Questions Regarding Siamesed Intake Ports (9 Port Head)

General engine tech -- Drag Racing to Circle Track

Moderator: Team

enigma57
Guru
Guru
Posts: 1216
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by enigma57 » Sat Mar 25, 2017 5:43 am

I am working up an intake setup for a 292 Chevy inline 6 that will be built for max torque and redlined at 5,000 RPMs due to long stroke. Its based on a sidedraught intake made for 3 Weber DCOE carbs. With the Chevy inline 6 siamesed intake port design, each carb has 2 throttle bores and feeds 2 cylinders, but unlike a true IR design, does so via a single (siamesed) intake port......

Image

Image

Would love to braze a center divider into headport and build a true IR intake, but CSA if split will not support this engine to designed redline. And a Sissel 12-port head busts budget. So will go the more traditional route with GM iron head (remove center headbolt boss from each of the 3 siamesed intake ports and add cap screw and lump port.

Have a couple questions regarding siamesed intake ports......

1. Runner length. As a single intake port feeds 2 cylinders and a true isolated runner design with this siamesed port head is not an option...... Is it correct to assume that runner length cannot be tuned and all volume within intake and head port upstream of where ports split in head should be considered plenum volume? If this is correct, then runner length cannot be set at tuned length and doing so only increases effective plenum volume?

2. Choke (main venturi) sizing. The charts below are published by Weber. Figure 31 is for 1 to 6 cylinder engines with plenum style intakes fed by a single throttle bore to 5,000 RPMs at redline, whilst Figure 32 is for IR type intakes......

Image

With a siamesed intake port feeding 2 cylinders, I believe choke sizing should be determined using Fig. 31. A 292 (4.8 litre) inline 6 displaces 800cc per cylinder. So each of the 3 siamesed intake ports feed 2 cylinders having 1600cc displacement in all.

This chart indicates a single throttle bore having choke size of 25mm to 28mm would feed 1600cc to 5,000 RPMs. And a single throttle bore having choke size of 17mm to 20mm would feed 800cc to 5,000 RPMs.

However, there is a note below Fig. 31 stating "If the engine has 2 cylinders, select a Venturi corresponding to twice the engine capacity."

So the joker in the deck is whether this applies only to a 2-cylinder engine such as a V-twin motorcycle engine...... Or to an inline 6 where the engine is divided by intake design into three 2-cylinder groupings. I am inclined to think that as the inline 6 firing order is such that cylinders 3 & 4 have no point at which both intake valves are open at the same time...... And cylinder groupings 1 & 2 and 5 & 6 have only a very minimal period where one intake valve is closing whilst the other is just beginning to open...... Selecting a choke (main venturi) size corresponding to twice the engine capacity would be a bit large in this instance?

Your comments greatly appreciated. I have only tuned Weber carburettors on IR intake and on a couple of 4 and 6 cylinder engines having all cylinders fed from a common plenum. This inline 6 siamesed port setup is a new wrinkle for me.

Thanks,

Harry

Brian P
Guru
Guru
Posts: 1202
Joined: Wed Jan 02, 2008 7:35 pm
Location: Toronto, Ontario, Canada

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Brian P » Sat Mar 25, 2017 9:17 am

From what I can surmise, Fig 31 applies when there is a large plenum between the venturi and the cylinders so that the flow through that venturi is essentially averaged, and Fig 32 applies when there is not and the engine has non-overlapping intake strokes. The comment relating to Fig 31 is because having two cylinders results in non-overlapping intake strokes.

If you use Fig 31 and pretend that you have a two-cylinder engine fed by a single carb and you double its displacement as recommended, you have 1.6 litres actual displacement, double it as recommended gives 3.2 litres, it's off the chart but heading for 40-ish mm diameter. Since there are two but they are fed in parallel, two 28mm would give the same cross-sectional area.

