Clutch hydraulic damper / delay valve, considering deletion...

[johnl]

So with further driving / familiarity with the now disabled CDV; the clutch action is now noticably somewhat lighter than it was before. The action now has a 'mechanical springiness', rather than a 'rubbery springiness', which is apparent on both the down and up strokes of the pedal. Pedal weight increases as the pedal moves toward the floor, but in about the last quarter of pedal travel it now decreases noticably, i.e. there is a now feeling of the pedal / clutch going 'over centre' that wasn't there previously. The shift action is cleaner, lighter and more consistent than before, and shift action is generally improved. This is all good.

However there is a slight down side. When 'rev matching' the car is now less tolerant of ham fisted throttle 'blipping' (ham footed?). If the rpm are a bit high or a bit low when I change into a lower gear, then the car will 'lurch' (momentarily accelerate or 'decelerate' slightly) as the clutch pedal is released. This wasn't as big a deal when the CDV was operating. It's obvious that when the CDV was unmolested it allowed a degree of extra slippage as the pedal was released, so if the rpm were not quite right then there was some 'blurring' of the clutch engagement that to some degree masked clutzy technique. Without this extra slippage the clutch engages more positively and strongly, so rev matching has to be better.

Clutzy technique is I think a by-product of becoming habituated to the original vagueness and erraticity in the clutch action, i.e. technique was hard to hone because the clutch action was less than consistent so it was harder to get the timing exactly right. I'm sure my technique will improve with more practice / familiarity, it's already getting better.

Regards,
John.

[johnl]

People seem to have lost interest in this, but I've become rather curious to know how these valves actually work and how this might affect the clutch action. I haven't found any specific explanations anywhere, so can only speculate based on what I found inside mine:

Firstly, there are no obvious valving arrangements inside the CDV, i.e. no springs or spring loaded components, or other obviously moving parts. It's a bit mysterious.

The CDV has a cylindrical body with a fluid inlet (from the MC) at the top of the side of the cylinder. At the very top is a centrally located outlet tube leading to the slave cylinder. There is a hard (ish) but still somewhat pliable largish OD (about 16mm) hollow 'rubber' tube inside the CDV, which fills much of the internal space (I initially thought this was made from a semi pliable plastic, but it is actually some kind of rubber-like polymer).

This rubber tube is closed at the end that is highest inside the CDV, but open at the other end. The open end is exposed to atmospheric pressure via a small hole in the base plate of the CDV. Fluid is pumped into / through the cavity between the rubber tube and the outer cylinder.

The outlet tube projects inside the top of the cylinder by approximately 1cm, and has an open end inside the cylinder. It also has at least one small orifice in its' wall very close to the internal roof of the cylinder (higher up than the fully open end). The length of the rubber tube seems to be such that normally its' flat internal end is pressed against the open end of the outlet tube, thus sealing the open end of the outlet tube.

When the pedal is depressed, pressurised fluid enters the CDV via the inlet tube. The fluid passes out of the CDV through the small orifice in the outlet tube wall (that is near the roof of the cylinder), and from there toward the slave cylinder. The clutch itself now creates a resistance to fluid flow, so fluid pressure increases above atmospheric.

I'm guessing that the rising pressure now causes the rubber tube to deform in a manner that causes its' effective length to become slightly shorter (i.e. effectively contract, 'crushed' by the pressure). This could occur because the rubber is somewhat pliable, the ID of the hollow rubber tube is only exposed to atmospheric pressure, but, the pressure on the tube OD tube is exposed to a substantially higher system pressure. If so, then this would cause the end of the rubber tube to move slightly away from the open end of the outlet tube, and so 'unseal' it. This would then permit a less restricted flow of fluid as the pedal is further depressed, once internal CDV pressure reaches X psi.

When the pedal is released, the internal pressure will fall, and the length of the rubber tube will again become slightly longer, thus resealing the open end of the outlet tube. Fluid flow backward from the slave cylinder will now be constricted to whatever flow is allowed by the restricted orifice at the top of the outlet tube, thus impeding the rate at which the pressure acting inside the slave cylinder drops, and so slowing the rate at which the clutch re-engages.

Well it's a theory...

