HD8 carb issue

Technical advice Q&A

christopher storey
Posts: 5698
Joined: Sun Mar 09, 2008 3:07 pm
Location: cheshire , england
Great Britain

#61 Re: HD8 carb issue

Post by christopher storey » Sun Feb 02, 2020 6:48 pm

John : I do not know where Tom gets his 120 degrees of intake duration from . The actual duration on these engines is 252 degrees for both inlet and exhaust and at least 2 cylinders are always sucking at a time, and as you say the balance passages mean that whether you have 2 or 3 carburetters , there is always continuous flow in the inlet tract - its intensity may vary cyclically ( because of course the 252 degrees encompasses the periods when the valve is only partly open - but that is all . ( I am also ignoring the effects of inertia, reflected waves etc which also tend to keep airflow going even in one direction or another even when a valve is not open ) . And - the plain fact is that the engine has to be supplied with a given volume of air , whether through 1, 2 or 3 carburetters and the degree of suction will vary in inverse proportion to the area of the smallest part of the inlet tract available . Hence the fact that 2 carbs suck harder at the venturis than 3 for a given engine size

Link:
BBcode:
HTML:
Hide post links
Show post links


Tom W
Posts: 697
Joined: Sat May 05, 2018 10:18 am
Location: UK
Great Britain

#62 Re: HD8 carb issue

Post by Tom W » Sun Feb 02, 2020 7:20 pm

Yes, the statement on the intake alternating between the front and rear carburettor applies in a scenario where the front 3 cylinders are fed by one carb, and the rear three are fed independently by the other.

With the firing order 1.5.3.6.2.4, you can see that the intakes alternate between the front 3 and rear 3. A standard XK engine has an intake duration of 252 degrees (I think), so there’s 12 degrees of overlap between induction strokes through any one carburettor on a twin carb engine. So in basic terms both carbs are drawing all the time as there’s always at least one cylinder they fill that’s on the induction stroke.

If your manifold is one large chamber, then the air will probably flow through both carbs to fill it. The air in the manifold will be below atmospheric pressure, and the air in the cylinders on their induction strokes will tend to be below manifold pressure. So there’s a natural flow from atmosphere to manifold and from manifold to cylinder. The induction stroke will still alternate between the front 3 and rear 3 cylinders, so the volume of air in the manifold isn’t necessarily at the same pressure along its whole length. Add to that, the effects on pressure of cylinders that are just about start their induction stroke, and those that are just finishing, and the pressure in the manifold becomes very complex. It’s further complicated by overlaying the pressure waves that are being and reflected around the manifold (organ pipe theory). I don’t pretend to understand enough of this to fully understand, let alone explain what’s happening. But you can rest assured both Jaguar and Austin Healey development engineers did understand this, so their choices of manifold are both likely appropriate for their intended engine, performance targets and cost parameters.

If you want some heavy bedtime reading that will explain it all far better than I can, there are a couple of books I can recommend.
Tom
1970 S2 FHC

Link:
BBcode:
HTML:
Hide post links
Show post links


Tom W
Posts: 697
Joined: Sat May 05, 2018 10:18 am
Location: UK
Great Britain

#63 Re: HD8 carb issue

Post by Tom W » Sun Feb 02, 2020 7:36 pm

I’d already written my reply to John when you posted your reply Christopher, so I’ll answer your points in a separate post.

120 degrees is for the purpose of explanation, not what an xk, or any other engine has for that matter.. I did state it’s ridiculously short.

As I explain above, on an engine with 2 carburettors and the 252 degree intake duration, each carb needs to be sized to flow enough to fill one cylinder at full valve lift, and 2 cylinders at partial lift during the 12 degree period where 2 valves are open at once.

