[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: Heat soak

In a message dated 97-06-17 18:54:02 EDT, DeWitt writes:

<< Scott, I'm not clear on the physical process(es) you call
 >"heat soak" and "over spin" on the turbo hot side. Would
> you please be kind enough to elaborate?  Is this some
 >sort of thermal transient or is it more in the domain
> of acoustics?
>>>>Scott writes here:  BOTH
>If I had a better picture of this concept,
> I would get more from your recent turbo post.Thanks,
THEN Ed Kellock writes:
>>I am very interested to hear more about the concept of short term boost
>>in relation to the 20vt motor. 

>>My 91 200qw has exhibited this behavior ever since I got the IA Stage 3
>>mod.  The best it's ever been was full boost through a gear and a half or
>>so and then a noticeable step down until the next gear.  It's kind of and
>>on/off thing.  Is this normal?  Is there any way to eliminate it?  Is it
>>to the IA mod?  I just accepted it but it seems to be becoming more

The latter answering the former?  Yes.  Heat soak is a common problem in a
bone stock k24, so no surprise a tweeked computer will kick that thing into a
heat gun.
HEAT SOAK:  Heat from the hot side of the turbo transfers to the cold side of
the turbo (or the head itself can heatsoak as well).  This increases the
Charge Air temps from the Turbo/supercharger (Charge Air Temp = Ambient Temp
+ Cold Side Temp + Compressed Air Temp, we normally assume Cold Side Temp =
Ambient Temp). When you heat soak to the cold side, you have effectively
created a turbo operating in a heater, this creates lower density > higher
charge air temps pre/post IC. This is also a natural phenomenon, when cold
and hot side temp differentials get large enough.   Water and oil in the
bearing housing absorb this in normal boost and driving conditions.
OVERSPIN:  Turbine spins faster than the design; of the MAP, cold side wing,
or hot side wing.  Also referred to as "stall" (hot side) or "vortex" (cold
side). This phenomenon sometimes creates a higher Pressure Ratio, but not
flow.  This easily creates Heat Soak.

K24 Turbochargers are designed to be fast spinning turbos at a low exhaust
velocity.  Increasing the velocity via computer mods, increases the potential
for Overspin, then the Heat Soak (HS) raises it's ugly head.  The k24 is one
of the few turbos I've seen that can do this in a matter of seconds.  That is
why k24 turbos don't do well with mods.  A larger Radiator and/or oil cooler
can help, but given the exhaust velocities we are talking about, not sure you
can get enough heat exchange in the small turbo bearing housing to avoid the
HS problem. 

Heat soak also happens when either side of the turbo goes "supersonic", this
is what creates the "vortex" or specifically, when the turbo cold side is
spinning beyond the speed of sound.  At this point, any increase in speed
creates only heat, not flow, and density of charge air decreases
exponentially.  Given the peak efficiency of a k24 is at 2000rpm on a 7000rpm
motor, Overspin has a really high potential.  Why most box tuners don't sell
'stage III' without the larger turbo associated with it.  Peak power on a
20vt with a stock k24 is 275hp prolly MAX, and that is peak HP, not constant.
 Get a larger turbo, with the same mod, and the HP can be into the 350+++
range, 400+ going to the larger than 26 series turbos, and more constant than
the 24 will give.  

Turbo "stall" can occur when the exhaust velocity is so high that the wing
design of the turbine effectively changes the high to low pressure
relationship.  IOW, the wing spins by high pressure underneath it, and low
pressure above it, spinning the turbo in a given direction.  When the exhaust
velocity gets high enough, you get high pressure into the low pressure area
of the wing "below".  This actually "stalls" the turbo, figuratively
speaking, more an aircraft wing term than a stopped turbo, since the turbo is
spinning at Max velocity, any more exhaust velocity will only create heat,
not boost OR flow.  

"Stall" on the cold side is the more common interpretation of the phenomenon.
 This is when you slam the throttle shut while at full boost.  The turbo is
spinning at max velocity, and a pressure spike wave goes from the t-body back
to the turbo wheel cold side.  This actually can "stall" or slow the turbo so
fast that the load on the bearing and wheel gets so high that the shaft (k24)
or wheel breaks from this load.  Not a good thing.   And why all 20vt k24
applications use a Bypass Valve.  This allows the turbo to freewheel,
relieving any load on the turbo from pressure spikes (and they are
significant folks, I've measured upwards of 50psi).  The lack of the valve
can create IC boom, lots of stage II 10vt owners are familiar with this
phenomenon.  Really, tho a PITA, it's better than picking pieces of turbo
parts from your motor.

Oil coolers will help, so will larger radiators, but only to a point.  Once
supersonic, a turbo creates heat, that's all.  Eventually, this will find
it's way to something not good.  Like the cold side of the turbo or the head
(combustion temps rise).  Larger IC will help to a point too, but really just
delaying the reality of heat.  Once the cold side of the turbo gets hot, the
increase in charge air temps over ambient increases, and air to air IC only
is efficient to increases over ambient.

Want a real example of turbo heat.  Take a look at your turbo after a hot run
some night, raise the hood...  See it glow?  That heat HAS to go somewhere.
 Sometimes faster than you think.

Again, formulas and nerdy stuff and terms deleted for concept.  


'87 5ktq/w
'84 Urq