How do teams choose exhaust diameter?

All that has to do with the power train, gearbox, clutch, fuels and lubricants, etc. Generally the mechanical side of Formula One.
Caito
Caito
13
Joined: 16 Jun 2009, 05:30
Location: Switzerland

How do teams choose exhaust diameter?

Post

Hi guys. I know that exhaust need to be tuned to specific frequencies.


In tuning to that specific frequency two parameters come into play, length and diameter.

I imagine a bigger diameter would be less restrictive, but would cause a long pipe. Meanwhile a longer pipe would generate more losses too.


How do they reach a final value? Since, by looks, all teams use the same diameter, it must be the best solution.
Come back 747, we miss you!!

bl79
bl79
0
Joined: 20 Jul 2011, 19:36

Re: How do teams choose exhaust diameter.

Post

I'm not sure how exactly it works at the F1 level, but Carrol Smith offers some insight to this in "Tune to Win." Basically my understanding is that things such as exhaust diameter, length, etc. are dictated primarily by the engine manufacturer. I'm sure the teams work with the engine manufacturer in order to reach the best compromise but on a whole I'm sure a huge portion of these parameters are set by the manufacturing companies and are followed pretty closely.

I know that doesn't really answer the question, just some insight. It seems more like the question now becomes "how do engine manufacturers determine exhaust diameter." I can't offer much on this point. Maybe look at some books on the subject? I hear "Internal Combustion Engine Fundamentals" by John Heywood is quite good.

riff_raff
riff_raff
132
Joined: 24 Dec 2004, 10:18

Re: How do teams choose exhaust diameter.

Post

Caito,

Exhaust header primaries are tuned to take advantage of the lower order (first and second order) acoustic frequencies, since they are the most energetic. Since the tuning is acoustic, it's the relationship between duct length and the exhaust gas sonic velocity that is of concern. The duct length being defined by the distance between changes in cross-section, or acoustic pressure wave reversion nodal points.

While pipe length is important for acoustic tuning, the actual diameter is not so much. Header primary diameter is selected to minimize any changes in flow area exiting the cylinder head's exhaust ports. The optimum header primary pipe diameter is a balance of maintaining sufficient exhaust gas flow velocities/inertias and minimizing exhaust flow back pressures.

The whole situation is much more complicated, but that's my layman's explanation of it.

riff_raff
"Q: How do you make a small fortune in racing?
A: Start with a large one!"

Edis
Edis
59
Joined: 31 Jan 2010, 16:58

Re: How do teams choose exhaust diameter?

Post

Caito wrote:Hi guys. I know that exhaust need to be tuned to specific frequencies.


In tuning to that specific frequency two parameters come into play, length and diameter.

I imagine a bigger diameter would be less restrictive, but would cause a long pipe. Meanwhile a longer pipe would generate more losses too.


How do they reach a final value? Since, by looks, all teams use the same diameter, it must be the best solution.
It's not really about frequency, or flow restriction for that matter.

When the exhaust valve opens a pressure pulse (compression) will form in the exhaust port. This pressure pulse will travel downstream the exhaust pipe with roughly the speed of sound. When the pressure pulse is subjected to a change in area, which is what happens in the collector, a reflection pulse will form and travel back to the exhaust port with roughly the speed of sound. Since the area in the collector (area of secondary pipe + area of the other primary pipes) is greater than in the primary pipe the pulse is coming from the positive pressure compression pulse will become a negative pressure expansion pulse. When the exhaust has the correct length, the pulse will travel from the port to the collector and back to the exhaust port and arrive when the exhaust valve is about to close to cause a low pressure in the port. In principle this is only dependant on gas properties and the length of the pipe.

The diameter of the pipes is a different issue which is related to how a pulse is affected by a change in area. The starting point is the port area of the exhaust (valve seat area minus valve stem area). If the primary pipe have an area that is smaller than the port area it will form a convergent duct and if the pipe area is larger than the port area it will form a divergent duct. Most engines use divergent ports, although there are some turbocharged engines using convergent ports. For most engines this gives a port-to-manifold area ratio between 1.2 and 1.6 where engines with higher bmep tend to use a higher ratio. The purpose of this area ratio is basically to prevent the loss of pulse amplitude and energy in order to maximize the pulse tuning effect. The secondary pipe usually have a diameter to provide an area ratio in the collector of about 6, which gives a good reflection pulse.

In practice the exact dimensions require a lot of optimization work using either engines running in test cells and/or engine simulation software such as GT-Power or Ricardo Wave.

