Bleed the ABS (C4)?
 

Getting ready to do a complete bleed, but I haven't found any procedure to bleed fluid from ABS circuits. I started thinking that fluid in these circuits could be very old if its not circulated somehow, short of hitting brakes hard in the wet. Then, after I change fluid, if the ABS activates the old fluid is injected into primary path.

Do the the normally closed solenoids open and pump run when ignition is turned on? Do I need the ABS tester to open solenoids and run pump?

Just curious about this.

Thanks for help,

Dan

+964
You are absolutely right, the service manual makes no mention of special procedures for renewing old fluid in the ABS circuits. I was curious and cut apart one of the nominally identical MB servo-valves to look at these issues.

There would seem to be no way that the built in test could effectively clear the return lines in the tiny amount of time the pump runs. Further, since we still have brakes during the test, the valves are not actuated long enough (if at all) to circulate the fluid.

Several Rennlisters (e.g. Spyrex and Yahoo!) have replaced their servo-valve assemblies and had a pretty difficult time getting the air out of the replacements. They describe just hammering the valves with the ABS test sets to exercise the valves and pump, allowing the power bleeder to force fluid through all the passages. It seems like we should expect much the same during the fluid renewal operation, at least we don't have the air.

The "Reduce" functions on the ABS test sets allows each of the three valves to circulate and return fluid with the pump running. With a power bleeder attached this should give you the best chance to move new fluid through those circuits.

Best regards,
Michael (cajonfan)

OK. But what's the best process, or order of bleeding. Would I do the ABS first, or last? Seems if I change fluid in reservoir, then initiate 'reduce' function, I'll end up with a mixture of new and old fluid in reservoir. I dont think I'd want to use that 'contaminated' fluid to bleed brakes, for example. Should I even worry about the contamination?

Maybe disconnecting the return at the reservoir (if possible) and collecting old fluid might work (if I'm understanding the ABS fluid flow correctly).

Given the threads describing how ABS pumps degrade from lack of use and old fluid I would think someone would have come up with a DIY by now.

I'll go take a good look at the ABS connections to reservoir and see what makes sense.

Edit: The ABS returns fluid to the master cylinder, not the reservoir, so the lines cant be disconnected. There are check valves on the input and output to the ABS pump. I am assuming these are passive. So there must be adequate pressure on the input side, that stores fluid in the pump reservoir, before the check valve will open allowing the pump to force it back to the MC. I'm thinking only a small shot of fluid will get forced back before pressure equalizes and input check valve closes. I'm not sure if that volume of fluid returned is enough to displace fluid in lines and MC. So you'd get a charge of old fluid in lines that has to be flushed out the brake bleeders, at least enough to get old pump fluid now in lines past ABS. Repeat process a few times.

This will only work, assuming input check valves are indeed passive, if pressure bleeder generates enough pressure in lines to fill pump reservoir and open pump input check valve once solenoid is opened. I have no idea how bleeder pressure compares to boosted pedal pressure. So for all the bother, nothing at all might happen.

Can anyone comment on my description of ABS function? Is ABS bleed possible with a pressure bleeder?

Dan

I can't comment on bleeding procedure but there is a way to actuate ABS in the driveway for extended periods of time by changing the resistance of one of the speed sensors (disconnecting one sensor might produce the same). The ECU will eventually throw a code, but I was able to run the ABS pump continuously for 10-15 seconds before ECU disabled it. Might help circulating the fluid more efficiently.

Thanks!
Andrew

Andrew can you explain how you did that?

By accident actually, one of my sensors broke (resistance out of range). This error returned wheel speed different than other wheels and so when slow moving in the driveway <5 km/h the ABS would continuously engage with brake pedal depressed lightly and car barely rolling.

I'm not 100% sure if this will work with the sensor disconnected in the frunk, but its worth a try.

Thanks!
Andrew

You can't actually circulate the fluid. There is never a complete path, only from MC to brakes with solenoid closed (Pressure building phase) and brakes through pump to MC with solenoid open (pressure relief phase). According to Adrian's book pump runs continuously in 'hold' (intermediate brakes isolated phase) and 'relief' (brake pressure relieved through pump) states which are triggered by slip condition.

Fluid would only move through pump in 'relief' phase if pressure across check valve was great enough to open it. I guess this could caused by combination of high brake back pressure and pump negative pressure. I would guess system is designed to not suck fluid from an unpressurized brake line, so you'd have to build pressure in line and then open solenoid and turn on pump to push a bit of fluid back toward MC until pressure normalized and check valve closed.

