1978 KZ100D Z!R Project

saxjonz wrote: Yes, the bubbles need to bleed through and for that to happen you need to tie off the lever.

This is a common belief, but I don't understand the logic behind it. When the lever is pulled to the bar, the front piston seal has moved past both return holes, so neither the fluid nor the air bubbles can flow from the line back into the reservoir. This has to be the case, otherwise the MC wouldn't hold pressure on the caliper piston when you pull the lever.



If you leave the lever in the resting position, the front piston seal is behind the forward return hole, which provides a (small) path back to the reservoir.



So why does blocking off the return ports help to make the bubbles go back into the reservoir?

My belief is that the process of bleeding the brakes tends to froth and aerate the fluid. Letting it sit allows the bubbles to collect at the top of the system. This will happen no matter what position the lever is in.

However, it is possible that pulling in the lever allows an air pocket trapped between the two seals in the piston to escape back into the reservoir, as it may be easier for this air to escape with the two holes exposed to the center section. I believe fluid passes from this center section over/around the front seal when the lever returns during the bleeding process. It's possible that removing air between the two seals allows fluid to properly flow into the main system.

So maybe, when people find they can't bleed the system at all, the real problem is that a big air bubble is trapped between two piston seals. On a standard (Non-Z1R) MC with the integrated reservoir it might be possible to intentionally pump fluid into this center section to purge any air. Maybe someone should try that. For science.

My opinion is that it is not about air. It's about rubber seal deflection.

By design, the caliper piston seals deflect with the piston when the piston moves out. Then when the pressure is relieved, the seals pull the piston back as they try to restore their shape. As the pads wear, the seals can't deflect enough due to the extended piston travel, and the piston slides through the seal a tiny bit to make up the difference. Then the seal pulls the piston back, but not as much as the piston had moved in the forward direction.

When everything is new and clean, the seals have a greater grip on the caliper piston and so they can pull it back farther when the pressure is released. This causes the next lever pull to have to travel farther to move the caliper piston back out. When you tie off the lever, overnight the piston seals lose their grip on the piston as they try to restore their shape and the seals retract a bit without the caliper piston retracting. So when you untie the lever and release the pressure, the caliper piston does not retract as far. The very next lever pull will be solid because it does not have to push the caliper piston out very far. Likewise the subsequent retraction won't be as far either, so the lever will continue to have good feel.

A similar effect can happen in the master, but is less likely, and won't have nearly as much effect due to the mechanical advantage of the master/slave piston area ratio.

(The movement and stretch are exaggerated in the animated gifs below.)

Normal lever feel and piston movement.

As the pad wears down, the piston will slowly advance through the seal, a tiny bit at a time so the lever will maintain a normal feel.


After a rebuild, the seal is stretched out farther than normal as the piston has to move a lot initially. Bleeding may not necessarily cure the issue, but will still get rid of air bubbles. Even after the air is gone, the seal may still be pulling the piston too far back.


Using the rubber band trick forces the piston to stay extended, but overnight, the seal slowly slides back partially. Then when the rubber band is removed, the piston only retracts a normal amount and the lever feel and piston movement are now restored to normal. In extreme cases, the lever may even feel more solid than normal, but the feel will return to normal as the pad wears down.

Wow, have to think about that one for a bit. The real question is: Did it work for you? Did you try it last night?! I’m excited to hear about it!

So, if a person was very careful during assembly to assure that each piston was installed so that it was sticking out from the caliper as far as possible, leaving just enough clearance to install the friction pads, the problem would be eliminated, or at least greatly reduced.
Seems worth a try.

TexasKZ wrote: So, if a person was very careful during assembly to assure that each piston was installed so that it was sticking out from the caliper as far as possible, leaving just enough clearance to install the friction pads, the problem would be eliminated, or at least greatly reduced.
Seems worth a try.

How? I’ve never had a problem with pistons being all the way in. If anything wouldn’t it eliminate the possibility of having air bubbles?
I’m still waiting on the verdict if that little trick worked or not.

Loudhvx, I see your theory there. But on my bike I never had the seals redone, I was just replacing that brake hose between the reservoir and the master. There was absolutely no movement on the piston whatsoever for that to happen. It’s interesting what you point out though. And yet I haven’t experienced that problem you speak of. I recent rebuilt all of my brake calipers (I powdercoated them) so as I reinstalled them I had no issues with brakes not pumping up.

If he installed the seals in the rings and lubed them and put the piston in master correctly he would still have that problem we are speaking with in that there is no pressure coming out of master. I don’t know how that works, I suppose I need to take a closer look at the innards of the master.

I’ve been thinking on that for awhile now. As long as there is pressure on the piston it can only retract so much. I think there is more push than retract value going on than retract. I remember cleaning my brake calipers doing a brake job last time and I pushed the lever in to have the pistons move out just a little so I could clean around the area (remove the dust) and the piston just popped out so I don’t know if that would be the case although I’m not exactly sure I just don’t think so.

TexasKZ wrote: So, if a person was very careful during assembly to assure that each piston was installed so that it was sticking out from the caliper as far as possible, leaving just enough clearance to install the friction pads, the problem would be eliminated, or at least greatly reduced.
Seems worth a try.

In my experience, it's pretty difficult to get just enough clearance to install the caliper over the pads without pushing the pistons in a little extra. But yes, if my suggestion is what's happening at least in some cases, that might be one way to prevent the soft lever. I think the rubber band trick would be easier.

I should emphasize, my "theory", is just purely conjecture. I haven't actually had this problem, but know several people (in addition to Saxjonz) who have described the rubber-band trick and it worked for them.

