Throttle "Glide" postion button/mod

wy not switch to N you have instant glide
or is it not good to switch like that when driving?

 

What we want is a "hold throttle position" function of some type. Obviously we need to be able to instantly override with our foot, and that is a major issue in implementation for safety reasons.

If we lift our foot completely off then we want to switch to regen. If we push deeper into it we want to proceed up the throttle range. The only way I can see to do this is by making the pedal "firmer" on either side of the set point so that it is less jittery to mild input changes. The problem with finding the magic spot now is that the tiniest bit of vibration/road surface aberration will shift you back and forth across it. If we could invoke this with a button on the shift selector/etc. and disable it the same way (and with any brake pedal input) then we might have a decent solution.

 

I realize this won't work @ =<25mph and we're still on the "List", but I use the cruise control +/- on the steering wheel, for fine adjustments of the throttle, on our Avalon. I don't know if it's a small enough change on the Prius. It's ~1mph on the Avalon. I know it may sound crazy, but I have gotten used to using CC anytime I can keep it above the 25mph cutoff for extended periods, even in traffic. Once I get to desired speed I use the +/-, unless it's not quick enough, to adj speeds. The speed adjustments are much finer than I can make with the throttle and I do get improved gas mileage over using the throttle. It's really easy once you get comfortable with using the CC over the throttle. Just though I'd pass it on. Don't have our Prius yet, so I can't try it.

 

It should be possible. I believe the throttle control is just a variable resistor. By using a switch to either add the right amount of resistance in parallel or series, you should be able to get the 0 throttle position to have the right resistance for glide. The fail safe part of it would be a little trickier. There should be a way to have the brake light switch disengage the glide switch though.

Rob

 

I'm no expert, but what little I know may tide us over until the really smart
posters come in.

If I recall corectly, starting in 2006, the accelerator,"go pedal," has two
outputs from Hall Effect sensors. In some way, the signals are compared to
determine exactly where the pedal is in its full range of travel, and whether it
is being depressed or eased off, and how fast that is occurring. I think it is
uncertain if the zero state, no pedal depression, is zero output or some small
reference values.

In some off-forum exchanges I have had on this, it was stated that hacking the
accelerator outputs would probably send the ECU(s) into an immediate state of
panic.

FWiW, I came to the belief that it would be far easier to use an external,
mechanical arrangement to fiddle the pedal itself, not the outputs. I have been
messing with a low pressure air arrangement with variable results for a few
months now.

 

I wish there was a way to do a true glide from 50mph down to 40mph instead of the 40mph to 30mph. It sure would be much easier to do it all the time without impacting the flow of traffic...

 

I know very little about electrical stuff, so I can't really help there.
Nonrtheless, here are some general comments:

The system has to deal with only a very small range of motion, in total less
than 1/4 inch. Finding the right actuator -- stepping solenoid? -- is only half
the battle. The input device must be highly sensitive to intended inputs, yet
not be subject to spurious inputs from bumps and such. When I was thinking
of an electrical solution, I decided a large rotary knob mounted in one of the
cup holders in the center console would do the trick.

But there is no need to restrict yourself to an electrical solution. A cable
actuated arrangement could also work. There are all kinds of fittings and
connectors available in bike shops.

As other posters have pointed out, whatever approach is used, the system
needs to return to zero whenever the gas or brake is used. This can be
accomplished in a mechanical system by spring loading the input mechanism
to go to zero -- think of the hand brakes on a bicycle.

To be really useful, the system should convert relatively large inputs to small
outputs. In my experience, 4 or 5:1 would be ideal.

Unlike the direct approach of messing with the go-pedal electrical outputs, a
mechanical solution accepts that there is many unknowns in how the ECU(s)
handle the signals, including safety related overrides and cutouts. So, a
mechanical solution is not a hack per se. I have come to think of it as a "finesse."

Regardless of how it is actuated, a go-pedal finesse should do more than just lock
in on the no-arrows glide state. In my experience, there is much to be gained if it
also allows you to selectively slip into HV-assisted glide or regen at <42 MPH.
Ultimately, it should also allow you to stay on the bubble when chasing the yellow
arrow HV assist when in SHM. In this latter case, I suspect that having access to HV
output via readings from ScanGauge, CanView, or the like is a near necessity.

 

Use an idle solenoid from an early to mid 80s carburettor 6 or V8. That is early emission control running lean idle mixtures. These solenoids don't have enough power to open the throttle but once open they hold at a pre-determined setting. A circuit that switches the solenoid off when the brake is applies would be simple. I don't remember if it was Ford or GM had a 1/4 inch hex adjustment. You can still apply power for pulse then lift right off to glide, touch brake to switch off.

