Honda Insight Forum banner

1 - 20 of 27 Posts

·
Registered
Joined
·
1,642 Posts
Discussion Starter #1 (Edited)
For those of you not familiar with Natalya's excellent force-autostop, check it out here:

http://www.insightcentral.net/forums/buy-sell-insight-related-items/93370-forced-auto-stop-microcontrollers-g1-5mt-only.html

I live in a cold area, and I had the idea to try integrating an outside-air temperature spoof into Natalya's autostop system. I don't want to just use a resistor in place of the outside air temperature sensor as it messes up the air temperature provided by the heater control, and it would be a lot less fun to simply put resistor and air sensor on a relay. So, I went looking through my diagrams to see if there were any really clever ways of doing it.




The outside air sensor only goes to the heater control, and does not split, as the VSS does to both ECU and gauge cluster. This means the heater control is reporting something to the ECU when outside temperature is within spec, rather than the ECU reading it. There are only two wires going between ECU and heater control:

1) ECU A23 (BRN/YEL) which is marked as "HTRS"

2) ECU A2 (BLK/YEL) which is marked as "ENGRDY"

When I see these two acronyms, I read them as "heater system" and "engine ready", but it's only conjecture. BLK/YEL (ENGRDY) goes to the gauge assembly, in addition to the ECU. Interestingly, the VSS (BLU/WHT) has a line that runs to both gauge assembly and heater control - why does the heater control need VSS?

My two interpretations of how it might work:

1) ENGRDY provides the "ready for auto-stop" signal. The gauge cluster sends voltage to the ECU when the car is below a certain speed or under certain deceleration conditions or if the engine is sufficiently warm (?), and the heater control panel also sends a voltage if the outside air temperature is above a certain level, and if this signal/voltage is absent from either, you can get auto-stop.

2) More likely, ENGRDY informs the heater control to shut off the cabin air blower, and for the gauge cluster to show the auto-stop light, while HTRS is the signal/voltage letting the ECU know whether conditions for auto-stop are appropriate - e.g. climate control isn't set to "auto", outside temperature is appropriate.

I'd love for someone else to take a look at this and help me puzzle it out. I'm strongly inclined to believe that one of these lines is a signal/voltage going to the ECU, and the other is a signal coming from it, and both are probably related to auto-stop.

It would be nice if it were simply a present/absent voltage that signaled if auto-stop could activate, rather than a signal of some sort. I could then put a relay with the FAS module, and enable auto-stop at the push of a button even if climate control is set to "auto" or outside air temperature is too low.
 

·
Registered
Joined
·
167 Posts
This means the heater control is reporting something to the ECU when outside temperature is within spec
Maybe heater takes outside air temp, as well as climate control setting (auto, econ, off, etc.) and just transmits a go/no-go signal depending on both of those states? Assuming that's the case, could you just report 34F/1C to the heater and let it pass that information on? I would assume it wouldn't screw anything up, and would let you use auto-stop in freezing conditions
 

·
Administrator
Joined
·
11,777 Posts
The proper workshop manual may give you a clue at to which unit is transmitting to which.
An open collector transistor at one end is the sending unit.

As you say just hook up an led or scope to the wires and monitor them in your various heater/ engine scenarios to build a picture.

Just checked in my manual and engrdy blk/yel signals to > the heater. Htrs brn/yel signals to > the ecm
 

·
Registered
Joined
·
1,642 Posts
Discussion Starter #5
Maybe heater takes outside air temp, as well as climate control setting (auto, econ, off, etc.) and just transmits a go/no-go signal depending on both of those states? Assuming that's the case, could you just report 34F/1C to the heater and let it pass that information on? I would assume it wouldn't screw anything up, and would let you use auto-stop in freezing conditions
This is one way to do it, and probably the method that requires the least thought. The only downside would be that for the brief period that I pressed the FAS button, the heater in the car would move the blend door to adjust for the change in temperature reading, which would be fine. I might do that, but I'd like to see if there's a more "elegant" way.


The proper workshop manual may give you a clue at to which unit is transmitting to which.
An open collector transistor at one end is the sending unit.

As you say just hook up an led or scope to the wires and monitor them in your various heater/ engine scenarios to build a picture.

