Two different manifolds to choose - which one?

Road-race engines and ancillaries - general discussion
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Re: Two different manifolds to choose - which one?

Post by Guy Croft »

Some good posts. But - can we just get this quite clear:

The chord radius is measured to the centerline of the pipe.

Please confirm the 2 x chord rads Tommi!


G
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Re: Two different manifolds to choose - which one?

Post by Guy Croft »

Brit01 wrote:

How much loss from a 90 degree bend I wonder.

Might be none if the chord rad/pipe dia ratio is 2:1 or more, ie: not adverse bends.

Mind you, length is a factor - the longer so the higher the frictional loss of the air moving over the runner surface. Internal area can compensate to an extent - the bigger (not all are of circular section) so the lower the frictional loss, though in saying that I emphasise that in design one can't ignore a commensurate reduction in air velocity - which is why some well-developed engines with huge ports and runners employ variable valve timing (eg: Honda Vtec). The worst kind of inlet manifold would be small internal area and long with adverse bends.

Inlet overall length from cylinder to atmosphere - as you surmised Chris, has a major influence on the torque curve. But for a given overall inlet length you can have long rampipes or long runners - makes no difference. For a given engine if you shorten the inlet the torque will come in higher up and vice-versa. However, lengthening the inlet to develop wave capture at lower speeds will naturally increase the viscous loss.

This is all pretty generalised stuff. Frankly it can only be proved out by accurate back-to-back dyno testing. That said I have done some of that and of course tons of flowbench work which so my assertions come from both sources.

G
TomLouwrier
Posts: 333
Joined: July 28th, 2010, 3:09 pm
Location: Leiden, The Netherlands
Contact:

Re: Two different manifolds to choose - which one?

Post by TomLouwrier »

hi Guy

I agree completely with you on chord radius, that is what we're after.
However it is not something you measure directly. It follows from either SSR or the opposite (Long Side Radius?) and half the port diameter. These can be measured easily.
That and some possible misunderstanding about definition of radius and chord radius led me to adapt the picture of the plasticine sample.

regards
Tom
GC_29
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Re: Two different manifolds to choose - which one?

Post by Guy Croft »

yes, + runner wall thickness...

To be honest the best way to assess the chord rad is to bend a piece of wire to approximately the chord shape and set it out on a piece of paper. Little point in chasing an olympian standard of accuracy.
G
Urbancamo
Posts: 317
Joined: August 8th, 2011, 1:04 pm
Location: Finland
Contact:

Re: Two different manifolds to choose - which one?

Post by Urbancamo »

Ok, i think i got it. Foreign language with technical phrases got me little wrong. I was measuring something that had nothing to do with hole thing...

I found that the chord radius is the same at both ends.
IMG_5588 (Custom).JPG
IMG_5588 (Custom).JPG (39.18 KiB) Viewed 6991 times
Cutted many different size of circles on the cardboad and this size fits perfectly on both ends. It forms similar curve than stiff wire i put on the channel. Thanks Guy for tip.

Unfortunately forgot to take pictures.

This is done by eye comparing the shape with outer casting. In the middle of course. I also compared this circle to plasticine mould and it fits nicely. I think it's reasonably close. I dont think that few millimetres would change anything in this situation?

So radius is 50 mm in both end i assume?

Port diameters are still:

- carb side 41-42 mm
- inlet side 35 mm

I also drawed both SSR's in paper with plasticine moulds. These are precice if someone is. These might give and idea of port shape.
Inlet side SSR. Its milder than casting shows. Scale is millimetres.
Inlet side SSR. Its milder than casting shows. Scale is millimetres.
IMG_5610.JPG (47.97 KiB) Viewed 6991 times
Carb side SSR.
Carb side SSR.
IMG_5595 (Custom).JPG (40.19 KiB) Viewed 6991 times
Attached manifold to spare head to demonstrate how it fits.
View from the back. Casting shows manifold's port shape very clearly.
View from the back. Casting shows manifold's port shape very clearly.
IMG_5614 (Custom).JPG (40.74 KiB) Viewed 6991 times
From the top. Notice little offset differences between cylinders. Of course the straight-shot manifold has an offset differences too, as seen in previous pictures.
From the top. Notice little offset differences between cylinders. Of course the straight-shot manifold has an offset differences too, as seen in previous pictures.
IMG_5633 (Custom).JPG (63.39 KiB) Viewed 6991 times
Fiat TC Alquati offset manifold. Firstly offset with cylinder head channels and then huge 90 degree bend. Can these compared to swan-neck at all?<br />These TC manifolds do work nicely in mildly tuned TC engines.
Fiat TC Alquati offset manifold. Firstly offset with cylinder head channels and then huge 90 degree bend. Can these compared to swan-neck at all?
These TC manifolds do work nicely in mildly tuned TC engines.
Kuva103 (Small).jpg (36.68 KiB) Viewed 6991 times

So, you tell me your opinions is this "bird-neck" manifold hopeless for flowing air or will it suit to street use. Thanks.

