Alfa Romeo Forum banner

1 - 20 of 38 Posts

·
Registered
Joined
·
465 Posts
Discussion Starter #1 (Edited)
Since I’ve owned this 147, I haven’t been all that happy with the brake pedal travel, or the pedal feel. The pedal was inconsistent with excessive initial travel before the pads contacted the discs, though not necessarily any worse than with many other cars that I’ve driven over the years (which is no excuse...).

I do like a brake pedal with minimal ‘free’ travel, and that feels firmly ‘solid’ underfoot (sure I’m not the only one). For me this isn’t such a big deal with an automatic car, but with a manual gearbox I want the most consistent and solid feeling pedal I can get, primarily to facilitate easy and consistent ‘heel / toe’ downshifts, but also for that warm inner glow of confidence each time I use the brakes (even if this feeling might be somewhat subjective, more than a reflection on actual braking performance). Note that my throttle pedal height and lateral position relative to the brake pedal has been modified to better allow easier ‘heel / toe’ use (I find the stock pedal positioning isn’t the best for ‘heel / toe’).

The brakes are properly bled (several times...). This isn’t a ‘spongy’ or ‘soft’ pedal, but something in the system ‘freely’ moving in a manner that increases initial pedal travel. Once the pads contact the discs this movement ceases and the pedal feels firm. Over time, looking at everything I could think of that might be causing this, it became apparent that the rear calipers were mostly to blame.

I know it was rear brake related because I’d earlier experimented with adjusting (rotation) the rear pistons outward enough that the pads were statically pressed somewhat firmly against the discs. This was only so much that some drag could be felt when rotating the wheels by hand, the wheels still rotating with relative ease (though not freely spinning). This ‘tightened up’ the caliper and eliminated unwanted piston motion. This worked a minor miracle; i.e. very little unwanted pedal travel, improved pedal feel, improved braking response, with no noticeable adverse affects (i.e. excessive brake heat, pads wearing dramatically, or impact on fuel economy). It all seemed quite normal, other than the improved pedal. This was until some minor pad wear did occur, so in a week or so the beneficial effects disappeared, with the pedal becoming ‘loose’ again (no surprise, I kind of expected this to happen).

On investigation, what I found was that (with the handbrake off), if the rear caliper bodies were manually ‘wiggled’ they could be moved significantly in directions other than the ‘in / out’ direction in which they are supposed to move (axially sliding on the pins), and this was even worse if the pads were removed. It looked like worn guide pins and / or pin bores, since it was easy to see that the unwanted movement was occurring at the pins (i.e. radial movement at the pins, with some slight axial movement that was in opposite directions at each pin, when all we really want is equal axial movement at both pins). The rear caliper bodies could ‘tilt’ up / down and backward / forward. The front calipers don’t do this to anywhere the same degree as the rear calipers did.

So it seemed plausible that with vibration and road shocks the rear caliper bodies could be ‘wobbling’ on the pins, causing them to erratically be changing orientation in a way that might be causing some degree of piston ‘knock back’ as they wobbled around. So, a working hypothesis.

Each rear caliper has two pins. One is all steel with a uniform diameter along its’ entire length (other than at the head of the pin), and is a close fit in its’ bracket bore (0.1mm clearance). The other pin has a short rubber (or plastic?) insert near its’ end, with the rest of the pin being quite significantly smaller OD than either the insert or the pin bore. The only part of this pin that is a close fit inside its’ bore is the short insert. This seems to be common with many cars, i.e. that the REAR calipers use two quite different pins, one being all steel and one with an insert (though not all do, e.g. my Accord rear calipers each used two identical steel pins with no insert, the calipers being quite similar otherwise to the 147 rear calipers).

On disassembly I measured both bracket pin bores at 10mm ID, and the all metal pin at 9.9mm OD, so minimal wear at this point (if any, considering at least some clearance will be provided to lessen the possibility of the pin seizing in the bore). The rubber insert on the other pin also measured 9.9mm to 10mm, so the insert didn’t seem to be worn either. This was the same for both rear callipers.

Yet it seemed that most of the unwanted movement was coming from the pins with the inserts. This was particularly obvious with the brackets off the car and the pins still in the bracket bores. The all steel pins would radially wiggle to a quite small degree (as you’d expect with 0.1mm clearance), but the pins with the inserts wiggled substantially. The pin wiggled a great deal at the point where the pin protrudes from the bore, only ceasing to move when the side of the pin contacted the lip of the bore in each direction of wiggle.

The movement of the pins in their bores was up / down / backward / forward, and since the metal part of the pin OD is substantially less than the bore ID (8.5mm in a 10mm bore), the pin movement was also substantial.
The front callipers are not like this. Both pins are identical, all steel and of uniform diameter with a close fit in the bores (again, about 0.1mm clearance).

