Well I checked the senors they are clean and appear to be in good shape.
The instant hard pedal is what I'm running into but I tried coasting and had the same problem. Any suggestion as to what coulf be failing?

 

Thinking about it now Lack of vacuum likely. I remember when people first started tracking they ran into rock pedal at the end of high speed straights due to there being boost still in the system and no vacuum in the manifold in which to assist the Booster. Aka quickly from gas to brake without letting it coast or other assists. There may be an issue there even if vac is present where the booster isn't doing it's thing.

 

Yes, even a small leak in the one-way check valve on the hose to the intake can cause a pressure build-up in the vacuum-assist plenum that is supposed to hold - wait for it! - some vacuum. That will get you the "boosted-hard" pedal. But I don't think that this has anything to do with the OP issue. Jeff: can you log the speed sensors in any way and make sure that all four agree when you drive straight ahead at relatively high speed? In my searching around, dropped counts at one or two wheels was the only high-speed issue that came up. And cleaning the sensors was only half of it; also make sure that what is being sensed, as it were, is also clean and even. (I'm assuming that you have verified that all sensors produce about 1.45 kOhms at rest. Not only do they need to be within about .20 kOhms of 1.45, but they should be close to each other, as well.)

edit: you want to measure the resistance of the sensors at the ECU, if you can, in case there's a wiring issue between the connector in the wheel-well and the ECU.

 

a dead booster/vacuum problems doesn't cause single wheel lockups. Wheel speed sensors do.

The service manual describes how to test them. it involves measuring the voltage as you slowly spin the wheel to make sure that the high and low voltage levels are in the appropriate range for the controller to recognize and that all the teeth give you consistent levels.

It's possible that a sensor is on the edge of proper tolerance and at high speed causes a fault that clears at lower speed. Find out which wheel is locking and swap some sensors and see if it follows the sensor or stays with the wheel.

 

Wheels aren't locking.... the minute I go for the pedal it's hard and increasing pressure only engages ABS causing a near death experience. Disabling the ABS solved the high speed issue but due to lack of experience I would lock up a wheel once speeds dropped to reasonable levels.

 

OP, its an 8/9 right? Any reason you dont just remove the ABS system?

 

Have you checked the valve between the intake and the brake booster? There's a chance that having boost in the plenum is confusing the ABS ECU via the pressure sensors in the ABS unit.

Of course, if it's the 3-channel system with no EBD, then you might be better off without it.

 

I have not checked that ..... and honestly I don't understand how it all works. Odd that only 100+ activity is affected by this.

 

I experience the "rock hard pedal" in a certain straight/turn at my home track all the time. I think mine is due to boost/vacuum now that I think about it. I go from full throttle (80mph) to instant hard on the brakes giving it no time to build vacuum.

 

Again, if there's a small reverse leak in the valve, rather than isolating the brake booster plenum from the pressure in the intake, it allows it to leak in slowly. After a long straight under (turbo) boost, pressure has built up in the brake booster and made the pedal hard. You're only feeling it at 100+ MPH because it takes some time under boost to get to that speed.

 

Alright .... what is the part number for that valve? I'll replace that without hesitation.

 

sorry I wasn't clear. It takes two things for the ABS to activate, the determination of excessive wheel slip and the brake light switch to activate. The ABS "thinks" a wheel is locking due to a malfunctioning wheel speed sensor and activates as soon as you engage the brake lights. When the ABS activates it isolates the MC from the ABS pump, causing the hard pedal. The ABS releasing the brake pressure on the wheel it thinks is locking creates a brake force asymmetry which causes the car to dance around.

If you were only getting the hard pedal alone with no dancing i would be the check valve. The dancing and hard pedal points to wheel speed sensor. you just got to figure out which one unless you'd rather just replace all 4 $300+ sensors instead of troubleshooting it.

I should also add, that if the check valve had failed, the likely hood of actually engaging ABS is very low because you'd have to push the pedal so hard to actually lock any wheels. early ABS engagement is definitely not a sign of a dead check valve.

 

That was very clear and makes it very doubtful that all of his problems (if any) are due to the check valve. I can explain the hard pedal, but not the squirrelly behavior.

With that said, if he had a consistently bad wheel-speed sensor, wouldn't the ABS fault light go on as soon as he'd rolled 15 feet? I.e., after the initial wheel-speed sensor check. If it's really the case that he's only dropping counts at high speeds (which is causing the ABS ECU to infer lock-up), then I'm not sure that replacing all the speed sensors is the next step. Definitely check them with a volt-meter at the ECU end of wire, first, at least. And make sure that there isn't any crap between the sensor itself and the axle/hub.

If the speed sensors on an Evo 7-9 are like those on a 2G DSM, then another possibility is the wire on the sensor or the next piece of wire from the sensor's connector to the ABS ECU. The actual wire inside was crap and would give faulty reading when wiggled around. Maybe only at high speeds does this happen. To test this would require reading the signal at the ECU end while driving. In my case, I was lucky and it was the part of the wire on the sensor, itself. My wife sat in the car with the volt meter on the wires from the speed sensors and I shaked and wiggled as much of the wire as I could. It ended up being the wire on the sensor from one of the front wheels. It took an hour plus to diagnose, but at least I only had to buy one new sensor.

 

Thank you all.... I'll break out the mulitmeter this weekend and start testing the sensors. If any of you have the ECU break out schematic that could be beneficial as I'm not even sure if its the main ECU or a special ABS ECU.

As for disabling ABS all together... that is an option but I'd like to fix the root cause rather than ignore it. I've never had an issue with ABS before.

 

I haven't downloaded the manual for an Evo 9, but here's the link:

http://www.lilevo.com/mirage/EVO%209...vice%20Manual/

You definitely want to verify this, but it's probably the following:

At the ECU, the left front sensor is 22 & 31, the right front is 30 & 29, the left rear is 6 & 7, and the right rear is 8 & 9.

The initial test is to verify around 1.50 kOhm on each pair, regardless of whether someone is shaking the wire in the wheel-well.