Just checked in my manual and engrdy blk/yel signals to > the heater. Htrs brn/yel signals to > the ecm
Right, so sometime soon I'll try to see what's being transmitted over HTRS (BRN/YEL). If it's a simple voltage / no voltage, I'll put a relay here. If I need to transmit something more, I may hint hint nudge nudge Natalya to build me a second FAS module capable of both the square wave for spoofing the VSS signal, and also the HTRS signal. Seems to me that would be extremely elegant.
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
I hope it's just voltage. I wanted to program it to pass through the VSS signal unless you were pressing a button, so you'd only need an SPST switch, but I haven't been able to figure out how to get PWM to work right.

PWM is configured by like 5 different registers, but I couldn't ever figure out the right combination of stuff in each of those registers.

There's a library I found that lets you generate a PWM signal, and that works, which means it is configuring the registers correctly, but I kinda have to reverse engineer it to figure out what those good register values indeed were, and then go from there. So as is with the PWM library I can make 1 signal but I can't make it copy the actual VSS signal because it's constantly changing. If this HTRS signal is some kind of varying signal like the VSS then I can't really help yet.
 

·
Linsight Designer
Joined
·
1,899 Posts
glad to see you're using the hardware PWM functionality. The 328p is fairly simple, but yes there are some knobs you have to turn to enable the counter mode on a specific pin, then configure the correct PWM mode you want to use. Feel free to post up what you're attempting and the actual result and I'll help out. Always love seeing someone learn microcontrollers. Mario will probably help out, too. Ask away if you have questions.
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
My other thread turned into a calpod switch development thread. But today I have some improvements on the forced autostop system itself, so I'm going to post in here. Also, this relates to Ecky's cold weather situation.

################
Item 1:
Code Update
################

PHP:
// Elegant Autostop
// Tells a G1 Honda Insight that conditions are right for AS
// Version 0.1

// #include <SPI.h>

int out_a = 9;
int out_b = A1;

void TimerOne_setPeriod(long OCRValue)
{
 TCCR1B = _BV(WGM12)|_BV(CS12)|_BV(CS10);   //  or use TCCR1B = 0x0d;            // CTC mode - 1024 prescale
 TCCR1A = _BV(COM1A0);                      //  or use TCCR1A = 0x40;            // Toggle mode 0C1A 
 OCR1A = OCRValue;                          // set the counter
}

void setup()
{
  pinMode(out_a, OUTPUT);
  pinMode(out_b, OUTPUT);
  TimerOne_setPeriod(3096);                 // set up and start Timer1 to blink at the same rate as the blink sketch
//  Serial.begin(9600);
//  Serial.println("setup");
}

void loop()
{
  digitalWrite(out_b, HIGH);
  delay(250);
  digitalWrite(out_b, LOW);
  delay(250);
}
(I'll post the code update in the other thread as well.)

What we've got here is the arduino pro-mini or a pro-mini clone is outputting the original EFAS signal on pin A1 but now it ALSO outputs a signal that's nearly the exact same via PWM on pin 9.

(This whole section gets long and boring, skip it if you don't care about how you get into Auto-Stop, and go until you see the next #### block.)

Why have two different signals at the same time and why are they produced with different methods?

Because the slight difference in their frequency leads to the car behaving differently depending on which signal you send to the ECM! This is kind-of exciting to me.

Note:
I may phase out the old signal entirely in the future and go PWM only, but for now I want to leave it in as a legacy option unless everyone prefers the new signal. I'll explain:

The old signal (produced with the alternating out high and out low code on pin A1) may be more suitable for vehicles which do NOT have a Calpod switch, or where you maybe have one but rarely use it.

The new signal (produced via PWM on pin 9) is more suitable for people with weak IMA batteries who usually have the Calpod switch on all the time and sparingly use assist.

What's the difference between them?

Old signal seems more finnicky and harder to get the car to Auto-Stop at times, especially in low-speed situations, but once you get AS from the old signal Auto-Stop will PERSIST -- the car will not leave it no matter what unless you tap brake 5 times or shift into gear.

New signal reliably gets you into Auto-Stop EVERY TIME as long as you follow an exact set of procedures, but it DOES NOT persist -- you have to hold the clutch OR the brake in, or have a clutch switch turned on, in order to keep coasting in auto-stop. This AS will acquire "persistence" if you hit 0 MPH, for example at a red light, whereupon now you can release clutch and brake and the engine will stay turned off until you tap brake 5 times or shift into gear.