I don’t know why the rear calipers are fitted with two very different pins, while the front ones aren’t (which seems to be a common arrangement with the front vs rear calipers with many other cars, i.e. two identical pins in the front calipers and two quite different pins in the rear calipers). I’ve heard a suggestion that it is to do with preventing the caliper from rattling and creating an irritating noise, but I fail to see how this would work to do that, and if so then why not the same for the front calipers?

Both front and rear calipers are very similar in concept and basic design, other than size, and for the rear calipers having a handbrake mechanism incorporated into them (working on the piston).

So, what to do? I sourced two all steel pins with a plan to replace the original plastic insert pins, so that both pins in each rear caliper would be steel and a uniform OD along their whole length (with close fit inside the pin bores). The hope being that the caliper body would be more securely mounted, and so wouldn’t ‘wobble’, and so eliminate any suspected piston knock back.

How did this work out? Back on the car, the calipers now have far less of the unwanted ‘floppy’ movement, while still sliding in and out as they are supposed to do. However, some much smaller degree of unwanted ‘floppy’ movement does still remain. The handbrake works perfectly; in fact the lever now feels more ‘positive’ (previously the lever also had some excessive ‘dead’ travel before the pads could be felt contacting the discs). The lever still has some initial ‘free’ motion, but it doesn’t feel ‘loose’ like it did before.

On the road, the pedal is now more consistent, but there is still some unwanted pedal travel, though significantly reduced. Previously, the pedal would unpredictably move maybe two or maybe three or maybe four cm, feeling like it was doing nothing with zero resistance until the pads contacted the discs. Now, the pedal moves less (maybe two cm), and feels like it is actually doing something before the pads contact the discs, i.e. there is some soft resistance prior to the pads contacting the discs (as opposed to the pedal more or less ‘free falling’ until the pads contacted the discs).

Previously, it was of course possible to give the pedal a ‘tap’ (or two sometimes needed) to bring the pedal ‘up’ before the ‘real’ pedal push, but this was inconsistent, sometimes it worked, sometimes it didn’t, unpredictably. This caused a crappy feeling brake pedal which was not very confidence inspiring. It also created a problem for ‘heel / toe’ downshifts, i.e. sometimes the brake pedal height (at the ‘working’ height) would be a bit above or equal to the throttle pedal height, making the ‘heel / toe’ action easy, but at other times the ‘tap’ didn’t work and the brake pedal ‘working’ height would be below the throttle pedal height, making it difficult to ‘heel / toe’ (especially as this was inconsistent).

Now, this is much more consistent, i.e. ‘tap’ the pedal once and the next ‘in earnest’ pedal push is much more reliably firm and reassuring. ‘Working’ pedal height is more predictably higher after the ‘tap’. In general, even without ‘tapping’ the pedal, the brakes feel to work better with lesser pedal travel and in a more reassuring manner.

So, I consider this simple and cheap modification to have been a very worthwhile success, if not quite everything I had hoped it might be, because some unwanted pedal travel still exists.

Regards,
John.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #2
So what about the unwanted pedal travel that still remains?

I’m fairly sure this is caused by the as designed piston ‘pullback’ that is created by the square section rubber piston seals and related groove geometry. With the handbrake off, it is possible to observe a significant clearance between the pads and disc by pushing and pulling on the rear caliper bodies. The bodies will easily move inward and outward by approximately 1mm, give or take. I assume that this is normal for these calipers (and most others, in variable degree), and caused by piston retraction associated with the elastic rubber seal. I can’t feel the same thing with the front calipers, which doesn’t mean it doesn’t happen, probably just a lot less than the rear calipers.

The groove in which the piston seal sits is of course located in the piston bore. It isn’t quite what it looks like, i.e. it isn’t a perfectly square (or maybe rectangular) section groove with both walls at 90° to the bottom of the groove, but is slightly asymmetric in cross section. This asymmetry is what causes the pad retraction desired by the caliper designer (typical practise for most calipers). The inner wall of the groove is ‘vertical’ (at 90° to the bottom of the groove), but the outer wall is angled slightly outward. The seal is the same width as the groove only at the bottom of the groove, above that the groove is increasingly wider than the seal (though not much wider, even at the top).

So, there is a narrow ‘wedge’ of ‘free space’ outboard of the seal. Fluid pressure causes the piston to move outward, and the seal to also FLEX outward into this free space, because the seal is also ‘pushed’ by the pressure, and also because the seal is ‘attached’ to the piston by friction (but the piston can still ‘slip’ on the seal to accommodate pad wear).

The seal flexes under this pressure (in unison with piston movement) until it contacts the outwardly angled outer groove wall (becoming in effect slightly conical in overall shape), at which point it can’t move further outward relative to the groove. When the fluid pressure is released the seal ‘relaxes’ back into its’ original unstressed shape, and so causes the piston (to which it is ‘attached’ by friction) to retract slightly back into the bore, not much but enough to ensure that the pads are not dragging on the disc (the pads can still lightly rub against the disc, but not cause actual drag).