Read Blairs "Design and Simulation of Four-Stroke Engines" for more detailed information.

Edit: In F1 it's the teams (or a supplier to the team) that make the exhaust manifolds to the dimensions given by the engine manufacturer.

ScudTop
ScudTop
0
Joined: 03 Nov 2011, 13:45

Re: How do teams choose exhaust diameter?

Post

I have been lurking around this forum for a while, but this is my first post.

I work with small, OHV motor development (less than 65 CI) and have been spending a lot of time recently on exhaust system development for these motors. This includes modeling several different primary configurations and then building and testing them on my in-house engine dyno.

The variables that I considered were as follows

1. Comparing 3rd Harmonic with 2nd Harmonic primary lengths
2. Primary Tube Venturi

Motor Specifications

OHV - 4 cyl Abarth
Cylinder head - Wedge head, valves inclined 11 deg. Single port entry into head feeding a resonant chamber feeding all four intake valves.
Intake - Single 36DCD Weber downdraft carburetor
Intake valve - 31mm
Exhaust Valve - 27.5mm
Bore - 65mm
Stroke - 74mm
Cam Duration - 254 deg @ 0.050 lift (156 deg @ .200 lift)
Cam Lift at cam - 8.43mm
Cam L/C - 108 Intake centerline 104deg.
Rocker arm ratio - 1.45:1


At the moment the "conventional wisdom" for exhaust systems for the motor in question has been to use primary tubes of 30mm ID, 940mm long. As the engine layout does not pose a packaging problem, this long a primary has not been a problem. As well the tail pipe, after the 4:1 collector is 50mm in diameter and 800mm in length.

In my research I came up with a combination that seemed to better optimized for the RPM range that I was concerned about (5800-8200). My modeling seemed to say that a 30mm ID primary of 560 in length would provide better pulse tuning in the desired RPM band (or at least a portion of it). In addition, length of the 50mm secondary was also reduced from 800mm oto 350mm. This would appear to be the difference between a 2nd and 3rd harmonic length.

The overall result of this was an increase in HP from 6800 RPM onward, but a slight decrease in torque below 6800 RPM.

I had read about "stepped" primaries as a means of extending the range of where a tuned system might be effective and tried several different iterations on my modeling software. This involved two increased diameter sections. On paper (read computer) this seemed to have little effect. I also tried an interation using a 75mm long "venturi" section of 28mm ID. I modeled this venturi placed both at the beginning of the primary and at the end, just prior to the collector. This second iteration had a result that I did not expect. While the overall HP curve was little altered above 6800 RPM, below this RPM the overall performance was significantly increased. Subsequently an exhaust system to this design was produced and tested on my Stuska engine dyno and the results of the "venturi" modified exhaust were confirmed. Overall torque was raised by 3-4 lb/ft in a range from 5200-6800 RPM, and an increase of 2-3 HP was seen at higher RPM levels.

The only explanation that I can offer for this increased performance is that gas speeds in the exhaust primaries would have been accelerated through the venturi sections, perhaps influencing the wave action intensity.

For a 3-main beaing engine, designed in the late-50s, this was a significant increase in performance.


I would be interested in any comments that board members might have regarding this "venturi" development.

User avatar
strad
117
Joined: 02 Jan 2010, 01:57

Re: How do teams choose exhaust diameter?

Post

I thought you had to get into piston speed
To achieve anything, you must be prepared to dabble on the boundary of disaster.”
Sir Stirling Moss

ScudTop
ScudTop
0
Joined: 03 Nov 2011, 13:45

Re: How do teams choose exhaust diameter?

Post

Strad

The average piston speed in this case is around 3981 FPM. The max piston speed occurs at 73.859 degrees ATDC and is 6561 FPM.

hardingfv32
hardingfv32
35
Joined: 03 Apr 2011, 19:42

Re: How do teams choose exhaust diameter?

Post

Here is a good paper on modeling a NASCAR system:

http://etd.fcla.edu/CF/CFE0000082/Dollh ... 7_MSIE.pdf

Brian

User avatar
strad
117
Joined: 02 Jan 2010, 01:57

Re: How do teams choose exhaust diameter?

Post

ScudTop wrote:Strad

The average piston speed in this case is around 3981 FPM. The max piston speed occurs at 73.859 degrees ATDC and is 6561 FPM.
Thanks but how do you arrive at those figures?
You know the rod length of say the Ferrari engine? The stroke? or are you working backwards from the displacement?
Not being funny,,,just trying to understand how you got the numbers.
To achieve anything, you must be prepared to dabble on the boundary of disaster.”
Sir Stirling Moss

ScudTop
ScudTop
0
Joined: 03 Nov 2011, 13:45

Re: How do teams choose exhaust diameter?

