Calculating effective intake runner length (siamesed intake port)

General engine tech -- Drag Racing to Circle Track

Moderator: Team

enigma57
Guru
Guru
Posts: 1296
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Calculating effective intake runner length (siamesed intake port)

Post by enigma57 » Sun Jan 28, 2018 5:39 am

Had a question regarding intake runner length. Engine in question is 292 Chevy inline 6 with 3 sets of paired (siamesed) intake ports, center headbolt boss cut out of each set of siamesed headports and lump port mod done to floor of siamesed headports......

When calculating runner length...... Am I correct in assuming that overall (effective) runner length from point where intake runner joins plenum...... Measured downstream through flow path to back side of intake valve resting on its seat is same as if ports were isolated...... Rather than siamesed in head and a portion of intake adjacent to head?

Or conversely...... Does effective runner length (for calculating tuned length) end where individual ports end downstream and air/fuel mixture enters shared (siamesed) area of headport upstream of valve pocket?

Appreciate your input. This is the first siamesed port head I have worked with and its definitely a learning experienced.

Best regards,

Harry

mk e
Guru
Guru
Posts: 4289
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: Calculating effective intake runner length (siamesed intake port)

Post by mk e » Sun Jan 28, 2018 8:36 am

enigma57 wrote:
Sun Jan 28, 2018 5:39 am

When calculating runner length...... Am I correct in assuming that overall (effective) runner length from point where intake runner joins plenum...... Measured downstream through flow path to back side of intake valve resting on its seat is same as if ports were isolated...... Rather than siamesed in head and a portion of intake adjacent to head?
That is how dynomation5 or 6 would want it calculated.

I guess there will also be waves that are reflecting between the seat and common runner point and also the common point to the plenum.. like what goes on in tri-y headers. I've never seen anyone talk about trying tune it, maybe because those engine design normally are cost not performance based, but it has to be there so probably could be used somehow.
Mark
Mechanical Engineer

hoffman900
Guru
Guru
Posts: 1709
Joined: Sat Feb 23, 2013 5:42 pm

Re: Calculating effective intake runner length (siamesed intake port)

Post by hoffman900 » Sun Jan 28, 2018 9:19 am

Dig through Sir Yun's blog:

"Mass Flow In The Siamese Intake Section of the 12G940 Head":
https://aseriesmodifications.wordpress. ... g940-head/

"Why Does Adding Manifold Length Does Something Different Than Adding More Trumpet":
https://aseriesmodifications.wordpress. ... e-trumpet/

He's out getting his PhD so he hasn't been here (or elsewhere) in a while, but again, dig through his posts here as well as his blog.

As shared by Aaron Kelly, who builds some of the best A-Series engines out there:
IMO, that further illustrates the reason to lift the valve more and gain more flow area. You're not going to quickly fill the cylinder in a 5 port BMC head.


I don't completely agree with that, IMHO the cylinder can be filled quickly with the a series heads, the weak point is the lack of velocity to achieve any inertia ramming at the end of the intake cycle. Aaron
Seems to me that is the trick with these engines, To use that pressure at the end of the cycle as best as possible before cracking open the intake on the adjacent cylinder and bleeding the pressure into that cylinder.That might need to be factored in when designing cam events and headers. I would think that pressure would be different that the pressure generated with wave tuning as well.

Sir Yun:
Image
Turns out I did some runs to test that a while back.

The two arrows refer to the 2 pressure taps (simulation data). The point to notice is that the pressure in the siamese section is increasing just before IVC and that is not a bad thing. I don't do these runs often as they are really time intensive as you need to do them twice so it might take anywhere from 1,5 to 6 hours.
It would seem to me a siamese intake port would respond well to a delayed opening intake (ala: Harold Brookshire).
-Bob

enigma57
Guru
Guru
Posts: 1296
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Re: Calculating effective intake runner length (siamesed intake port)

Post by enigma57 » Mon Jan 29, 2018 9:13 am

Thanks, Mark! Thanks, Bob! That makes sense. I read over Sir Yun's commentary several times as well.

I have thus far worked up a 2-piece intake that will allow me to adjust overall runner length a bit by placing a machined spacer piece between the inboard and outboard halves of the intake. Clearance will be tight to master cylinder and steering column, though. So any added runner length over what I have now will be somewhat limited.

At present, runner length without adding a spacer (including siamesed portion of intake and headport) is 11.5" measured through flow path from where runner joins plenum downstream to back of intake valve.

