Flow vectors & Peugeot 205GTi inlet manifold

[benlilly]

Dear Guy,

I was interested to read that you found flow losses when fitting the 205 inlet manifold to a head you had modified.

Did you manage to make any modifications to reduce these losses?

If I was to look at fabricating a new manifiold, would I be correct in thinking that the matching of the profile and straightness at entry are the most important factors to get right if the flow loss is to be reduced?

Regards,

Ben Lilly

[Guy Croft]

Ben, hi

are you referring to the OE manifold or the aftermarket sidedraft one?

GC

[benlilly]

Hi Guy,
Thanks for your reply. I was refering to the O.E. manifold on the 1.9 8V engine.
Ben

[Guy Croft]

OK, when a flowbench says 'loss' you're ill advised to try and prove otherwise, I refused to let that particular client run with it, as I was sure it would negate all the gains I had achieved.
Was I wrong for sure - will never know.

But insofar as the loss I saw is concerned - it must be to do with a mismatch between the axis of the main exit flow vector from the OE manifold and the ideal vector in the port. That's what I call them anyhow, won't find it in any book, see below. The vector being the fastest airstream, it grabs the trace.
If I was fabricating a substitute I'd want a horizontal run into the port as you describe. Get that right and there will be no loss at all.

GC

[benlilly]

Thanks very much, Guy.
Ben

[benlilly]

Hi Guy,

Do you have a record of the flow loss you found with the OE manifold in place vs the bare head and can you give a rough idea of the bhp loss this would result in?

If it is only a couple of horsepower then it would be difficult to justify fabricating a tubular plenum manifold.

Regards,

Ben

[Guy Croft]

Hi Ben,

Although the OE manifold (below) flowed well on its own, the loss on the head was near 20% compared with almost zero with finished head and the reworked s/d. The loss was so catastrophic I'm not sure I even wrote it down and I certainly refused to let the owner fit it (still got it..). it shows, I think, the supreme importance of doing flow development WITh the inlet manifold.
The loss, I assumed to be a result of mismatched entry flow vectors (for want of another name) betw head and manifold and I have seen it many times on downdraft manifolds on Fiat /Lancia 8V and 16V TC heads (see more of them than most). Now, of course, it may welll be true that the flowed head with OE manifold might well be better than un-flowed head with same, but that did not overmuch interest me in that case. I don't like to see head gains wiped out by the wrong type of manifold if you follow..
Air (fluid) just doesn't like changes of direction or section and every one adds to the total fluid flow loss.

GC

[benlilly]

Hi Guy,
Thanks for your fast reply.
Am I right in thinking a 20% loss in flow will result in a 20% loss in power?
Ben

[Guy Croft]

hi Ben

the peak bhp power loss - compared with an optimised sidedraft setup will not be so much, maybe 12-15% less.

(This is loosely based - working backwards - in my exp - that 'head flow gains of 10% bare port flow - no valve - will typically yield about 7% power increase').

Sorry I can't be more specific Ben, that's where dyno tests back-to-back come in.

GC

[benlilly]

Even though it is less, this is still alot.
I'll get a drawing of a new manifold drawn up in the next few days.
How much of the runner do you think I should keep in line with the heads port flow before bending down to clear the underside of the bonnet?
I was thinking 100mm would be ok as long as the bend starts slowly and the radius is as large as possible.

Ben

[benlilly]

Something like this...[/img]
Do you have any recommendations for a company that could make this for me?