CV Carburettors - how do they really work

Really struggling with this ipad right now....

 

Courtesey of jh haynes publishing, a rather old book.

 

Now, I hear you say, what's this got to do with our xj carbs? Quite a bit really. The big difference is that Mikuni and Hitachi decided they needed more accurate low speed/load mixture control, so added an idle circuit.
To make this work a couple of changes happened - you may see that the SU piston travels to the bottom, effectively closing the venturi off. Our carbs keep it off the bottom. This necessitates a port facing forward on the SU, the Mitachi on the bottom. (I know, couldn't face typing both names over and over).
The piston seal/diaphragm thing you can ignore, it is a mere detail.
So the Mitachi carbs have a different starting circuit, then the idle and progression circuit has all it needs to deliver idle fuel via the idle mixture screw, and, as the throttle plate is opened, the progression holes are uncovered, the manifold vacuum,sucking more and more fuel out. It's interesting to see that both companies only added two holes - Dellorto and Weber, both performance type carbs added 3 or more.
So the plate opens (slowly), and the fuel is added gradually, with air added to the fuel to emulsify it to prevent neat fuel being sucked in.
Eventually the airspeed under the piston (allong with manifold depression) creates a partial vacuum under the piston, drawing fuel out, and lifting the piston, but at the same time maintaining the vacuum signal balanced according to the spring strength. Full throttle and peak rpm should be achieved before the piston gets to the top. What happens to the mixture if the piston goes too high too soon? Weak fuelling, since the relationship between the vacuum and the needle is lost - you get more air, but no more fuel...
Lastly, snap opening throttle. I've allready said that idle and progression fuelling is accurately controlled, and that this happens before the piston lifts off the bottom. Someone somewhere here @ chacal? Has worked out how little fuel the idle jet gives per pulse so I won't go into it further, except to say it's tiny.
So when you snap the twist-grip, what happens?
Firstly, the two progression holes which may have been partly uncovered now see "full" engine vacuum. I say full, because remember our piston? It's not fully at the bottom is it, so these holes only see a partial vacuum at best, but better than nothing. So they add what tiny amounts the idle jet will allow. Next allong the carburettor the piston sees the engine vacuum, but, remember our port in the Mitachi? It's at the bottom, so needs air travelling under it before it lifts. Add to this someone only drilled a small hole in the bottom. Together these two mean that the piston takes a fraction or two of a second before lifting. This delay means that for a short time the vacuum signal below the piston peaks, and this peak is just enough to drag the extra fuel needed for accelleration out of the main jet.
There we have it - accelleration fuelling, without a pump.

 

next up "How a I-Pad really works"

 

Mitachi!

 

I'll answer @chacal s question here, maybe one day someone will read this drivel..


b) EACH of these "progression" holes is thus much larger than the pilot fuel jet that feeds them; of course, the forward-most hole (that the mixture screw would fit into during normal operation) is restricted by the taper of the screw tip.

I found this arrangement to be surprising, and don't quite understand why you would have such a large "total" opening dimension, that could theoretically flow much more fuel than the pilot circuit could provide (although, it seems that fuel would only flow thru the "nearest" exposed hole; but then, why have multiple holes? Once you uncover the "first" (not the mixture screw hole, but the one closest, linearly, to it), what is the purpose of the rest of them?



The simple answer to this is that it's not neat fuel that comes out of the progression and idle mix screw holes, but emulsified fuel - that is fuel mixed with air. The idle air jet actually has a function..

 

Sorry, missed the first bit of this - the "holes" are passed by the edge of the butterfly as it opens, each one adding more fuel than the previous one, until the secondary circuit starts working, and adding to the air and fuel.

 

So...they serve the same function as an accelerator pump. Interesting...

 

You really must stop putting your words to my replies...
Read it again, I described the plate passing the holes and allowing emulsified fuel out, at no point did I say this happened quickly, or at a speed that might be approaching the need for acceleration fuel.
I struggle with your attitude @k-moe , you are an educated person (I believe, yet you are as closed minded as anyone I have ever conversed with - if it doesn't suite you then it can't happen..)
Rather than putting words to my text, tell me why I'm wrong?

 

At what speed would you reckon that vacuum exerts a force upon a fluid (especially in a channel that is already full of fluid, due to the idle speed screw port already being active)? Would it be:

a) more than
b) less than
c) or about the same as other universal forces, such as magnetism, gravity (which ain't, but that's another story..), etc.

Inertia surely plays a role, but in this instance, inertia is already working in "favor" of the fluid.

