Allpar Forums banner

Brake bleeding

10K views 45 replies 8 participants last post by  John Wood 
#1 ·
OK, I fixed a leaky wheel cylinder on the left rear by redoing everything: new drums, shoes, hardware, wheel cylinders, rear rigid brake line on both sides; reman calipers, new pads and rotors in front. Time to bleed.

I've done this before with no particular issues, other than my truck took 2-3 iterations to bleed fully. So last night my wife worked the brake pedal while I bled each wheel. Started with RR, LR, RF, LF. I never let the master cylinder get low. Procedure: connected a rubber hose from bleeder to a bucket of new fluid, submerged. Told her to press and hold pedal when I say "down" and only release when I say "up." She followed this precisely.

At each wheel I called out "down", then opened the screw, and bubbles from the hose itself came out. Closed screw and called out "up." Maybe once or twice more, got rid of the bubbles. A tiny bit of discolored fluid came out. Didn't really seem like enough fluid or air blew through to account for empty wheel cylinders and calipers. Finished up and the pedal was rock-solid with engine off.

I started up and it did travel a tiny bit more than normal. Cautiously stopped a few times in the driveway, then tried level road. Each time I hit the brakes somewhat hard. The pedal sighed a little, felt mushy but not much longer travel than usual, not as firm as usual. The car nose-dived and stopped, but there were two things I noticed:

1) When I hit the pedal hard, I did not get that hard impact feeling that I should, of the pads making contact. I could not lock up the wheels at all. No ABS.
2) When I released the brake while still rolling, instead of feeling it release and roll more, for about a second I still felt braking action, even with the pedal released.

When I got back home, the rims were very warm to the touch, having hit the brakes somewhat hard for about 15 times within a mile.

I didn't change the flex hoses because they are rusted to the upstream rigid lines and the rubber is in great condition - no cracks or defects at all. If I change them, I have to replace every rigid line in the car. And they were fine just before this job.

Consensus? I believe there's still trapped air, and that when I released the brake pedal, the air was slow to decompress, and so it held the brakes on for just a second. I don't believe the flex hoses are bad and acting as a one-way valve. I just didn't see as much air blow through as I expected. So I will try again.

BTW, it IS a dual diagonal braking system. Should I bleed RR-LF and then LR-RF? I did furthest to nearest. I'll also leave all 4 screws open to ooze, for about 15 minutes before I start. I might then step lightly on the pedal before bleeding, to eject air and fluid from each screw first.
 
See less See more
#3 ·
That's not the correct way to do it. Stepping on the pedal first tends to force the air bubbles closer to the bleeder screw before opening it, so as to purge them better and faster.
 
#5 ·
It's not likely, since I didn't let the fluid get low, but I will try that, thanks. I'm going to open all bleeder screws and let it ooze for awhile, then tap the metering block lightly, press the brake pedal slightly, then close the valves and bleed again, this time in diagonal order. Rain tonight postponed it.
 
#6 ·
Repeated the process. I opened all four screws, and all but the RF oozed. I closed them up and bled LR, RF, RR and LF. When I did, I got good flow out of all but the RF. The rest ejected bubbles and then a reasonable amount of fluid. The RF oozed some bubbles slowly, then very little fluid. I went back and did it last. Topped off the master cylinder, drove it. Same as before. I can't lock up the brakes, and when I stop on a hill, then let go of the brakes, the car doesn't move at all for 1-2 seconds, then slowly begins rolling. Drove around the block a few times. Front rims were warm, rear were cool.

Could it be trapped air still? It brakes in an absolute straight line, even when letting go of the wheel, feels even all around. I'm getting to the point of taking it somewhere to have it bled, as much as the thought sickens me, since it should be an easy task. I can safely drive it around at low speeds, it stops fine, just slow to release.
 
#7 ·
Any way to undo the RF brake line at both ends and just try blowing it out with compressed air, Bob? If you can do this, then you'll know it's most likely a problem with the caliper.
 
#8 ·
I'll probably undo the flex hose at the caliper and see if fluid can blow through the hose when I step lightly on the brake. These are (new) reman calipers from Centric.
 
#9 ·
I once had a problem kinda like this on the RF on my wife's now retired 92 Grand Caravan - I thought at first the caliper was seized cause it would drag a bit always after brakes were applied. Replacing caliper did not help. What I believe it turned out to be was a rusty mount on the flex line - the little bracket that goes around the flex line and attaches it to the strut. The flex line looked fine but the bracket had corroded on the inside of the piece that holds the tube and was pinching it from the swelling due to the rust so the pressure would not release after braking. I replaced the flex line to fix it but I always wondered if I had just bent that bracket a little (it had an open end it was not fully closed) to relieve the pressure if that would have worked. I did that on the other side cause it was rusty too.
 