Whatever, that the hollow inner tube is demonstrably made from a pliable (and non-reinforced) rubber, is exposed to hydraulic pressure on its' OD, and only exposed to atmospheric pressure on its' ID, means that the rubber tube must suffer some degree of deformation when the clutch pedal is depressed. This explains quite a lot.

Today I went for quite a long drive, and was noticing that with the CDV now disabled (i.e. the rubber element deleted) the engagement and disengagement points in the pedal travel were now consistently at the same height, whereas before disabling the CDV they generally weren't.

It would also explain changes in the engagement and disengagement points that seemed to be temperature related (the pedal becoming mushier on warmer days and after longer periods of driving). The rubber must get softer the hotter it happens to be, which must be affected by ambient and under bonnet temperature.

It would explain why the pedal action is now generally less 'mushy' and more 'defined'. It would explain why the pedal 'weight' has become noticeably lighter (the rubber no longer abuts the open end of the outlet tube at any time, so can never obstruct it).

It would explain why deleting the rubber would increase the pedal height at which the clutch engages and disengages, as it has done, i.e. pedal motion is no longer 'lost' due to the tube deformation.

Having this relatively soft tube in the system is analogous to having a rubber hose a portion / length of which that has at least partially failed, and so swells with internal pressure (causing a mushy pedal), except that instead of expanding with pressure, the tube partially collapses. This must surely be a cause of 'mushiness' in the system, as my experience with this supports...

The more I drive the car without this BS valve in operation the more I'm liking how the clutch is now working, and the improvement in gear change quality that this has created. If you have a manual gearbox then your car will have one of these damned things, my advice is get rid of it, I don't think you'll regret it...

Regards,
John.

[johnl]

No comments or questions, or even interest? Maybe if I say that disabling the CDV has turned out to be one of the most effective modifications I've yet made to this car (and I've made some other very effective mods...). It's been even more effective than a similar modification I made to my old Honda Accord. And, it only took a couple of hours and cost zero dollars (if I don't count half a welding rod and having to buy a bottle of brake fluid...).

I'm now on an anti CDV crusade, in case you hadn't noticed...

Regardless of the car to which they are fitted, these CDV devices are seemingly always generically described as having zero restrictive effect on the pedal downstroke, and to only 'work' on the pedal upstroke. Assuming that my newfound understanding of how these things might actually function is correct (at least the ones fitted to the 147), then this just can't be true.

According to my understanding, in a static state (pedal undepressed) the internal rubber tube abuts and blocks the largest orifice at the end of the outlet tube (its' internal open end), leaving only a more restricted orifice for fluid to pass through. As the pedal is depressed and fluid pressure rises to X psi this changes, with the largest orifice becoming opened by the rubber tube becoming compressed and 'withdrawing' from the open end of the outlet tube, and flow capacity thus increasing (i.e. flow out of the CDV to the slave cylinder, as well as back into the CDV from the slave cylinder).

So, as the pedal is depressed there is an initial flow restriction which then 'de-restricts' with further pedal motion. I can't imagine that the pliable rubber tube moves all that far away from the open end of the outlet tube, so it's likely that though the end of the outlet tube becomes 'unsealed', the proximity of the closed end of the rubber tube to the open end of the outlet tube may still tend to be at least somewhat restrictive (which would explain why with the rubber tube deleted my pedal now feels at least a little bit lighter throughout the pedal strokes, i.e. there is no restriction at all at any time).

This all happens in reverse as the clutch pedal is released. The internal CDV pressure falls to X psi and so the rubber tube expands (de-compresses), and so the end of the rubber tube reseals the open end of the outlet tube (through which fluid is now flowing in reverse direction), and creates a more restrictive path for the fluid pressure to release from the slave cylinder, but only after the pedal has been partially released.

So as I understand it, these valves create a significant increase in restriction in both directions, but only when the pedal is in the upper part of the pedal travel. When the pedal is lower than this (i.e. closer to the floor), then the restriction decreases (though doesn't necessarily disappear completely), and this is the case whether the pedal is being depressed or being released. So, the 'softening' of the clutch re-engagement only occurs after the clutch has already become partially re-engaged, not from the beginning of the re-engagement.