I actually have the opposite problem. On one of my bikes the brakes have a very solid lever, but the brakes feel like they are dragging a bit. I think the seal is not retracting the piston as it should. It may be the seal is too loose and not grabbing the piston. Another clue is that it occasionally, while sitting for an extended time, loses all pressure. Then I have to bleed air out of them. But no fluid leaks out anywhere. I think air is getting sucked in though the loose seal during temperature changes.

So what else could cause the soft lever? As DOHC points out, when the pressure seal of the master moves forward, the entire pressure portion of the brake system is sealed off. If all the seals hold, no air can get in or out. This is the case during the rubber band trick. That is why I suspect it's a rubber-deflection.

But before going into the rubber, let's look at air again. Even in my animation, you can see there is a small crevice behind the seal that air may get trapped in, and might not move away from right away. If a bubble was stuck there, maybe the rubber band trick would push it out and somehow fluid would not also leak out. There may also be a situation where air gets past the pressure seal of the master, and exit into the re-capture area between the two seals. (The rear seal is never under pressure. It is only there to capture the leakage of the front seal and from there the fluid, or air, returns to the reservoir.)

In order to properly bleed a master, the clyinder of the master needs to be angled in a way that the high point is near small hole just in front of the front seal. That's only place where air is intended to exit the master cylinder. If the high point is elsewhere, then not all of the air would get out. If the high point is near the seal, but the master is angled so the bleed hole is not at the top, then the rubber-band trick might get that air out. If the master is angled so the high point is away from the seal, then the air would never get out.

Getting back to the piston seal. One big, obvious flaw in my suggestion is that the caliper should be designed so that excessive seal movement is prevented by the shape of the chamfer on the seal groove. In other words, the seal is meant to deflect all the way up to the chamfer every time.

Another possibility is a distorted seal in the master. If the rubber part is distorted, it may take some lever movement to restore its shape before building pressure. Eventually, over time, the rubber will return to its intended shape and the lever will regain a solid feel. The rubber band trick may just speed up that process.

New brake lines may also need some pressure cycles to achieve their final shape, whether it's from expansion before the reinforcing weave gets tight, or if there is extra expansion do to the hose being curved or kinked in storage.


So there are clearly quite a few possibilities to explore, and probably many I'm not coming up with. The seal theory may be a weak one.

saxjonz wrote: I suppose I need to take a closer look at the innards of the master.

It's actually quite simple on the Kz's. If I get a chance in the next couple days, I'll try to make an animation of that.

This is just the normal operation during braking.


The way the master bleeds itself is that little bleed hole in front of the main seal. When air is at the top of the cylinder, near the seal, the air bleeds out as the seal starts to close off the port. When there is no air, it creates a little jet of fluid. If the reservoir is a remote type, then you won't notice the little jet of fluid.

In order for the air to bleed out, the bleed hole must be the high point in the system. If the cylinder points upward, air will get trapped in the end away from the seal. If the master is rotated so the bleed hole is not at the 12-o-clock position, air can get trapped near the seal, but won't bleed out. In this case, the rubber band trick may get the air to leak past the seal and go into the fluid re-capture area (area between the seals). From there the air will eventually migrate into the reservoir.

loudhvx wrote: This is just the normal operation during braking.

Your animations are fantastic!

I've been thinking more about the idea of an air bubble trapped between the two seals in the master cylinder.

First, vacuum bleeding the system will not remove a bubble between the seals. It just pulls fluid from the reservoir past the front seal.

Second, if you picture your master cylinder animation with the lever pulled all the way to the bar, when you release the lever, there is a fair amount of return travel before the front seal passes behind the front bleed hole. I believe during this part of the return travel fluid can pass around the front seal, transferring from the center section to the main system. If there is a large air bubble trapped here, you'll instead pump air into the system each time you pull the lever. And air right in front of the MC piston should make the lever feel pretty dead. And in order to bleed this out you'll need to pull the bubble all the way down to the caliper. And the next time you pull the lever it happens again. Which would be as frustrating as it sounds.

I'm pretty confident that fluid is supposed to flow around the front seal on the return stroke. The factory master cylinder on the Z1R has holes covered by a check valve in the solid ring behind the front seal. The K&L MC piston does not, but it does have fluting in the front seal that looks like it might aid fluid passing over the seal.



And third, as you said, the rear seal is never under any pressure. There is no pressure at all that would help to push air out of the center section. Gravity is all you have. Those holes are pretty small. and decades of gunk just makes them smaller.

I could believe that tying the lever to the bar and exposing the two bleed holes to the center section of the piston would help a bubble get out faster. Of course, as with my KZ650, just letting it sit for long enough could allow the air to slowly bubble out.

Either way, when you come back and the center section is filled now with fluid, the next pull of the lever pumps that last little bit into the system and all of a sudden you have a firm lever.

I do see how a bubble could get trapped between the seals, but I cannot even imagine how that would make a soft lever, or one that travels too far, since the pressure is supposed to build down stream of the primary cup.

That's an interesting idea... air migrating from re-capture chamber to pressure chamber during the return storke during bleeding. It makes sense that fluid must go past the pressure seal (or through the check valve if so-equipped) on the return stroke otherwise there would be a vacuum in the system (and the spring may not be strong enough to overcome it). So air could presumably travel the same way. But I think eventually that air should get bled out the normal way. That is, through the small bleed hole, as long as the bleed hole is the high point of the pressure chamber. I don't think air in the master can get entirely bled out through the caliper without some radical orientation of the master assembly. Some of it gets trapped, so it must bleed out through the small hole.

I agree what you are saying is possible, but it just seems to me that pumping it over and over should eventually get the air out through the small hole. It certainly should be added to the list of why the rubber band trick works.

Vacuum bleeding should pull fluid from the reservoir into the pressure chamber through the bleed hole assuming the master piston is all the way retracted. I can watch the fluid go down the small hole during a vacuum bleed but not much happens with the larger re-capture hole.