Kingsong S20/S22 (Confirmed)

[Rawnei]

54 minutes ago, Brendan "nog3" Halliday said:

For anyone curious, with the latest firmware it is possible to tweak the rear light colors and animations (for idle state).

The standard animation is the single color breathing, marked as 'Auto Mode' in the Kingsong app. This can be set to any RGB color in the app using the color wheel, it takes the first color palette you set in the app for the animation.

There is also a breathing color mode that is the same breathing pattern as above but rotates through the RGB spectrum .

And then just because, there's also a rainbow mode:

I've asked Kingsong to add another option to this list, a solid single color always on mode. This means people could have a solid rear red light.

Nice, glad to hear, red all the way.

[Eucner]

5 hours ago, techyiam said:

Average speed is commonly calculated by dividing the total distance traveled by the total elapsed time. (The speeds are time averaged) I am putting forth a hypothesis contending that the common average speed metric is more misleading than another average speed metric I have in mind when used to gauge range for scenarios where much lower speeds are also used for significant distances during the range test. Average speed is a single speed at which when travelling at, would result in the same elapsed time as in the actual trip travelling at various speeds.

Average speed is very good metric of average speed . For energy consumption we have different metrics, like Wh/km.

5 hours ago, techyiam said:

Consider a hypothetical scenario where the first half of a trip is ridden at 5 kph. And the latter half at 40 kph on a 20 km trip. Thus, the first half of the trip will take 2 hours, and 15 minutes for the latter half. A distance of 20 km traveled over 2.25 hours translates to an average speed of 8.9 kph. (Time averaged)

Whereas, in my proposed average speed, where speeds are average over distance, the calculation becomes as follows:

( 5 kph x 10 km + 40 kph x 10 km ) / 20 km = 22.5 kph.   

My hypothesis is that the 22.5 kph average speed would more accurately reflect the true battery energy consumption. By this, I mean the battery drain will be closer to the actual trip of going at 5 kph for the first half, and 40 kph for the last half, with the euc going the full distance traveling at 22.5 kph than at 8.9 kph.

Let's check you hypothesis with an other example. Consider the first half of a trip is ridden at 0,00001 km/h. It would take 114 years to finish this part. Still your distance weighted average speed would be about 20,0 km/h. This is not an accurate metric of average speed or energy consumption.

[techyiam]

1 hour ago, Eucner said:

Average speed is very good metric of average speed . For energy consumption we have different metrics, like Wh/km.

At the outset of my discussion, I clearly spelled out the context of which I was referring to, that of gauging range. 

If you prefer to use time-averaged average speed to gauge range, that is your prerogative. 

1 hour ago, Eucner said:

Let's check you hypothesis with an other example. Consider the first half of a trip is ridden at 0,00001 km/h. It would take 114 years to finish this part. Still your distance weighted average speed would be about 20,0 km/h. This is not an accurate metric of average speed or energy consumption.

If you feel riding at a speed of 0.00001 kph is useful in your range test, knock yourself out. BTW, you have extraordinary balancing skills, never mind patience. 

Again, it is your prerogative to go outside the range of applicability. Personally, I don't see anyone will find it useful in a range test to ride below 5 kph for any significant distance.

"My hypothesis is that the 22.5 kph average speed would more accurately reflect the true battery energy consumption. By this, I mean the battery drain will be closer to the actual trip of going at 5 kph for the first half, and 40 kph for the last half, with the euc going the full distance traveling at 22.5 kph than at 8.9 kph."

Perform the test and report back. You may able to prove that 8.9 kph would be more accurate in this scenario. 

[mrelwood]

I think the formula proposed by @techyiam makes sense. We all know that there can’t be a simple formula that would be anywhere near truly “precise”, but I’m sure the proposed one is still much more precise than simply the avg speed.

[Grimm10]

3 hours ago, Brendan "nog3" Halliday said:

For anyone curious, with the latest firmware it is possible to tweak the rear light colors and animations (for idle state).

The standard animation is the single color breathing, marked as 'Auto Mode' in the Kingsong app. This can be set to any RGB color in the app using the color wheel, it takes the first color palette you set in the app for the animation.

I've asked Kingsong to add another option to this list, a solid single color always on mode. This means people could have a solid rear red light.