It's less noticeable and has less affect on the pedal down-stroke because the pedal pressure is not limited, other than by the speed with which the pedal can be depressed (so the pedal will be a bit stiffer the faster you push it, but fluid will still flow through the restriction). However, on the up-stroke the clutch springs (diaphragm) can only push against the fluid pressure with X force, and if the fluid has to pass through a restriction then the rate at which the clutch itself can fully re-engage will be impeded, but only after the pressure has dropped below X psi...

This would be much easier to explain with accompanying diagrams...

Whatever, these things add a rubbery elastic element into the system, which compresses and de-compresses with changes in pressure. This causes mushiness in the pedal. If that's not bad enough, the elasticity of the elastic element will be variable depending on the temperature of the elastomer...


Regards,
John.

[johnl]

Further (over of this yet?), the atmospheric air cavity inside the CDVs hollow rubber tube must in effect act like there is a large air bubble in the system, one that is at best only partially isolated from the hydraulic system by the stiffness of the wall of the rubber tube. The rubber is kind of 'hardish', but I can squeeze the tube walls significantly closer together between my fingers, so it's not all that rigid.

If there were an actual air bubble trapped in the system, at least its' pressure would rise (and fall) in step with system pressure, rendering the bubble increasingly less compressible as the pressure increases, but the pressure in the rubber tubes' cavity never rises above atmospheric...

Did I mention that the clutch engagment and dis-engagement points in the pedal travel are now not only each at the same consistent pedal height, but also are now significantly higher off the floor than they were with the CDV in operation? It's no longer essential to fully depress the pedal all the way to the floor to fully release the clutch (as it often was before, though this was something of a variable...). I'd say that removing the compressible rubber tube probably has a big part to play with this, i.e. less deformation of the tube equating to less 'lost' pedal motion, and a more consistent action.

Pedal feel has been substantially improved, I can now feel with my foot the engagement and dis-engagement points in the pedal travel, which was previously quite 'numb'.

Regards,
John.

[Citroënbender]

No comments or questions, or even interest?

Well, I had two days off for a cultural occasion.
I await the next instalment of automotive retrogression. Will it be a power steering delete?

[johnl]

CB,
From a position of having actually experienced back to back how the 147 clutch behaves both with this device operating and without it, and how this affects the driveability of the car, I assure you that disabling / deleting it is a fairly substantial step forward into the past, and not in any important way a downgrading of the engineering of the car. Not every manufacturer uses these, even today.

That these newfangled gizmos have been adopted only relatively recently by some manufacturers doesn't ipso facto make them a good thing. They aren't the solutuion to a significant problem, rather they are a band-aid for a potential problem and the creater of a significant real problem (or chain of problems). I strongly suspect that they were forced upon drive train engineers from above as a crude means of reducing drive train shock if drivers abuse the clutch (i.e. avoidance of potential warranty claims...).

I can imagine the way it might have gone; the 'powers that be' inform the engineers that there have been some drive train claims under warranty, and that therefore they are instructed to come up with some means of mitigating the possibility of more claims, and while you're at it, make it cheap to implement. The engineers grumble, but comply...

I do think I've put forward a reasonable case for why CDVs are not such a good thing to have. I've detailed my personal experience, and develped a cogent explanation (I think) as to how exactly they must do what they do, based on a reasonable analysis of the internal components that exist inside them (at least the ones in the 147). I've described the problems created by having an elastic element in a hydraulic system, and how this must adversely affect the dynamic behaviour of the system. If this isn't enough then I don't know how I could convince you, unless you could drive my car...

PS, and off topic, if even fairly small modern cars weren't so damn heavy (kerb weight), then manual steering would be fine by me. Over the years and numerous cars that I have driven, by far the most communicative, lively and fun to use steering that I have experienced has been with non PS systems (though not all manual systems are that good). It's all smiles, until you want to park it (but then I have very strong arms and shoulders...).

