So, We have been using West coast fiero's Saturn vue power steering kit. As anyone who has done this swap knows, it kind of interfers with footroom. Given my Client 's is in his 70s, he has an even harder time getting in and out of the car. So we did some research and found that this 2007 Mazda RX8 EPS rack has the motor contained in the rack, and the rack is approx the same 24 inch length between the inner tie rods in the center. We then made plastic pieces to mock up how we would make the brackets to hold it, and we came up with this design that we had waterjet just out of 1/4 steel. I got the steel parts back from water jet and tacted them into place. You can see the old Bracket on the Driver side gets repurposed, and the passenger side one gets eliminated.
Here is what we have so far. I am going to explore controlling it with the Saturn EPS controller. Will update when i know more. ------------------ 86 SE FASTBACK T-TOPS 88 FORMULA 4.9 87 GT MEDIUM RED METALLIC stockish
EMAIL : Anaverin@gmail.com
[This message has been edited by anaverin (edited 10-05-2020).]
I love the idea of an electronic assist rack. If I do any more power steeing systems in a Fiero, I will look into an electric rack. I am not sure about the electronic controls of it tho. I will keep an eye on this thread. Thanks
I am just wondering if i can use the Torque sensors in the Mazda rack, (which i have all the factory service information for) and soder them into the saturn controller. Alternatively I could modify the mazda controller to work, but since we already have the saturn controller working, i would like to try that first. I am going to ohm out the sensors on the saturn rack, to see if they are similar to the 14 ohms of the mazda sensors.
Rest assured though, after spending money on getting waterjet cut parts, I will be figuring out a way to make this work!
thank you for your interest, i will try to upload the service manual stuff for the mazda in case you are curious. I have actually done more research then i was able to convey with the post.
If you have questions about that, I know a thing or two about EPS.
You can always post pics of the connectors and where they go, and we can attempt to figure things out.
[This message has been edited by anaverin (edited 10-05-2020).]
If the Mazda controller already normally works with the Mazda motor and the Mazda torque sensor, then it seems like it would be easier to work with the Mazda controller, but that's just my first impression.
To make the controller work, it needs to think it's in the Mazda. I think it would be easier to make the Mazda controller believe the lie that it lives in a Mazda. This would be telling it the proper lies via the CAN bus.
The other choice is to make the Saturn controller work with "major organ transplants"... the thing is that the Saturn controller can't reject its new friends!
If the Mazda controller already normally works with the Mazda motor and the Mazda torque sensor, then it seems like it would be easier to work with the Mazda controller, but that's just my first impression.
To make the controller work, it needs to think it's in the Mazda. I think it would be easier to make the Mazda controller believe the lie that it lives in a Mazda. This would be telling it the proper lies via the CAN bus.
The other choice is to make the Saturn controller work with "major organ transplants"... the thing is that the Saturn controller can't reject its new friends!
I guess i am concerned about the mazda controller wanting to talk to the pcm via can bus. The Saturn controller has already been rigged to work via a dial that is nicely installed into the AUX pod. The owner would like to keep that if possible. As far as usability goes, the setup is amazing other then the motor intruding into the footspace and looking ugly. We will see how this goes. Anything is possible, but we will be working until the solution is satisfactory.
First tests are as follows: Mazda RX8 TQ Sensor at rest =12 ohms Saturn TQ Sensor at rest = 5 ohms
currently we use the PWR Steering at less then 50 % assist on the Dial. So being that 12/5= 2.2 maybe that means that we would get the same assist (assuming the motors draw equal power, which is a big if) at 110% on the dial. We will be doing some practical experiments this week but farther research indicates that it has proven very complex to replicate the Can signals needed to make the RX8 module work. so for now, we will be trying to make the saturn module (which is a Denso / KOYO product anyway ) work. will let you know how it goes.
[This message has been edited by anaverin (edited 10-06-2020).]
Originally posted by anaverin: First tests are as follows: Mazda RX8 TQ Sensor at rest =12 ohms Saturn TQ Sensor at rest = 5 ohms
Huh, cool.
If both torque sensors are variable-resistance output, that could be plug-and-play.
quote
Originally posted by anaverin: We will be doing some practical experiments this week but farther research indicates that it has proven very complex to replicate the Can signals needed to make the RX8 module work.
Usually what works to get up and running (at least half the time) is to simply record the CAN bus traffic of an idling vehicle onto a PC (using a CAN to USB interface). If you play back the recording, oftentimes that's enough. If you're lucky, you can sit in someone's idling car and get a recording from the OBD-II port.
Then, to determine which CAN messages are essential, you play back the CAN trace, but with one less message each time. Once you're at the point where taking away any single message causes the EPS to stop functioning, you know you're at the bare minimum.
Here are the patterns for the water jet service to make the parts for the RX-8 mounts. NOTE that the bolt holes in the Part 1 graphic ARE NOT to size, so only a pilot hole, maybe 1/4" dia, should be water jet cut.
[This message has been edited by Getafix (edited 10-07-2020).]
Jig for aligning the mounts for the RX-8 EPS to the '88 Fiero suspension. Two 1/2 aluminum "L" rails with alignment jigs are taped into place. Those jigs locate the notches in parts 2, 3, and 4. After those parts are tack welded into place, the "L" rails are removed and the two parts 1 are tack welded onto the parts 2 (x2) and parts 3+4. The part 1 that has the 1/4' milled out goes onto parts 3+4 with the "milled" relief facing up to clear the RX-8 EPS.
All EPS racks provide assistance when they get steering input which would cause them to seem like steer by wire if they were getting their steering input from a wheel not connected to the steering shaft.
Working on the mods for putting 88 suspension into the 86, so that is what is holding us up. We took a lot of pictures so let us know is we need a separate thread about that.
i'm told it 16:1 but, it seems to be able 2.5+ turns lock to lock iirc.
Next up we needed to modify the column to add the torque sensors ONLY, which should not interfere with footroom, making the entire project viable.
I started by using the guide in //www.fiero.nl/forum/Forum2/HTML/138106.html and then modified the process so we just cut at the collar instead of disassembling the entire column. It should be noted that we also cut down the length of the torque sensor input spines, but the new length has the same amount of bite , just allows for less collapse. So i guess, don't get an accident where you need full column collapse.
[This message has been edited by anaverin (edited 11-12-2020).]
just a heads up... this is in the car and working well. It isnt very picture impressive though being under the car and the column being hidden. Feel free to ask questions. Alot of the RX8 guys doing Ls swaps want to know how we did it. I have walked a few of them through it.
It should be noted that we also cut down the length of the torque sensor input spines, but the new length has the same amount of bite, just allows for less collapse. So i guess, don't get an accident where you need full column collapse.
People don't generally get a chance to choose the circumstances. I don't know if it's such a good idea to be reducing the effectiveness of an important safety feature... but as long as everyone knows what they're up against. And hopefully not literally!
[This message has been edited by Patrick (edited 05-30-2021).]
