The rationale behind "feathering the throttle"

[Trex]

Well, don't feather brakes and have your passengers bend reverently as the car slams to a halt, do feather them and they will remain peacefully reclined in their seats...
I'm not quite convinced truckers are the ultimate authority on sensitive ( or should that be sensible ?) driving.

:lol:

No the context for the truckie's comment "feathering the brakes" was they could go down steep grades 2 gears or more higher than they should by "feathering the brakes" and not get excessive wear
on the brake linings or lose their brakes altogether.

They were wrong because I had to help fix those brake linings and they were worn quicker than the truckies that came in 10 min slower on the same routes.

I used to drive the old "Bulahdelah Mountains" route a notorious "black spot" ( means a lot of deaths on a bad road section here in Australia ) and there would be regularly semi-trailers over the side of the cliffs on the side of the road with dead drivers (or at the best badly injured and trapped in their cabs) that had lost their brakes through "feathering".

Thankfully they fixed that section of the road.

Some truck drivers are incredibly sensible and I would like to think that includes me although I am only part time when I can not get another driver.

As the owner of a engineering business I have to be able to do any job when people are away.

That is my burden. :lol:

Regards Trex.

 

[Claymore]

It seems to me that there needs to be a definition of 'feathering' as it relates to brakes or any other function to ensure that everyone is singing from the same hymn sheet....which does not appear to be the case at present.

Only saying!

A definition of Hysteresis, as applies to the PHEV drivetrain, glued up there, would be bleedin' handy as well!

:idea:

JimB

 

[Trex]

Not sure if this adds anything to the discussion as you all probably know this stuff far better than me but all controls exhibit hysteresis. Old-fashioned throttles probably more than many. When I was a young child working large machine tools (it was a hard life) I got used to setting controls and backing them off to get consistent performance - not necessarily better performance in that case just more consistent. These days similar machines have much less hysteresis thankfully. I thought then and I still think it is odd that hysteresis is not on the school curriculum given that it is such an important and ever-present phenomena. Should be taught in maths I guess. Anyway; I suspect that electronic controls on devices as complex as the Outlander have a fair bit still and for reasons I could only guess at.

Hi Phevourite,

Hysteresis can be made up of a few different things. "Lag" or hysteresis where you move a control and the output of the control (say RPM of motor) catches up to the input (say movement of the accelerator ) later certainly exists.

Then you have hysteresis loops that exist naturally in some materials, electronic controls etc

An example are thermostats for heaters or pressure controllers for air compressors of hysteresis loops that are used intentionally to not have them start and stop at the same temp or pressure ie unwanted frequent switching.

A good example of a intentional hysteresis loop is on my Phev Parallel mode cuts in at 70kph but cuts out at 65kph. Otherwise with no hysteresis loop it would try to cut in and cut out at 70kph.

But hysteresis is not really suitable for this discussion I think.

Regards Trex.

 

[Trex]

One way of driving is this: We can push the pedal in a certain position and the car will accelerate. At some point power generated by the engine is matched by (air)resistance and the car does not accelerate any further. But that is not how we typically drive. At least, not me.

We accelerate and when we reach the desired speed we lift the throttle a bit to prevent the car from going even faster. I assume this is not what we mean with feathering. Feathering would, if I understand correctly, be to lift the throttle a little bit further, after acceleration has already stopped. Right?

Now somebody explain to me, if I can, while driving at a constant speed, lift the pedal from (e.g.) 18% to 17.8%, without loosing power and thus without loosing speed, then why could I not do the same thing again a few seconds later and lift from 17.8% to 17.6% without loosing power or speed? And so on?

Hey anko,

Its just "VOODOO MAGIC MON". :lol:

Regards Trex.

 

[Titan]

I play with the throttle a lot more in the PHEV than any other cast I've driven. If I'm driving along notice the road is flat or slightly downhill I'll lift off and coast on B0. If I'm in Charge mode the little lift offs help the battery get some juice which, at 60+ it might not otherwise get. I believe it's these little lift offs that are the fathering referred to.