Question I have is whether pressure bleeder can create enough brake back pressure in relief phase to open check valve. I can't see how you could ever tell as you'll never see the fluid moving in the tubing.

???????

The specific question is about understanding ABS mechanism and the way to perform the ABS unit bleed. This is not covered in RicardoD 13 point bleed article.

Thanks!
Andrew

Dan's description pretty well sums it up - we don't really know how much fluid is moved each time the "reduce" function is actuated, and we can't directly see it.

But let's not lose sight of the big picture -

If the fluid in the system has been periodically replaced, then any circulation of fluid through the valves will be an improvement over the stale fluid that would be present if the valves were not actuated. There are likely cars that have 25 year old fluid in their servo-valves and their ABS functions could be useless.

So - periodically exercise the ABS system and change the fluid.

Michael (cajonfan)

So here's my strawman process for getting new fluid into ABS reservoirs and pump integrated with a brake bleed. I am assuming one cycle of a 'reduce' test will not push ABS fluid past the MC.

1) Change fluid in reservoir.
2) Bleed the bomb.
3) Bleed the dual solenoid valve block
4) Prepare to bleed rear brakes.
5) Draw some fluid through RR brake, enough to flush MC to ABS path and beyond. Your guess is as good (or better) than mine. The idea is to get new fluid in line past the ABS as this is fluid that will be pulled back when solenoid is opened.
6) Using the ABS tester, perform Test 5 part 3. This will start pump and open rear solenoid flushing the rear ABS circuit path.**
7) Bleed the rear brakes.
8) Set up to bleed RF brake. Draw some fluid as before.
9) Using ABS tester perform Test 5, part 2.**
10) Bleed RF brake
11) Setup to bleed LF brake. Draw some brake fluid.
12) Using ABS tester perform Test 5, part 1.**
13) Bleed LF brake.

** During ABS Test 5, you are instructed to press the brake pedal. At this point the pressure bleeder is still attached and pressurized. (1) I am not sure of advisability of pressing brake with pressure bleeder attached and pressurized. What can go wrong? (2) I don't know if pressing brake is even necessary, i.e., to pressurize brake line to high enough back pressure.

After above procedure, all new fluid should be throughout brake system.

You might think you can get away with doing Test 5 before your brake bleed thinking your current fluid is old, but not that old. However think about what happens. When you do part 1 of Test 5, you push old ABS fluid upstream towards MC. It stays there. When you do part 2, that old fluid gets pushed back into ABS and into line, only to get sucked into pump reservoir when you open solenoid. Best practice would be once you've flushed one solenoid, then bleed that circuit to get rid of old fluid before going on to next solenoid.

Make sense??

Now if I could just get my ABS connector off without super human strength... I cringe when force levels go up.

Dan

Looks like a solid plan,

One comment -
Your brake pedal intuition is right on, in the test you are verifying that the reduce function fully separates the MC from the caliper, but in this case you just want to open the valve. There is no need to actuate the brake pedal.

EXCELLENT thread guys!

What ABS tester did you use? Will it work on a 1999 C4?

You do not need a tester to bleed the ABS pump itself.
Several years ago, i found this here on rennlist and this helped me after my last two complete rebuilds of the brake system to bleed the system properly:

>>
The ABS pump is made up of a motor and 3 solenoids, which control the right front, left front and both rear brakes. Using 2 sets of jumpers, you can power up the motor and the solenoids.
The first step is to remove the screw on the 12 pin connector on the pump and disconnect it. The solenoids are pins 1, 3 and 5, with pin 4 as the common. Pin 9 is the motor.
Use a pressure bleeder in the normal fashion. To activate, tie 1, 3 and 5 together and prepare to hook to the positive side of the battery. Wire #4 to ground. Run a jumper from #9 and prepare to connect to the positive side of the battery.
Now, you can power up, For A Short Period of Time, both the solenoids and motor while bleeding each caliper. Note that the solenoids and motor will get hot so don’t keep them powered for too long!
hope it works for you!


>>

You just need some wiring with two clamps for the battery connection and some round crimping sockets, that fit the pins from the ABS... Nothing fancy there, the pins are just the connection to an electro-motor and trigger the valves in the ABS block...