 

I'm sorry, I don't understand this question? The "speed" I mentioned was the speed of the plate passing the progression holes - and this is a function of how fast you twist the grip. In "normal" riding, say tootling allong in traffic, nomsnap accelleration etc this would be a gentle progression - hence the names for those holes, and actually, where the bike spends most of it's time, unless it's doing trans desert courier or racing work?
Anyway, I would suggest that given the nature and weight of anything coming out of those holes not to be affected by inertia in any way - Bernoulis would be the laws most at work I suspect, after gravity of course.
Forgive me if this doesn't answer your question, I'm assuming it was a serious one?
Edit to add - OK, first scenario, as above, the plate passes really slowly, those little holes do exactly as intended, the add fuel to the air going past the plate.
Second scenario, open the throttle instantly, or as instantly as possible. The plate lifts past both holes in a flash - what comes out of them? The flow rate will be tha same, but the time between no holes, and all holes seeing that vacuum will be minuscule, then they see the remains of engine vacuum applied to the piston - and we know this isn't closed down tight. So right when we want our squirt of fuel to actually increase, we will get less coming out of these holes. Bugger. Tell me something different then?

 

I'll stick to my adjustable SU accel pump.

 

These work in exactly the same way, but don't have the adjustability of the su - perhaps that's a good thing, the number of times I've had old guys telling me that 3 in 1 oil is the stuff to put in them, then saying they have a stumble on accelleration...

 

Yeah, there's a huge amount of lore and bullshit attached to SUs.
Ignore it and crack on. You have to, because all the bollocks will lead you up the garden path.
I'm now at the stage where it's easily knocking out the same mid-range torque as the factory-fit bank of four.

 

Trying to upload a video image here, but apparently it's too large. How big can it be?

 

You have to use a hosting site like Imgur and attach it as a direct link.

 

It's never big enough.

 

Hopefully this works, and my ancient ipad actually manages it you'll have the proof that answers the question - where does the accel fuel come from?
Huw

 

Excellent.
Thank you.
My opinion has been changed.
I was wrong.

 

Not to argue, but isn't that a bit meaningless with no load?

 

Same as the SU - extra suckage caused on the annular fuel ring around the needle, the weight of the piston and the spring above it cause a delay in the piston rising and give some accel pump effect. In the SU, this is further added to by altering spring rate (which used to be just the weight of the piston, but that changed in the 1930s) and the choice of damper oil.
Hitachi / Mikuni added in some drillings and jets to augment that, probably because of patent considerations, and quite likely they thought they could do it better anyway. Oddly though, the nearer CV car carb to the Japanese design is the Stromberg CV, which also used a diaphragm, but like the SU it was dependent on simple spring, needle and oil choice. Seemed to do well enough on millions of cars though.

 

To be fair I think the bike carbs did improve low end fuelling control, the SU allways runs better slightly rich, and probably wouldn't cope very well with the pulsating nature of one carb per pot. The downside is millions of people have successively failed to understand how they work. Keihin went one step furtherbwith this and added an intermediate jet - I sold my Kawasaki GT750 before I got my head around this.

 

there used to be a video like that while the guy was riding but damm if i can find it

 

There are several in the Information Overload Hour - search "- CARB VACUUM PISTON OPERATION:"

https://www.xjbikes.com/forums/threads/the-information-overload-hour.27544/

 

There is one major factor in the design of the Mik-Hins and that is the lack of oil. That, more than anything else may have been the reason to dump it entirely. SUs were designed in an era when owners had some mechanical knowledge and were expected to do routine checks (and even their own servicing if they weren't all that well-0ff) so the ritual of the dashpot oil achieved some level of arcane knowledge know only to those High Priests of the SU of the Third Order of Magnificence.
Suffice it to say, Honda and others knew their bikes weren't being sold into a market where owners would appreciated having to firstly recognise the symptoms of a dashpot getting low, and secondly get their hands dirty by doing something about it. In tandem with that, they'd also managed to create banks of four carbs which, once set, stayed set for thousands of miles and that was plenty for most people.
It's simply not true that a properly set of SUs drift out, but a knackered old set, set up by somebody less cluefull, certainly does.
They avoided all that stuff. As a result, they created a reliability image for their product which now results in a situation where 4-some banks of carbs are rescued from bike breakers and fitted to Ford Zetecs and others for some cheap fuelling, if the owner can't be arsed with Megasquirt, etc.

My next SU project will be the RX-7 engine in the Scimitar, but fuelled by a single 44mm SU - via an Eaton M45 blower.

 

Yes, google bog brothers in the UK - https://www.boggbros.com/
have sold hundreds of cv carb kits fitted to race cars. I built a class 4 autograss engine (semi downdrafted inlets, big valves, big cam, extractor exhaust, on Peugeot/citroen alloy block) - it screamed like a bike engine, competetive with the Nissan 16v cars, and with a K&N airbox too!
Without doubt, the japanese really improved the SU/Stromberg CD design.