In the other thread there was a convo explaining how the two different modes work. Here Ecky describes the how OLD signal works:

In my US 2000 5MT, activating the relay that connects your FAS board (I'll call it my "FAS button") only sometimes puts the car in auto-stop. To reliably initiate auto-stop, I have to tap the brake (or brake button) while holding the FAS button. Once in auto-stop, it stays there regardless of whether or not I press the clutch or brake, which is actually different than the behavior I get when it goes into auto-stop normally when slowing down. Normally, the car will auto-stop if I brake in neutral at or below 19mph, but releasing brake or clutch generally causes it to exit auto-stop if I'm still above maybe 2-3mph.
Here I describe how the NEW signal works contrasted with what Ecky said:

To enter AS (With NEW signal):
- Hold AS button
- Press Clutch
- Shift into Neutral
= Car Auto-Stops Here, or tap brake to get AS =
- Release AS button
- Keep holding clutch in OR have clutch-switch on and release clutch

(With OLD signal):
- Hold AS button
- Press Clutch
- Shift into Neutral
= Car Auto-Stops Here, or tap brake to get AS =
- Release AS button
- Release clutch (irrespective of whether or not the clutch switch is on)

So there's an extra step (with the NEW signal to get into AS)

...

I still haven't reprogrammed the chip that's in the car with the old code, but I have learned a little more about the current behaviour.

It seems that every time the car actually hits 0 like when you're stopped, and assuming you spend at least 1 second at 0 WITHOUT holding down the FAS button, something gets reset.

So you start driving again, the first time you want AS you press the FAS button and go in neutral and it won't AS unless you tap the brakes. But every subsequent time it WILL go into AS without the brake tap, until the car goes back down to 0 mph, and then the next time you want AS you're going to have to hit the brakes again.
Blah blah so anyway point is there's two options now but I might decide to remove the old option if everyone agrees that it kinda sucks and holding the clutch down while coasting in AS isn't a high price to pay.



################
Item 2:
Circuit Update
################


Okay I've said a hundred times now if you want this Auto-Stop button you also want (even more than my FAS unit) a "Calpod" clutch switch. But clutch switches interfere with Auto-Stop, and so does outside ambient temperature sensor in most vehicles. So I built some boards with relays to solve the clutch switch problem but I never addressed the OAT sensor because I couldn't find a cheap TPDT relay and fitting 3 small SPDT replays had problems.

But I've settled on a solution!

Clare LCC110 SPDT Solid-State Relays

(Blah blah skip to next ## if this part about the relays is too obvious.)

These tiny things have almost no current draw. The DPDT relay I used in the past had a coil resistance of 167Ω while the new solid-state ones each only have 35Ω resistance. This means the voltage drop isn't so big you have to work around it.

The second plus is there are no moving parts, so they'll probably never have to be replaced and will probably never malfunction.

Third benefit is tiny form factor, each one has half the footprint of the old relay.

Only negative to these relays is they cost more. The old mechanical DPDT relay costs $2.75 Canadian a unit (and goes down quickly when ordered in bulk) whereas the cheapest I could get the LCC110 is at $2.16 USD each. Remember, they are SPST so you need (at least) two per board, so think of them as $4.32 for a direct replacement. Also, $USD > $CAD for almost the whole time I've been alive.

Up there I said you need at least two per board, we can build a board with 3 of the SPDT solid-state relays where each relay does a different task but all are controlled by one EFAS button:

Relay 1 switches out the VSS signal with the spoof signal to ECM
Relay 2 opens the clutch switch circuit
Relay 3 switches the OAT sensor wire to have a resistor in series with it which tells the climate control the outside temp is above 0°C//32°F.

## Okay stop skipping ##

Last night I built a new test PCB with 2 of the LCC110 relays and a board with the new software and a little switch to choose between which signal you want sent to your ECM -- old or new.



Turning both off will obviously make this not work, turning both on would have unpredictable consequences. I wired the test board so that when you turn 1 on you get the old signal sent to ECM when your actual auto-stop button is pressed, or if you turn 2 on you get the new signal, so you get to choose. But again I might just forego the old signal in the future anyway.