This is good news.  I'm happy with this function.

[Forwardnbak]

3 hours ago, Brendan "nog3" Halliday said:

For anyone curious, with the latest firmware it is possible to tweak the rear light colors and animations (for idle state).

 

Awesome, thanks Brendan. 

[Rollin-on-1]

1 hour ago, mrelwood said:

I think the formula proposed by @techyiam makes sense. We all know that there can’t be a simple formula that would be anywhere near truly “precise”, but I’m sure the proposed one is still much more precise than simply the avg speed.

Would it be even better simply record the average wh/mile reported by EUC world in set intervals such as @ 120v, 115v, 110v, 105v, 100v, 95v, and 90v?  This would smooth out the starts and stops, and other variations.  Then other riders can compare the wh/mile they normally expend for their riding weight, style, and terrain to get a sense of what their expected range may be.

Edit: I haven't used the other apps so I'm not sure if they have the same wh/mile reporting as EUC World, but I'll assume they do until someone corrects me.

Edited March 1, 2022 by Rollin-on-1
Added more to my original thought

[Eucner]

55 minutes ago, techyiam said:

At the outset of my discussion, I clearly spelled out the context of which I was referring to, that of gauging range.

If you prefer to use time-averaged average speed to gauge range, that is your prerogative. 

Again, it is your prerogative to go outside the range of applicability. Personally, I don't see anyone will find it useful in a range test to ride below 5 kph for any significant distance.

We already have a metric called average speed. It's definition is commonly understood and used. You should't create a new metric with same name. Your "time-averaged average speed" is universally called "average speed". On the other hand your "average speed" should be called "distance weighted average speed" or something similar.

You didn't specify any limitations for your metric. An example isn't such.

1 hour ago, techyiam said:

If you feel riding at a speed of 0.00001 kph is useful in your range test, knock yourself out. BTW, you have extraordinary balancing skills, never mind patience. 

Didn't you find anything else to defend your metric against the limitation I showed?

1 hour ago, techyiam said:

 

Perform the test and report back. You may able to prove that 8.9 kph would be more accurate in this scenario. 

You have demonstrated this new metric and have the burden to show its usefulness and limitations.

I understood that with your distance weighted average speed you try describe the effective wind drag effect on the received range of EUC. Using the values of your example, the effective wind speed would be 28,5 km/h. Your proposed metric underestimates this speed by 21%. Every ride at least starts and stops at low speed. This kind of metrics should be compatible with low speeds.

[Seba]

26 minutes ago, Rollin-on-1 said:

Would it be even better simply record the average wh/mile reported by EUC world in set intervals such as @ 120v, 115v, 110v, 105v, 100v, 95v, and 90v?  This would smooth out the starts and stops, and other variations.

But energy consumption doesn't depend on battery voltage. Riding style, wind, rider weight, tire pressure, terrain profile - these are the factors impacting energy consumption. In fact, to reliably compare wheels, you have to perform tests in controlled environment, strictly adhering to estabilished test procedure that is the same for all the wheels tested. This is not the case when riding outdoors, on different routes, in different weather.

[Chriull]

22 minutes ago, Eucner said:

understood that with your distance weighted average speed you try describe the effective wind drag effect on the received range of EUC.

This @techyiam's "distance weighted average speed" is a how really a little bit as the  rms value of the speed. 1(

But this could be more precise for faster speeds as consumed power consists of some constant part (self balancing of the rider, more or less speed independend and friction). 

1) As average speed is the sum of v * Delta t . If one sums v * Delta s with Delta s=v* Delta t this equals to sum over v²*Delta t.

Just the square root of the result is missing after the division by the total distance.

No idea if the principle of the root mean square to abtain an effective average value for energy consumption in electronics can be similary used for energy needed to overcome air drag?

[RagingGrandpa]

Regarding measuring the efficiency of one EUC vs another...

1 hour ago, Chriull said:

energy needed to overcome air drag

As a function of speed, that metric is "power". 
I think efficiency (wh/mi) will be a simple linear relationship when mapped to power.
This gets us away from the nonlinear pitfalls of working with average speeds.