With a 3 carb engine considering the centre carb first. The induction strokes it feeds are 360 degrees apart, but the induction stroke is only 252 degrees, so there’s a 108 degree period where this doesn’t feed its cylinders. It may still see a draw from the balance pipe. The front and rear carbs both feed cylinders that are separated by 240 then 480 degrees. So these see the same single cylinder induction draw followed by the 12 degree overlap as the carbs on the twin carb engine. Difference is, they then have a period of 228 degrees when neither of their cylinders are on the induction stroke.
Tom
1970 S2 FHC

Link:
BBcode:
HTML:
Hide post links
Show post links


christopher storey
Posts: 5698
Joined: Sun Mar 09, 2008 3:07 pm
Location: cheshire , england
Great Britain

#64 Re: HD8 carb issue

Post by christopher storey » Mon Feb 03, 2020 10:17 am

Tom ; the theory is fine but the fact is that with the SU 3 carb systems all carburetters play a part in feeding at least 3, and in the case of the centre carb , at least 4 cylinders. Your theory would be fine if it were not for the fact that there is considerable interlinking between the inlet tracts of each carburetter. It is also a fact that at any speed above idle the inertia of the incoming air once it has been accelerated is quite considerable, and contributes to a ram effect, so that the air does not stop and start in the way you postulate with SU setups on Jaguars at any rate

The difference between the two methods of induction can be heard with Weber side draught systems, when the separate sucks resulting from discrete inlet tracts for each venturi can easily be heard at idle ! Some love it but others think it is dreadful ....

Link:
BBcode:
HTML:
Hide post links
Show post links


Tom W
Posts: 697
Joined: Sat May 05, 2018 10:18 am
Location: UK
Great Britain

#65 Re: HD8 carb issue

Post by Tom W » Mon Feb 03, 2020 2:15 pm

I don't disagree with you. In my first post, I did qualify by saying not exactly true, rather than wrong.

Really, this comes down to the semantics of how the balance pipes are considered in the explanation. In mine, I discount the balance pipe, which isn’t correct, but makes the explanation of intake pulse duration and carburettor draw more straightforward. In your explanation, there is capacity in the balance pipe for any carburettor to feed any cylinder. In reality, the truth is somewhere between the two, so whilst flow through a given carb won’t stop completely, it will reduce during periods where it’s primarily cylinders aren’t on the intake stroke.

This can be demonstrated by the greater influence on mixture strength seen in the cylinders fed directly by any given carb, when adjusting that particular carb, compared to adjusting one of the others.

There’s also the consideration of the duration over which you measure flow. On average, over one cycle of the engine (2 revolutions) then three carbs will flow less per carb than 2 carbs. 6/2 is greater than 6/3. But measured instantaneous, then the difference might not be so great.

Regarding flow stopping and starting, this depends where in the manifold you measure. At the inlet valve, flow must stop and start. At the carb, yes it won’t stop dead, but it’s not necessary constant, particularly with 3 carbs on a 6cyl. There’s a school of thought that suggests the outer carbs should have a slightly different mixture to the centre carb due to the difference in fuelling provided by a more pulsed flow vs a more constant flow. I don’t have my book with me, but will look up later the details on this.

As an aside, I’ve looked at an XK140 manifold. Whilst it looks like one large chamber from the outside, there’s a web inside with a hole drilled through to balance between both chambers. The diameter of this hole is much smaller than the diameter of the manifold chamber.

I haven’t found anything in my reference books yet that gives any calculations on the size and flow of balance pipes beyond the recommendation that for development, an adjustable plate that allows different hole sizes to be tried is a good way to go. I guess when the books were written, it wasn’t an exact science.

Regards,

Tom
Tom
1970 S2 FHC

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

PeterCrespin
Posts: 4561
Joined: Fri Mar 05, 2010 8:22 pm
Location: Gaithersburg, Maryland.
Contact:
United States of America

#66 Re: HD8 carb issue

Post by PeterCrespin » Mon Feb 03, 2020 3:33 pm

“As an aside, I’ve looked at an XK140 manifold. Whilst it looks like one large chamber from the outside, there’s a web inside with a hole drilled through to balance between both chambers. The diameter of this hole is much smaller than the diameter of the manifold chamber.“

I’m glad you pointed that out, as it saves me typing. In the XJ6 manifold I once inspected with a probe made from a wire coathanger believe it was a solid barrier, making two three-cylinder manifolds. If there was a tiny hole that I missed I suspect it must have been cast-in as I couldn’t see any obvious drill access.

