Speak EV - Electric Car Forums banner

1 - 20 of 62 Posts

·
Registered
Joined
·
225 Posts
Discussion Starter #1 (Edited)
I was thrilled to find a recent article that explores a hyper-mile technique they are calling "pulse and glide." You can find it easily enough with Google if you wish to learn more. "Can Pulse and Glide work to Improve Electric vehicle highway range" I have been doing this for a few years now. Pulse and glide is just that- you pulse up to speed and then pull back the throttle a little and glide or, coast as long as possible without getting deep into regeneration which would constitute a loss of inertia. I built an automatic timer to low frequency pulse modulate the throttle. I did this to prevent the negative effects of voltage sag. This allows the car to accelerate in pulses (3.5 second) and gives the battery a 1 second rest every 3.5 seconds. There is a brief recovery of the voltage and the current has dropped significantly lower at cruise speed of 60mph. The timer can pull back the throttle as much as 25% or less depending on the throttle position. I swear that this is working well in my car despite screams from many that I am violating the laws of physics or the laws of thermodynamics. Hyper-mile techniques apply well to any vehicle with or without circuits to automate them. I think we need to explore all methods that will help our cars go the distance. Many of these techniques are unconventional but, so is the electric car. I am one of those freaks that converted his own car to electric just so I could experiment and explore all of the possibilities. For anyone who thinks that resting the battery for 1 second is not very long, well, it amounts to ~13 minuets in a 1 hour period of driving. The seconds add up to a very big savings in terms of time and energy. It means that for almost 1/4 of the time on the road, the car is coasting instead of guzzling energy. I named my circuit the "Pot box Manipulator / Performance Enhancer." Others called it "snake oil." Finally, others are discovering the principle behind it.
 

·
I'm not crazy, the attack has begun.
Joined
·
25,200 Posts
There is possibly truth to this because the operation of electric motors is generally more efficient at higher loads. So if you load the motor over a shorter period of time to accelerate and then eased off then, if all else were equal, it'd be more efficient than a continuous lower level load, providing the period of higher load puts you in the max efficiency spot for the motor.

The problem here is that the battery works differently so the most efficient point may be a little less than the peak efficiency point for the motor. As you have identified, it is also a battery-dependent phenomena.

Just take a look at the efficiency plot for a motor and you'll see why this is. However, as most EV motors operate in a very narrow range of efficiencies (i.e. they are pretty broad in their peak efficiency) the effect may be as much observer bias as real. Yes there should be a slight improvement, but that doesn't necessarily mean you'll see it. I would say it is not possible to see improvements of, for example, 10% or such with current EV motors. However, normal day to day driving varies naturally by 8% so anything less than that, which I think it would be, would be swamped in statistical noise.
 

·
EV Convert
iPace HSE / 2019
Joined
·
583 Posts
Be very cool to build this idea into the cruise control. So a computer does this for you, minutely adjusting as it goes, learning and adapting its algorithms to the actual battery condition and temp.
 

·
Registered
Joined
·
1,231 Posts
I read about this when we had a Prius. I tried it out a few times but can't say I noticed any particular difference.
 

·
I'm not crazy, the attack has begun.
Joined
·
25,200 Posts
I doubt you would in a Prius due to the particular manner in which it implements its strategy. Basically, it already does this to a certain degree.
 

·
Chartered Engineer
Joined
·
1,855 Posts
The Japanese hyper milers used this technique on the Prius with apparent success although I never managed to get any noticeable benefit when I tried. I got bored too quickly I suspect. They also went to the extremes of driving barefoot to get more precise feel of the throttle - definitely can't be bothered to try that one.
 

