Cylinder Barrel Coatings, Alloy Castings -- Heat Build Up??

cliffa said:

This is an interesting thread to which I have a couple of questions:

Quote from @mdt-son "In reality, most of the heat is generated in the upper part of the bore, or transferred from the cylinder head. Evidently, the heat conduction of cast iron is low"

• Presumably a copper head gasket better in terms of thermal transfer in each direction than it's composite equivalent. In a real world everyday bike could this be the reason for any common issues we see such as blown head gaskets, soft heads or loose guides etc.?
• How is the laminar layer affected when you have a hot surface dissipating heat but in reasonably close proximity to another trying to do the same? Is it better to have less but larger fins for example?

Click to expand...

Copper head gaskets are superior heat conductors and will lower the cylinder head temperature. Remember heat will flow from hot to cold domains (where domain can be a solid, a liquid, or a gas).
Copper also offers higher tensile strength and ductility ("malleable property").
"Soft" heads has to do with the material stiffness parameter K. It is an inherent material property, albeit dependent of the temperature. If the cylinder head gets very hot and K drops with T in an unfavourable way, the head will be prone to deformations and contact pressure at the gasket may drop, which could result in leaks or a blown gasket.

Knowing the Norton twin has "hot heads" generally, using a composite gasket is not a good idea, IMHO (unless you live in Alaska or Siberia).

I think the NHT will benefit greatly by using an aluminum cylinder barrel, which has a much higher heat conduction rate than cast iron (aluminum: 237 W/m/K ; cast iron: 55 W/m/K).

Whether there will be a lesser problem with valve guides when using a copper head gasket is hard to say. It depends on ambient conditions, amount of gas burnt, oil flow for cooling, ignition timing, gas/air mixture, how well the engine is cooled by air (ducting, shielding, etc)., and other factors. It's very hard to make general assessments. Discussions and specific actions need ta assume engine is in top nick, so that's where your hunting starts.

- Knut

"Knowing the Norton twin has "hot heads" generally, using a composite gasket is not a good idea, IMHO (unless you live in Alaska or Siberia)."



I'm not sure about this statement do you mean for racing?
I've never noticed any heat difference on any road going Norton between using copper or composite head gaskets ?

baz said:

"Knowing the Norton twin has "hot heads" generally, using a composite gasket is not a good idea, IMHO (unless you live in Alaska or Siberia)."

I'm not sure about this statement do you mean for racing?
I've never noticed any heat difference on any road going Norton between using copper or composite head gaskets ?

Click to expand...

I mean any running condition which creates a high thermal load and poor heat dissipation (low speed, high ambient temperature), typically encountered at steep hills as in the Alps. Racing may not be the worst case actually.

Making a qualified judgement requires use of a termocouple and continuous readout.

- Knut

mdt-son said:

I mean any running condition which creates a high thermal load and poor heat dissipation (low speed, high ambient temperature), typically encountered at steep hills as in the Alps. Racing may not be the worst case actually.

Making a qualified judgement requires use of a termocouple and continuous readout.

- Knut

Click to expand...

Agreed. I know of a number of overheating events caused by being held on the line for too long at the races; i.e., little to no work being done, but zero airflow.
IMHO radiant heat dissipation is almost irrelevant when compared to airflow over the body.

Slick and Community,

We seem to have a similar background, and I concur with your phenomenological explanation (#18). That is, *if* the surface had been very smooth, as for an airfoil. However, fins are very rough, and the concept of two adjacent laminar layers doesn't apply.
Instead, we have a flow field as found in rough iron or concrete pipes.

We seem to agree the incoming flow is fully turbulent. Out of curiosity, I computed the Reynolds number. The disturbance caused by the forks alone will be in the order of 1.0E5 @ 100 km/hr (62 m/hr) @ 15 degrees C, and additional turbulence caused by mudguard, wheel and frame tubes will increase this number further. I will assume that Re-number is in the order of 1.0E6 for this discussion. In fluid mechanics, a Reynold number this high is associated with turbulent flow.

Next, we recognize the heat transfer problem is inherently coupled to the fluid (air) flow field along and transverse to the fins. As you mention, we have a very small regime next to the fins governed by heat conduction (the "no slip" boundary layer) and a comparatively very large regime outside of this governed by turbulent flow, this includes the so-called buffer layer. The reason the buffer layer is turbulent is due to the friction caused by the rough peaks with heights exceeding the thickness of the laminar layer, and probably the corresponding buffer layer thickness of a perfect even surface as well. The corresponding friction factor for flow between two fin surfaces is in the order of f = 1/10. For such a high friction factor, the boundary layer flow structure is predominantly turbulent (starting at Re = 1.0E4) and so will the heat transfer be. Citing the experiment referred to in #17, it is clear heat convection by far exceeds heat conduction in the case of turbulent flow along a rough surface.

I would like to maintain that the exact movement of air due to swirls (large and progressively smaller Eddies) within a rough surface is NOT well understood actually. By "well understood" I mean to a degree where someone could present a coupled 3D flow and heat simulation, thus research is confined to experimental studies.

A phenomenologic description of a related but still different problem doesn't really answer the question about the near wall flow field and the effective heat conduction rates off a rough m/c cylinder fin.

Here is a nice presentation which explains laminar and turbulent flow in rough pipes (no heat conduction involved, but it's easy to understand calculation of the effective heat transfer is not a walk in the park).




"This makes analysis of turbulent flow very complex, to the point that it is probably the most significant problem facing the field of Fluid Mechanics".

- Knut