Be it there or not, judging (sorry Chris!) by both relative cross-sectional areas and by volume of through-flow, any such hole would be insignificant at anything over quarter throttle, at a guess. In practice, such a hole would probably operate only at idle when vacuum and time separation between individual pulses is highest, with no ram or standing wave filling. Jaguar would have paid a lot of attention to idle refinement, as saloons were their bread and butter.

“I haven’t found anything in my reference books yet that gives any calculations on the size and flow of balance pipes beyond the recommendation that for development, an adjustable plate that allows different hole sizes to be tried is a good way to go. I guess when the books were written, it wasn’t an exact science.”

I have always regarded the connecting pipe as much, if not more, for tapping a steady vacuum source as balancing carb flows.

When Jaguar first fitted triple Webers to the ‘53 C- and ‘54 D-types, they included a large bore balance tube between adjacent carbs but soon dropped it, never to reappear. Since neither car had servo brakes it could only have been carburation-related and nobody cares much about idle on race cars. So we may never know if those cutting-edge induction systems were thus equipped for sound theoretical reasons, or because it appeared to be a logical and helpful addition? We do know that, theory or no, the idea was quickly dropped because it didn’t come back even on the Lightweight Es, AFAIK. XK140s were launched the same year as the D-type connected design a d the XJ6 obviously in the later non-connected era?

Since an intact vacuum tapping by definition does not throughput fuel mixture, or even air, so much as lower its pressure in connected sealed servos, capsules and other blind spaces, I struggle to visualise where and how these supposed mixture flows occur to any great extent?
1E75339 UberLynx D-Type; 1R27190 70 FHC; 1E78478; 2001 Vanden Plas

Link:
BBcode:
HTML:
Hide post links
Show post links


Tom W
Posts: 697
Joined: Sat May 05, 2018 10:18 am
Location: UK
Great Britain

#67 Re: HD8 carb issue

Post by Tom W » Mon Feb 03, 2020 8:12 pm

If anyone wants any further reading on this, I can recommend “The scientific design of exhaust and intake systems” by Philip H. Smith and John C. Morrison. It’s not specifically Jag focussed, but XK engines do feature in a few examples.

Notably there’s an illustration of a Jaguar 6 cylinder inlet manifold. Presumably it’s from a 2.4l mk2 as it has downdraft carbs, though may be from a pre XK 6cyl. The illustration shows the web between the forward and aft chambers, and the bypass hole drilled through it.

There’s also a photo of a sectioned inlet manifold from an E-type. With this comes a description of how the centre cylinders have even ducting, whereas the outer pairs of cylinders have longer ducts for the outer cylinders to compensate for their uneven aspiration period.

There’s also discussion on how with 3 carbs (not specifically Jag) it’s usual to adjust the outer carbs to a weaker setting, on account of the irregular pulses drawing more fuel than the regular pulses on the centre carb.

Well worth reading, and far more detailed and informative than anything I’ve found on the web.
Tom
1970 S2 FHC

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

PeterCrespin
Posts: 4561
Joined: Fri Mar 05, 2010 8:22 pm
Location: Gaithersburg, Maryland.
Contact:
United States of America

#68 Re: HD8 carb issue

Post by PeterCrespin » Mon Feb 03, 2020 10:00 pm

My copy is the third edition.

The 6-cyl manifold is from the pre-war pushrod 2.5 engine and does show a single short drilking between the very close #3 & #4 ports. The basic engine parameters were based on late 1920s Triumph technology so I would be even less inclined to regard it as necessarily representing best practice.

The point about outboard carb runner length and asymmetry is moot, since packaging issues make it practically impossible to fit triple carbs lined up with the front and rear pairs of ports. Aftermarket triple Weber manifolds show this clearly and position the carbs at different distances. The most closely-spaced versions suit the E-type better because there’s so little room for the front carb behind the inner wheel arch and on RHD carbs the rear carb is also short of space.
1E75339 UberLynx D-Type; 1R27190 70 FHC; 1E78478; 2001 Vanden Plas

Link:
BBcode:
HTML:
Hide post links
Show post links

Post Reply Previous topicNext topic