Today I installed this new test PCB into my friend's 2001 Insight (the Canadian one I formerly owned) and it ALMOST worked, except there was a critical problem: The +5V line from ECM connector C pin 28 that was supplying power to the whole thing got too much current drawn because there was only 70Ω resistance between it and ground. This caused the instrument cluster to flicker every time I pressed the Auto-Stop button.

To solve this problem I wired in a 220Ω resistor because that's all I had laying around, and it solved the problem and everything works great. There's probably a better Ω value to use for the resistor.

Anyway, after test driving and verifying it worked I had my friend drive around and try the new board with the NEW signal going to her ECM and she really liked how predictable it was to get into Auto-Stop. Her IMA is pretty much dead (it was already basically dead when I had bought the car) so she almost always has her clutch switch turned on anyway, so that's going to mean she won't have to have the clutch pressed in every time she goes into AS and the new signal won't cause a problem for her.



(Final section is in next post, there is a 10000 character limit apparently.)
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
################
Item 3:
Towards a Final Product
################


Okay so new hardware has been settled on, which is great. Next step is to determine whether or not anyone wants to keep the old code, or if the new PWM signal is inherently superior in all use cases. If it's decided that the new signal should be used no matter what, like even if you have a good IMA and hardly use a clutch switch, then I ought to design a PCB that just has an Atmel 2542 processor instead of using a whole Arduino clone because the signal can be generated via PWM.

Then again, someone else might want to have the Arduino in there because it can easily be programmed to do other stuff on top of faking a signal for the ECM, and you could just stack my code onto whatever else you're doing with it.

So maybe an interim step is design a PCB that uses the Arduino clone plus 3 of the solid state relays, an appropriate resistor to prevent electrical issues every time you press the auto-stop button, and a resistor to fool the OAT sensor.


Question:
What is the preferred software to use to draw a circuit diagram, and the software or perhaps file type to design a PCB that I could get printed? There's lots of stuff out there to do this, I'm wondering like what do actual electrical engineers use//prefer.

Another Question:
Anyone out there with the old code want to have their chip flashed with the new code for the new signal? You'll have to re-wire your unit to use pin 9 instead of pin A1. I can either show you what you need or maybe you guys can mail to me and I mail it back? We can work this out however. I won't charge for updating the code, just $3 US for postage if we go that route. The code (as you saw above) is very simple and free to anyone who wants to use it.

Final Question:
Would something like this possibly work in a 6MT CR-Z? I think it might.
 

·
Registered
Joined
·
1,642 Posts
Discussion Starter #11
If I'm understanding correctly, the new code is functionally different in that releasing the clutch (switch) will take the car out of auto-stop? Or am I sleep-addled and missed something?
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
If I'm understanding correctly, the new code is functionally different in that releasing the clutch (switch) will take the car out of auto-stop? Or am I sleep-addled and missed something?
Yes, it will do that, so just hold clutch down or keep the clutch switch on.

But you are missing something:

With the new signal it reliably goes into Auto-Stop every single time as long as you follow the steps correctly, even at low speeds.


*EDIT*
You can use one of the switches I had the picture of on page 1 which only cost 39 Canadian Cents to switch between the old signal and the new signal if you want to try it out and see what you think. If you mail your chip back to me for reprogramming I'll throw one of those switches in for free.
 

·
Engine-Off-Coast
Joined
·
1,765 Posts

MB2411A1W01-FA

I've settled on this exact switch. It's made by NKK, is of solid construction, and is rated for way higher duty than this job requires -- 3 amps at 125V AC. It also makes a satisfying (but quiet) click when you press it. The switch is smaller than this photo, the black cap for example is only 5.1mm in diameter.

It's a SPDT switch so it can be used either with the FAS chip as a standalone, or with any FAS+relay setup.

I'm going to see if I can design a small bracket to hold this on the side of the shifter. I took some measurements today, the shifter is 1cm in diameter where the boot meets the knob. The button rests nicely when positioned on the side of the knob 5cm above the boot.
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
PHP:
// Elegant Autostop
// Tells a G1 Honda Insight that conditions are right for AS
// Version 0.1B