Regarding Sir Yun's mention of the open (siamesed) portion of the intake tract causing the engine to tune to a lower frequency of several hundred RPMs as compared to same length isolated (non-siamesed) runner intake tract...... Based upon something we learned when working with Dan Miller and Gene Adams' hemi at EMC...... I believe that the increased volume of the siamesed port area has something to do with this.

That is to say, the engine sees the additional volume of the siamesed area as added runner length. After head work is done and lump port installed...... I will measure (cc) both the individual intake runners and the siamesed area of this manifold and headports to determine what that equates to in terms of a longer, isolated (non-siamesed) intake runner of equal internal volume.

Then I can look at Helmholtz calcs and see how close I can get to optimal runner length for the operating range I have in mind. Space limitations underhood being a factor, of course.

Many thanks,

Harry

mk e
Guru
Guru
Posts: 4289
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: Calculating effective intake runner length (siamesed intake port)

Post by mk e » Mon Jan 29, 2018 10:45 am

enigma57 wrote:
Mon Jan 29, 2018 9:13 am

Regarding Sir Yun's mention of the open (siamesed) portion of the intake tract causing the engine to tune to a lower frequency of several hundred RPMs as compared to same length isolated (non-siamesed) runner intake tract...... Based upon something we learned when working with Dan Miller and Gene Adams' hemi at EMC...... I believe that the increased volume of the siamesed port area has something to do with this.

That is to say, the engine sees the additional volume of the siamesed area as added runner length. After head work is done and lump port installed...... I will measure (cc) both the individual intake runners and the siamesed area of this manifold and headports to determine what that equates to in terms of a longer, isolated (non-siamesed) intake runner of equal internal volume.
I think your on the right track. In DM, the runner is assumes to be a taper top to bottom and they say is comes out pretty close even though normally neither the intake or exhaust are actually tapers.

On the intake, a larger stack opening raises the tuned rpm without changing the peak hp # much, but increasing the min port size, assumed to be at the seat lowers the tuned rpm a bit and the hp # quite a bit....a siamesed port probably acts a lot like the larger seat simulation....although on 4v setups which look a whole lot like Siamese setups they say use the area on a single seat so the confused me a bit.

Then I can look at Helmholtz calcs and see how close I can get to optimal runner length for the operating range I have in mind. Space limitations underhood being a factor, of course.

Many thanks,

Harry
Is the manifold 6 or split into 2x3?

I've not messed with it yet but everything I read and looking at oem setups says there is pretty limited effect available if you have more than 4 cylinders, 3 is optimal. I keep thinking on my v12 project i want to split the airnox into 4 groups of 3 to get it to ring....but as I day I've not messed with it beyond calculation....but the way I was calculating the runner length isn't in the math, just plenum and feed tube volumes, so I may be doing something wrong.
Mark
Mechanical Engineer

enigma57
Guru
Guru
Posts: 1296
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Re: Calculating effective intake runner length (siamesed intake port)

Post by enigma57 » Mon Jan 29, 2018 12:47 pm

Mark, I'm building a 292 Chevy inline 6. It has 3 sets of siamesed headports and each of them feeds 2 cylinders. I managed to score a couple of identical intakes off the HAMB a while back. These are the Brazilian made intakes made to fit 3 Weber DCOE sidedraught carbs to these engines.

I ran into clearance issues with sidedraught carbs, though. So I decided to convert 1 of the intakes to take some Weber DCNF downdraught carbs which I have here. This intake is now the outboard half of a 2-piece intake...... The other (unmodified) intake being the inboard half. The halves will bolt together at what once was their DCOE mounting flanges.

There are some pics here in this earlier thread where I was working on mods to what is now the outboard half of the 2-piece intake. Should give you an idea as to what it will look like when finished......

viewtopic.php?f=1&t=49431&start=45

After I got the pieces fitted up, I sent them to my friend Nick Smithberg and he took care of the welding and machining. Did a great job. :D

The intake and engine are on the back burner for now. I have finally located the emulsion tubes, jets and choke tubes I need for a good baseline setup. Plan on rebuilding and reconfiguring the DCNF carbs in the next month or so and getting back on the build.