At least part of the issue with heavy (manual) steering is the trend to zero scrub radius (or very near to it), especially with FWD. Along with high kerb weight, this can make PS more or less a necessity (unless you can live with maybe five or so turns lock to lock). With zero SR, at low speed as the front wheels are turned the contact patches must be 'scrubbed' around the point at which the steering axes intersect the ground, which adds substantial 'weight' to the steering (more so at low speed, the effect diminishes as speed increases). If the SR were to be e.g. about 100mm or so, then instead of the contact patch 'scrubbing' around the steering axis, the entire wheel will roll along an arc centred on the offset steering axis, with very little 'scrubbing' at the contact patch, and so the low speed steering effort will be a lot lighter. Unfortunately this isn't all that good for FWD cars, as substantial SR is a major contributor to 'torque steer' affects.

I once owned a BMW 2500 with PS and a fair dose of SR. One day the PS died, but this made very little difference to the steering weight. I happily deleted the PS pump rather than spend the $s to replace it, and my wife didn't complain about the steering becoming heavy......

Regards,
John.

[johnl]

I await the next instalment of automotive retrogression.

CB,
Maybe if we approach this from the other end. Rather than me continuing to attempt explaining why I think CDVs are inherently a bad thing that significantly degrade drivability and therefore ought to be ditched, can you say what it is that a CDV actively does that improves the clutch functionality? Why exactly is it that deleting it is an act of "retrogression"? How does deletion make the clutch function and driving experience worse? Why are we better off with a CDV than without?

I accept that a CDV will to some degree protect the drive train from abusive treatment (other than the clutch itself, which is not unlikely to wear at least a bit faster because of additional slippage), so let's assume that the driver isn't an idiot, has at least some mechanical sympathy and won't use the clutch in a brutally stupid manner...

Regards,
John.

[Citroënbender]

I have indeed, not driven a 147 with the clutch damper removed.

The present manual car's clutch bears a slightly disconnected feeling, it is no particular concern to me. How effectively the gears actually engage is of more interest, as syncros are never cheap to buy, nor easy to replace. (Excluding surprises, I try to not brake from more than 30km/h in a manual car - preferring to read the conditions, use my throttle and gears.)

If I had reason to consider the damper as suspect I would first try a genuine new item, and bleed the system fully.

[johnl]

I have indeed, not driven a 147 with the clutch damper removed.

I thought it unlikely. Respectfully I suggest it's untenable to criticise this particular modification if you haven't experienced the results of it.

The present manual car's clutch bears a slightly disconnected feeling, it is no particular concern to me.

So you recognise that some problem exists, but you don't find it to be an issue for you. That's fine, but it might not be for others, it certainly wasn't any longer acceptable to me. Your car may be less affected by the CDV action than was mine, maybe down to the state of the clutch itself, or some difference between brands of clutch, who's to say...

How effectively the gears actually engage is of more interest, as syncros are never cheap to buy, nor easy to replace.

My main issue wasn't so much the actual feel of the pedal (though I didn't like it much, it just didn't feel 'right' or anywhere near as good as many other clutch pedals in my experience, even ones that used a cable). It was more how the erraticity in the hydraulic system adversely affected the shift quality in a rather random fashion. It was obvious that sometimes the clutch was fully disengaging, and sometimes not quite fully, the gears would sometimes be easy to shift, sometimes a bit resistant. This was not synchromesh related (which feels different). However, the feel of the clutch pedal and the quality of the shift action are interrelated. 

Disabling the CDV resulted in a complete cure for both the pedal feel, inconsistency of clutch action, and the resultant shifting issues. Changing gears is now a predictable pleasure, whereas before it just wasn't.

Keep in mind that my car doesn't have a stock 147 shifter, but a very modified Honda shifter that shortens the lever throws and is much more precise in its' action. It also has modifications at the gearbox end which further reduce the shift lever throws (my cars' shift lever movement is very 'tight', by which I don't mean stiff). This reduction in lever throw means that the lever has substantially less leverage on the gearbox internals associated with shifting gears, and so any resistance to changing gear that might be caused by say a slightly dragging clutch will be more obvious with my car than a car with the OE long throw shifting mechanisms.

I'm suggesting that your shift mechanism will have a lot more leverage than mine, so where I can feel untoward resistances when shifting gears, you may not be able to feel them? If so then this doesn't necessarily mean that such resistances never occur...

If I had reason to consider the damper as suspect I would first try a genuine new item, and bleed the system fully.

FWIW, there were no signs inside my cars' CDV that it was in any obvious way worn or damaged.