 

Continuing the "Wouldn't it be great if..." approach, I wonder why the next version can't have a cruise control where you set two speeds, the minimum glide speed and maximum pulse speed desired so the computer can handle the glide pedal position, and maximize mileage through the most efficient pulse acceleration. You could set speeds close together for the sanity of following traffic, and spread the speeds when you're on your own.
I am still trying to decide on the pulse part. I've used zero-arrow acceleration pedal pressure, and I've used my scangauge to do the long, slow IGN 14 acceleration, and I think the former is better.
Surely the engineers can use software to take it out of our hands...er...foot and make the cruise efficient rather than the floor-it device that it is now.

 

what about instead of messing with throttle , mess with the signal going to/from ECU, say theres a specific signal that comes from the ECu that tells the car to enter this glide mode, if you can replicate it then you will have glide.

 

I have to agree with an earlier post, why not use neutral? It is essentially forced glide. I've been doing this for years.

 

Has anyone hacked the software of the main engine computer yet for a Gen II prius?

Attaining a wider glide pedal range could easily be done by making the pedal output less sensitive for a small range then more sensitive for acceleration towards the rest of the range. All one would have to do is reprogram the calibration equations between the throttle position sensor voltage and the throttle-plate actuator position.

I have not dug any deeper into the prius' controllers, but this is a logical controls solution.

A second alternative (already mentioned?) is to use a non-linear pot for the tps sensor (not really available).

Final alternative (as mentioned) is to use circuit to remap the position to tps sensor voltage, or use a small digital micro-controller.
Two important things to know when designing this circuit:
-- the resolution of the analog-to-digital conversion of the pedal position (tps sensor) (prob 4096 bits / 5 volts)
-- the total voltage range of the tps sensor (prob 0-5) and the voltage output from the sensor at the edges of glide mode (has anyone datalogged this?)

You could probably build it w/ a small microcontroller for $50, but I must warn that this is adding a less reliable element in single-redundancy to an extremely reliable drive system, and could at a worst case result in a dead throttle when you need it or a wide open throttle when you dont want it!!!

So- I must recommend the computer hack as the only safe solution (and I dont think we're getting the prius source-code anytime soon!) But- this is a very easy change for Toyota Engineers to consider! Are you listening?!

A final thought- if you redesign the pedal linkage/ cam for the throttle pot, if there is one, then you could change the position-to-voltage curve there. After looking- unless you open up the integrated pedal-pot assembly, this is not possible, and probably not possible on the throttlebody actuator assembly either.

 

In this discussion of a possible go-pedal finesse there are clearly two
alternaives; hacking the electrical signals, or a mechanical approach.

As to the electrical possibilities, it looks like there are more questions and
uncertainties than workable ideas. At this point in time it comes down to
speculation, and a suggestion that it might be doable at some time in the
unspecified future... well, maybe... but first we have to crack the codes...
This sounds too much like wishful thinking, chasing butterflies.

On the other hand, you can create an inexpensive, reliable, and safe
mechanical system right now -- or at least in the near future.

The mechanical "problem" can be deconstructed into four parts:

1. How do you create the manual input? Three possibilities; you squeeze,
twist ,or move something.

2, How do you get the signal from the input device to the go-pedal? Four
possibilities: rods and levers, an encased cable, pneumatics/hydraulics,
electrical wires.

3. How do you move the go-pedal? Now this requires some creativity,
mostly in finding how and where to anchor the arm, cable end, actuator, or
solenoid that will push or pull on the go-pedal.

4. Whatever the arrangement, it should be easily/quickly demountable.

I guess the real question is, How bad do you want it? You can wait until
somebody, somewhere out in the ether figures out an electrical solution.
But when it arrives, you can bet there will be a massive hew and cry about
reliability, safety on public roads, anti-social behavior, liability, and possible
warranty issues.

Or you can start thinking about, and then creating, a simple mechanical
solution. With diligence and even some luck, the result could be super
simple -- elegant even -- fail-safe, and easy to use.

 

What about a back door approach through the CC?

 

I find that each and everytime I use neutral, for gliding, I get considerably worse mileage. I intermix D,B and neutral to keep the charge at about 4 bars. Why this should be is very puzzling, as the mileage improvement should boost to 60- 70mpg. The only thing I can figure is " The engine is running"!!!! This has been concluded by some of our Techies before. The Prius has its own logic, don't fight it.