// #include <SPI.h>

int in_a = 3;
int out_a = 9;
//int out_b = A1;     //  This pin was used for the original version
//int out_c = A3;     //  This pin was used for debugging

bool pressed = false;

void TimerOne_setPeriod(long OCRValue)
{
 TCCR1B = _BV(WGM12)|_BV(CS12)|_BV(CS10);   //  or use TCCR1B = 0x0d;            // CTC mode - 1024 prescale
 TCCR1A = _BV(COM1A0);                      //  or use TCCR1A = 0x40;            // Toggle mode 0C1A 
 OCR1A = OCRValue;                          //  set the counter
}
void Interrupt_Setup()
{
  DDRD &= ~(1 << DDD3);                     //  clear pin 3
  PORTD |= (1 << PORTD3);                   //  enable pull-up on pin 3
  EICRA = B00001000;
  EIMSK |= (1 << INT1);
  sei();                                    //  turn on interrupts
}

void setup()
{
  pinMode(out_a, OUTPUT);
//  pinMode(out_b, OUTPUT);
//  pinMode(out_c, OUTPUT);
  TimerOne_setPeriod(3096);                 //  start blinking -- 3096 usually gets autostop instantly
  Interrupt_Setup();
//  Serial.begin(9600);
//  Serial.println("setup");
}

void loop()
{
/*  digitalWrite(out_b, HIGH);
  delay(250);
  digitalWrite(out_b, LOW);
  delay(250); */
  if (pressed == true)
  {
    On_Pressed();
  }
}
ISR (INT1_vect)
{
    pressed = true;
}
void On_Pressed()
{
  pressed = false;
//  digitalWrite(out_c, HIGH);
  delay(500);
  TimerOne_setPeriod(3896);
  delay(2500);
  TimerOne_setPeriod(3096);
//  digitalWrite(out_c, LOW);
}
This isn't finalized yet. What I'm trying to do is get the car to always go into Auto-Stop AND have it persist by slowing the fake pulse down half a second after the button is pressed. I haven't gotten to test this yet. If someone else wanted to test they'd need a 10uf capacitor between pin 3 and ground and they'd need a line from the +5v right BEFORE the switch to pin 3, preferably with a 220Ω resistor in series between the +5v and the switch//pin 3.

There's an interrupt from INT1 on the processor which is mapped to pin 3 on the arduino pro mini board. When you press the switch it gets triggered and it sets a boolean to true. The arduino code sees that and runs the code block at the bottom changing the PWM speed. I have the PWM and the interrupt at like the processor level, not relying on logic from the chip itself. If this works as expected then it will be good for every 5MT Insight, whether or not they have a clutch switch. At that point maybe I can re-write the code to have everything done right on the processor without the arduino language.

I'll update once I've gotten to test it.
 

·
Engine-Off-Coast
Joined
·
1,765 Posts


All right, here we go. I've got a PCB. This one is for the hybrid-output programming in my previous post. I put one like this in Techno Trash and tried it out, and everything seemed to be working fine.

You'll see the three LCC110 solid-state relays on the left, one for the VSS signal, one to disable your clutch switch if you have one, and one to trick the climate control ambient temperature sensor if you live in a cold climate.

I'll have the PCBs at Maryland on Saturday along with parts to build them out.


#########
Functionality
#########



Know how you get Auto-Stop in a bone-stock 5MT sometimes if you're in 3rd and braking for regen then you see the bars disappear and you throw in neutral, but you're still at like 17 mph but decelerating? Know how if you let off the brake and the clutch before you get to 0 the engine restarts?

With the current setup, this device performs exactly like that, so you can hold clutch or brake to stay in AS. **BUT** now if you hold the EFAS button for about ¾ of a second or longer then you can release everything because the car will think it went slow enough and it won't turn itself back on. I called this "persistence" in previous posts. Now we've got both an instant every-time Auto-Stop AND a persistent Auto-Stop if you hold your EFAS button for over half a second.

I think at this point I just need some new people to try these and give me feedback on them, but it could very-well be a finished product at least with how it functions and is programmed, if not the actual physical implementation. (Wires, installation location, etc.)
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
I've been thinking about increasing the first delay near the end of the code from 500ms to 750ms. Sometimes I'll try to go into AS and it won't immediately go because it wants me to tap the brake, but by the time I do that I've passed the 500ms threshold and it will go into a simulated 0MPH AS instead of the decelerating AS. With clutch switch on I usually want the decel AS because maybe I'm going to come out of AS and then want to go back in later without a brake tap.

########

Somewhat related question:
Does anyone know if it harms the transmission or the clutch to have the clutch pushed in all the way and hold it in while the engine is turned off and you're in neutral but you're coasting?
 

