Back in July, I drove through a bad storm. When I was backing out of a driveway that afternoon, the power steering went out when I turned full lock left. Recalibration (switch to on; lock to lock, then start) was usually able to get it working again when it failed a few more times, always when turning full lock. A few weeks later, recalibration stopped working completely if the car was on the ground. I tried soldering the torque sensor connector at the radiator junction and replaced the whole harness (again, soldering the torque sensor connector) with a junkyard harness with no change. Cleaning the connectors at the rack with the CRC cleaner didn't help either.

If I calibrate while the car is off the ground, it will reset until it encounters any steering resistance (either full lock or any steering input with wheels on the ground). It has consistently thrown a B1342 code. Is there any additional troubleshooting I can do to narrow it down between wiring, rack, and PS control unit? The control unit is obviously the easiest part to replace, but most people's B1342 codes seem to be related to wiring.

FIXED!! The repair for mine was cutting off the two spade connector terminals that feed power to the EPS computer. These are different from the grey connectors everyone else talks about under the airbox that get soaked by the overflow tank. The spades were corroded and loose. Rack/computer were fine. Photos of the culprits are in posts #19 and #23.

 

I pulled off the 4 connectors on the control unit today. All contacts in there were shiny, but I bent the female ends on all the power connectors to get them a bit tighter. A-B and B-C on the torque sensor connector both measured ~15 ohms which is what it should read at the sensor itself (12-15 according to the shop manual). Power cable to rack measures around 8ohms. I cleaned the rack power connector again too. No change--PS light still comes on with any steering resistance (wheels on ground or full left/right in the air).

 

I suddenly was also greeted by 2023 with no power steering. I have Code C1099. I did all diagnostics, everything checks out... even the sensor wires were previously soldered. They are dry anyway. Curious what CODE you might have?

I even swapped EPS modules, same issue. Last thing is a new rack, but I don't want to be wrong.

 

I pulled a rack and computer from a junkyard over the weekend; replacing both is likely my next step.

 

I missed the Code B1342 part... thanks. Mine says Motor, which means a new rack.... great. I am picking up a used rack and EPS computer this week as well.

I checked all the wiring like 10 times.

 

Turns out my scanner can display live data from the EPS system, so I was able to see what the system is outputting. I also changed the computer; no effect. Torque sensor seemed like it was operating correctly. The motor current did bounce around between 1A and 5A as I was turning the wheel before it threw a B1342 again. It also gave me a C1099 once, but I haven't been able to replicate it. The torque sensor still outputs data even if a code is present.

The last thing I tried was to put 12V directly to the rack. On my old harness that was off the car, I de-pinned the connector that would have been at the radiator, hooked it back into the rack, and MOMENTARILY TAPPED those exposed pins against the terminals on a spare battery. The new rack, off the car, moved smoothly, but the one currently installed seemed a little choppy. I don't think I have any other options at this point besides swapping out the rack.

 

I was able to pick up a used rack and another EPS module today. I will try to test the new rack by extending the wires to it and look for different results.

I never get 12V to the rack by measuring the connector, but I don’t know if it needs to physically connected for the module to recognize it’s there and send 12V. It’s a mystery how the EPS operates.

 

12V is always present at the feed to the computer, even with the ignition off. The computer then uses input from the torque sensor to decide how much power (or at least which direction) to apply to the rack motor. The polarity given to the rack determines which way it turns. Unless the computer is supplying power to the rack, you won't measure any voltage there. You can, however, measure the resistance at the output connector (disconnected from the computer) and through the noise filter with the car off to at least see if you have continuity through the motor.

Now how exactly it determines what codes it's going to throw... that is the mystery.

 

I had the same issue as Op years ago. Ended up being the boots around the inner tie rods were not sealed and my rack was full of water, almost a full year after the drive! These boots likely need replaced if you haven't installed your new one yet.

 

No water in mine, but even changing the rack didn't fix my issue. Not sure where to go from here.

 

New computer did not fix C1099. I have a new rack electrically connected, no codes... but then again I haven't put any torque on it since its not connected to anything. Guess I need to simulate torque now.

 

Any progress?

I tried 3 racks, 3 EPS Modules, 2 PCMs, 2 Instrument Clusters, Bypassed all the wiring harnesses, Reset NVM and Fuel trims/etc., still have no power steering. At a loss as well.

 

Dumb question but you check the fuse? Not just visually but actually replace it?

 

Not sure if you are referring to me or OP but I did check and test that fuse, it's totally fine. First thing I did. Of course, easy never happens to me and my RX8 currently sits uselessly in my driveway as I ran out of ideas on what the issue could be.