EUC World reports "average power" for EUC models where the app can measure or estimate battery current.
Perhaps the only remaining concern is "when is the average-power calculation active?" (@Seba could you confirm it?)
While Riding (and not while stationary) is the answer we're hoping for; otherwise stationary time would drag down the average and defeat our purpose.

This "Average power while riding" would be a good metric for an efficiency comparison, performed like below:

• Test two EUC's for the same riding condition: Same-rider, same-route, same-day (same wind). And attempting to sustain the same cruising speed.
• Use a short route that consumes only ~20% battery, and repeat it 4 times for each EUC in an alternating fashion. (Creates a distribution of average power values)
• From all 8 tours, plot their resulting Average Efficiency (wh/mi) against Average Power (W) in an X-Y graph. (Average speed is ignored)
• The more efficient EUC will show a trendline higher than the other.

Sadly this won't be a valid comparison to "other riders on other routes"... because of too many variables in the riding condition.

.02

[Tawpie]

2 hours ago, Seba said:

Riding style, wind, rider weight, tire pressure, terrain profile - these are the factors impacting energy consumption

This why power efficiency is so difficult to estimate and why estimates simply don't extend well to individual results. Auto fuel economy is lumped into "city" and "highway" classes and if you want to know what city or highway driving entails, you have to look it up. Adam Savage says your fuel economy depends on your mood—and he's very emphatic about it... having spent a fair bit of time poring over my EUC World .csv files, I tend to agree.

The range you get on any particular wheel is going to be your range, looking at my range is nothing more than entertainment.

Speaking of range and estimates and Wh/km, I started a topic to try to gather some information from other riders https://forum.electricunicycle.org/topic/26948-whats-your-whkm/. So far we have me, who evidently uses magic to keep my consumption down, and @RagingGrandpa who shared data showing how you can "ride like Ian of SpeedyFeet" and drain a battery rapidly. Two riders, different wheels, different conditions... radically different results.

 

Edited March 1, 2022 by Tawpie

[SanDiegoGuy]

Is there any way King Song will let people choose mph instead of kph with a new firmware update?

[Tawpie]

4 minutes ago, SanDiegoGuy said:

Is there any way King Song will let people choose mph instead of kph with a new firmware update?

There's no reason they couldn't, other than it takes time and money. My money is that they will add the option in the future but I'm not holding my breath.

[Rollin-on-1]

3 hours ago, Seba said:

But energy consumption doesn't depend on battery voltage. Riding style, wind, rider weight, tire pressure, terrain profile - these are the factors impacting energy consumption. In fact, to reliably compare wheels, you have to perform tests in controlled environment, strictly adhering to estabilished test procedure that is the same for all the wheels tested. This is not the case when riding outdoors, on different routes, in different weather.

Perhaps I should have explained why I listed the voltages.  It is only to provide some granularity for the rider's demand on the wheel at easily repeatable intervals - more data points if you will instead of a single value for the entire ride.  I never intended to imply that the battery voltage was a variable that impacted energy consumption, although throttling at lower voltages will by default limit the energy consumption because the allowable speed is reduced by the firmware.  

By looking at the wh/mile used in any given interval (126v to 120v for example), it provides a convenient yet relative tool for comparing different rides.  If we could see Ian's wh/mile average for each interval for a range test that got 30 miles and compare that to someone else's range test that got 60 miles, a rider could then make some educated guesses about where they may fall along that spectrum based on the wh/mile they tend to get for their particular weight, style, terrain, etc.  It isn'rt meant to be a precise calculation, but rather a way to help estimate.

 

Edited March 1, 2022 by Rollin-on-1
Added further clarification.

[mrelwood]

3 hours ago, Rollin-on-1 said:

Would it be even better simply record the average wh/mile reported by EUC world in set intervals

If EUCw (or DB) is running, it can already measure the overall power consumption as well as the wh/km (wh/mi). No need for other metrics.

 

3 hours ago, Eucner said:

You have demonstrated this new metric and have the burden to show its usefulness and limitations.

We all know by now that you like to be very scientific about all things numbers and physics. But this isn’t a scientific debate, and he isn’t trying to create new peer reviewed measurement methods. The formula was never suggested as either. It was just a thought and a suggestion, which incidentally does give better results than the also non peer reviewed method of calculating consumption from the raw avg speed.