But, this also creates additional initial piston motion as the pedal is depressed, and so increases pedal travel. The caliper designer may well have his / her reasons to prioritise pad retraction over minimisation of pedal travel, but a driver who enjoys driving may well prefer to have these priorities the other way around...

There isn’t much that can easily be done about this, but there might be some things. Theoretically it would be possible to re-machine the seal groove to make it symmetrical with a ‘vertical’ outer wall. This would eliminate or substantially lessen the degree to which the seal can be pushed outward by the fluid pressure, and so eliminate or reduce piston retraction when the pressure is released (at the expense of some probably slight pad drag). Practically this would be a difficult thing to do, machining the groove would be tricky, and since this would widen the groove all the way to the bottom, a wider seal would need to be sourced and used (I assume).

Some calipers (mostly intended for racing purposes) use a simple spring inside the caliper, behind the piston. This spring pushes against the back of the piston in a manner that prevents the piston from being pushed inward (‘knock back’, typically caused by axle flexure), or if the piston is pushed inward, quickly pushes it back into place. The spring causes the pads to be lightly pressed against the disc even when the brake is not being used. This eliminates unwanted piston movement and substantially reduces unwanted ‘dead’ pedal travel that is caused by pad knock back, helping to ensure a ‘solid’ pedal. But, I think it would probably also address a pad to disc clearance problem created by the seal retracting the piston into the bore.

Such springs can be fitted to some calipers not originally so fitted, but in this case I’m fairly sure that the handbrake mechanism inside the 147 rear calipers would prevent their fitment. Even if the springs could be fitted, finding the correct spring strength is a matter of trial and error (a number of different strength springs are available from companies such as Willwood etc), which would mean caliper disassembly each time a different spring were experimentally installed and tried (though it’s possible to be lucky first time, what chance though?). Some degree of pad drag is associated with using these springs (not necessarily a problematically bad thing, depending on just how much drag).

Perhaps a ‘residual pressure valve’ could be fitted between each rear caliper hose and caliper body. Such valves might retain enough ‘residual pressure’ in the caliper cylinder to possibly overcome the hysteresis of the rubber seal and therefore prevent seal retraction (the seal would at all times be pushed against the angled outer groove wall by the entrapped pressure in the cylinder, and so piston retraction wouldn’t occur). Again this would create some pad drag.

A 2psi residual pressure valve will ‘entrap’ 2psi of pressure in the cylinder, which quite possibly wouldn’t be enough to overcome the hysteresis of the seal, and so be of no use (?). A 10psi valve would probably be more than strong enough, but may also cause excessive pad drag...

If I ever do decide to seriously chase this remaining unwanted pedal motion further (and I might), I think I’d try fitting a 10psi residual pressure valve to each rear caliper, and see what happens. The valves could always be easily removed if they caused any significant problems.

Pad drag is in itself obviously not a good thing, it must increase pad wear in some degree, put at least some heat into the pads, and can’t be enhancing for fuel economy. Car makers go to great lengths to minimise fuel consumption, each small saving adding up to a significant saving (fuel efficiency helps sell cars...), so my guess is that this is the main reason for creating pad retraction. I doubt they care all that much about pad wear, so long as it’s within reason.

So a question might be; just how much pad drag would have a significant impact on fuel economy, pad wear, heat? I can’t really say with any certainty, but considering my relatively brief experience when I tightened the rear calipers (by rotating the pistons until some significant drag existed), and that I didn’t notice any adverse affects from doing this (including no significant impact on economy), I suspect that some relatively small amount of drag is probably not particularly problematic, and any slight affect on economy and pads a small price worth paying for the joy of improved pedal travel and brake function...

Regards,
John.
 

·
Registered
Joined
·
788 Posts
Interesting stuff. Perhaps the front/rear differing pin arrangement is down to cost? The heavily loaded front callipers need two closely fitting pins while the lightly loaded rears can make do with one and a ‘guide’. The smaller pin with more clearance would not need to be machined to as tight a tolerance as the larger.

Some cars (Renaults?) have swivelling callipers with trapezoidal section pads. They look odd but the calliper is probably more stable (and less likely to stick). I wonder if they give better pedal feel?

Mine has a bit more free travel than I would like but not excessively and not inconsistent. Have you checked the disk run-out?
 

·
Registered
Joined
·
2,556 Posts
147s and GTs have the same ABS systems as 2002-on 156s. It is known to create longer pedal travel. There are differences (in 156s) from around that time which are different rear pads, discs, calipers and wheel bearings. I doubt they are to blame though.

I've tried similar stuff in other cars to reduce pedal travel but there is not really much which can be done realistically. I have no idea about the pad pin modification though but interesting anyway.

