Advice on header connections.

[Luddite]

I found some pictures of Sil 900SS headers, which might have the same fixing method as yours.



These seem to have a small collar with the same ID as the OD of the header held within the exhaust port by the flange (just like Dukedesmo describes). The flange has two holes for the springs to hook onto.

These pictures show the collars and flanges in place on the headers and it looks like both loops for the two flange springs are on the same side, like you describe.







So it sounds like your suspicion that you may have some pieces missing might be correct.

Hope that helps.

[utopia]

Thanks for that, mate.
Helpful and informative, as ever.

Yep, that 900SS system is exactly like mine is at the engine end.
The loops for the springs are in exactly those positions.
So it does seem that I am missing a pair of stubs.
Well, who'd have thought it.
When I started this thread I had no idea that my system was incomplete.
I guess there's something (beyond the obvious) in the saying .. "you don't know what you don't know".

[Darren69]

A bolted up stub with springs would be the best option IMHO. Both of my 4 Valvers use that and Ive never have any leak issues and there is some inherent flexibliity to get things lined up easier.

If you've ever had to fit a 3 or 4 cyl inline system to a bike with just the bolted up headers with half rings you will know where I'm coming from! Getting them to all line to the collector up is a right skinned knuckles pain.

[Dukedesmo]

Originally Posted by utopia

Yes please, that would be spot on.
I think I can visualise the stubs and how they're clamped from your description but a pic or two would definitely be helpful.

The system appears identical to Luddite's pics above.











As you can see the short pipe bolts solidly into the head with a gasket and the pipe slips in to be held in place by the springs. This allows some movement for heat, vibration etc, it also makes it easier to remove as you only need to unhook the springs and the pipe pulls out - OK rear exhaust is a bit tricky due to access issues but the principle is good.

If you want any measurements let me know or if you want to get your eyes or hands on it for any inspiration, you know where I am.

[utopia]

Brilliant, mate. Thanks a lot.

So, it seems that I need to make some stubs.
I'm still uncertain whether to make these to be welded on to the system and bolted to the heads in the std manner, or whether to use springs .. but the springs are gaining ground as the likely option.
One issue that concerns me somewhat is the change in diameter from the exhaust ports themselves (around 38mm) to the 42.5mm bore of the header pipes.
I would like to arrange for the bore of the stubs to be tapered, to smooth this transition.
I don't yet know whether this will be possible while still being able to use springs.
I may have to design stubs that will fit inside the header tubes rather than outside .. and this may require the stubs to be welded to the headers. I dunno yet.
Maybe it can be done by just using some tapered reducers welded into the bore of the headers, and then simply copying the MADASL /SilMotor system.
Or better still, I might be able to take account of the reducing function in the design of the stubs and avoid the need for any welding at all (in that area at least).

I have now got my eye on an offcut of 316 stainless round bar on ebay.
Its 45mm dia ,, which I reckon is probably just big enough to machine some stubs from.
It might be useful to know the outside dia and overall length of your stubs Tim.
I can probably work out what's required, but some dims would clear the fog a little more.

[Darren69]

That MADASL system looks sweet. If I had a 2V thats what I'd be fitting. The issue with 2 into 1 systems is that you may lose some midrange, high level may get some back as the twin hi level does appear to improve the midrange torque.

I was reading recently some article or other that when Ducati used a 2 into 1 on the 999/749 bikes they experimented with pipe diameter on the rear cylinder to get the same tuned length on both. IIRC I think they ended up with 50mm front and 57mm rear in order to achieve the same tuned lenght for the Desmo valves to work at they're optimum. Food for thought, maybe?

[Dukedesmo]

Originally Posted by utopia

It might be useful to know the outside dia and overall length of your stubs Tim.
I can probably work out what's required, but some dims would clear the fog a little more.

Stubs are 20mm long whilst the exhaust port 'hole' in the 900 head is 18mm deep and gaskets are around 2.5mm so the stub sits 4 - 4.5mm proud of the port.

OD of the stub is 48mm as is the ID of the port (give or take for loose fit clearance) and stub ID is 45mm - same as the OD of the pipe so that it sits nicely inside.