#10 ·
Thing is, it did NOT do this before the brake service. Brakes grabbed fine and released immediately, there was no dragging at all.

Weather has not cooperated, won't get to it until later this week.
 
#11 ·
OK, so now it's gotten more complicated.

Last night I verified that the brakes don't release for a few seconds, by having my wife press and release the brake, and not being able to turn the wheel at all for about 3 seconds. Then I had her press and release the brake, and I immediately opened the bleeder valve, and the brake released right away. Nuts. That means that the new caliper is not seized, since the piston would still stick whether or not the bleeder is open. So the trouble point to the flex hose. BUT....

When I bled the caliper, I went to fill the master cylinder and both sides were down. Filling one side brought both levels up. :huh:

I bled again, and again, both chambers were down, and I filled one side and both filled up, as if they are no longer separate. This did NOT happen the last two days I worked on the car, after the brake hardware was replaced. This was a new development last night, after the brake dragging appeared. The brake pedal still feels firm.

So, is my master cylinder gone? What is the mechanism of failure? Looks like I should replace it before touching the brake hoses, which will be just about impossible to replace without destroying the rigid line upstream. Advice appreciated. I've done a dozen brake jobs, NEVER had an issue like this before.
 
#12 ·
Man, I've been tinkering with cars for 40 years, Bob and have never seen anything like that. I sounds like somehow one of the seals in the MC must have a tear in it and is bleeding through to the other side. I don't understand how this could happen, but a defect in the MC seems like the obvious culprit here.
 
#13 ·
[sub]The master cylinders reserviors used on these cars do have connected chambers so thats normal what you saw. [/sub]

[sub]If you need to change a flex hose you could always cut the rubber hose and then unscrew the metal stub from the flare nut so as to not mess up the line trying to turn the flare nut and just twisting the line up. I have done that many times to save a brakeline.[/sub]
 
#14 ·
I don't think that's true about the master cylinder. It never behaved that way before. It's a dual diagonal system, and the LR an RF were on the back chamber, and LF and RR on the front chamber.

Problem is, the flex hose has the female thread, and it doesn't have a hex, it has 4 straight sides and the other two are a curve, and it fits through a bracket, then locked in with a U-clip. The only way to unscrew it from the rigid line without turning the rigid line's nut, is to remove the clip, push the rigid line through the bracket hole, grip both fittings with vise grips, and unscrew the flex connection. And there are three problems with that: the flex connector has a thin flange to grip; the rigid line must be held perfectly motionless even though it's now unsupported; and the strut assembly is in the way of all this.

And there's undercoating sprayed all over the connection, making it difficult to dissolve, wash away and get penetrating oil in there. This is the original 19-yr old hose, factory overtightened.

I need to run the first test tonight. I did the second test, and it did not spurt out, but the brakes did release immediately.

"If no external leaks are found, clean and remove the reservoir cap. Check the fluid level and condition. Then have a partner push the brake pedal as you observe the fluid. When the pedal is released, a spurt of fluid should be observed in the reservoir. No spurt of fluid indicates a clogged replenish port, and the only repair is a new cylinder. Unless the fluid can release into the master cylinder reservoir when the pedal is released, the pressure cannot be relieved and will hold the brakes on only partially. Another check for this is to apply the brake, released it and open the bleeder screw. A clogged vent or replenish port will cause fluid to spurt out of the bleeder."
 
#15 ·
I don't think that's true about the master cylinder.
Bob not to be an a** here but I know its true because I have physically seen it when I replaced the MC on the Shadow I took the original one apart to recycle the aluminum and saw clear as day the reservior does have an opening towards the top to allow fluid to go between either side if its high enough. You may have not noticed it before if you didnt pour enough fluid in to get one side significantly higher than the other. Like an ice cube tray you can fill one end up and keep filling the rest as the water spills over the top between them, same idea here. If I can find one today I'll cut it open on the band saw and take a picture if you think I'm full of it.

kerosene or brake cleaner will dissolve undercoating.
 
#16 ·
I understand what you're saying, however, both levels were down more than an inch, and both sides began to rise immediately as soon as I added fluid to one side. That's never happened before, and did not happen last month, when I bled the brakes on two different days. Only one side dropped, and only that side rose when I filled it. I'm saying that if it's constructed as you say, the fluid was lower than any barrier between them, and it still replenished both sides at the same time.