Now, if only someone could market kits whereby series 1 156 ABS system could be fitted instead which would solve the later cars' reluctance-to-brake-whilst-turning fault too.

Caliper piston springs could work (there is space inside) but they'd need a long free length to accommodate pad wear.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #5
Interesting stuff. Perhaps the front/rear differing pin arrangement is down to cost? The heavily loaded front callipers need two closely fitting pins while the lightly loaded rears can make do with one and a ‘guide’. The smaller pin with more clearance would not need to be machined to as tight a tolerance as the larger.
I doubt it. The pin with the insert is a more complex component with more parts to it, must be more expensive to make. Besides, economies of scale would make the all steel pins very cheap, and probably cheaper to just order say 20,000 all steel pins than say 10,000 all steel and another 10,000 with inserts. They'd be made on a CNC machine, which would just spit them out by the thousands, to a tightly computer controlled tolerance.

A clearance of 0.1mm isn't a particularly tight tolerance, I'm sure the pins could be a bit either way and be 'good enough' (not that a CNC machine would have any difficulty consistently producing near exact 9.9mm pins in huge numbers, or near exact 10mm pin bushes).

Have you checked the disk run-out?
The discs are near new, and the run out is good (or lack thereof).

Regards,
John.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #6 (Edited)
147s and GTs have the same ABS systems as 2002-on 156s. It is known to create longer pedal travel. There are differences (in 156s) from around that time which are different rear pads, discs, calipers and wheel bearings. I doubt they are to blame though.
Is it definitely the ABS itself that causes a "longer" pedal, or could it be that there was a coincident change in the rear calipers, from a caliper with relatively little piston retraction to one with relatively more? Such a change would be hard to spot...

I'm fairly confident that I've correctly identified seal induced piston retraction as the probable culprit. It explains all the symptoms, and the slight in / out looseness of the rear caliper bodies (when there has been no possibilty of the caliper bodies having been vibrated or shocked out of alignment, ie. when the car hasn't been driven since the last brake application), and why tightening (the rotation of) the rear pistons eliminated most of the problem.

What little unwanted pedal travel is left remaining is probably explainable by a similar retraction issue with the front calipers (though the front ones don't appear to be as bad in this respect, i.e. the front pistons don't appear to be as 'enthusiastically' retracted by the piston seals).

I'm 99.9% sure piston retraction will have been intentionally designed into the caliper. It is with pretty much any caliper these days, though racing specific calipers maybe not. Some very early calipers used a piston retracting spring to physically pull the piston back, until it was realised that the seal could do much the same thing much more simply if the groove was made asymmetric as described above.

I've tried similar stuff in other cars to reduce pedal travel but there is not really much which can be done realistically. I have no idea about the pad pin modification though but interesting anyway.
What "similar stuff" (if not the same stuff?). Not sure what else could be tried, other than what I've described.

To be painfully pedantic, they aren't "pad pins". These are found on 'fixed' calipers and retain the pad in the caliper (but I'm sure you know this).

Now, if only someone could market kits whereby series 1 156 ABS system could be fitted instead which would solve the later cars' reluctance-to-brake-whilst-turning fault too.
I wasn't aware of such a fault with the ABS. Can't say I've noticed anything like that with my 147. Not saying it doesn't ever happen, just never noticed it with my car (but then I tend not to brake whilst turning, unless racing the kart, trail braking into nearly all corners...).

Caliper piston springs could work (there is space inside)
Maybe. I've only had one of the rear calipers fully apart some time ago, and was just recalling that it seemed a bit crowded in there, so just assuming there probably wouldn't be the space for a spring...

but they'd need a long free length to accommodate pad wear.
This would be at least a theoretical issue with the piston springs, i.e. the pressure exerted by them would diminish as the pads wear, starting off strong and ending up weak (or at least weaker). I have no idea how significant this might or might not be.

I think I'd try residual pressure valves first, much easier to fit for one thing, and just as easy to remove. And, pad wear wouldn't affect the entrapped residual pressure, so the piston drag, however much, would be more consistent from new to worn pads.

Regards,
John.
 

·
Registered
Joined
·
2,556 Posts
On 147s, GTs and 2002-on 156s, even pulling the handbrake on so the pads contact the discs, then pumping the footbrake before assessing the amount of travel still results in an inferior brake pedal feel and travel when compared to an early 156 which has the handbrake fully off.

Front calipers are the same (unless very early 1.8 -UK- which has solid discs and alloy calipers.

All sliding calipers seem to have quite a lot of piston retraction but that is likely because they are sliding calipers which must account for both pads (obviously).

The stuff I tried years ago to reduce pedal travel was orientation of caliper pistons (if they didn't have flat mouths) and linkage adjustment and modification. In general, I had reasonable success.
 