 

[anko]

I do that to. Lift the gas in B0 in order to gain SOC in Charge mode. Especially during town trips. Whether it actually increases efficiency needs to be seen, but that is a totally different discussion

I got the impression that people are saying that you can "feather" without impact on power or speed. When coasting downhill, probably speed is not reduced when lifting, that is true. But speed would have increased if you had not lifted. So there is still an impact on speed of the lifting. And indeed, reducing your speed or not increasing it saves fuel. But that is common sense, that IMHO has nothing to do with feathering.

 

[jaapv]

Feathering the throttle is easing up the pressure on the throttle to find the minimum at which the car maintains its speed.

 

[anko]

I do that to. Lift the gas in B0 in order to gain SOC in Charge mode. Especially during town trips. Whether it actually increases efficiency needs to be seen, but that is a totally different discussion

Suppose you need to cross some other road via an underpass (is that a word?). So, first you have a decline, then an incline. On the decline you loose potential energy as your altitude decreases. Now this potential energy needs to go somewhere. I think there are roughly three options:
- heat from the friction brakes
- increased speed (or less reduced speed)
- increased SOC

On the way up, your potential energy increases as so does your altitude. You have to add energy. There are again three options:
- reduced speed
- use more fuel
- use battery power

I like to think that increasing your speed on the decline and decreasing again on the incline is more efficient than temporarily storing energy in the battery, as it does not involve to many additional conversion processes with.

 

[Titan]

I like to think that increasing your speed on the decline and decreasing again on the incline is more efficient than temporarily storing energy in the battery, as it does not involve to many additional conversion processes with.

But there are times when or conditions might not allow an increase in speed

 

[Claymore]

But hysteresis is not really suitable for this discussion I think.

Regards Trex.

Thanks anyway........I was beginning to think it had a more elaborate meaning in relation to the PHEV.

JimB

 

[Phevourite]

But hysteresis is not really suitable for this discussion I think.

Regards Trex.

Well only in as much as it is essentially the technical word for feathering or at least for why feathering is possible. Hence, even with old technology, putting your foot down till you hit the desired speed means you can then back-off slightly before your speed decreases again. That margin between the end state on the way up and the same speed state on the way down is caused by the hysteresis in the control loop - albeit via flaps and valves and stuff in early combustion engines.

So with the PHEV if you end up driving at say a steady 50kph, using the petrol engine, but get there by accelerating to 50 the ratio on the gearbox and the tuning of the engine is going to be different than if you get there by decelerating to the same speed. Will it be enough to matter? That's another question.

 

[anko]

So with the PHEV ... the ratio on the gearbox...

:shock:

 

[Phevourite]

So with the PHEV ... the ratio on the gearbox...

:shock:

Ratio - the number of input revolutions counted relative to the number of output revolutions.

 

[Trex]

But hysteresis is not really suitable for this discussion I think.

Regards Trex.

Well only in as much as it is essentially the technical word for feathering or at least for why feathering is possible. Hence, even with old technology, putting your foot down till you hit the desired speed means you can then back-off slightly before your speed decreases again. That margin between the end state on the way up and the same speed state on the way down is caused by the hysteresis in the control loop - albeit via flaps and valves and stuff in early combustion engines.

So with the PHEV if you end up driving at say a steady 50kph, using the petrol engine, but get there by accelerating to 50 the ratio on the gearbox and the tuning of the engine is going to be different than if you get there by decelerating to the same speed. Will it be enough to matter? That's another question.

Hi Phevourite,

"Hence, even with old technology, putting your foot down till you hit the desired speed means you can then back-off slightly before your speed decreases again. That margin between the end state on the way up and the same speed state on the way down is caused by the hysteresis in the control loop - albeit via flaps and valves and stuff in early combustion engines."