Since the intake strokes are non-overlapping, you can use Fig 32 pretending that it's one cylinder fed by one throttle, the fact that there is another cylinder happening in what would otherwise be "dead space" doesn't matter. Again, you are off the chart but it's heading for 40-ish mm diameter. Since there are two but they are fed in parallel, two 28mm would give the same cross-sectional area.

Intake port "ramming" will still happen, it just won't be as effective.

Newold1
Expert
Expert
Posts: 823
Joined: Fri Apr 08, 2016 9:50 am

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Newold1 » Sat Mar 25, 2017 11:39 am

ON your setup with a definite budget constraint I think using your intake manifold a set of Weber DCOE's of proper mm venturi/bore size and lumping the ports is going to be just about the best you can do. As for runner lengths, the trumpet length used on the Webers will allow that tuning. All seems doable as its been done many times before with success.

wyrmrider
Guru
Guru
Posts: 6932
Joined: Thu Jan 10, 2013 10:52 pm

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by wyrmrider » Sat Mar 25, 2017 11:49 am

map out the degrees separating the firing of 1-2 3-4 5-6
they are not the same
If I remember 3-4 will have to be richer
3 jag 2" SU s work (as would 235 Corvette style YH Carters or Corvair turbo sidedrafts) if you can't find webbers
service procedures which appear in the 1953 Passenger Shop Manual
lump port the head
trying to divide does not work
spend your time on getting some quench with the CR you need
I've done the longer rod lighter piston way but that's pricey
head is the same as a 250 so cam will be quite different (it's a different core anyway- PM me if you need a core)

User avatar
4sfed
Pro
Pro
Posts: 297
Joined: Sun Mar 13, 2011 11:32 pm

4sfed

Post by 4sfed » Sat Mar 25, 2017 12:11 pm

If there is no overlap in the port, use fig 32 and divide cylinder displacement by two because you're feeding one cylinder with two venturis. Are these 40 DCOE's? 28mm to 32mm would be appropriate ... 28 for street, 32 for max HP with high compression and an aggressive cam.

BTW I have not found long airhorns effective regardless of runner length. Short airhorns always seem to make more power. Spacers under the carbs are effective ... at least on individual runners.

PackardV8
Guru
Guru
Posts: 5036
Joined: Sun Jul 30, 2006 2:03 pm

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by PackardV8 » Sat Mar 25, 2017 12:45 pm

I am working up an intake setup for a 292 Chevy inline 6 that will be built for max torque and redlined at 5,000 RPMs due to long stroke.
With modern materials, there is no reason a 4.120" (104.6 mm) stroke has an inherent physical limit of 5,000 RPMs. However, the one thing not mentioned thus far is higher horsepower higher RPM I6 engines often have a problem keeping the flywheel on. Maybe, do some homework on front balance damper and flywheel retention.
Jack Vines
Studebaker-Packard V8 Limited
Obsolete Engineering

User avatar
Sir Yun
Expert
Expert
Posts: 527
Joined: Fri Feb 17, 2006 7:16 am
Location: Near ze germans
Contact:

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Sir Yun » Sat Mar 25, 2017 2:17 pm

If I'm right the order is 1-5-3-6-2-4, then it's basically not a Siamese port problem. I would tend to just using 3rd harmonic length intakes as predicted by pipemax by adding extra length before the fork with a good amount of taper. I also would opt for a short horn and maximize intake length.
“There is no authority who decides what is a good idea.”
― Richard Feynman

Pro ? Me ? ...whhhhaaahahhaahahaha

User avatar
panic
Guru
Guru
Posts: 1914
Joined: Sat Dec 11, 2004 12:04 pm
Location: Ecbatana
Contact:

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by panic » Sat Mar 25, 2017 2:48 pm

The RPM limit isn't the stroke length, it's the point where the VE falls on its face.
Yes, the center carburetor (with equal intake cycle intervals) needs to be slightly richer than the end pairs with their offset cycles, per Jaguar research.
Get Santucci's book. http://tinyurl.com/n674pht

Tuner
Guru
Guru
Posts: 1890
Joined: Sun Apr 01, 2007 5:26 am

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Tuner » Sat Mar 25, 2017 2:52 pm

For a significant gain in air flow, remove the head bolt column in the port, use a stud, and countersink the nut in the floor of the port. Because the pulse rhythm is not the same, don't be surprised to find that you need to calibrate the center carb different than the ends to get the same color on the plugs in all six cylinders. If they were playing music or dancing, the ends are 1-3 rhythm and the center pair are 2-2. The end pairs are more Harley like, with irregular intake pulses.

