Jump to content

Intake Manifold Experiment


Recommended Posts

(consider this a 'thinking out loud' post .... if folks have data points on the subject, feel free to share)


Often - when setting the low speed tuning* on a Weber 2x38 carburetor (*idle jets, idle volume screw and idle mixture screw) - I've noted a difference between the settings for the 'right' and 'left' side / barrel. This phenomenon remains true after confirming the throttle plates are in the same position, and on both new and used units. Sometimes the difference is a 1/4 turn on the volume (aka mixture) screw, other times the difference is substantially greater.


This got me thinking about whether / if / how the hogged-out "NASCAR" oval opening impacts the low speed circuit, and beyond. The best I can determine, the BMW factory switched between a two-hole and peanut shaped hole for the Solex 32/32; what were they chasing? In theory, I think, the plenum would distribute the charge from each barrel equally; but, when looking at the design of the manifold and where the barrels drop down (one barrel closer to the 'outer' runners, the other closer to the 'inner'), I wonder. Hence, the experiment.


To start, I made a 'two-hole' manifold for the 38/38, with testing to follow after I finish prepping it. -KB




(second photo depicts original two-hole type for 2x32 Solex next to hogged-out manifold for 2x38)

Link to comment
Share on other sites

Remember that the 32/36 Weber is progressive--on a stock (not Sync-Link) setup the second barrel doesn't begin to open until the primary barrel is 3/4 open, so the two hole manifold wouldn't be a benefit.  And with a symetrical manifold, no matter which barrel (or barrels) are open, airflow is gonna be more or less the same to all four cylinders.  Even with a synchronous 38/38 both barrels will be feeding all four cylinders at the same rate.


Let us know whatcha discover.



Link to comment
Share on other sites

I thought that the factory Solex DIDTA manifold always had two separate ports. Hogging out the manifold was necessary when switching to a Weber carb because the ports between carb/manifold didn't match up (flow was restricted). The porting for my Weber conversion was done mostly to the phenolic spacer plate and not the manifold itself.

Link to comment
Share on other sites

I thought that the factory Solex DIDTA manifold always had two separate ports.


But then there's the famous 'peanut' manifold, which - to the best I can determine - came on some cars from the factory.

That's also what got me wondering what BMW was exploring....



Link to comment
Share on other sites


And with a symetrical manifold, no matter which barrel (or barrels) are open, airflow is gonna be more or less the same to all four cylinders.

Yeah, agreed, Mike, airflow will be equal.


But fuel-in-air is noticeably different- in a 'tight' corner, the fuel doesn't follow the same path that the air does.

The 'heavier' fuel (yes, it's supposed to be an atomized mixture.  No, it never is.  Look at your pump jet's output)

will be less able to change direction, and end up going 'straighter' than the air does, and thus, the more convoluted

path will get less fuel, and the straighter path will get more...

So it would follow that 1 and 4 would run richer than 2 and 3 when you're on the primary, and then change when

you got the secondary open.


The 'Peanut' manifold's an attempt to mitigate that by reducing the sharp corner that the fuel/air mix has to

negotiate at the bottom of the 2- hole's primary to feed cylinders 2 and 3...  basically, it evens out the 'plenum'

so that all 4 paths are more similar no matter what the carb is doing.  It probably makes the plenum volume larger

than optimum for what BMW was trying to tune it for, but the more- even charge was worth the torque hit.

The waves inside intake plenums are pretty neat- and even now, not trivial to model.  Especially with a 'wet'

charge in the intake.


...according to something I read on the internet once...



Link to comment
Share on other sites

Lots of good information here. I remember having a similar discussion a while back... If only I could find that thread. 


Toby and Mike have just about hit the nail on the head. I'm curious what you discover... this is not something that is easily experimented or modelled.


The progressive vs synchronized throttle progression also plays a huge part as said before.  As far as equalizing mixtures goes, the peanut style manifold would definitely benefit using a progressive carb, while the two hole version works better with a synchronized carb. 

Link to comment
Share on other sites

Much has been written about laminar flow and plenum/combustion chamber design.  Often times the exception becomes the rule??   Rumor has it that dimpling and rifling may be just as influential as shape and size.   :unsure: 













Link to comment
Share on other sites

  • 2 years later...

A million years ago I read about porting manifolds on the metric mechanic website and other materials.

I rebuilt my motor with over-sized pistons with full blueprinting and balancing and got the 32/36 Weber and Petronix. I cleaned up the inside of the intake manifold a little bit matching the peanut to the carb holes and runners to cylinder holes etc.  For the exhaust I matched the ports and went a 64th bigger on the manifold. The were a lot of bumps inside the exhaust manifold from where some sensor went in and was able to grind these down and radiused  a few sharp edges. I did not touch the head. I did the intake first and a year later did the exhaust.

I didn't notice much difference from doing the intake manifold.  I think it's something that anybody with a steel exhaust manifold should consider doing if they ever take the exhaust manifold off. I did not measure (no G analyst) but I could feel the difference.

Link to comment
Share on other sites

You have my attention, Ken. As FAQ's 38-38 guru, I know you will discover some interesting info (be careful your racing competition isn't paying too much attention). 


As Toby alludes to, the dynamics within the intake manifold must be quite complex. The unequal-length runners 1&4, opposed to 2&3 must certainly affect flow and turbulence within the shared plenum space.  I suppose one could tap the manifold at several points around the peanut and measure flow to witness any variances.  We sidedraft users have completely different issues with intake harmonics, though variances in idle mix adjustments across the four barrels is common with sidedrafts, as well.  


You might try two different carbs during your experiment to rule out internal orifice imperfections?  Certainly, no two carbs will run exactly the same... I know my Solexes don't, despite my best efforts to set them up identically.  


Good stuff.. I look forward to hearing your findings.


Ed Z

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

  • Upcoming Events

  • Create New...