[Kutvis]

Not quite on the topic anymore, but this could be attained with a gyro. Just set wheel speed propper and enough weight on it + let it roll empty, do this with enough different speeds. Then do this in realworld riding, in calm weather with a few different average speeds on rather flat terrain, throw it all into excel and voela, rough numbers we have.

And back into the subject. I got info from an EUC shop. They expect the wheels to further delay, 2-3mnth. No, not going to go in more depth as the info wasn't solid, but one of the biggest EU shops anyway.

Speculation: The train to move the EUC from China to EU is going through .. Russia and Belarus => Because of the War, all the wheels have to be shipped by sea. Good catch @Rolzi

Edited March 4, 2022 by Kutvelo

[Seba]

2 hours ago, RagingGrandpa said:

Perhaps the only remaining concern is "when is the average-power calculation active?" (@Seba could you confirm it?)
While Riding (and not while stationary) is the answer we're hoping for; otherwise stationary time would drag down the average and defeat our purpose.

I can confirm - power averaging is only active during a ride (or to be more precise, when wheel speed is at or above 2 km/h).

[Paradox]

Just posted.

 

[Rollin-on-1]

4 hours ago, mrelwood said:

If EUCw (or DB) is running, it can already measure the overall power consumption as well as the wh/km (wh/mi). No need for other metrics.

I think you misunderstand what I'm saying, but it isn't important.

 

[Tawpie]

3 hours ago, Paradox said:

Just posted.

 

Average riding speed of 24 mph is moving right along. His route must have very few stops (slowing down to stop and speeding up really cuts your average riding speed).

[Eucner]

7 hours ago, Chriull said:

No idea if the principle of the root mean square to abtain an effective average value for energy consumption in electronics can be similary used for energy needed to overcome air drag?

Thanks, you are on the right track.

The challenge with range estimates is the huge number of variables. No two rides are equal. If we want to compare wheel effectiveness, a bench test would be the best way. On the other hand, if we want to have comparative range estimates, we should define standard conditions. Car consumption tests are done this way nowadays. However there would still be a problem. It would tell hardly anything about your specific ride.

For energy consumption the Wh/km (Wh/mi) value is otherwise good, but it is very speed sensitive. To overcome this the average speed is quite often also mentioned with it. Energy consumption doesn't linearly depend on speed. This creates situations were same average speeds can give totally different energy consumption values. @techyiam created his own metric to overcome this problem. It has it values, but also limitations. It is not derived from physics or experimental data.

Let's see what physics could give as. At the first, problem need to be simplified. I shall omit all other resistances but air drag. It is the most important energy consumer.

The air drag force is related to speed². 

The amount of consumed energy is force times distance, so the energy consumption by distance (Wh/km) is related to speed².

Then the effective wind drag speed shall be related to square root of Wh/km value. This can be also calculated from speed RMS values per distance (wink, wink @Seba).

From here we can see that actually @techyiam was quite close with his metric. Just exponent 2 need to be added to all speeds and a square root taken from the result.

5 hours ago, mrelwood said:

We all know by now that you like to be very scientific about all things numbers and physics. But this isn’t a scientific debate, and he isn’t trying to create new peer reviewed measurement methods. The formula was never suggested as either. It was just a thought and a suggestion, which incidentally does give better results than the also non peer reviewed method of calculating consumption from the raw avg speed.

Why to settle in something half-baked, when we can do better?

[Flying W]

54 minutes ago, Tawpie said:

Average riding speed of 24 mph is moving right along. His route must have very few stops (slowing down to stop and speeding up really cuts your average riding speed).

It sure is. On my commute I'm beeping once up to speed the entire ride, but I also have 5 stop lights that I almost always get red. My ave riding speed is usually 22 to 26 depending on how many I get green. 

This is on an RST. It beeps around 36.5 mph when I have a full pack to change into dress attire at the office, riding weight on commute is aprox 200lbs.

[Finn Bjerke]

Backlight = RED please....  No bloody rainbows here ffs. 

[Finn Bjerke]

Range is a laff ...... Ill wait for Inmotion V13 ...... What exactly was the s20 HYPE about .... fancy sidepads in nice colors?