Anyone pulled their power steering fuse?

[Zebra]

Curiosity got the best of me today so I pried the 50 amp fuse out of the fuse panel and gave manual steering (aka: armstrong steering) a try. I have to admit that I was impressed with the improved handling on the highway. Keeping the Yaris going straight is full-time job with the EPS, it's much better without.

BUT...... Under 30mph it was just way too much work and in my opinion, somewhat unsafe. Manual steering in just about any other car needs less effort and yes the steering wheel is too small for manual steering. Still, it would take more than a big steering wheel to make it tolerable on a daily basis.

Having a system to attenuate the EPS at higher speeds might be the answer. I really like the effect on the highway but that's where it ends..

[marcus]

zebra: now you know why im planning a switch option..but worried about turning this on and off while driving...

[Zebra]

Originally Posted by marcus

zebra: now you know why im planning a switch option..but worried about turning this on and off while driving...

Why are you worried? Worried that a power spike would damage the EPS? Worried someone else will trip the switch and have an accident? Worried that your wiring job won't work at a critical time? Worried that an accident could result in a lawsuit for negligence?

I wouldn't worry about the power spike....

[toy_toronto]

Originally Posted by CTScott

You won't hurt anything having it pulled.

I have done some experimentation with fooling the P/S ECU and will eventually have a product to offer which allow configurable disabling of the EPS. Where, you'll be able to select options like having it on-proportional or on-full at low speed (<5, <10, or <25 MPH) and off completely at speeds above the configured threshold.

CTScott, i would pay someone to have this option

-toy

[rningonfumes]

I like this thread!

Marcus and CTScott.

What is the unit that adjust to speed levels...ie more power when you're not moving vs less when you're at speed? Is there any way to just affect that unit? Like telling it to give no power at all after a certain speed?

As for the switch, where in the line would you put it, between the unit and the fuse or do it in the fuse box?

Lastly...CT...I know you have ideas! Where are you at on this subject?

[Bluevitz-rs]

There's a torque sensor built into the steering column that senses how much force is being applied to turn the wheels and adds more or less power to the motor to assist.

I dont know if CT has experimented with adding resistance to the output ifthe sensor to tell the system not to add as much power.

[CTScott]

From what I have found in my experimentation, fooling the PS ECU in regards to vehicle speed is not enough. I actually have to also override the torque sensor data. The "highway wander" is a good example of why this is necessary. Even at high speeds, there is enough power assist that small changes in the torque sensor feedback result in steering adjustment.

One of the simplest implementations may be to add a DPDT relay in line with the power steering motor leads, and control the relay based on vehicle speed.

[CrankyOldMan]

If you want to just add an on/off switch that won't mess with the existing wire harness you could get an external fuse holder, 50A relay and a toggle switch. The toggle switch would be connected to the battery or accessory line and the coil of the relay, and the switch of the relay would enable/disable the power steering fuse. I'm assuming that the power steering is still 12V, so pretty much any automotive relay with 50A contact rating would work. The toggle switch can be anything that can handle 12-14V and whatever the coil current of the relay is. You would have to pull a pair of wires through the firewall, but that shouldn't be too hard either.

[CrankyOldMan]

Originally Posted by CTScott

One of the simplest implementations may be to add a DPDT relay in line with the power steering motor leads, and control the relay based on vehicle speed.

Is this something that could be based on the "vehicle speed" wire on the radio harness? I've only got a basic electronics knowledge at the moment, but I'm sure there's an IC that can read the pulses from this line.

[CTScott]

Originally Posted by CrankyOldMan

Is this something that could be based on the "vehicle speed" wire on the radio harness? I've only got a basic electronics knowledge at the moment, but I'm sure there's an IC that can read the pulses from this line.

If you wanted to do it without a microcontroller, you could use a frequency to voltage converter IC and then feed that through a comparator.

[CrankyOldMan]

Originally Posted by CTScott

If you wanted to do it without a microcontroller, you could use a frequency to voltage converter IC and then feed that through a comparator.

I've located some potential parts for these, but I'm not sure how I'd go about measuring the pulse per mph. Do you by chance already know what it is?

I'm also guessing that the signal is 12V, so you'd have to use a regulator or voltage divider to get down to the 5V input that the F/V chip is expecting. I'm still looking through the datasheet to see exactly how the F/V converter works. Any suggestions on what kind of comparator would be the right type to use?

[CTScott]

Originally Posted by CrankyOldMan

I've located some potential parts for these, but I'm not sure how I'd go about measuring the pulse per mph. Do you by chance already know what it is?

