exhaust manifold advice

[fingers99]

I'm looking at building a small run of Fiat Cinquecentos with the 1.4 16v normally aspirated engine. There's nothing original in the build which is very well documented. With inlet and exhaust manifold and aftermarket ECU, these will put out 110bhp with factory reliability. In these shells this results in close to 160bhp/tonne which is enough to be very interesting, but to present no insurmountable problems for uprated brakes, tyres and suspension.

Only two manifolds fit this build (everything else puts the primaries through the slam panel). One is a £900 4 into 1, the other a £400 4 into 2 into 1. Naturally, I picked up a second hand 4 into 2 into 1. I recon I can make these for a little less than the retail price and to a better quality. Space, however, is so limited that I doubt the flow characteristics can be significantly improved and, even if it were my intention, these are fairly tricky engines to tune, because of hydraulic valves, cam in 1 piece head design and a reported predisposition for throwing rods if pushed past 140bhp.

So, this basic design of exhaust manifold is acceptable because it fits! But no-one wants to throw away power.

Sizing up the manifold, the primaries are 35mm, the secondaries are 45mm and the tertiary is 45mm. This last dimension caused me some concern as 2 45s into 1 45 is a huge restriction. Off the top of my head I'd have expected 50mm, especially as most folk seem to run 2" exhausts.

Any comments or ideas welcome.

[Guy Croft]

Let's have a photo of the engine and the header please.

G

[fingers99]

I'll grab some -- they'll be of other people's cars, and other people's photos, so I'll need permissions.

Back as soon as I can.

Thanks.

[fingers99]

Some photos.(Guy, I've emailed you the permissions seperately.) I'm still waiting for the owner of the most telling photo to get back to me, but these
should give some idea of just how tight the packaging is.

I ought to mention that one well respected builder of manifolds and exhausts attempted to produce an alternative manifold for this kind of build, but
ultimately failed. In the end, I think the issue was one of equipment. I think I have someone that can do the bends, but if that fails, I may have to get down to
hygenic bends or hydroforming.

The engine is the 843 A 1000 motor, as becoming almost omnipresent in mid range FIATs, found in Stilo 1.4 16v, Panda 100hp, Grande Punto, 500 but also in some
Idea and Multipla models.

Plenty of discussion about the 1.4 16v Sei and Cinquecento builds on Fiat Forum and on the Polish forums, too, if anyone fancies a look.







These 3 photos thanks to Arc (Mark Simpson). This manifold has been treated; although, perhaps because, they're in 304 stainless, they look pretty unpleasant after a couple of years of road salt. I want to use 316.

Dan's (Daniel Hollingworth) photo begins to show just how tight the fit is:



My thanks to Arc and Dan for kind permission to use their photos.

[Guy Croft]

Can't see anything about the 4-2-1 that concerns me. Can't see much of the 4-1. Nice of the owners to let you use the photos and thanks for going thru the process of asking - which I know you did because you kindly wrote me by email..!

Insofar as what you said about sizes:

"Sizing up the manifold, the primaries are 35mm, the secondaries are 45mm and the tertiary is 45mm. This last dimension caused me some concern as 2 45s into 1 45 is a huge restriction. Off the top of my head I'd have expected 50mm, especially as most folk seem to run 2" exhausts"

45mm tailpipe is going to be adequate for your 16v 1400 atmo units in any state of tune. Using a bigger tailpipe would not, in my opinion, yield more performance, more likely lose it.

My rule of thumb is sy should be 1.25 x bigger on dia than py so the py-sy dia ratio is fine. Not completely sure about the py diameter, read on.

You could alter the design if you want to, I can give you some good guidance - but first let me know the py and sy chord lengths. Study the diagram below that tells you what these are. Also tell me the ex valve size and cross-sectional area of the exhaust port.

[fingers99]

Thanks Guy, will do. Waiting in for an engine to arrive as I write this.

[fingers99]

I've been unable to find a photo of a 4 into 1 system -- it looks like no-one has bought one!