·
Registered
Joined
·
9,190 Posts
Some cars (Renaults?) have swivelling callipers with trapezoidal section pads. They look odd but the calliper is probably more stable (and less likely to stick). I wonder if they give better pedal feel?
Compared to my '98 156, the Clio dCi brakes are "vicious" ... stab the pedal and it feels like it wants to flip over!!! I don't know what make the pads are but they have lasted over 60k miles and don't looks like they need replacing yet!


My 156 brakes always felt a bit soft from day 1 .. the first replacement pads/discs were Delphi and definitely felt better than the OEM pads (I'd had the 156 from new!). The originals had done around 72K but were by no means down to the backing plates.

The next change at 157k was to braided hoses (failed MOT on front hoses - cracking) and a move to Castrol Response fluid (a complete flush!). That make a big difference. Even with the old Delphi pads on the front the "feel" was much firmer. I know the dealer never changed the brake fluid in the first years ... so maybe that helped. I have read people moving to braided hoses say the pedal feels firmer. I also replaced the rear pads and discs as the hand brake was a bit low on the MOT test .. then the cable snapped! So new cables and I had a hand brake again.

(I don't have any notes of changing the front pads and discs again so did they really last 120k??? If I did they didn't have a huge impact on me or the braking! They would have been Ferodo/Mintex/QH/etc)

Finally at 192k I switched the front pads to Ferodo Performance and to MTech drilled discs. Braking was still as firm and, once bedded in, much, much better than OEM pads.

So maybe the hoses need a look or the fluid is getting old (remember brake fluid is hygroscopic and attracts water which compresses easier that the fluid)?
 

·
Registered
Joined
·
465 Posts
Discussion Starter #9
On 147s, GTs and 2002-on 156s, even pulling the handbrake on so the pads contact the discs, then pumping the footbrake before assessing the amount of travel still results in an inferior brake pedal feel and travel when compared to an early 156 which has the handbrake fully off.
I have tried this (with the handbrake partially on, just enough to feel the lever go a bit tight), and thought the pedal felt much the same (other than there being less initial 'free' motion). This is just my recollection of an 'experiment' I did some while ago. I'll try it again later, next time I drive the car.

Since fitting the new all steel pins, I think the pedal feel is pretty good, especially after a 'tap' to close the multiple gaps between all the pistons, pads, discs. Once properly engaged, I don't feel that the brakes are unsually soft or 'spongy', they don't feel like there is any air in the system, or excessively expanding hoses, or something mysterious and unwanted happening in the ABS magic box.

The best pedal I've ever felt on any vehicle was in the 'Nota Sportsman' I owned on my youth (Google should show some pictures of similar cars, an Oz baked interpretation of a Lotus 6 / 7). The best pedal feel on this car was when it had Cortina GT front disc brakes ('fixed' calipers with opposing pistons), and Morris Minor 1000 rear drum brakes (provided the drums were properly adjusted, which they nearly always were). The pedal travel was near zero, feel exceptional, heel / toe easy as and blissful. The brakes were unboosted, which gave a very 'pure' feel of what was happening, at the expense of being fairly 'heavy', but worked very well in such a featherweight car.

Later I fitted Fiat 125 rear disc brakes (don't know why I thought th car needed something 'better' than what was working very well already, probably a case of 'upgradeitis'...). The pedal feel was never quite as good, with noticably more travel. I assume that pad retraction in the Fiat calipers was partially to blame, but now also think that the caliper bodies may well have been 'wobbling'. The Fiat calipers were very similar to those on the 147...

All sliding calipers seem to have quite a lot of piston retraction but that is likely because they are sliding calipers which must account for both pads (obviously).
Not sure, but personally I think it's quite variable. My front caliper bodies don't move in / out (when pushed and pulled) nearly as much as the rear caliper bodies do, I can't really feel or see them do it to more than a barely perceptable degree. The rear caliper bodies, it's quite obvious to both see and feel (even after fitting the new pins and substantially reducing the body 'wobble'). This is also my recollection of the quite similar (to 147) front and rear calipers on my old Accord, and my son's old Prelude.

Regards,
John.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #10
So maybe the hoses need a look or the fluid is getting old (remember brake fluid is hygroscopic and attracts water which compresses easier that the fluid)?
Gazza,
The hoses are fine. Brakes are fully bled and fluid is new (well a few months old now).

You misunderstand the affect of water in the brake fluid. Water is at least as incompressible as the brake fluid is, so water contaminated fluid won't compress any more than new fluid, so won't affect system pressures or pedal travel or feel. What water in the fluid does is to lower the boiling point of the contaminated fluid mixture, which means it will take less heat to cause the fluid to boil, and form gas pockets in the calipers when the brakes become quite hot. Of course the resulting gas is highly compressible, and sudden complete brake failure is possible...

Water contamination also damages internal system materials.

Regards,
John.
 