Post

Strad,

These numbers were for my own small motor (from the previous post).

I own any number of software programs that can compute this information, including

PipeMax
Car-For
Engine Analyzer Pro 3.5
Lotus Engineering Software

No I do not know the rod length of the current Ferrari motor, but it probably would not be too hard to make some educated guesses.

Paul

User avatar
strad
117
Joined: 02 Jan 2010, 01:57

Re: How do teams choose exhaust diameter?

Post

So F1 engines would have different numbers.
Each engine is different.
I used to have a formula for this back when I was racing, but it's lost in the piles somewhere...thought you might be able to help.
Stroke and rod length affect piston speed and you have to know that I believe to compute both the diameter and length,,,especially of the primaries, and then another computation for collector length.
I will divulge how we used to gauge collector length.
We would paint the collector and run the engine,,,the point where the paint quit burning off was invariably the best place to end the collector.
I'm sure you guys will laugh at such a simplistic solution,,,but that worked very well.
Last edited by strad on 08 Nov 2011, 21:37, edited 1 time in total.
To achieve anything, you must be prepared to dabble on the boundary of disaster.”
Sir Stirling Moss

hardingfv32
hardingfv32
35
Joined: 03 Apr 2011, 19:42

Re: How do teams choose exhaust diameter?

Post

I would say the collector or last section of the system is the least sensitive to length changes. That has been my experience on the dyno.

That could be why your technique seemed to work for you.

Brian

User avatar
strad
117
Joined: 02 Jan 2010, 01:57

Re: How do teams choose exhaust diameter?

Post

We found that it moved the torque band around and had a real effect on how our car launched
To achieve anything, you must be prepared to dabble on the boundary of disaster.”
Sir Stirling Moss

Art Smith
Art Smith
0
Joined: 01 Dec 2014, 02:15

Re: How do teams choose exhaust diameter?

Post

ScudTop wrote: I had read about "stepped" primaries as a means of extending the range of where a tuned system might be effective and tried several different iterations on my modeling software. This involved two increased diameter sections. On paper (read computer) this seemed to have little effect. I also tried an interation using a 75mm long "venturi" section of 28mm ID. I modeled this venturi placed both at the beginning of the primary and at the end, just prior to the collector. This second iteration had a result that I did not expect. While the overall HP curve was little altered above 6800 RPM, below this RPM the overall performance was significantly increased. Subsequently an exhaust system to this design was produced and tested on my Stuska engine dyno and the results of the "venturi" modified exhaust were confirmed. Overall torque was raised by 3-4 lb/ft in a range from 5200-6800 RPM, and an increase of 2-3 HP was seen at higher RPM levels.

The only explanation that I can offer for this increased performance is that gas speeds in the exhaust primaries would have been accelerated through the venturi sections, perhaps influencing the wave action intensity.

For a 3-main beaing engine, designed in the late-50s, this was a significant increase in performance.


I would be interested in any comments that board members might have regarding this "venturi" development.
fascinating post. based on similar work for a 1600cc Kent for FFord, I'd guess the answer is on the acoustics side of the solution and not the mass transport side. a careful check of the local speed of sound all along your exhaust may be in order. "venturi" suggests an area change and expansions lower the gas temperature and the local speed of sound. how much overlap do you have after TDC in crankshaft degrees (the union of intake and exhaust cam profiles, lobe separation angle, installed camshaft timing, and both intake and exhaust lash) and where does your current tuned length put the inverted pressure wave from the collector entrance from the next cylinder in your firing order? based on your installed camshaft timing, where is exhaust valve close with respect to maximum piston velocity?

Art
artesmith@earthlink.net

Facts Only
Facts Only
188
Joined: 03 Jul 2014, 10:25

Re: How do teams choose exhaust diameter?

Post

If you are referring to 2114 and F1 teams then they don't choose the exhaust diameter.

The engine suppliers design and make the pre-turbo exhaust. As for the the tail pipe there is a maximum diameter and a very narrow box that the exit can be placed in, the diameters are to the maximum value and the length is set simply by how long it needs to be to go between the turbine exit and the tailpipe position box. There will be some parameters such as bend radii set by the engines suppliers that will be given to the teams to make their tailpipes.
"A pretentious quote taken out of context to make me look deep" - Some old racing driver