Every one has their preference, mostly chosen from the material expansion properties (cracking), what they have for tools to work with the material, corrosion properties, what they heard on the web, etc.

This is more about the thermal properties which relate to HP loss and spool time.

Mild Steel thermal conductivity: 50 W/(m*K)
Cast Iron thermal conductivity: 43 W/(m*K)
409 Stainless thermal conductivity: 25 W/(m*K)

Modeling a turbo header/manifold as a 2.25" diameter by 1.5 foot long pipe, estimating the exhaust temperature at 1200 *F (650 C) and engine bay temperature at 200 *F (93 C).
Stainless and mild steel manifold is .065" (16 gauge) thick
Cast iron manifold is 3 times thicker (3*.065") = .200"

The power loss due to heat conduction is:
conductivity * temp_diff * pipe_surface_area / pipe_thickness

Mild Steel thermal conductivity: 50 W/(m*K) * (650-93) * (3.14*2.25*25.4*10^-3)*(1.5*12*25.4*10^-3) / (.065*25.4*10^-3) = 135,982 kW
Cast Iron thermal conductivity: 43 W/(m*K) (650-93) * (3.14*2.25*25.4*10^-3)*(1.5*12*25.4*10^-3) / (3*.065*25.4*10^-3) = 39,981 kW
409 Stainless thermal conductivity: 25 W/(m*K) (650-93) * (3.14*2.25*25.4*10^-3)*(1.5*12*25.4*10^-3) / (.065*25.4*10^-3) = 65,271 kW

In terms of HP lost due to heat convection:

Mild Steel thermal conductivity: 135,982 kW / 746 = 182 HP
Cast Iron thermal conductivity: 39,981 kW / 746 = 52 HP
409 Stainless thermal conductivity: 65,271 kW / 746 = 87 HP

What this says is that cast iron wins in terms of max HP from the engine and quickest spool assuming flow characteristics are the same for all pipes. Stainless 409 is a good choice in terms of price, HP loss, and corrosion resistance. Mild steel wastes a lot of power. A whopping 182 hp loss. That is 130hp lost comparted to the cast iron.

The cast beats the stainless only because of the thickness. If the stainless was just as thick it would be better than the cast iron.

 

Very interesting! Curious, where did you find the calculation for the power loss due to heat. I'm taking some classes (Statics and strengths) and am into that stuff. Again, Very cool!

-Dan

 

Seems interesting but not sure i buy it lol

 

I know his kW to hp is right. Im just wondering about/where the other figures/standards came from.

 

Orr,
I would love to see some calcs showing the calcs are incorrect. I know you are more than capable of verifying.

Z28,
The calcs use the conduction heat equation at static flow.

Why are the calcs posted? I have a project where doing an engine mount turbo is tight and a remote mount would be much easier. I wanted to see the benefits of using stainless pipe to the turbo if I did a remote mount. Due to the stainless having a 50/25 = 2x better heat loss, the remote mount will get 409 stainless feeding the hot side. The 409 is relatively cheap and a bottle of 98/2 gas and some .035" 409 wire in the MIG will weld it.

Why not use 304 stainless which has a 25/16 = 1.56x better heat loss than 409 and 50/16 = 3.125x better than mild steel? The reason is cost and the expansion rate. 409 expansion rate is close to mild steel.

More HP stuff:
Suppose a built turbo engine produces 1000 bhp with a bsfc of .55. That is, it burns .55 lbs of fuel per hp per hour. In an hour, it burns 1000*.55 = 550 lbs of fuel which is 550/5.92 = 93 gallons of fuel. A pound of fuel has 19K BTUS in it. Converting to total hp generated by the fuel this is 550*19000 / 2545 = 10450000 / 2545 = 4,100 HP.

The engine burned 4,100 hp worth of fuel and produced 1000 hp at the crank. Where did the rest of the hp go? The 3,100 hp went into heating the exhaust, air, and coolant. Is it surprising to see that the HP loss in the headers stated in the first post?

 

What I am getting at is that I would like to see the exhaust manifold loss heat (HP) go through the turbo instead of being lost due to conduction before the turbo. This heat (HP) will drive the turbo (more boost) and exit out the tailpipe where it is desired.

 

care to do those smae figures with .065wall 316L stainless since thats the most affordable good stainless to use for a turbo system??

in all honesty 321 stainless is what u should use but most cant afford it and 316L is much cheaper and works just as good

 

Yeah your standard internal combustion engine is only 20-30% efficient. Most of the thermal energy provided by the combustion process gets wasted as expelled heat. Some of the fuel mass used in combustion is just to keep chambers cool

Sure the calculations may give some indication of whats lost as far as heat energy goes, but when talking about the grand scheme of things, most hp lost is going into the coolant system. Or countered by overcoming frictional losses of bearingns and rings. Exhaust is miniscule. Turbine drive energy of the system is not entirely related to heat of exhaust stream either, you still can have good gas energy/velocity with cooler stream as long as there is some pressure differential and pipes are sized right for higher velocity.