·
Registered
Joined
·
225 Posts
Discussion Starter #7
Be very cool to build this idea into the cruise control. So a computer does this for you, minutely adjusting as it goes, learning and adapting its algorithms to the actual battery condition and temp.
I think that this would be very easy to incorporate into the motor controller. An algorithm that adjusted the controller to find the lowest value of Kwh per mile at any speed is what I am thinking of. I am certain every automaker is putting some kind of economy mode feature into the cars that they produce. As for my own car, I am not yet certain of the results. It takes a lot of honest testing and data collection. I don't have a test track to drive on to find the answers. I can go 62 miles in a mix of city and secondary roads with many stops and starts but, the maximum highway range is unknown. I will have to drive at constant 63 mph and to the point that the car will not maintain 63 mph. This can be difficult on a highway with other drivers going 90+. I know that I will strand myself for a while to find the answer. I carry a generator in the trunk so I can charge up where ever I end up. This Summer I might find the time and the courage to do this.
 

·
Registered
Joined
·
225 Posts
Discussion Starter #8 (Edited)
The Japanese hyper milers used this technique on the Prius with apparent success although I never managed to get any noticeable benefit when I tried. I got bored too quickly I suspect. They also went to the extremes of driving barefoot to get more precise feel of the throttle - definitely can't be bothered to try that one.
There are several reasons why this technique might not work. If the car goes heavy into regenerative braking as soon as you let up on the pedal, this will make coasting or gliding impossible and this will kill the desired effect. Also, if the car is already extremely efficient, it may be difficult to realize a significant gain in range. And, then of course, there is the driver and what that driver does could reduce the range if they are doing it wrong. It seems to be a delicate matter to get exactly right. Finding the best speed to maintain is part of it. Timing is also important.
 

·
Registered
Joined
·
225 Posts
Discussion Starter #9
There is possibly truth to this because the operation of electric motors is generally more efficient at higher loads. So if you load the motor over a shorter period of time to accelerate and then eased off then, if all else were equal, it'd be more efficient than a continuous lower level load, providing the period of higher load puts you in the max efficiency spot for the motor.

The problem here is that the battery works differently so the most efficient point may be a little less than the peak efficiency point for the motor. As you have identified, it is also a battery-dependent phenomena.

Just take a look at the efficiency plot for a motor and you'll see why this is. However, as most EV motors operate in a very narrow range of efficiencies (i.e. they are pretty broad in their peak efficiency) the effect may be as much observer bias as real. Yes there should be a slight improvement, but that doesn't necessarily mean you'll see it. I would say it is not possible to see improvements of, for example, 10% or such with current EV motors. However, normal day to day driving varies naturally by 8% so anything less than that, which I think it would be, would be swamped in statistical noise.
There are many variables to take into account, load, speed, voltage, current, battery temperature, etc. I believe that it is fair to say that when all operating conditions are perfect, peak efficiency will result. I tend to think that the battery has much to do with it. I have run motors until they stop running and will not run any longer on the battery. I then disconnect the battery from the motor and charge a capacitor in parallel with the same battery for a short time. Next, I re-connect the motor to the battery and capacitor. What happens? Answer: the motor runs at full speed for a short time. The process can be repeated over and over again such that there is a cycle of on and off times (run and rest / recover). Finally, the battery will be truly discharged and the motor will not run. To me, this is proof that, how energy is applied is as important as how much total energy is available. It also can be a demonstration of what happens to a battery when it is truly dead and can never be made to work again. This last part is something to avoid where the electric car is concerned. So, in my car I am careful not to drain the very last bit of energy from the Lifepo4 battery. But, I am determined to go the very last safe mile.
 

·
Chartered Engineer
Joined
·
1,855 Posts
There are several reasons why this technique might not work. If the car goes heavy into regenerative braking as soon as you let up on the pedal, this will make coasting or gliding impossible and this will kill the desired effect. Also, if the car is already extremely efficient, it may be difficult to realize a significant gain in range. And, then of course, there is the driver and what that driver does could reduce the range if they are doing it wrong. It seems to be a delicate matter to get exactly right. Finding the best speed to maintain is part of it. Timing is also important.
The hypermilers apply just enough pressure on the throttle pedal to coast i.e.: all powertrain idling without delivering or absorbing power. The idea of driving barefoot is that it's supposed to make this more precise. On a Prius the main region comes in with the brake pedal anyway.
 