This picture shows what you're talking about, and I believe the fluid level was BELOW the plenum when I started filling:
http://www.tegger.com/hondafaq/mastercylinderreplace/howworks.html

This is a good article:
http://www.agcoauto.com/content/plugins/p2_news/printarticle.php?p2_articleid=202

I know I can dissolve the undercoating. I don't see a way to keep the dissolved solution from flowing into the crevices that require penetrating oil, and re-hardening there.

Tonight I'll do the 'spurt' test with the master cylinder caps off and determine if the ports are clogged. If not clogged, it's probably the hoses.
 
#17 ·
OK, here's the result of checking the master cylinder:

With both caps off the master cylinder, my wife pressed the brake pedal. A small gesyer of fluid shot up from the front reservoir when the pedal was pressed down, enough to get on the air cleaner. The rear did not react at all. I put the cap on the front reservoir and had her press and release the brake again. The rear reservoir very slightly went down, then up on release. Maybe a few mm. This was repeatable - a geyser from the front only, whether or not the rear was capped.

I then repeated this on the 84 Daytona. With both caps off, there was no geyser at all. The fluid levels simply went down a tiny bit when pressing, and back up again when releasing.

I should note that with both cars, the fluid came back up right away, there was no delay. The front reservoir of the 93 Daytona, however, did not go up and down in level, it just shot the geyser as if something is blocked. That reservoir should be for the left front and right rear wheels. Both front wheels drag, the right one a tiny bit more.

Thoughts?
 
#19 ·
No, but since I can't tell which is at fault, I ordered a new master cylinder and flex hoses from Rockauto, and some of the parts just came in. Brake flaring kit is on order for next week. I plan to change the master cylinder first, because it's easier to do without destroying the lines. If that makes no difference, I'll change the flex hoses. Could be up to two weeks before I have the time and weather to complete this. In the meantime, I'm spraying the connection between rigid and flex hoses, hoping to separate them without damage to the rigid line.
 
#20 ·
#21 ·
Yes, I looked at those last time. Wonderful tool, but $250 - $400, couldn't justify it for my use. But it makes a great double flare, perfect every time.
 
#22 ·
Well, I bench-bled the new master cylinder thoroughly, installed it, and tried again. When my wife pressed the brake and released it, it took several seconds for the calipers to release the rotors. When I had her press and release it and I opened the bleeder screw, it released right away.

So it must be the flex hoses. Never had an issue with them right up to the moment of this brake job. It's not going to be easy to replace them, but I'm going to try, without damaging the rigid line. If I do, I have a brake flaring kit on order, should be here Tuesday.
 
#23 ·
Well.

Got the right flex hose off without damaging the rigid line. Installed the hose, got that side all back together.
Left side, the rigid nut started to unscrew from the flex hose, but the line was twisting with the nut. And the nut at the metering block end was so tight, I had to use a 15" copper pipe on top of the tubing wrench as a breaker bar. When I did, the line broke at that end. I had anticipated this and bought a 12" new pre-terminated rigid line. So I bent this into shape (complicated shape, has a small, small crease, so I might re-do this line) and connected it and the new flex hose. The flex hoses are Raybestos professional grade, made in USA from RockAuto.

One thing I noticed was that the right hose had brass ends, the left one, steel-colored.

I used new copper washers on the banjo bolts, topped off the master cylinder and stepped on the brake. No apparent leaks. So I had my wife operate the brake while I began bleeding. That's when the NEW problem showed up.

I have bubbling and hissing and fluid loss at the left banjo connection. I double-checked the orientation, put two more new washers in, cranked it tight. Same thing. I bled the right side, all was tight, good and dry. For the first time, both front hoses flowed the expected amount of fluid instead of a trickle.

I also checked the release of the pads, and when my wife let go of the brake pedal, both front calipers release immediately and I can turn the front wheels by hand right away. No more 3+ second delay in the caliper releasing. So I solved that problem. But now I have a new one.

Either the new left flex hose or the reman left caliper is defective, in that I don't have a flat, tight connection at the banjo bolt. I will have to replace one at a time to know which, before I can return any parts. So the car is still off the road for another week.
 
#24 ·
Unless there was an extremely apparent defect in the hose banjo or caliper mating surface I have found sometimes you just gotta crank the banjo bolt especially on these newer replacement hoses where there going cheaper by using steel instead of the softer brass that takes a seal much better.
 
#25 ·
If I go any tighter the bolt will break or round off. I've never had a leak there in 34 years of working on cars. As I said, the right side was brass-colored, the left (leaky) one, steel.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top