But theory is somewhat correct, you should see higher exhaust temps with cast iron vs steels. Whether or not holding heat in stream longer is helpful or not depends on the system. From my experiments, my car spools via certain rpm rather than temp of exhaust stream. Wrapping mild steel didnt help much but may have kept engine bay parts cooler. I've run my car hot and cold and spool up seems based on exhaust gas mass flow via certain rpm. Alittle hotter gas may have a hundred rpm difference but its hard to notice in my setup.

 

Some good articles on stainless vs mild steel:

409 vs mild steel in high temp applications (a wood fire burns at 1,100 *F) which is about same temp as gas engine turbo temps. The graphs were verified against datasheet values. Note, that 409 expands at slightly less than mild steel rate. No cracking due to unequal header tube lengths.
-- http://www.crownroyalstoves.com/409-...eel-comparison

Good article on mechanical and temperature properties:
http://www.azom.com/article.aspx?ArticleID=1175

Datasheets:
http://www.pennstainless.com/stainle...ainless-steel/

Good info (post #11)
http://www.camaroz28.com/forums/forc...eaders-302217/

Orr,
I agree with most of what you wrote. Measurements have been done showing that at 3000+ rpm, exhaust energy can be up to 40% of the total hp generated. This is significant. More energy went into the exhaust than the coolant.
You mention sizing pipes, with more energy in the pipe then the pipe can be sized larger which leads to more hp overall. This is the classic remote mount problem discussed in other threads. A remote mount designer will complain about poor spool up and limited boost with a xx A/R turbine when using headers. They switch to a set of cast manifolds and gain 400 rpm spool and 2-3 PSI of boost. Some will run a single 2" exhaust to the turbo to try and get the spool rpm lower and more PSI...........but this is a mistake. Using something like 409 stainless with the larger pipe would be a better solution. Of course, using cast manifolds or stainless headers is the best way to go to.

Overall,
After looking at datasheets, 409 has the best expansion rates to match mild steel. 409 is good for the oxidation temperatures. 409 is cheap. Only $18 for a 4 foot straight pipe 2.5" diameter. Heat retention is 2x better than mild steel. The only downfall I see is welding it. A small bottle of 98/2 and some 309 or 409 wire will do it.

 

316L has a 16 W/(m*K) conductivity. Using post #1 equations results in 58hp loss. Very close to the cast manifold that is 3x thicker. Keep in mind the mechanical expansion rate when designing the header. Suppos the cyl 8 tube is 14" and the cyl #2 tube is 2". This 12" difference when the exhaust temp goes from 200 to 650 *C will give a length difference by 16um * (1/3)*450 = 2.4mm or .090".

 

the headers on my v6 car are built out of 321 flanges with 316L Primarys and are unequal length few years and no issues with cracking though i did use a mild steel crossover

 

The cracking tales of stainless are the least of my worries regarding stainless. The real world is that most of the turbo cars with stainless headers are garage queens that barely see the road and barely see the exhaust temps get into the critical range. Even a "real" race car that gets raced every weekend does not have many high temp heat cycles to cause fatigue.

I think the best application that has real heat cycle and cracking issues are wood stoves and daily drivers that see full boost for a while each day. A typical outdoor wood boiler will see a few high temp heat cycles every day. A race car sees them once a week.

No doubt 321 is the best for turbo applications. No doubt 409 is the cheapest and the worst conductivity of the stainless material. No doubt mild steel works. I wanted to look at the trade-offs of using each material if I had to go remote mount turbo. The car is a WS6 and the build constraints is I am not cutting up the front to move the radiator and keeping the A/C. This doesn't leave much room. The turbo will be either an MP T70 or Ebay GT45. I would really like to keep it an engine mount turbo, but the remote is tempting.

Since 409 is only a few bucks more than mild steel and holds in 67% more heat it seems like the best material to go with. Probably will ceramic coat it to make it even better.

What is the favorite DIY ceramic coating around here?

 

4xx series stainless steels are not meant to be welded.
3xx series are your weld ables.
I made the exhaust for my diesel suburban out of scrap yard .062 inch 304 piping, its worked flawlessly and looks the same as the day I installed it. It was cheaper to buy the scrap yard stainless than it would have been to use new mild steel pipe. The diesel engine is always on boost unless its idling.
I have some scrap yard .062 inch thick 316L alloy that was used as steam pipe in a food grade wash down environment. I want to make some exhaust stuff out of it.

 

Junk... I've been saying this for years but no one has been listening, you can see it in a regular exhaust just noticing how hot things get with stainless vs mild steel, especially headers.

4xx isn't weldable? Then why do they make welding wire specifically for it (though wire specific for welding 409 is usually flux core that needs a shielding gas). 409 welds _really well_ with regular mild steel (ER70S{whatever} wire) and welds OK with 309(L) in a pinch.

 

CrownAlloys 409nb comes in small spools for 20 bux or so.

 

I welded 409 with regular flux core and has been holding fine for few years on my downpipes