10 Attachment(s)
ABS Flush with Test Cable
After discovering a brake fluid leak from an overflown ABS brake fluid reservoir goldnsx, britlude and Kaz pointed towards air in the accumulator and a non-closing inlet valve of the ABS. This is something of a speciality due to the inner workings of the Nissin ABS system.
To understand, first a little technical background: An ABS system prevents wheel lock-up during braking. To do so it utilizes a wheel speed sensor at each wheel. To make a locked-up wheel turn again a modern ABS pushes brake fluid from the back to the master cylinder by means of a pump, valves and a small accumulator which causes the wheel pressure to drop until it's no longer locked-up. This causes the typical pedal vibration during ABS braking.
The Nissin ABS of the NSX uses a different approach. The wheel speed sensors work the same way but wheel pressure is controlled differently.
As soon as the car reaches a certain speed (~20 km/h) the ABS pump starts to charge a high pressure accumulator. A well known sound of 90's Hondas, typically right after leaving the garage. The pump stops once the pressure switch signals sufficient charge. If it runs for more than 120 seconds an error is generated, the ABS lamp lights up, the pump is switched off and the ABS system is non-operational.
Attachment 13501
Let's assume everything is OK and the accumulator was charged without problems. The outlet valve is open, the inlet valve is closed and the pressurized brake fluid in the accumulator is kept ready for usage.
A leaking inlet valve can cause the reservoir level to rise after the car was parked and causes the pump to run more often (e.g. every time you leave the garage). If the inlet valve seals well, pump runs are quite rare and the reservoir level is stable.
If a wheel locks up during braking, the outlet valve is closed and the inlet valve opened. Pressurized brake fluid from the accumulator flows into chamber C and pushes the slide piston up. As the slide piston rises, the cut-off valve disconnects the master cylinder from the wheel (caliper) and simultaneously enlarged chamber B which reduces the pressure at the wheel.
If the wheel pressure is supposed to be kept, both valves (inlet and outlet) are closed. To increase wheel pressure the inlet valve is closed but the outlet valve opened. The fluid in chamber C is discharged into the reservoir, the sliding piston moves down, chamber B becomes smaller and the wheel pressure rises. When the sliding piston hits bottom the cut-off valve re-connects the wheel to the master cylinder.
The typical pedal movement is created by volume change of chamber A, by the way.
This mechanical construction connects the master cylinder and the ABS system only by means of the sliding piston. Neither fluid nor air is mixed between the two systems.
How does this explain the reservoir overflow? Well the accumulator uses a rubber diaphragm that never 100 % airtight (nitrogen from below very slowly penetrates the rubber). When this air runs through the inlet valve it bursts into millions of small bubbles, creating a kind of "brake fluid foam". This foam has a totally different characteristic than pure brake fluid.
Normally, small amounts or air leave the accumulator when the ABS is active and don't cause too much trouble. Unfortunately, when the ABS is not used for long periods of time the amount of air can become significant and the foamy fluid might even be sucked right back from the reservoir into the accumulator, further worsening the problem and can even prevent the ABS from working at all (!).
The official method to release the air is to use a special Honda tool called the "T-Wrench" (part number 07HAA-SG00101, pricey and difficult to obtain). It does the trick by connecting it to a bleeder screw near the accumulator. If there's no access to the tool, if air is trapped past the inlet valve or in case of sticky or leaking valves it's not of much help, unfortunately.
So how to release the air and flush the valves? By means of a self-built ABS test cable. It's very similar to the famous "Danoland" flush procedure: http://www.danoland.com/nsxgarage/brakes/abs/abs.htm but with a little extra.
Attachment 13502
A four-way-switch (Joystick) individually energized each inlet valve valve. The "outlet valve extra" switches energize inlet and outlet valve simultaneously if activated. This causes the sliding piston to be moved up just like during regular ABS control. Any remaining air in chamber C should be pushed out that way.
Attachment 13503
An additionally fitted push switch is used to manually active the ABS pump. Please note that the test cable bypasses the pressure switch and therefore please keep the run time below 30 s and allow for cool-down.
Attachment 13504
The connectors needed are Sumitomo HW090 (all three-pin variants, including male and female) and might be hard to get. The website CycleTerminal was of great help in this particular case.
Attachment 13514Attachment 13515
Overall costs (including all parts, switches, housing, etc.) summed up to about 60 Euro and an evening of work.
Using it at the NSX is rather simple:
- Connect the battery terminals
- Open the vehicle fuse boy in the front near the firewall
- Remove the ABS relay
- Connect the pump switch to the terminal pair towards the large block fuses (marked red in below image)
Attachment 13505
- Connect the individual valves (note that these are the cables with a green sleeve, it's possible to connect the to the ABS computer, too but that's probably a bad idea..)
- Ready :-)
Attachment 13507
First activate one inlet valve by pushing the four-way-switch into one of its directions. In my case, huge amounts of air were released into the reservoir. It seemed the accumulator was mostly filled with with air at that point. Now remove the excessive brake fluid/foam by means of a turkey baster, large syringe or similar.
Once the accumulator is empty, new brake fluid is added almost to the neck of the reservoir. Running the pump sucks the new fluid into the accumulator as pressure is build up. Repeat the procedure until almost no air/foam is discharged from the valves when activated. I was not able to completely avoid it but it was only a few drops of milky white fluid in the end.
Attachment 13506
Now enable the "outlet valve extra" switches to flush the area past the inlet valve. Note that no fluid or air is releaed while the valves are avtivated but you can hear the fluid pushing up the sliding piston.
Once the four way switch is released, the fluid is pushed out into the reservoir:
https://www.youtube.com/watch?v=e6oYr_qVloo
This is an alternative, improved design. Easier to use and build:
Attachment 13516
PS: This is a technical summary of the corresponding thread: http://www.nsxcb.co.uk/showthread.ph...-Bottom-of-ABS