T
GC_25
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Re: Two different manifolds to choose - which one?

Post by Guy Croft »

Clearly the chord radius - runner diameter is adverse in this case, far less than two. The thing has a certain length which might - for a given cam and header primary pipe length work reasonably well but the bend losses will really rob performance vs the 'straight' manifold you also pictured at the beginning of this thread.

Here is the additional point about flow direction I mentioned earlier. Air likes to take straight lines (unlike water) and flow straight for the center of the inlet valve. Literally! I've observed this many times on flowtest by inroducing a cotton trace - the trace is always pushed into the fastest - low pressure - air and on a poorly laid-out port it will show a marked reluctance to go anywhere else. A port with good flow distribution ( = a high flow port) that simple test gives a very different result.
The ideal port shape is round and straight!

Of course you can constrain air and make it follow a particular direction (usually to get the manifold to fit into the engine bay etc) but you will get losses. These losses are caused by turbulence at the bends, 'drag' if you like, just like on the trailing edge of an aircraft aerofoil section if the contour is too agressive to ensure adhesion of the airstream. This is where the bend/dia ratio comes in. If the ratio is beneficial (which is not the case on the swan-neck item) there is no bend loss at all and I have tested such types.

By losses - what I mean is if the head - for example - flows 80cfm @10" bare port flow (no valve in) and you bolt the manifold on it is going to generate a bolt-on loss. Any loss of flow on test will mean a reduction in engine power, believe me; that is one immutable rule of flowbench testing.

In the case of the swan-neck the exit path of the manifold does point right down the downdrafted port but whilst the air will try to flow along the blue line the changes of direction forced on it by the casting will completely disrupt that airstream - yellow shading indicates turbulent regions. The tighter the inside bend (short radius) the worse the problem - air will not follow the radius contour it will separate from it. If that resulting turbulence (as it's known) is violent enough (and it gets worse with speed) it can completely disrupt the flow in the manifold - up to a 25% loss of flow - or more! You see this in the ports themselves on some heads - especially sidedrafted units with a big drop from the port floor to the valve and thus a very long curved short radius - like the Peugeot SOHC series (205/105 etc).

In the case of the IDF manifold illustrated above it's a slightly different matter as the pictures below illustrate. This model (and yes they vary) has a generous short radius and given the chance - as here - the air will follow it quite well. Near the center the bias of the airstream is at an angle - if allowed, the air does not enter the manifold perpendicular to the carburettor mounting flange. Thus when you bolt on an IDF carb you are constraining the air to flow in a 'non-preferential' direction!

Trust the pictures of the flow directions and bolt-on losses tell the story well enough..

G

G
Attachments
IDF flowtest_bare manifold easily outflows the 110cfm head.jpg
IDF flowtest_bare manifold easily outflows the 110cfm head.jpg (114.8 KiB) Viewed 7080 times
IDF flowtest_at short side air follows generous contour.jpg
IDF flowtest_at short side air follows generous contour.jpg (110.44 KiB) Viewed 7080 times
IDF flowtest_awayfrom the short side the air wants to head straight for the valve.jpg
IDF flowtest_awayfrom the short side the air wants to head straight for the valve.jpg (108.75 KiB) Viewed 7080 times
IDF flowtest-bolt on a 40 IDF and the constraint of on the airstream shows up markedly.jpg
IDF flowtest-bolt on a 40 IDF and the constraint of on the airstream shows up markedly.jpg (112.85 KiB) Viewed 7080 times
IDF flowtest_bigger carb & bigger choke but the loss is still there.jpg
IDF flowtest_bigger carb & bigger choke but the loss is still there.jpg (115.17 KiB) Viewed 7080 times
IMG_5534.JPG
IMG_5534.JPG (51.71 KiB) Viewed 7080 times
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Re: Two different manifolds to choose - which one?

Post by Guy Croft »

Oh - and re your question as to 'OK for street use?'

absolutely no idea!!!!

G
WhizzMan
Posts: 459
Joined: August 13th, 2010, 8:05 pm
Location: Amsterdam, the Netherlands

Re: Two different manifolds to choose - which one?

Post by WhizzMan »

If this was my car, I'd insulate the exhaust to keep the carbs and air intake cool. It may be more work initially, but from the outcome of the discussion, I wouldn't want to lose so much potential flow with the swan neck. If at all possible, you could consider making a bend in the intake trumpets to lengthen them, if you want them longer for low RPM efficiency. It won't be easy to find those trumpets, but I know they exist because I've seen them used on racing Alfa Romeo cars.

This is the point where you have to decide to take the "easy way", or do the extra work to get more potential benefits. As Guy has said already, the only way to find out what actual benefit there is to be had, you need to flow bench both setups. With the combined common sense and experience of the forum, you can only get a (very good) estimation of where you will lose flow.