But Jules Toye ( "Pirus" on Fiat Forum) kindly sent me a few photos of another build. These show up the clearance issues better than any of the others I have.







The area of the exhaust port I calculate as 401 sq. mm -- the shape is a rectangle with circular ends, something quite strange to one used to peering down 8 valve exhaust ports!

The exhaust valves are 22.5 mm (on some models, fractionally smaller at 22.4 according to parts manuals).

SY lengths are consistent at 145mm (o/d is 45 mm)

PY lengths are not as consistent as I'd like to see, 295mm would be an average, but they vary between 308 and 278mm. I think I'd like to use a measure length, bend, equalise volume, weld method if I ever do make some (o/d for PY is 35mm).

Thanks again.

[Guy Croft]

The area of the exhaust port I calculate as 401 sq. mm -- the shape is a rectangle with circular ends, something quite strange to one used to peering down 8 valve exhaust ports!
The exhaust valves are 22.5 mm (on some models, fractionally smaller at 22.4 according to parts manuals).

Those valves will have a throat dia about 3mm less than valve size giving a combined valve throat area of 597sq mm. This is the ex controlling section - the ex port is a bit on the small size at exit by comparison. Of course one would have to measure the valve throat. Some might - in this case - make the py bore equivalent to the port exit area but actually It is often a help to have the ex primary bigger than the port exit region. I do not know the precise reason but it's true. So maybe go for a primary pipe diameter 1 or 2mm bigger all the way round. You should sketch that up and see how the size looks. A single pipe bore of 597sq mm would have dia of 27.6mm. If that leads to something that's actually smaller than the port - then it needs to go bigger.

SY lengths are consistent at 145mm (5.7") (o/d is 45 mm).
PY lengths are not as consistent as I'd like to see, 295mm (11.6") would be an average, but they vary between 308 and 278mm. I think I'd like to use a measure length, bend, equalise volume, weld method if I ever do make some (o/d for PY is 35mm).

OD means nothing - it's the ID we're interested in!

Your approx combined py & sy pipe length is too short for my liking at 17.3". If you can do it lengthen the py pipes to 17" and keep the sy at 5.7". Once you have determined the py bore dia make the sy 1.25 x bigger. Keep the tailpipe the same dia as the sy whatever that is. The chord lengths really DO need to be equal to within about 5millimeters.

G

[fingers99]

Thanks Guy.

OD means nothing - it's the ID we're interested in!

Very true, my brain must have switched off! I don't want to section this mani, but remembered I've still got an old, damaged 8v manifold from the same manufacturer which I had to section to get the dings out (should have tried a hydraulic pump, but we live and learn). I think it's a reasonable assumption that they use the same thickness tube in their manifolds, at least, in naturally aspirated manifolds for similar capacity engines. So, on that basis, secondary i/d is 42mm, primary i/d is 32mm.

The rest shouldn't present any insurmountable problems (or, rather, the theoretical considerations you raise shouldn't), although the practical ones really come down to being able to get the bends done correctly and being able to orient them in the right position: very tricky!

I'll weld up a jig to make sure that this manifold cannot enter into already occupied space and that the tertiary pipe exits in the right direction and at the right angle, then order up some parts.

I'll keep you informed of any progress, but won't post up lots of photos of part completed manifolds. Nothing could be more boring than looking at a series of TIG welds, and there are better welders out there than me.

Many thanks for all your help.

[Guy Croft]

"Nothing could be more boring than looking at a series of TIG welds, and there are better welders out there than me"

a perfectly reasonable disclosure, but you should find the best coded welder you can to do the full-form weld, though you can tacking it for him in your jig.

Otherwise well-made headers often fall-down due to inexpert welding and lack of flatness on the flange. A header with those features impaired is, well, a piece of junk, really.

G

[fingers99]

Oh, I'm good enough to get the flanges flat (and wise enough to pressure test), no worries there! ;)

It's just that I look at the stuff done by people who are really good and know I'll never get there. Mind you, I look at stuff done by people with fairly high reputations and wonder why people pay money for such tat.

I'm mister in-between. :)