Best regards,

Harry

User avatar
modok
Guru
Guru
Posts: 1720
Joined: Sun Jun 06, 2010 1:50 am

Re: Calculating effective intake runner length (siamesed intake port)

Post by modok » Mon Jan 29, 2018 9:05 pm

The extra volume at the valve end lengthens the period slightly, but other than that tuned length is the same. Remember....you can't get too carried away with duration because the two intake valves overlap so much, so compared to "normal" I think you will end up with a larger runner and less duration.
Glen Urban

DrillDawg
Expert
Expert
Posts: 980
Joined: Tue Jul 06, 2010 11:03 am

Re: Calculating effective intake runner length (siamesed intake port)

Post by DrillDawg » Mon Jan 29, 2018 10:45 pm

Weld a length of alum. plate with a tube for the head bolt to each of the intake manifold ports that would isolate/divide the port in the head if you want to make the port longer. Just have to make it sturdy enough that it won't fall apart. Bolt on the manifold then add the head bolts in those ports.
BORN RIGHT THE FIRST TIME

enigma57
Guru
Guru
Posts: 1296
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Re: Calculating effective intake runner length (siamesed intake port)

Post by enigma57 » Wed Jan 31, 2018 4:28 am

Thanks, Glen. I was thinking of it in pretty much the same way. Hard to quantify the effect of increased volume on effective runner length where individual ports end and larger siamesed port serves both intake valves. Just trying to see if there is a way to calculate it so as to make it work for me rather than against me.

I do plan on keeping duration conservative and will build this engine for torque due to intended operating range and long stroke. Was thinking about late opening of intake valve to allow for better cylinder filling, as I reckon port velocity will diminish noticeably when air/fuel mixture enters larger cross section of siamesed port. Will have to balance that out against whatever IVC I need to get DCR where I need it. Might look at tight LSA with this setup as well. Need final cc of chambers once head is done to figure all that, though.

Best regards,

Harry

DrillDawg
Expert
Expert
Posts: 980
Joined: Tue Jul 06, 2010 11:03 am

Re: Calculating effective intake runner length (siamesed intake port)

Post by DrillDawg » Wed Jan 31, 2018 7:19 am

With the three separate manifolds you could treat it just like three stroker Shovelhead Harley V-twins (4-5/8" x 3-5/8"), those engines didn't like much above 5-5500 rpm and have a siamesed intake port, having said that you might think about calling S&S cycle and talk with the Shovelhead expert there about some insight on cam timing, I remember a lot of 98-99 ICL on the better running bikes. Just a thought.
BORN RIGHT THE FIRST TIME

joe 90
Guru
Guru
Posts: 2645
Joined: Tue May 27, 2014 4:02 am
Location: The land of the long white cloud

Re: Calculating effective intake runner length (siamesed intake port)

Post by joe 90 » Thu Feb 01, 2018 2:38 am

It's not something you can get perfect due to the firing order.

You might be able to come up with a number for the middle 2 cylinders but the end ones will be different.
There isn't 360 deg between events so one of the pair will probably end up running rich while the other runs lean.

A staggered cam might help?

DrillDawg
Expert
Expert
Posts: 980
Joined: Tue Jul 06, 2010 11:03 am

Re: Calculating effective intake runner length (siamesed intake port)

Post by DrillDawg » Fri Feb 02, 2018 7:41 am

I don't know how much lift you will have at the valve, but if it's .450" or so I would develop a 30 degree valve job. It will flow more than a 45 or steeper seat at very low lifts and exposes a larger window sooner.
BORN RIGHT THE FIRST TIME

enigma57
Guru
Guru
Posts: 1296
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Re: Calculating effective intake runner length (siamesed intake port)

Post by enigma57 » Sat Feb 03, 2018 2:57 am

* Thanks, DrillDawg. Good ideas. I wanted to install a port divider in the siamesed portion of intake and headports to create a true IR induction setup and had worked out a way to do it. But even with the center headbolt boss removed and without adding a lump port installation...... There is insufficient CSA in the siamesed area of headports to accomplish this.

Standard bore is 3.875" and stroke is 4.120". Will keep overbore to minimum needed to clean up cylinder bores. Cam will be fairly mild, as these 292 inline 6 engines have issues with harmonics when spun over 5,000 - 5,500 RPM range. Lift at valve will be somewhere in the 0.500" - 0.550" range. I will definitely look into 30 degree valve job when ready to do the head.

With the engine divided into three 2-cylinder groupings on intake side, displacement of each 2-cylinder grouping is very close to that of a Harley V-Twin such as the Twin Cam 96. Initially, I did look into S&S and other motorcycle carbs. Considered working up an intake to fit 3 Mikuni 42HSR carbs to this engine. However, I have a good set of rebuildable Weber 42 DCNF downdraught carbs here and after it became apparent that sidedraughts would present clearance issues with my master cylinder and steering column in this car...... I decided to modify the intake(s) I have here to fit the DCNF carbs to this engine.