 

There aren't that many places to look. Its either a wiring issue, the steering controller, or the rack
From what i have seen... the wiring is the. Mist common forrlowed by the controller and then the rack. I have seen a couple that worked with different controllers that the original controller worked in a different car... so i figured from that you have to check the wiring from the controller end as well as thats likely the issue

 

It's frustrating. 3 EPS modules, 3 racks, 2 PCMs, 2 Instrument Clusters. I verified PIDs and Canbus functionality. Grounds, opens, voltages. VSS, Steering Wheel Angle sensor, RPM signal.

I only get 3+ volts out of both black wires from noise filter to rack, so essentially 0. Nothing I do makes that voltage differential change. I have no ideas left.

 

The same issues with my rx8 only started after I reflashed my ecu with a stage 1 tune. Have soldered the connectors, and ran all continuity and ground tests as well. The weird part is that it worked for me for a solid second until it shut itself off. Going to try a replacement module soon but I doubt that would change anything. My best guess at this point is that there's a calibration or something that we're missing.

My current attempts have been at trying the neutral position setting that's described here: https://rx8handbuch.de/esicont/en/sr...html#wp1025486
There was one thread where he described nothing working until he raised the front tires off the ground during maintenance and it fixed itself randomly. (it's here on rx8club somewhere) At this point with everything else you have described and also what I've tried that has to be the only possibility left.

I've attempted using forscan to trigger the TRQ_S_CAL command, and I've also seen your username on the forums over there too, but even following that procedure listed above to a T hasn't changed a thing for me. While logging the torque sensor too it seems maxed out every time I turn the wheels, though the sensor isn't faulty I believe as its resistance was within 12-15ohms. Is there another reset or calibration that must be performed? If not any ideas? I'm stuck same as you.

 

I tries every calibration procedure documented, although forscan fails the neutral procedure. But that usually is for calibrating the left right balance of the power.

My steering 'seems' to have died right after I changed the fuel pump and emptied half my gas tank by jumping the pin in the fuse box. So I checked the entire fuel system to see how that could possible interact with power steering, but nothing.

The torque maxes out at +/- 7.2 when the engine is off, and there is 0 amps for the motor. When the engine in ON, I get the same exact results, where when my friends car is on, you can see motor amps going to 20-40 when torque is only like 1-3 (no where near max of 7.2)

There is some magic undocumented interaction that we just don't know about, which is extremely frustrating.

 

There are two additional power connectors that I have never seen anyone else mention that have fixed most of my problem. I have a Keisler LFX swap, so these may be in a different spot if stock, but for me they were fairly close to the two grey connectors that all the other threads talk about. These are heavy gauge wires that feed and ground the PS ECU. The connectors are poorly waterproofed single-spade in big chunky housings. One photo shows how badly corroded one of the removed spades was. The other photo shows one of the connections with one side of the housing already removed. I cut off both ends of both wires (positive and ground) and soldered them directly together. Ground was way worse than positive.

The new rack now can put out the full 85A at full lock according to my scan tool, but it's "choppy" when turning, and it even took off on its own once. The more I moved it, the better it got though. I read some other accounts of folks having this problem right after startup, then it goes away until the next cold start. I'm going to try cleaning/tightening the torque sensor pins at the rack one last time, but if that doesn't fix this rack, I'll throw my old one back in which I know works. The new one shows a small torque (.75 ft-lbs?) when the wheel is centered with me not touching it (even on a lift), so I suspect there's extra resistance from a bad connection.

 

delturious - is this the ground from the battery to the connector to the 2A input to EPS Module?

 

There are TWO of these connector pairs--one that's between the EPS fuse (positive) and the EPS computer as well as one that is between the EPS computer and chassis ground. Mine are likely in different locations than yours from my motor swap, but they are the OEM connectors. Both were awful.

 

delturcious Holy S! It Works! Even though I had measured +12V and Ground a hundred times on the 1A/1B input to the EPS module, I never tried to jumper it and bypass it altogether. BAM, I have power steering! Thank You for the GROUND cable idea. I need to now look closely at that wire and see where it falters and fix it.

 

Yeah!!!

After some more cleaning/inspecting on my replacement rack, I decided I couldn't trust it (torque indicated by the computer with no steering input spooked me) and put the previous rack back in. All is good now! These are the plastic housings from the horrible connectors that were at fault on my car if it helps:

 

So for me, it ended up being 60amp fuse block assembly. Obviously it looks corroded, but it still passed all the tests. Eventually I smashed it all open and I found that there is a terminal in there that fell apart. I couldn't just unplug it cause all the plastic was bonded together. So now I have some parts on order to make a bypass.