User avatar
MadBill
Guru
Guru
Posts: 12624
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by MadBill » Sat Mar 25, 2017 9:20 pm

Unencumbered by any actual 6 cylinder experience, a few musings:

o Yes, it's like almost two different engines: The center pair fire at 360° intervals and so could possibly benefit from a tuned length suitable for ~2X the actual RPM (keeping in mind the multiple tune points of any given length and the different resonance characteristics resulting from the second, closed-end runner.)

o With 240°/480°intervals, the end pairs wouldn't appear to have any easily-targeted tuned length.

o As far as mass flow rate during the induction cycles, since the pulses are completely separated, I would think the center pair could use throttle bores and venturis sized for a single cylinder.

o Taking the 0.050" duration as the effective induction interval, the outer pairs probably have only very slight flow overlap and so should also be sized for single cylinders, but as mentioned jetting might be expected to be leaner.

o Is it for certain that even with the bolt bosses removed there is no room for/benefit from 'intruder' dividers welded into the intake manifold? (even if imperfectly sealed to the ports) Sure would solve a number of problems...
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.

User avatar
modok
Guru
Guru
Posts: 1300
Joined: Sun Jun 06, 2010 1:50 am

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by modok » Sat Mar 25, 2017 11:10 pm

It's pretty simple.
If there is 3 or more cylinders fed by one barrel, they assume flow is continuous. (not really true, but that IS the assumption)

If there is two cylinders, size for a one cylinder, as it is like two one cylinder engines sharing one carb, but vent ends up slightly smaller since the volume of the manifold dampens slightly.


Size the venturis using the one cylinder "racing" chart. Use 400cc cylinder displacement since you are using two barrels.

Size them by the multi-cylinder chart at 800cc (halving the CC since there is two barrels, then doubling it since there is two cylinders)

Size them by making "venturi area" slightly larger than "port MCSA"

Should all get the same answer.
If not, then something is off, post the info and we'll figure out what is off. :D
Glen Urban

User avatar
modok
Guru
Guru
Posts: 1300
Joined: Sun Jun 06, 2010 1:50 am

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by modok » Sat Mar 25, 2017 11:25 pm

You can tune the runner length, I'm sure it does something, but it is far from the ideal design to take advantage of it, so don't sweat the calculations there.
Glen Urban

Geoff2
Expert
Expert
Posts: 580
Joined: Mon Nov 09, 2015 4:36 pm

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Geoff2 » Mon Mar 27, 2017 5:48 am

The most important thing to remember with this intake manifold is that one cyl can draw from 2 carb barrels. Therefore, the carb should be sized on the small[er] side compared to a true IR system. The Australian 265 cu in Chrysler Valiant inline 6 cyl in E49 form came with a true IR manifold, individual intake ports in the head & came with triple 45 DCOE Webers. With a 4 peed manual trans, it would pull smoothly from 20 mph in high gear, ran a 14.3 sec quarter mile & pulled to 6000 rpm. Not bad for a 6 cyl a car available to the public from the showroom floor....
In your case with the lower red line & shared ports, I think 40 DCOEs would be best. It is just TOO easy to over carburate....