·
I'm not crazy, the attack has begun.
Joined
·
25,200 Posts
.... I then disconnect the battery from the motor and charge a capacitor in parallel with the same battery for a short time. Next, I re-connect the motor to the battery and capacitor. What happens? Answer: the motor runs at full speed for a short time. The process can be repeated over and over again such that there is a cycle of on and off times (run and rest / recover). Finally, the battery will be truly discharged and the motor will not run. To me, this is proof that, how energy is applied is as important as how much total energy is available......
err... well you're actually demonstrating here the changes in internal resistance with SOC, and that it is better to draw a lower current over a longer time, which is the polar opposite of pulse-and-glide.

The battery doesn't favour pulse and glide, but at low overall power levels the efficiency gains in the motor trump the reduced efficiencies of the battery.
 

·
Registered
Joined
·
3,837 Posts
Being a somewhat trusting person, since the ideas behind 'pulse and glide' are not only well documented but also not new, would not this control method have been already tested by EV manufacturers, and built into the control mechanisms for their motors to the extent tht it works. (at the split second level) ?
I would suggest that if you need to drive in bare feet to obtain better milege control your right foot needs training to be more gentle in its actions. From what I can read 'pulse and glide' seems to be a leftover from old style diesel engines , where they were driven as if the accelerator was an on/off switch. Maybe in an old somewhat basic diesel lorry with slow human control of the accelerator through big boots it made some sense, but in modern electrics with state of the art electronics it has no place.
 

·
I'm not crazy, the attack has begun.
Joined
·
25,200 Posts
It is new for EVs.

In fact, PHEVs basically implement the scheme. They run the engine at high load and store the 'excess' power until the storage is full, then switch off the engine.

This is quite different for EVs and is a different bag-o-fish altogether. It is much more nuanced and is going to produce very little gain because in an ICE the difference in efficiency is a big band between 0% and 35% whereas for electric motors it is more like 85% to 97% efficiency.

As you are already using primary convertible energy, electricity, there is nothing left to 'store' the energy in, except the kinetic energy of the vehicle itself. Once you know the efficiency map of you car you can drive it a bit more efficiently for a given target average speed, but the most efficient is still more likely to be 'as slow as possible'.
 

·
Registered
Joined
·
410 Posts
I didn't realise this was a thing, but it's essentially how I drive in EV mode. Seems obvious to me as it seems to keep a motor running at constant speed, you need to continually drive it, if you apply e.g. double the energy you get a big kick.

For example, I might need 12kWh/100km to maintain a constant speed, or could apply 40kWh/100km for 5 seconds to get 10mph above my current speed and then coast at 0.3kWh/100km for the next 15 seconds until I'm back to my original speed.
 

·
Registered
Joined
·
3,837 Posts
For example, I might need 12kWh/100km to maintain a constant speed, or could apply 40kWh/100km for 5 seconds to get 10mph above my current speed and then coast at 0.3kWh/100km for the next 15 seconds until I'm back to my original speed.
I don't know what sort of info display the Golf has, but in the LEAF there are several possible displays of power in use, a couple of which give an instantaneous readout which can be presented in ways 'pleasing' to the driver.
Using the power info pages on the satnav screen there is a screen to show current energy useage, by accessories, heater and aircon, and more importantly the engine. When driven at a constant speed, using speed limiter, on a flat road in constant conditions the power used display settles to a constant output, when the speed is low the power used often becomes a pulsed readout, with a short burst of energy used and a shorter time of no useage displayed. I would suggest that the electronics in the LEAF already use a pulse and glide technique, built into the motor controller, and already makes the optimum useage of 'fuel'. (and all without annoying other drivers with constant changes of speed).
 