·
Registered
Joined
·
1,961 Posts
I've recently fitted new disks, pads and fluid. At the same time I fitted new braided hoses to replace the 10-year-old OE rubber. And had to fit new rear calipers as well, thanks to the bleed nipples stripping the caliper threads when undone. Handbrake cables also replaced. I wouldn't say pedal free movement is excessive now (though I initially panicked about just that before driving and bedding in the new pads). The brakes are now really rather good (bog stock 284mm fronts, Ferodo FDB1052 & FDB1039 pads). I've not experienced any fade, but my hooligan days are past.

I suspect the biggest improvement was braided hoses, as the feel and progression is much improved, and characteristic of getting rid of old hoses that balloon slightly. They give a vague feel to brake application, from my experience of doing the same on other vehicles. The ones I got were a bit of a bargain - from hydraulichosesupply on eBay, with stainless fittings and securing grommets, £59.95 incl post. Just as good as HEL and Goodridge, which I've used previously on other cars and bikes, but much cheaper, and fit perfectly.

TBH, if your rear pads are backing off excessively, I'd have a good look at wheel bearings (fronts as well) before blaming seal or pin design. The new calipers don't seem to do it any more or less than seems right.
 

·
Registered
Joined
·
9,190 Posts
Gazza,
The hoses are fine. Brakes are fully bled and fluid is new (well a few months old now).
My hoses appeared fine .. I'd changed the front pair but the braided still made a big difference ...
 

·
Registered
Joined
·
2,221 Posts
I stumbled across this thread almost by mistake as I’m normally found lurking in the 916 forums as I own a Spider and GTV. I have exactly the same problem as the OP with the brake pedal on the GTV but not on the Spider. I believe I have the same front calipers as some 147s and GTs. In the last couple of years I have fitted new MTEC discs, Ferodo DS Performance pads and replaced the fluid with Motul RBF600. I think I need to check the adjustment of the rear calipers but next on the list are braided hoses and newly re-built calipers (sand blasted, painted red, new bleed nipples, seals, sliders etc and the all important Alfa logos). Each of the changes has made improvements to the car stopping but I’m still not 100% happy with the pedal feel.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #14
I suspect the biggest improvement was braided hoses, as the feel and progression is much improved, and characteristic of getting rid of old hoses that balloon slightly. They give a vague feel to brake application, from my experience of doing the same on other vehicles.
I suspect your old rubber hoses may well have been suspect, so the change to new braided hoses might have been a substantial improvement. Keep in mind that 'rubber' hoses are heavily reinforced with fibres that are highly resistant to stretching under pressure, they don't 'ballon' nearly as much as some people seem to think.

Note too that 'braided' lines are not so resistant to 'ballooning' under pressure because of the braided wire sheath, but because they are essentially a fairly stiff plastic (nylon?, something similar) tube that is itself very highly resistant to expanding under pressure (a bit more so than typical 'rubber' hoses). The braided sheath mostly just protects the plastic tube from damage.

I wouldn't mind fitting 'braided' hoses, I had them on the Nota and think they were likely to be at least one of the things which contributed to the exceptional brake pedal on that car (feel, travel, no matter how hard the pedal was pushed). But, my issue with my 147 isn't to do with pedal motion once significant force is being applied to the pedal, it's to do with pedal travel when only a very modest force is being applied, i.e. 'free' travel.

Braided hoses will improve pedal travel / firmness when exerting significant force on the pedal, i.e. they will reduce 'sponginess' with heavier brake application, but not have a significant affect at lighter pedal pressures. My brake pedal isn't especially 'spongy' even when braking fairly hard, so I don't think braided lines will much improve the issue I have with unwanted travel at lighter pedal pressures.

This isn't to say that 'braided' lines are not on the wish list. My pedal is firm enough when braking hard, no reason to complain (and I'm fussy about such things...), which isn't to say that it couldn't be firmer still with braided hoses...

TBH, if your rear pads are backing off excessively, I'd have a good look at wheel bearings (fronts as well) before blaming seal or pin design. The new calipers don't seem to do it any more or less than seems right.
The bearings are fine, there is no detectable play whatsoever, they are silent, they spin freely.

The piston retraction can be felt and seen (by pushing / pulling on the caliper body) even before the wheel has rotated after a solid push on the pedal. That is, with the car stationary the brake pedal can be pushed to fully engage the pistons against the pads and the discs, then without the car moving at all, if the caliper body is manually pushed inward and then pulled outward the body will move by up to one millimetre in and out. There is an obvious clearance that can't be attributed to any load induced flexures, other movement of the discs, or disc runout. The only remaining culprit is seal related piston retraction.

I can't significantly replicate this with the quite similarly designed front calipers, so all I can suppose is that the seals in the front pistons have lesser retraction built into the seal grooves.

Regards,
John.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #15 (Edited)
Thinking thinking thinking, over-thinking...