Because of pedal 'sponginess' I had already bled the clutch previous to disabling the CDV, which made no difference to the "disconnected" feeling and the issues associated with this. I used a pressure bleeder, and bled it a lot, several times because the pedal didn't improve, there were no bubbles coming out. After disabling the CDV I bled it again (of course), but due to a limited supply of fluid on hand only did a very quick job, i.e. I didn't put much fluid through the system, about one third of a bottle. Regardless, the pedal feel, clutch function and shift quality were immediately much better than with the CDV in operation.

I'm just trying to share this very effective and virtually free modification with others. Maybe someone has similar issues to the ones my car had, will try it, and reap the multiple benefits of my pioneering work...

Regards,
John.

[Citroënbender]

It is not a "problem" in my eyes.

[johnl]

It is not a "problem" in my eyes.

CB,
If you're happy then I'm happy for you to keep your car totally stock, whatever works for you.

I've argued my corner on this ad nauseum because you seemingly want to dismiss this simple modification out of hand as being in some unclear manner and in principle worse than pointless, when my direct experience of it is that it is in reality a very worthwhile improvement.

The simple inescapable fact is that my car is now hugely more enjoyable and easier to drive with this 'thing' disabled. The more I drive it the more I get used to the improved characteristics and the more I like it. It's 'before', and it's 'after', no other changes. The positive affects are way more than placebo, much too distinct for that...

Regards,
John.

[bazzbazz]

Now look you two . . . . . don't make me come back there . . . . . !   

[johnl]

Now look you two . . . . . don't make me come back there . . . . . !   

Bazz,
I'm not looking to have an unfortunate argument with anyone, just a disussion. I've learned to rate and respect CBs opinion on most things Alfa related, and am sure there are areas where he knows more than do I. But, in this instance I'm struggling to understand what seems a stubborn possibly ideological recalcitrance to consider that there are obviously very positive aspects to disabling or deleting the CDV, and no real downsides. Just a bit frustrating...

Regards,
John.

[johnl]

I do feel a bit like a 'prophet in the wilderness' on this, considering the apparent lack of interest. It's not just me who has found CDVs to be a sugnificant issue, I've lifted these reported experiences from a Honda site, they match my experience with my Alfa:

https://www.civicx.com/threads/clutch-delay-valve-delete-prl-ss-clutch-line-install-review.21587/

"After deleting the CDV, it made a world of a difference! It honestly feels like I have a whole new clutch inside the car, almost making it feel like I have a Stage 1 clutch...That noticeable of a difference. :yes: The Clutch bites instantly with no slips during normal driving and aggressive high RPM operation MUCH better and RPM's drop between gears and shifting motion became much more fluid and natural. Best part is, it cost me $0. Only thing I bought was a bottle of brake fluid to top of the reservoir during the clutch bleeding process."

"Few pumps of the pedal and you notice a huge difference. Drove the car for a bit and it's amazing having that natural clutch feel again."

"I notice a big difrence going from stop->1st->2nd. Alot smoother. Shifts are quicker."

"pulling away in 1st on a normal start is much quicker
pretty sure I chirped the tires at 2k rpm
if I didn't know any better, the pedal feels more "direct" and slightly stiffer
you can definitely feel the difference where the CDV "assisted you" vs without it."

"Just removed the delay valve and it was highly worth it. Much faster response on the clutch and even better at higher RPM."

It's easy to find plenty of similar reports from drivers of Hondas, Bimmers, VWs, Subies, and other marques. The CDVs in these cars (including Alfas), are all similar in what they do (if perhaps not in the detail of how they do it?). If they commonly cause issues with other cars, then Alfas won't be immune either (mine wasn't).

Have I become obsessed with these things? Maybe a bit, but I am rather annoyed that they are fitted in the first place, and that I've had to go to the trouble of disabling mine just to have a clutch that works properly, as it should have done from the factory. I'm not specifically targeting Alfa for this, it seems like a pandemic infecting more and more cars...

If you have a manual car, then IMO it is causing problems for you, even if you have become so habituated to crappy 'modern' clutch action that you aren't aware of it. Disable it and you will notice a positive difference, I promise...

Regards,
John.

[Vne165]

How do I know if I have one fitted? Would my 2002 916 TS Spider have one?