* Thanks, Joe. Point well taken. Yes, it is my understanding that just as with the Jaguar inline 6...... Cylinder pairings 1 & 2 and 5 & 6 will tune similarly, whilst the center cylinders 3 & 4 may require slightly different jetting or adjustment due to firing order. As I will be running 3 carbs, each isolated to these 3 sets of 2-cylinder groupings, I believe I can work that out during tuning.

I have Leo Santucci's book on the Chevy inline 6s here and he noted that the late Jim Headrick in the late '70s had worked out a revised porting approach along with grinding each cam lobe on different centerlines (some advanced, others retarded) for these engines. His factory iron reworked lump ported siamesed heads were flowing 319 cfm on intake side in 1978. I would love to take advantage of the differing cam lobe centerline approach on my engine even though it is for a road car, but Leo indicates in his book that Headrick did not make make the details of his approach public knowledge. Will ask Mike about this when I am ready to do the cam. He may have some insight into this approach.

Here is a tribute page to Cotton Perry and Jim Headrick. They campaigned a 292 inline 6 powered '67 Chevy II in those days (H/MP)......

Image

http://www.drwebman.com/perry_headrick/

Best regards to all,

Harry

hoffman900
Guru
Guru
Posts: 1709
Joined: Sat Feb 23, 2013 5:42 pm

Re: Calculating effective intake runner length (siamesed intake port)

Post by hoffman900 » Sat Feb 03, 2018 10:24 am

I have Leo Santucci's book on the Chevy inline 6s here and he noted that the late Jim Headrick in the late '70s had worked out a revised porting approach along with grinding each cam lobe on different centerlines (some advanced, others retarded) for these engines. His factory iron reworked lump ported siamesed heads were flowing 319 cfm on intake side in 1978. I would love to take advantage of the differing cam lobe centerline approach on my engine even though it is for a road car, but Leo indicates in his book that Headrick did not make make the details of his approach public knowledge. Will ask Mike about this when I am ready to do the cam. He may have some insight into this approach.
This is something David Vizard and David Anton worked on to produce the 4cyl BMC A-Series "scatter pattern" camshafts. They got the idea from Harry Ratcliffe (UK engine builder) who supposedly had tried it in the 1960s sometime.

The idea was to grind the camshaft with varying lobe centerline angles to reduce the charge robbing of the siamese intake valves. With the scatter-pattern camshafts, the middle two cylinders had slightly advanced centerline angles from the outer two.

For example, the popular Kent 310SP camshaft:

Intake: 310* adv, 1&4: 106* LCA , 2&3: 103* LCA
Exhaust: 310* adv, 106* LCA

A couple ways I can see helping with the charge robbing :

Kill low lift flow. This would mean 30* seats would be a HUGE no-no. 50* seats may be worth experimenting with.
As much rocker ratio as you can run. This will help with getting lobe lift area but reduces the seat-to-seat requirements to obtain that area.
An assymetrical camshaft (like Harold's designs). These designs delay the intake opening and act like different LCA camshafts.

Lots of BMC A-Series racers winning racers with non-scatter pattern camshafts though. If you have the cam lobe profiled, you could see how much the valves would be open at the same time.

I like the assymetrical cam idea the best, but please share how Mike approaches this problem.

Also, this is not an unique issue. Any 4 cylinder with only two carburetors acts similarly. Here is a quote from Rick Parent (tech head with SVRA (vintage road racing sanctioning body)).
Concerning the David Vizard comment about the scatter pattern cam, I have been thinking about that for a long time because the 1300 Spitfire SU set up is almost the same as the single intake port 1275. While I have not tried that yet I realized the other day why it probably does not matter. I was taking a motor off the Dyno and I glanced at the 3 and 4 intake ports. Both valves were open about .300!! What I realized was there is no way 3 degrees is going to do anything when the valves are both open that much! 1 degree is worth about .004 of lift, so 3 degees would be about .012. I have always looked at that concept at the the seat opening and closing event, I had no idea what the valves looked like during the actual overlap period, I am amazed that I never saw that before.
-Bob

DrillDawg
Expert
Expert
Posts: 980
Joined: Tue Jul 06, 2010 11:03 am

Re: Calculating effective intake runner length (siamesed intake port)

Post by DrillDawg » Sat Feb 03, 2018 1:33 pm

28/44 x 52/20
BORN RIGHT THE FIRST TIME

Post Reply