Ks Fats
Pro
Pro
Posts: 227
Joined: Wed Feb 12, 2014 12:39 pm

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by Ks Fats » Mon Mar 27, 2017 1:32 pm

Harry,
Do a search for Holden cylinder heads; Twisted 6, and 12 bolt .com both have bolt in lumps for the intake ports. I'm not the best at computers but tried to attach a couple of pics done by Knightengines on this forum for ideas; they are Holden but are similar. The short side radius on these heads is very limited and the lump inserts help address this.
You do not have the required permissions to view the files attached to this post.

enigma57
Guru
Guru
Posts: 1216
Joined: Sun Oct 07, 2007 11:59 pm
Location: Galt's Gulch

Re: Questions Regarding Siamesed Intake Ports (9 Port Head)

Post by enigma57 » Mon Mar 27, 2017 6:23 pm

Thanks to all for your comments and for setting me on the right track here. Much appreciated, I assure you.

1. I have ordered a copy of Leo Santucchi's book.

2. And I am studying buildups of the 202 Holden red engines 'down under' with similar 9-port head design to learn what I can about them and apply it to this 292 Chevy engine. Beautiful portwork to the Holden 9-port head, BTW.

3. Regarding my choosing to build this engine for torque and set redline at 5,000 RPMs...... This is not an engine for a drag car that sees full power runs for under 15 seconds each run and will be torn down, inspected and freshened up at frequent intervals. It is for a road car that will see occasional long periods of sustained RPMs much as an inboard boat engine would. And from time to time, I will pull a trailer.

I was told many moons ago by an engine builder who did his share of engines for boats and transporters that had to stay together for long hauls at sustained RPMs to "use good parts, balance the bejavvers out of it and keep sustained piston speed below 4,000 ft. per minute."

"Use good parts" meant forged crank, rods and pistons, well prepped and ARP rod bolts and fasteners. "Balance the bejavvers out of it" meant to balance the entire reciprocating assembly within 1/2 gram including torsional dampner and flywheel with pressure plate indexed to flywheel.

My intent is to build this engine using OEM rods and fully counterweighted nodular iron crank with 1/2" flywheel bolts. Well prepped, balanced and with ARP fasteners. Probably forged pistons, possibly hypereutectic.

4,000 ft./min. piston speed with 292 crank stroke of 4.12" equates to 5,800 RPMs but I think it wise to hold engine speed to 5,000 RPMs, as torsional issues with crank and flywheel retention are a concern. Perhaps Santucci's book will touch on that.

4. Very interesting comment regarding redline being limited by the point max VE is reached, rather than being limited by the long stroke. Reckon it depends upon which is reached first (max VE or 4,000 ft./min. piston speed). I will look into that.

5. Will also look at jetting of center carb feeding cylinder pairing 3 & 4 once I have it up and running. I did look up firing order for 3.8 and 4.2 litre Jaguar and it is same as Chevy inline 6, although Jag cylinders are numbered from rear to front rather than front to rear. So will expect similar jetting issues.

6. I have looked at choke size based upon comments here using both Fig. 31 and Fig. 32. Looks like 28mm to 30mm chokes should do it. Will be careful not to over carb this engine.

7. Regarding adding a center divider to the 3 siamesed intake port groupings after grinding out headbolt boss to create a 12-port head so a true isolated runner intake can be used...... Wish I could, but cannot, as each siamesed intake port measures roughly 3.5 sq. in. cross section and in order to support 292 cu. in. inline 6 to 5,000 RPMs...... Minimum of 2.2 sq. in. cross section would be required for each runner if siamesed runners were split...... And that is without subtracting for thickness of divider and (if used), a lump port added to floor of port(s). Just not enough material in head castings to increase port size that much without hitting water. Center divider might be possible with smaller port matched to smaller displacement engine such as 194 or 230, but not the 292.

8. As to tuned length runners with siamesed intake ports...... At this point, I am thinking that everything within the flow path downstream of throttle blades and upstream of port dividers in each of the 3 groupings of siamesed headports...... Are more akin to plenum volume than anything. However, I might be able to tune what I will call for lack of a better term, 'effective runner length' a bit if I can apply something I learned whilst working with Dan Miller on his and Gene Adams' hemi engine for the 2010 EMC.

Many thanks to all and if you should think of anything else that might steer me in the right direction with this build, please post it here.

Best regards,

Harry

Post Reply