·
Registered
Joined
·
410 Posts
Probably. TBH, I don't like the display options in the Golf GTE. Consumption options are limited to mi/kWh, km/kWh and kWh/100km (which turns to kWh/h (!) when stationary). I'd honestly just prefer kW the entire time and a kWh estimate on the battery, but we're not allowed to see anything like that.

At some point I plan to stick my OBD2 reader in to see if I can more more interesting stats that way, but it's low down my priority list at the moment.
 

·
Registered
Joined
·
225 Posts
Discussion Starter #17 (Edited)
Being a somewhat trusting person, since the ideas behind 'pulse and glide' are not only well documented but also not new, would not this control method have been already tested by EV manufacturers, and built into the control mechanisms for their motors to the extent tht it works. (at the split second level) ?
I would suggest that if you need to drive in bare feet to obtain better milege control your right foot needs training to be more gentle in its actions. From what I can read 'pulse and glide' seems to be a leftover from old style diesel engines , where they were driven as if the accelerator was an on/off switch. Maybe in an old somewhat basic diesel lorry with slow human control of the accelerator through big boots it made some sense, but in modern electrics with state of the art electronics it has no place.
I don't know because some people claim that pulse and glide works in their Prius. What I find ironic is that the electric car is a very old idea that was replaced by a state of the art ICE. I have been told many times over the years that if electric cars worked and were practical, they would have built them using state of the art technology. But, the thing about state of the art is that in a few years it is no longer state of the art and the whole picture changes. Ideas both old and new work together. The idea that, if it could be done they would have already done it is a common notion that has been around for as long as I can remember. Still designs are always changing and what is old becomes new again.
 

·
Registered
Joined
·
225 Posts
Discussion Starter #18 (Edited)
err... well you're actually demonstrating here the changes in internal resistance with SOC, and that it is better to draw a lower current over a longer time, which is the polar opposite of pulse-and-glide. The battery doesn't favour pulse and glide, but at low overall power levels the efficiency gains in the motor trump the reduced efficiencies of the battery.
I first started pulsing the throttle long before before I knew what pulse and glide was. On several long trips the battery voltage would drop very low until acceleration was poor. I found that acceleration with higher voltage pulses was working better to get the car moving than was holding the pedal to the floor while the battery died. In fact, if I held the pedal to the floor, the car would not go any faster but, the motor was drawing very high current as the battery voltage dropped sharply and the battery discharged rapidly. A motor supplied with higher voltage will typically run faster and draw less current. Conversely, the same motor supplied with lower voltage will tend to work harder, run slower and draw more current.
Something good is happening. I can accelerate up to 60 mph using no more than 200 amps at 120 volts. I can cruise at 60 mph on less than an average of 100 amps.
 

·
Status of my member is private
Joined
·
262 Posts
They also went to the extremes of driving barefoot to get more precise feel of the throttle - definitely can't be bothered to try that one.
As a middle aged gentleman, +1 vote for proper driving shoes, which are made to give you barefoot-like precision.
 

·
Registered
Joined
·
3,344 Posts
My objection to pulse & coast is the aerodynamic fact that windage losses go up with the square of your speed. So if you do x miles at 50 (coasting) and x at 70 (applying power) your windage energy loss is proportional to (5*5)+(7*7) =74. If instead you did it all at 60 it's proportional to (6*6)+(6*6)=72. Not a huge difference and I doubt you would vary your speed so much, but the principal remains. My dad tried this to the limit on fosse way circa 1960 having read a newspaper article saying this was best, so he accelerated to 70 overtaking lorries & cars, then coasted to 30 as they overtook him, then he accelerated to 70 again, overtaking, etc. Before long the others were flashing & tooting him furiously!!! After about 50 miles of this he pulled into a cafe for a sticky bun ! he declared himself satisfied with the experiment, to the immense relief of his passenger, the North Leicestershire coroner!!! He never repeated this tactic.. :)
 
1 - 20 of 62 Posts
Top