Broke exhaust valves and 4ktq's (fwd)

[Jorgen Karlsson]

> >  We finally got Seans car taken apart tonight and yet anouther broken
> > exhaust valve.  This time #3.  This makes 3 confirmed valve breakages in
> >  4ktq's that I know about.  All seem to happen the same way.  2
> of the 3 had
> >  good intercoolers 1 even being a stock 5kt intercooler.

I wouldn't say that a stock 5kt intercooler is a good intercooler, have you
measured the intake temperature?

> >  So it begs the question why?  They were all running less than 15lbs of
> > boost
> >  which 5kt's dont seem to have a problem with.  The other 2
> were in cylinder
> >  4.  They all apear to have broke near the base of the stem.  So does
> > anybody
> >  know if the 10v turbo's in 5kt's have had problems with valve
> breakage?  If
> >  they do I have never heard of it.
> >
> >  I do have a couple of theory's though.

> >  1:  Fuel volume.  One of the things they all have in common is they are
> >  using stock 4k fuel pumps.  Is it possible that these fuel pumps arent
> > supplying the volume to keep things cool.

I don't think that they are not good enough.

> >  Sean did have an EGT guage but
> >  sadly was not watching it when it gave up.  There does not seem to be
> >  problems with knocking and it does not show up on a oxy sensor
> reading as
> >  lean.  Orin seems to think that they typically do not show
> enough detail to
> >  be worth much valuable information.

I aggree with Orin there, lambda meters are very sensitive to exhaust
temperature. On lean mixtures they show leaner then it really is. But on
rich mixtures they show a richer mixture then it really is. I would say that
they are useless as a warning system.

I have noticed that the exhaust temperature increases very rapidly after the
engine starts to run lean. A lot faster then for example the VDO EGT meters
can keep up with. VDO's fast probe combined with an Autometer Gauge is a
very fast combination. I can't understand why VDO's meter is three times
slower then the autometer gauge when using the same probe...But that's the
way it is. I have noticed that when I hit the limit the exhaust temperature
increases with more then 100oF/second and that measurement is after the
turbo. My highest note is around 1850oF, when I checked the meter three
seconds earlier it was stable at 1500.

> >  2: Transmission.  With the stock 4kq transmission we are
> spending alot of
> >  time right near redline.  This is true on PIR, Bremerton, and SIR.
> >  Could this be the problem.  I have honestly not been watching 5ktq's
> >  tachometers on the track to see where they are running but I
> would guess
> >  they could run a bit lower in the rev band for 2 reasons.
> Longer gearing
> >  and better aerodynamics which helps them pull a little better at higher
> >  speeds without stressing the engine as hard.

The highest fuel demand are at high rpm, I got problems above 5800rpm or
something like that.

I am pretty sure that the CIS can't supply fuel for more then 220hp even
under ideal circumstances. That would be around 13psi with stock exhaust and
air box. With a freeflowing exhaust and cool air 10psi would be more like
it. Limits can be stretched quite a bit if you allow the engine to lean out,
but if you want safety you must get all the fuel the engine demands.

My theory is that every time you use the last 1000rpms of the range you get
somewhere around 1700-2200oF exhaust temp until you get into the next gear.
Then the exhaust temp drops to 1500-1700oF again. This is before the turbo.
This cycles the exhaust vavles to death.

You could take a look at water injection, that should take care of the
problem. But proper fuel injection is a better alternative. Or you could
check http://members.tripod.com/turboracer/kits/, Joe has methanol injection
kits that could supply you with internal coolant and the additional fuel you
need.

Jorgen Karlsson
Gothenburg, Sweden.