Original Ducati pipework is 40mm OD and the ID of my 45mm pipe is 40mm so perhaps if the Sil is the same you could use that to help with a 'slip' fit using a combination of parts from the Sil and standard exhaust?

Regarding a tapered system, with Desmoquattro Ducati it is generally considered the done thing to start out small and go larger down the pipework rather than fitting larger headers: Hence the 45mm - 50mm system on my 916, as large headers kill the back pressure and lose power, at least in the lower/mid rev range which is where Ducatis are tradionally strong.

As such big systems are normally only used in race conditions where top end is more important - everything else being equal 54mm systems make less power on the older bikes.

Presumably similar applies to the 2v and that maybe why most aftermarket systems for them appear to be no more than 45mm?

Also smaller displacement engines might suffer more from 'over-exhausting'? That said going to a single outlet may well redress the balance? Don't know how much research has been done into the pros/cons of size vs. number of pipes on 2v Ducatis?

[utopia]

Brilliant.
I have all the info I need for now .. now.

I was off out to the shed to have a further tinker but I had a slight distraction.
The high winds had blown two collared dove chicks and their nest clean out of the overgrown leylandii which towers over my sheds.
Yes, thats right .. two chicks in january.
Damn near fully fledged too .. they must've been born around christmas (sorry Gazza .. solstice) time.
Randy buggers, those collared doves.
Anyhow, out with the ladders and I managed to wedge the nest back in what seemed like a suitable position and replaced the chicks, still nice and warm, therein.
Fingers crossed.

Anyway, yes I agree with much of that thought on header design.
Though equally, I'm not sure exactly which way it cuts in my particular situation.
I was wary of the big bore headers and I was considering (and still can) fitting std size headers and blending to the bigger bore further along the system.
I may still do that in the end but, once I had the system in my hand, it seemed sensible to do minimum cutting on it for now, thus leaving maximum potential for later on.
Kind of "suck it and see while burning the fewest bridges".
In practice, it turns out that's the easiest way of adapting the overall dimensions of the system to fit my 750 anyway.

I'm gradually becoming convinced that, whatever the theory, there is a lot more than just theory involved in exhaust design.
Packaging on the bike is obviously one serious constraint.
Aesthetics is another, for the manufacturers at least, and in my case also (although I tend strongly towards the idea of form following function).
I've already mentioned issues around ease of maintenance.
Another thought might be that, given the 4kg weight saving by losing one of the cans, the overall performance of the bike might be increased despite the engine making less than its maximum performance potential.
Either way, the lighter system will score in the handling dept and also wins in the simplicity stakes.

The plan is to dyno test the first design and take it from there.
Tbh, I might start by refitting the Termis and dyno testing that as a benchmark.
Then after the exhaust is done the FCRs will go on .. so more dyno testing will be needed for those.
Its tempting to do exhaust and carbs at the same time but, although it would cost a lot more, I will learn a lot more from testing it at each stage.

[Flip]

Jeff, I 're have a pre-Ducati Performance system made by Gia.Co.Moto and that too is all held together with springs using little stubs that bolt into the head in the normal way with the remaining sections spring linked.

It's off my bike at the moment so if you want me to send you up the stubs to use as a pattern you're more than welcome.











I did think a while back about using it as a pattern to get a couple of systems made up to sell if there was enough interest.

Oh and what you might possibly loose by using the 2 into 1 system I think you will gain back by doing away with the restrictive squashed 'X' collector. By doing away with that on a 900 motor it certainly feels more peppy in the mid range and can apparently add around 5bhp .

[utopia]

Excellent stuff, Flip.
And thanks very much for the offer.
I'll see how things go.

Your system is interesting actually.
Am I seeing it right ? ... It looks like your stubs fit INSIDE the headers rather than outside .....?
That's the direction that my thoughts are going at the moment.
In fact I'm currently having a bash at machining up some adaptor sleeves that will enable me to do a mock-up assembly using internal-fitting stubs which are basically the cut-off ends from std headers (with further sleeves to adapt their 40mm OD to the 42.5mm bore of the Sil headers).
These are being made from the aforementioned rusty scaffold tube.
With any luck, using these I might even be able to fire up the motor briefly with the system reasonably securely fitted.