In engineering terms I think I would describe that phenomenon you describe there as backlash or "play" in bushings, pins etc .

Backlash leads to what we call in motion control engineering as a " deadband " .

Hysteresis is more a "memory " based effect . ie like friction in a throttle cable for instance or time based " lag " as I described before.

But either way look at what was written originally in the first post:

I found that the venerable practice of "feathering the throttle" (which means pushing the accelerator to obtain a certain speed/acceleration and then easing up marginally without losing the power - I've been doing that since the stone age on cars varying from an A Ford to a Morgan, so it is an automatism for me) works well and reduces racing the ICE significantly.

Note the easing up marginally without losing the power

If it was hysteresis, backlash, deadband or whatever you are doing nothing but making extra work for yourself.

Why with your foot would you try to find the deadband of a system.

But that is just my opinion.

Regards Trex.

 

[anko]

So with the PHEV ... the ratio on the gearbox...

:shock:

Ratio - the number of input revolutions counted relative to the number of output revolutions.

Sure. But you wrote:

the ratio on the gearbox and the tuning of the engine is going to be different

I was just curious which gearbox you were referring to ....

 

[Claymore]

Draft post deleted.

 

[Claymore]

I'm not an engineer or any sort of technician but I understand the basic theory of hysteresis.

My fan heater, set to 21 degrees will produce heat until its sensors indicate that the target temperature has been reached, then it will cut out, hysteresis based programming will ensure it freezes the hell out of me by allowing the temperature to drop to 19 degrees before restarting its hysteresis cycle and thus eliminating the possibly more annoying trait of constantly switching on and off between 21 degrees and 20.9 degrees.

Turning to motor vehicles, ever since my Lambretta GT200 followed by my Austin A35 and every vehicle in between then and now...if I was travelling along (let us assume a never ending smooth level road with no wind) at 50 mph (chosen to equal 80kph) and wish to increase my speed by 10% I would increase throttle and watch the speed increase until the target was achieved and then reduce power to maintain the target speed.

That, in my mind, is nothing whatsoever to do with hysteresis but may well be an example of 'feathering'.

 

[ian4x4]

The competence area of the man-machine interface.

The man-machine interface is becoming increasingly more important for today's telematics and automation systems to make complicated technology manageable. The human being is often the limiting factor for the overall system performance. A system can only be used with maximum efficiency and optimum costs if the human being - as the user and the operator of the system - is capable to understand and to utilize the behavior of a system through the man-machine interface.

I came across this quote while investigating the possibility that someone had scientifically studied the the optimum control of EV /PHEV systems.

It seems that when 'modeling' control systems, every effort has to be taken to help the human operator carry out the action as efficiently as possible. I think sometimes (and in my experience) this has to be done in spite of the operators actions.

Can't we all just agree to work the controls in a sensitive manner and thereby save fuel.

 

[Regulo]

Agree? On here? Are you mad?

 

[Neverfuel]

I've got to be honest - I think there is a whole load of nonsense being promulgated here.

I'm not an engineer or any sort of technician but I understand the basic theory of ...........................

My fan heater, set to 21 degrees will produce heat until its sensors indicate that the target temperature has been reached, then it will cut out. .........programming will ensure it freezes the hell out of me by allowing the temperature to drop to 19 degrees before restarting its.........cycle and thus eliminating the possibly more annoying trait of constantly switching on and off between 21 degrees and 20.9 degrees.

Turning to motor vehicles, ever since my Lambretta GT200 followed by my Austin A35 and every vehicle in between then and now...if I was travelling along (let us assume a never ending smooth level road with no wind) at 50 mph (chosen to equal 80kph) and wished to increase my speed by 10% I would increase throttle and watch the speed increase until the target was achieved and then reduce power to maintain the target speed.

That, in my mind, is nothing whatsoever to do with............but may well be an example of 'feathering', for all we care.

Hi Claymore

I like your exam question - the answer is "hysteresis" :lol:

Have I passed?