When the master cylinder (more specifically the fluid reservoir attached to the MC) is mounted above the caliper cylinder, some hydrostatic pressure must exist (due to the 'weight' of the fluid), increasingly greater the higher the 'head' of fluid is above the cylinder. This pressure represents a 'natural' residual pressure in the caliper cylinder. A ruler tells me that this height is approximately 500mm, give or take.

An online pressure calculator thingy tells me that 500mm of fluid 'head' will create 0.711167165 psi in the caliper cylinder (being anal with the decimal places). This 'natural' pressure obviously isn't great enough to prevent piston retraction, given my observations. It's less than half the residual pressure created by a 2lb residual pressure valve, i.e. a 2lb RPV will more than double the 'natural' static pressure that already exists in the caliper cylinders (it's more like 2.8 times the natural pressure).

With the rear caliper pistons being 38mm diameter, the residual pressure entrapped by the RPV still only equates to about 3.5 pounds (1.59 kg) of outward linear force being exerted on the piston (if my suspect maths are close). Not a lot, but substantially more than the 'natural' pressure would create, and maybe, or maybe not, enough to hold the seal against the angled wall of its' groove and so prevent / inhibit piston retraction, or maybe not...

However if a 10 lb RPV is used (they appear to only be available in 2 and 10 lb ratings) this number jumps dramatically to 35 pounds (15.9 kg) of linear force acting on the piston. This could easily be enough force to cause a very significant pad drag, or maybe it isn't (chances are not good though, I suspect).

Anyway, this is suggesting that a 2 psi RPV would be the place to start this experiment, which might or might not be a strong enough residual pressure, and at the risk of being a waste of time and some money. If it doesn't work, then it might be worth a crack trying the 10 lb RPV, at the risk of excessive drag and all that may follow, and another waste of time and more money...

Regards,
John.
 

·
Registered
Joined
·
788 Posts
RP valves are new to me so I looked them up. Apparently the 2psi ones are for disk brakes and the 10psi ones for drums, but what I can't find is anything about compatibility with abs. During operation there must be pressure waves running up and down the pipes and presumably the abs control is tuned to match. Will the RP valve mess this up?
 

·
Registered
Joined
·
465 Posts
Discussion Starter #17 (Edited)
RP valves are new to me so I looked them up. Apparently the 2psi ones are for disk brakes and the 10psi ones for drums,
The 2 psi valves are intended to be used with disc brakes if the master cylinder is mounted below the height of the calipers, such as with a floor mounted MC. With a low mounted MC, without an RPV the fluid 'head' will be negative, i.e. the hydrostatic pressure will be slightly higher in the MC than it is in the caliper cylinder, the 'weight' of the fluid will tend to cause the fluid to flow from the caliper cylinder toward the MC (instead of the other way around), which can apparently retract the piston too far and cause a lot of pedal free play (especially with shock and vibration). This would also make bleeding the system quite difficult. The 2 lbs of residual pressure stops the fluid from flowing downward into the MC (and from there into the reservoir), and so the pistons don't retract too far into the cylinder.

Given that with a floor mounted MC, the MC can't really be mounted all that far below the caliper cylinders, the hydrostatic 'suction' involved can't be very much at all, so it surprises me that the negative pressure in the caliper cylinders can be great enough to 'suck' the pistons inward to any significant degree, but apparently it can happen. If the MC is mounted above the caliper cylinders (of course far and away the most common arrangement), then there is automatically a hydrostatic pressure already existing in the caliper cylinder (i.e. higher than the pressure in the MC), so an RPV isn't necessary.

The 10 lb RPVs are intended to be used with drum brakes, to prevent the slave pistons from retracting too far into the slave cylinder bodies (the pistons would no longer abut the shoes), which would also cause a lot of pedal free play (note that this nothing to do with the brake shoe adjustment, it can happen even if the shoes are correctly adjusted).

MCs intended for use with drum brakes usually (always?) have an integral RPV built into them, or if intended for a front disc / rear drum system, an RPV only on the port leading to the drum brakes. These aren't always a 10 lb RPV, sometimes they are more than 10 lbs, sometimes a bit less, but always significantly more than the 2 lb valves used with disc brakes. The shoe return springs have no problem pushing the pistons back into the cylinder against the 10 psi of residual pressure, but the pistons can't retract further than the shoe springs 'push' them.

but what I can't find is anything about compatibility with abs. During operation there must be pressure waves running up and down the pipes and presumably the abs control is tuned to match. Will the RP valve mess this up?
With the rapid cycling of ABS operation, I'm sure there must be pressure waves in the fluid (when the ABS is enacted). I'm not sure how strong they would be, or if they would significantly affect ABS operation, or if changing them in some way would 'confuse' the ABS. My gut says no, but I'm not an expert on this.

