Plenum size

[zetecspit]

I am (finally) going to make  a start on changing from ITBs to a single TB/plenum. 

I eyeballed a std ST170 plenum (sadly totally the wrong shape for the spitfire) and was surprised by its size. 125mm diameter and the main part is around 400mm long, giving a volume of approx 5L plus the inlet runners and other "bits". This seems to be much larger than other ideas I have seen suggested (often approx the size of engine displacement, 2L in my case)

I thought I understood that a smaller volume was desirable for fast response, too large and teh volume is too bid and things get vague. 

Can anybody shed any light on this? I plan to buy a 500mm length of tube, but not sure if 100 or 125mm would be best. Or should I go for a "double plenum" of 2 lengths of 75mm tube, on a "shotgun" pattern joined by a long slot. 

Guidance/ideas/discussion very welcome.

 

[egret]

It's a complicated relationship and one that involves aspects of noise, inlet air temperature, and probably many other compromises.  I wrote a load of stuff, but when attempting a quick refresher decided I wasn't sufficiently confident in how it works to be explaining it!

I believe the general rule of thumb states that a larger plenum should give more even flow to all cylinders and possibly even better performance at high rpm, but will lower throttle response time.  These basic rules that can be designed around by clever OEMs with the right expertise and computational resource (hence their use of variable inlet lengths).  You can look at helmholtz resonators and the potential for tuning manifolds for specific RPM bands, but it's very complicated and I suspect that you'll run into issues with packaging.  The gains are probably marginal for what we see on our cars too.

Your previous post on the subject suggests that a smaller tubular manifold with a larger conical one in the shotgun configuration you mention could be a good solution for ensuring equal flow to all cylinders, should package fairly well, and will give you something to chat to people who are taking an interest in what's under your bonnet!

I'm afraid I'm really not sure how sensitive something like this would be.  You're obviously restricted with space, inlet port size, and throttle body size, but I'm not sure if the difference between the smallest sensible plenum and the largest sensible plenum would actually be noticeable when driving.  If you can work out a way to build something where you can relatively easily change plenum volumes then you could get a stopwatch and do some timed runs!

I recon I'd just build something that looks proportionally right and fits nicely into the space available, similar to this, that looks big, but not unreasonably so: https://www.g19engineering.com/view.php?cat=92&ViewProduct=206&page=1

I'm sure others will have experience of over/undersized plenums and how they affect drivability and performance.

[zetecspit]

Thank you for the reply. It puts my mind at rest. 

Yes, I think I will try a "dual" plenum thingy, see how it pans out. I happen to have some 3" stainless tube going spare, which will be handy.

[JohnD]

Zetec,

I'm sorry, I can't recall if this is for a straight six, a four, or something not Triumph at all.

In any case, and within reason, the largest plenum possible is desirable AFAIK.    By Bernouille, the  slower the velocity of a fluid, the higher it's dynamic pressure.   If it's rushing across the carb/injector inlet then its pressure will  be lower than if it was still, and the harder it is to be sucked in on induction.

On the straight six, some have found evidence - rich plugs - of air starvation in  bores 5 and 6, and presumed that as the intake is at the front, that 1,2,3 and 4 have taken more than  their share.    Others disagree, but with that, and the Bernouille lesson in mind, I've expanded my plenum to 6" from the original 3" diameter.  Here's a comparison:

Flow down a tube is subject to the Fourth Power rule, as resistance is inversely proportional to the fourth power of the radius.    So doubling the radius will reduce resistance and increase flow SIXTEEN times!    (2x2x2x2= 2^4 = 16)    So flow to the more distant bores is easier.

The inlet I left at 3", to simplify fabrication and avoid the need to source a filter with a very large attachment ring!  The difference can be seen here:

And, the intake ring approximates to an orifice (Length<<<Radius) in which flow obeys a Square law, so only four times the flow by doubling the radius.

John 

Edited January 6 by JohnD

[Gt64fun]

When I was researching airbox design, I came across a 'rule of thumb' that the volume of the airbox should be equal to or greater than the engine capacity.

I have also seen designs which have a built in taper from front to rear, presumably in proportion to demand.

Bit more difficult to fabricate!

Ian

[zetecspit]

1 hour ago, Gt64fun said:

When I was researching airbox design, I came across a 'rule of thumb' that the volume of the airbox should be equal to or greater than the engine capacity.

I have also seen designs which have a built in taper from front to rear, presumably in proportion to demand.

Bit more difficult to fabricate!

Ian

That is what I saw, yet Ford have used one 2 1/2x the engine capacity! I bet they did some research, but they also used variable inlets too, and other clever stuff. 

I think using a "shotgun" setup will work pretty well, the tube with the throttle body tapering,  and a slot 15mm wide between the two tubes so air gets evenly distributed between the cylinders. I shall pilfer the "Jones minor" technique for creating the taper. (slit the tube, then jubilee clips to squeze one end to a smaller diameter)

[Nick Jones]

What they said already really. I did loads of reading on this before building the Vitesse one. Ultimately, real world packaging constraints usually dictate.

[Escadrille Ecosse]

23 hours ago, Nick Jones said:

What they said already really. I did loads of reading on this before building the Vitesse one. Ultimately, real world packaging constraints usually dictate.

This ^^^^

Based on my admittedly limited knowledge of fluid dynamics from university 40+ years ago (and before CFD came along) I am unconvinced of the merits of a taper design from what I have seen, particularly on normally aspirated engines. And whatever benefits there might be, it seems to be far easier to get it wrong and have something that reduces performance. Whilst also being harder to make that a simple design.