there, so i had a good run on the motorway, and fingers crossed it behaved itself, didn't miss a beat, hurrah!

    the engine idle has been been getting a little higher over the last couple of weeks, and when we got back last night it was hunting between 1500-2000, so investigation needed......

    the internet search puts the favorite as the cold idle air controller being at fault.... if it doesn't seal properly when the car is hot, then the air bleeding past it raises the revs. if sufficient revs are seen by the ecu, with no throttle, it decides you're obviously coasting for the engine revs to be high with no throttle input, so it cuts the injectors, no fuel, no revs, so the engine speed drops... once it reaches the idle level expected by the ecu it then re-fires the injectors and tries to control a stable idle..... but the bleeding air means the revs raise too far, and the cycle continues...... or something like that anyway!

    first i removed the filter cover and the rubber connecting hose to the throttle body, just to check there was nothing holding the throttle butterfly open, and that the throttle cable was slack with no pedal depression, but that's all fine.

    on the generic honda throttle body internet threads, the fast idle control and electronic idle control had separate filter-side ports feeding them, and the video examples showed simple fault finding by covering each port in turn to isolate each air valve circuit. unfortunately on my early JDM throttle body i only have one common port however covering that did stall the engine, so i know the main throttle butterfly has a good seal! it also shows there's no air leaks elsewhere!!! mines that early i don't even have the baseline throttle adjusting screw everyone else has!!!!

    no throttle baseline adjusting screw above the butterfly
    WP_20171230_12_30_18_Pro by jon sutherland, on Flickr

    anyway, taking the endplate off the cold idle control valve, and starting the engine, i could raise the revs by depressing the valve plunger as expected, but couldn't seal it to confirm how much it was bleeding air in the hot/closed position. spraying carb cleaner onto the valve plunger did alter the engine speed, so there was some air getting thru......

    so i unscrewed the valve seat, and removed the plunger to give it all a clean.....
    dirty
    WP_20180205_16_29_20_Pro by jon sutherland, on Flickr

    clean
    WP_20180205_16_31_29_Pro by jon sutherland, on Flickr

    the actual seal is right at the base of the brass bit, where it's still brown, so it got a little more cleaning just to make sure! blowing thru the closed valve showed it pretty much sealed! i fitted a new o-ring on it too, as the one on it was like a solid strip of plastic!

    clock fuse out and back in to reset the ecu, the car was started and let to sort itself out.... fingers crossed, it now idles a lot slower, 850-900 on the dash when hot... i just need it all to cool down so i can see what the water temp idle control valve decides the cold idle is, and adjust to suit!

    right, now back to the hotrod......
    Last edited by britlude; 05-02-2018 at 06:34 PM.
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  7. #37
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    another quick fiddle as it had all cooled down.... revs still seem a little high when i re-tried it.... re-checked cold fast idle valve... that seems closed with the carb cleaner test, so i'll have to back that off once i i try it properly cold!

    took it around the block to warm it up, and when i returned it did have a little hunt so i disconnected the ecu controlled valve just to see what the actual baseline is, and it settled quite happily at 650rpm on the dash.... so the ecu is controlling the revs...

    reconnected everything, and started up, it fired up at about 1800rpm then slowly dropped to 900... more monitoring!
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  8. #38
    Join Date
    Sep 2008
    Location
    Bucks, UK
    Posts
    2,231
    Blog Entries
    2683

    Default

    Next time when the engine is cold, please start the engine and monitor the rpm after 10sec.
    Should be around 1,500rpm.
    If the fast idle valve is faulty, normally you would see really high fast idle rpm such as 2,500rpm.
    It seems this is not the case with your NSX.

    Once fully warmed up, please check the resistance of the TW sensor.

    On our NSX, there are 3 coolant temperature related sensors.

    The Rad Fan control sensor, Coolant temperature sender unit and the TW sensor.
    Rad fan one is on the thermostat cover and it controls the on/off as well as mode of the radiator fan.

    The sender unit is purely for the dash gauge temperature display/dial and it sits below/near the TH body.
    It's the vertical body sticking out of the water passage.

    The TW sensor is the one for ECU EFI control.
    It's located on the front bank below the front VTEC spool valve.
    The connector body is covered in black sleeve.
    It's thermistor and when fully warmed up (about 83degC at idle rpm), it would be about 300ohm (200 – 400ohm).

    There were several TW sensor failures reported in the past resulting in high idle rpm even after fully warmed up although it would be much higher than your case (900rpm).

    By the way, the rpm gauge could be showing the wrong reading than the real state.

    There is a small rpm signal terminal at the INJ resistor block.
    If you can measure the frequency, it's 6cyl engine so 3 pulse per 360deg.
    At 800rpm, it is 40Hz so you can compare it against the rpm/tacho gauge.
    Target idle rpm for AT model in Neutral is bit lower than the MT and it's 750rpm so 37.5Hz.

    Kaz
    Last edited by Kaz-kzukNA1; 05-02-2018 at 11:39 PM. Reason: font

  9. #39
    Join Date
    Sep 2005
    Location
    Garden of England
    Posts
    2,772

    Default

    thanks once again for your input Kaz, much appreciated!

    i did realise, after i'd logged off, that the 650rpm baseline with the ecu controlled air valve disconnected shows that particular valve isn't letting air through, and it IS the ecu lifting the revs.

    starting up cold, the revs on the dash are 1650rpm. i'll have to wait until the weekend to see if the temp controlled valve is dropping, letting some air past when cold! after driving home from work and parking up, the revs are still about 1500rpm, front radiator getting warm..... letting it run on the drive the revs eventually, like 5 mins later, dropped to around 1000rpm....

    i did investigate the water temp sender that the ecu uses, but they didn't make its position easy to access! you can see it, but getting to it is another matter!!! after almost burning my hand i gave up trying to get the cable connection clip off.... i have ordered a replacement anyway, and shall swap it in at the weekend when i have more time, light and a cold engine!!! no doubt with snow thrown in!

    i've got to track down the rpm signal terminal at the INJ resistor block....... but its dark and cold out there now!!!
    aka Jonathan!!

    '92 charlotte green auto.... as a daily
    '37 Ford Y street rod......... something for the weekend!

    ...... if a photobucket pic is foggy, click it, and it'll take you to the clear version, yes, it's a clicking faff....

  10. #40
    Join Date
    Sep 2008
    Location
    Bucks, UK
    Posts
    2,231
    Blog Entries
    2683

    Default

    TW sensor is for EFI control so you can measure the resistance at the ECU connector.
    For your NSX spec, it's located on the 22pin connector.
    The VBu (Backup memory voltage) is also on the same connector so if you disconnect it, it will reset the ECU, same as removing the 7.5A CLOCK fuse.

    Not required but to be on the safe side, just disconnect battery GND terminal, pull out 22pin ECU connector and measure the resistance between the terminal 'TW' and one of the 'SG (Sensor GND)'.

    As your idle rpm was 1,500 – 2,000rpm at one point but now it seems to be much better at 900rpm, very unlikely to be the TW sensor but at least, by checking it, you can eliminate one possibility.




    The rpm measuring port is the tiny 2pin connector held by a small rubber cap near the INJ resistor.
    The bracket is actually on the Eng bay cooling fan unit.
    I thought about taking photo of my NSX but I already removed the cooling fan long time ago so not relevant and I believe your NSX is AT spec so you should still have the cooling fan in the eng bay.
    Although it's tiny 2pin connector, there is only one wire used.


    Kaz



Page 4 of 5 FirstFirst ... 2345 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •