2012-07-15

Replacing the fuel lines (1)

As is obvious from the photos, the rusty fuel lines are my top concern on this car. They look as if they're about to disintegrate any moment, and although there's no visible leaks, sometimes it smells of gasoline inside the car. Part of the feeder line had already been replaced with a hose, probably when the filter was changed.

While replacing the lines, I'm also trying to use E85 (ethanol fuel) compatible parts, so that it's possible to run on cheaper (or, just existent...) fuel in the future. Stainless steel, and hoses and O-rings specified for gasoline and ethanol should be sufficient.

Pipes

The pipe layout looks like this:

Pontiac Firebird underbody fuel lines


There are 3 underbody fuel pipes:
  • Fuel feed to the TBI unit (outside diameter 9,5 mm)
  • Fuel return from the TBI unit (outside diameter 8 mm)
  • Fuel return from the evaporative emissions canister (outside diameter 8 mm)

For the feed pipe, I've purchased 4 m of 10mm outside diameter steel pipe w/ 1,5 mm wall thickness.

For the return pipes, two 3m 330 cm pieces of 8mm outside diameter steel pipe w/ 1,0 mm wall thickness.

Total cost for the pipes was € 89,35.

IMAG1086

Fuel Level Sending Unit


In the tank end, the pipes form part of a device called the Fuel Level Sending Unit (or Assembly) that also holds the fuel pump and fuel level sensor. The current one is in as terrible condition as the underbody pipes, so I've acquired a new one.  The unit is a Spectra Premium FG20A from RockAuto, cost € 100.46 ($127). Unfortunately I don't know what material this unit is, i.e. whether it's E85-compatible. At least it's new.


New Fuel Level Sending Assembly
Fuel Level Sending Unit
New Fuel Level Sending Assembly
Fuel Level Sending Unit - hose connectors


Counting from the left:
  1. Vent (breather) pipe (?)
  2. Evaporative Emission System  line (?)
  3. TBI return line (?)
  4. Fuel feed line


IMAG1040
This is how the current sending unit's pipe ends look like.

The hose in the leftmost pipe connects to this mushroom-like device.

IMAG1060

2012-07-13

Fixing an SJ-269SE LED Daytime light


 
This is how it should look like (thepontiactransampage.com)
Normally, the Firebird's front turn signals double as "parking" / daytime running lights, but on this side of the Atlantic, these lights must be white. This rule is enforced quite randomly though; this time the car's previous owner got the short straw, and had to install these ill-suited aftermarket DRLs.

Of course, one of the aftermarket lamps had to fail. As Mr. Previous Owner wasn't one to waste his time on storing receipts or other documentation, warranty replacement wasn't an option either. The device itself is insanely simple (apparently a power regulator, some resistors and 5 LEDs), but (on account of ECE standards compliance?), they tend to cost around 50-100 eur.

Hence, the need to fix the MyCARR SJ-269SE, in as ECE-R7 compliant a way as possible, and to document it here. The device's conformance inspection report says it should emit 30 cd of light with about 20-30 degree horiz./vert/. angle, i.e. 6 cd per LED.

I tested three different LEDs:
Of these three, the latter one is a nearly identical match for intensity, emission angle and color tone. The flat ones in the pic are the unit's original LEDs, the 33BCWK5A is the tall one in the middle.






I later found what looks very much like the original LED, http://uk.rs-online.com/web/p/visible-led/7134876/

    Three of the original LEDs were still working, and as they fit the reflector better than the common bulb type, I kept those in the center.

    They're connected in series with current limiting resitors, in two parallel groups, like this


    (+)-----------+--------------\
    14V           |              |
                R1 =           R2 =
                52 ohm         270 ohm
                  |              |
                 _|_            _|_              
                 \ /            \ /
                 ---            ---
                  |              |
                 _|_            _|_              
                 \ /            \ /
                 ---            ---
                  |              |
                 _|_             |
                 \ /             |
                 ---             |
                  |              |
    (-)-----------+--------------/

                orig.           new
                LEDs            LEDs

          14V - (3 * 3,8V)
    R1 = ------------------  = 52 ohm
               0,05 A


          14V - (2 * 3,7V)
    R2 = ------------------  = 246,66.. ohm
               0,03 A

    The LEDs' voltages and max. currents (3,8 V / 50 mA and 3,7 V / 30 mA) come from the datasheets and seller's specs.

    After experimenting, I ended up with 270 ohm for R2 to get similar brightness from both LED types.

    There's more on LED circuits at http://www.kpsec.freeuk.com/components/led.htm



    DSCN9606

    DSCN9607


    I've yet to test how it looks like next to the original light.

    2012-05-13

    More underbody horrors




    The hatch door leaks.



    Left rear wheel

    Left rear wheel

    Right rear wheel












    Part of the steering, I believe 


    Examining the Fuel Lines

    (backdated post, 2012-05-13)


    While changing the coolant, I had a chance to see what condition the underbody, chassis and various pipes are in. 

    Of course, there's rust everywhere and the fuel and brake lines look like they're about to break any time. Someone has even replaced part of the fuel feeder line with a rubber hose. 

    This means I won't be filling up with E85/Gasohol anytime soon, but OTOH, I'll have a reason to replace the fuel lines properly instead of taking stupid risks.





    Fuel lines going to the Fuel Level Sender Unit in the tank






    2012-05-01

    Heat problems continue

    (backdated from 2012-05-01)


    After driving around in the city for a few hours yesterday, the coolant started _boiling_ in the expansion tank. It was  intensive enough to - apparently - have blown off the expansion tank's cap. The boiling continued for about 15-20 min after shutting off the engine, bubbling out perhaps a litre or so of reddish coolant.

    Even though the coolant was supposed to have been changed in April, it still looks like a mixture of Soylent Green and red rust particles.