If your stubs are fitted inside the headers, could you measure the bore for me please ?
I'm wondering if they've attempted to blend the diameters of exhaust port and pipe as I mentioned before.

And yes, I think you have a point about the benefit of losing the X collector.
I'd forgotten about that.
Putting it all together, I reckon I would be unlucky to end up with a worse system than my current one .. which leaves the weight loss and simplicity benefits as gains.
Best case scenario might even be a few more bhp and better mid-range torque .. but tbh, my holy grail is only a meagre 70 bhp.
That's enough on the right roads .. but I do like light weight and a nimble chassis.

Where's the fingers crossed smiley face ??

[Mr Gazza]

Flip, we talked about your exhaust system quite a while back and you mentioned making a few sets then. Just to say I'm still interested if I am flush when you are ready to have some made. They look just the ticket, but I'm wondering why they are made in sections? surely they could be fitted in one piece, especially the front pipe.

Jeff, It might be useful to know about Every Exhaust Part, they might have some get out of jail free cards for your project? http://www.everyexhaustpart.com/cart...nds-and-elbows

[utopia]

Thanks, Gazza.
Though they were already on my radar.

Well, a 100mm long offcut of 50mm dia stainless steel popped up on ebay this morning.
That gives me a bit more room to play with, so I've nabbed that to machine some stubs from.

Last night I successfully machined one temporary adaptor stub from the rusty scaffold tube.
That's on the bike now, all clamped up to the head and with the header installed on springs.
Its fairly pukka.
Today I will attempt a slightly more ambitious version which will include the tapered blending between port and pipe diameters.

Turns out that the same rusty scaffold tube will also yield some temporary connector sleeves to reunite the system where I have cut it in half, to deal with the damage.
Furthermore, said damaged section turns out to have the same bend angle (and pipe dia/gauge) as one of the old link pipes that came from Capo's S4R .. so that will probably be drafted in. Thanks again, mate.

This is turning out to be a proper, forum based project now.
Which makes it even nicer.

[BigOz]

If your going for the weld on stubs it would be worth thinking about where they are going to be welded. If you weld around the back of the stub you could compromise the area the flange/clamp clamps down on and that could give uneven pressure on the sealing face.
Welding round the inside should be possible though tricky if the weld is too far up the tube.

Oh and 316 from your e-bay find to unknown possibly 304 tubing you want to specify ER309 or ER312 filler for that, if it is known 304 then go for ER316 filler.

[utopia]

Thanks, Oz.
For the moment, it looks like it'll be spring-fit stubs, at least for the first build and probably the final one too.
If I do go for welded stubs, they'll possibly be made from stubs cut from std monster headers, then sleeved (ie telescoped) up to full dia using a short, bridging tube of interim dimensions.
This will require two welds, but the end one will still be around 20mm from the flange, so hopefully far enough away.
And with luck, although I don't know exactly what material the tubing is, it should hopefully be all the same .. and weldable using normal exhaust-welding kit.

The ebay offcut is actually described as F51 grade, which is apparently a "duplex, 50/50 ferritic/austenitic grade.
Bit of a new one on me, that.
I'm not expecting to have to weld it though, as I plan to use it for the spring-fit stubs.
But I may change my mind. Do you know if this stuff is easily weldable ?

[BigOz]

Google tells me F51 is "UNS S31803 F51" and it is "readily weldable" looking at the composition it is similar to 2205 for which information is quite widely available but mostly covers it being welded to the same grade.

Welding to austenitic stainless such as 304 I have found no references yet so if I had to go blind I'd recommend Hastelloy W or Inconel 625 as a filler as these are good mismatched or unknown steel/stainless metal fillers. The issue with those is you really don't want to pay for either of them in small diameter filler, the price is excruciatingly high and they are difficult to get in small quantities in the UK.

Just checked 1lb packs are available in the US.