Regarding the Alfa boxer 90 degree inlet, there is another thing to consider here. These inlets are similar to the VW beetle inlets. They use the reflective port waves from the opposite cylinder inlet to push air into the inlet port. Replacing the manifold with straight velocity stacks may improve high RPM power, but will rob a lot of power from lower regions. For street use, I would not be considering this. The inlet manifold of those boxers is well calculated and designed and you would need to put a lot of effort in it to really get a better setup for track use if you decide to replace them with velocity stacks. You could look into polishing the inside with a flapper sanding bit and a flexible axle, as far as you can reach from both ends? For a street setup, I don't see a lot of benefit in this, but if you have a spare inlet, you could try just because.
Book #348
Urbancamo
Posts: 317
Joined: August 8th, 2011, 1:04 pm
Location: Finland
Contact:

Re: Two different manifolds to choose - which one?

Post by Urbancamo »

WOW!

This sums everything I could ever imagine when i started this topic. And maybe more. And all this info in very easily readable form.

So I'll bought the straight one for sure. Even my head is not superb, I don't want to ruin it. You never know what kind vortex there will be with swan-neck...

Guy; thanks. You are the true master. And I mean it.

Please feel free to discuss about these things.

Happy Tommi
GC_25
Brit01
Posts: 825
Joined: June 28th, 2011, 4:54 pm
Location: Uruguay

Re: Two different manifolds to choose - which one?

Post by Brit01 »

Regarding the Alfa boxer 90 degree inlet, there is another thing to consider here. These inlets are similar to the VW beetle inlets. They use the reflective port waves from the opposite cylinder inlet to push air into the inlet port. Replacing the manifold with straight velocity stacks may improve high RPM power, but will rob a lot of power from lower regions. For street use, I would not be considering this. The inlet manifold of those boxers is well calculated and designed and you would need to put a lot of effort in it to really get a better setup for track use if you decide to replace them with velocity stacks. You could look into polishing the inside with a flapper sanding bit and a flexible axle, as far as you can reach from both ends? For a street setup, I don't see a lot of benefit in this, but if you have a spare inlet, you could try just because.


Yes I have the Alfa boxer with the 90 degree outlet and I don't want to lose the power at lower rpm. Long trumpets won't fit under the bonnet unfortunately.

If these Alfa air intakes will be any use to you I know someone in UK who might have some. Let me know.

This was another design which gives more surface area to suck in air as opposed to the panel filter I saw but would suck in warmer air.
Attachments
Modified filter set up.
Modified filter set up.
alfaenginewithkns.jpg (31.86 KiB) Viewed 7053 times
my current set up with K&amp;N panel filter inside.
my current set up with K&N panel filter inside.
airbox.JPG (108.18 KiB) Viewed 7053 times
Urbancamo
Posts: 317
Joined: August 8th, 2011, 1:04 pm
Location: Finland
Contact:

Re: Two different manifolds to choose - which one?

Post by Urbancamo »

Will be back to swan neck a little bit.

I did some "flow testing" with home made "bench" where i used vacuum cleaner. I don't know if this is right or wrong. It makes some vacuum, though very little. I tested only the carb side.
IMG_5678.JPG
IMG_5678.JPG (37.9 KiB) Viewed 6887 times
There is only one little spot in the middle where the string goes straight to center. Shown in picturen
IMG_5677.JPG
IMG_5677.JPG (34.52 KiB) Viewed 6887 times
All other directions, especially little closer to SSR side, violent vortex. String bounces and flaps around like mad.

I'm not real familiar with these, but this obviosly shows that this manifold is not really good as shown earlier in this topic.

I heard a local guy who has real flow bench, it would be nice to see how much this manifold affects the overall flow. Maybe i go to him, or not. Will see.

T
GC_25
Urbancamo
Posts: 317
Joined: August 8th, 2011, 1:04 pm
Location: Finland
Contact:

Re: Two different manifolds to choose - which one?

Post by Urbancamo »

We are back with manifolds.
IMG_5717.JPG
IMG_5717.JPG (57.95 KiB) Viewed 6867 times
So i forgot swan neck-type and bought this straight one. It's 11 centimeters or 4.3 inches long. Pretty decent length.
IMG_5720.JPG
IMG_5720.JPG (39.52 KiB) Viewed 6867 times
It has little more downdraught angle than head has. Measured 35 degrees. Cylinder head inlet port angle is approx. 18 degrees.
IMG_5718.JPG
IMG_5718.JPG (31.67 KiB) Viewed 6867 times
IMG_5724.JPG
IMG_5724.JPG (22.57 KiB) Viewed 6867 times

This has VERY light curves in both ends, definetly much better than swan neck.
GC_25
Post Reply

Who is online

Users browsing this forum: No registered users and 31 guests