I'm also guessing that the signal is 12V, so you'd have to use a regulator or voltage divider to get down to the 5V input that the F/V chip is expecting. I'm still looking through the datasheet to see exactly how the F/V converter works. Any suggestions on what kind of comparator would be the right type to use?

The VSS signal is actually a 5V signal. If memory serves me correct the frequency is 100 Hz at 40 MPH and about 20 Hz at 10 MPH.

For the comparator, a general purpose one, like the LT1011 would be fine. It is amazing how much more complicated this gets trying to do it with analog, rather than digital. I should just build you up one to experiment with using a microcontroller. I could make a simple little black box that connects to ignition switched power and then has an output to drive the relay coil. I could have a pot on the box to adjust the switch over speed.

[CrankyOldMan]

Originally Posted by CTScott

It is amazing how much more complicated this gets trying to do it with analog, rather than digital.

Why would this be the case? There are a few additional passive components that need to be added around the F/V and comparator, but it seems to be fairly straightforward. Is it a matter of getting a comparator that has enough hysteresis to prevent noise around the switching point from making the output fluctuate? Sorry if those aren't the right terms, I just looked some of this up.

I looked at the EWD and the power steering is already throttled by the ECU based on speed and the torque sensor. According to p 176 of the EWD:

* In order to reduce driver’s steering force, the assist current is calculated from the torque sensor signal and vehicle speed
signal of the meter or CAN communication to output to the motor.

∗ When abnormality occurs in the system, the assist is prohibited by shutting off the power source relay and motor relay in
ECU, and the warning light (P/S) in the meter is lit up.

So if we were to ever crack the ECU, we could in theory change the control algorithm. It seems strange that removing the fuse doesn't turn on the (P/S) light.

[Jason@SportsCar]

I like elec power steering. It rocked in my RX-8, made it one of the most pleasurable cars to autox - past cars I raced with traditional systems always struggled to keep up, our Corvette use to puke power steering parts regularly.

I think the system in the Yaris works great. In the GT/B-Spec cars it works seamlessly, I had no problem on the track with it.

I hated the H Prod car when we first started racing it... Due to some hacked wiring the system never worked, and with a 10" wide slick the car was a beast, always fighting you in the turns. However, with time we were able to work out some of the steering bind, making it easier to drive. Ditching the elec motor and the stock column made it much easier to turn (still not as good as the EPS system), and it actually dropped a good bit of weight.

No brainier for me, in a street car I would keep it.

[CTScott]

Originally Posted by Jason@SportsCar

I like elec power steering. It rocked in my RX-8, made it one of the most pleasurable cars to autox - past cars I raced with traditional systems always struggled to keep up, our Corvette use to puke power steering parts regularly.

I think the system in the Yaris works great. In the GT/B-Spec cars it works seamlessly, I had no problem on the track with it.

I hated the H Prod car when we first started racing it... Due to some hacked wiring the system never worked, and with a 10" wide slick the car was a beast, always fighting you in the turns. However, with time we were able to work out some of the steering bind, making it easier to drive. Ditching the elec motor and the stock column made it much easier to turn (still not as good as the EPS system), and it actually dropped a good bit of weight.

No brainier for me, in a street car I would keep it.

It actually sounds like it is better for track than street. The primary complaint about the EPS is how it induces wander at highway speed.

[CTScott]

Originally Posted by CrankyOldMan

Why would this be the case? There are a few additional passive components that need to be added around the F/V and comparator, but it seems to be fairly straightforward. Is it a matter of getting a comparator that has enough hysteresis to prevent noise around the switching point from making the output fluctuate? Sorry if those aren't the right terms, I just looked some of this up.

I looked at the EWD and the power steering is already throttled by the ECU based on speed and the torque sensor. According to p 176 of the EWD:


So if we were to ever crack the ECU, we could in theory change the control algorithm. It seems strange that removing the fuse doesn't turn on the (P/S) light.

I just mean based on the overall number of components and the little bit of math required for configuring the comparator, etc.

The P/S light id turned on by the combination meter when the EPS reports an issue. If the EPS isn't alive it can't complain.

[CrankyOldMan]

Originally Posted by CTScott

I just mean based on the overall number of components and the little bit of math required for configuring the comparator, etc.

Understood. For an entry level EE student like me, the more cost effective solution would be using components and a breadboard instead of a PLD. Most of us DIY-ers don't have access to a programmer. The only microcontroller I've been exposed to so far comes on a $150 dev board, and that's a bit too expensive to cloister in a box under the dash. =)

[rningonfumes]

...it's like watching soft core.

Keep going!