·
Engine-Off-Coast
Joined
·
1,765 Posts
Bumping this with a video. From talking in a couple other threads (one here, one on Ecomodder) I don't think I'm effectively communicating what this does.


Just watch the tachometer and Auto-Stop indicator, on-demand I put the car into actual Auto-Stop whenever I want at whatever speed I want. In video I do it at 33 MPH but you can do this at literally any speed. Maybe I can get a passenger to film a better version where you can see the shifter.
 

·
Registered
Joined
·
824 Posts
A couple of questions as I ponder adding EFAS, clutch and brake switch features.

Sitting in the fast food line with headlights on and a low battery and normal autostop, the engine turned on while I had the clutch depressed. I think the battery got too low.

I charged it a little, moved forward, and it autostopped again. This time I left the clutch in but also put it in gear.

<EDIT> There is a problem with my memory of this "test". I'm still new to Insights but if this was auto-stopped and I had the clutch pedal down, placing it into gear should have caused it to start?</EDIT>

It eventually started up again on its own.

This was an informal test and not well documented and I was tired so before quoting this it should be properly retested. I recall putting it in first and pressing in the clutch to see if it would possibly start in gear should the clutch switch erroneously report clutch disenagaged(but with clutch disengaged.)

If I had the Calpod mod and had turned it on after entering auto stop normally, then released the clutch pedal, but left it in gear, would the car have restarted and launched into the car ahead of me? Does the car autostop if the Calpod switch is on?

For an EFAS installation, where there is also a Calpod, does activating EFAS also disable the Calpod (actual clutch switch operates normally?)

Finally, if you simulate autostop by dropping the VSS pulse rate to the ECU to 1 MPH (while coasting at 30 MPH) what happens when you return the VSS pulse rate to its 30 MPH equivalent? Does the car restart? Or does it remain stopped unless you press the clutch/put it in gear?

Finally finally, I am wondering what would happen if one were to vary the VSS in real time with an external pot providing a simulated speed of between 0 and 30 MPH? What if you're coasting at, say, 40 and vary the VSS to the ECU from 30 MPH to 0 and then back up to 30, gradually? What if you're parked, waiting for your burger?
 

·
Registered
Joined
·
1,642 Posts
Discussion Starter #19
There is a separate sensor that detects whether or not the car is in gear.

There are two variants of EFAS programming, one that stays in auto-stop even once speed increases and the clutch switch is no longer pressed, and one that restarts the engine. I used to be a fan of the former, but I've come to feel the latter is better.

I believe what actually decids if the car restarts or not after speed increases is just how low you take the VSS. For instance, in normal driving if you get auto-stop at 15mph with the clutch in and then let it back out, the car restarts. However, if you starts on top of a hill at a dead stop in auto-stop and begin rolling down the hill, and let the clutch out, the car will not auto-restart until you throw it in gear. Presumably this is so you can still have auto-stop if you're sitting at a traffic light and don't want to hold the clutch the whole time, but also so it's less disruptive if you're rolling in stop and go traffic.
 

·
Premium Member
Southern California
Joined
·
801 Posts
I believe what actually decids if the car restarts or not after speed increases is just how low you take the VSS. For instance, in normal driving if you get auto-stop at 15mph with the clutch in and then let it back out, the car restarts. However, if you starts on top of a hill at a dead stop in auto-stop and begin rolling down the hill, and let the clutch out, the car will not auto-restart until you throw it in gear. Presumably this is so you can still have auto-stop if you're sitting at a traffic light and don't want to hold the clutch the whole time, but also so it's less disruptive if you're rolling in stop and go traffic.
A small addition/correction to this - if the car is moving and it auto-stops, then it will stay in autostop as long as you have the clutch OR brake pressed (while moving). Once the car comes to a complete stop, you can let off the brake AND clutch and it will stay auto-stopped, even if the car starts accelerating again (like going down a hill). In this scenario, if you want to restart the engine, you can either put it in gear, or hold the clutch in and press the gas, and the car will start the engine again.

This is the behavior I observe in my 2000 Insight. If I want to stay in autostop without coming to a stop and without having to hold and pedals, I flip my calpod switch on so the car thinks the clutch is held in.
 
1 - 20 of 27 Posts
Top