Ignoring any pressure waves in the fluid, I tend to think that the ABS probably wouldn't be affected by fitting an RPV. Certainly I don't think that it would be any sort of issue when the ABS isn't actually doing anything.

Does the ABS actually 'care' about the detailed nuance of pressures in the brake lines? Or is it a complex on / off switch that only 'cares' about whether the brakes are applied or not, and what the rotational speed of each wheel is relative to the other wheels, so whether or not to rapidly open and close valves if these parameters are outside of the 'norm'? The ABS wouldn't need to 'know' what the individual line pressures were to do this, I don't think (but could be wrong...).

Regards,
John.
 

·
Registered
Joined
·
1,961 Posts
I suspect your old rubber hoses may well have been suspect...
That's why I changed them. They were apparently OE from 2008.

I'm unconvinced by your 'piston retraction/seal design' diagnosis. My brake pedal has a consistent approx 15-20mm free movement before the brakes start to work. There's no significant piston retraction, nor drag. Both old and new calipers were TRW.

The only time I've had excessive pedal travel (>50mm!) on this car was when my vacuum pump was failing (it's a diesel) due to a leak.
 

·
Registered
Joined
·
2,556 Posts
Seems like pedal travel is a personal observation, possibly. That said, I've never been as satisfied with pedal feel in a 2002-on 156 or a 147 when compared with an early 156.
That becomes dire with a 2.4 JTD. The already lengthy pedal travel really has barely enough travel to slow it at high speed and possibly won't even do that if brake temperatures are (too) high.

Stainless braided brake hoses remain a tried and trusted remedy but I'll still have a closer look at the cylinder retraction.
 

·
Registered
Joined
·
465 Posts
Discussion Starter #20
I'm unconvinced by your 'piston retraction/seal design' diagnosis. My brake pedal has a consistent approx 15-20mm free movement before the brakes start to work. There's no significant piston retraction, nor drag. Both old and new calipers were TRW.
As already described, my rear caliper bodies do have an easily detectable in and out movement (with the handbrake off and after the pedal has been depressed and released). I'm confident that this is caused by piston retraction. I can come up with no other explanation for this, and I've tried hard to find one...

I wouldn't be at all surprsised if yours do too (have you actually investigated?). I've noted this with several similar style rear calipers on several other cars as well (not only Alfas, all were rear calipers that used a pin which turned out to have an insert). If so then I suggest that the piston retraction is more than is 'required' just to prevent pad drag.

The movement (on my cars' rear calipers) is about 1mm, which doesn't sound like much, but the MC piston is significantly smaller in diameter than the wheel cylinder pistons are, so the area of the MC piston is substantially less than the piston area of the caliper cylinders. So, the MC piston will have to move significantly more than 1mm for this gap to be closed up, X 2 because there are two rear calipers, plus, whatever MC piston movement might be needed to close the seemingly smaller piston gaps at the two front calipers. Note too that the front caliper pistons are substantially larger in diameter than the rear caliper pistons, so it will take as much or more pedal stroke to move the front pistons to a lesser degree than for the rear caliper pistons.

And don't forget the pedal ratio, about four or five to one is fairly typical. This means that if the MC piston has to move (e.g.) 5mm to take up any clearances between all the caliper pistons / pads / discs, then the pedal pad will have to be depressed by 20 to 25mm before the brakes actually 'grab' (on top of the slight free play that must exist at the MC push-rod). It doesn't take much 'unnecessary' caliper piston clearance for this to start getting out of hand...

It's my suspicion that the rear calipers may have inherently somewhat more piston retraction than do the front calipers. I suspect that the tapered gap between the seal and the seal groove (which is what creates the retraction) may well be a bit wider in the rear calipers than in the the front calipers, but I've made no measurement of this to prove the notion.

What I know for sure is that replacing the rear calipers pins (those with rubber inserts) for ones more uniformly closely fitting in the entire length of the bores (i.e. all steel pins without inserts) has resulted in less erratic clearance between the rear pistons / pads and discs (with the old pins, if the car was driven on a bumpy surface then coasted to a halt, the rear calipers bodies could be moved in and out by more than the 1mm that occurs after the pedal has been depressed and released).

So now, with the new pins, the free pedal travel is significantly less and much more consistent than previously, and what pedal free travel still exists can reliably be 'taken up' by a single 'tap' on the pedal, after which the working brake pedal height is un-problematic (which wasn't the case with the original steel / rubber insert pins). The brakes are now far more pleasant to use, and more confidence inspiring. 'Heel / toe' gear shifts are now easier, more consistent and more enjoyable, with almost no 'mis-blips' as a result of inconsistent working brake pedal height (i.e. 'blipping' too much or not enough). In general, braking peformance is just 'nicer'.

This doesn't mean that I'm 100% happy with the remaining free pedal travel...

Regards,
John.
 
1 - 20 of 38 Posts
Top