Begode EX20S 3600wh: 100V, suspension, 104lb

[meepmeepmayer]

17 hours ago, EMA said:

https://ecodrift.ru/2022/05/03/begodeex20s-vneshnij-vid/

First sentences:

"Another new product from Begode in stock. And it seems that this is just an intermediate solution that will remain in history."

Looks like many think this wheel is already  halfway to being replaced by a successor?

[Nick McCutcheon]

1 minute ago, meepmeepmayer said:

First sentences:

"Another new product from Begode in stock. And it seems that this is just an intermediate solution that will remain in history."

Looks like many think this wheel is already  halfway to being replaced by a successor?

It must be dealers not carrying it which is killing its hype. I can't see why many people wouldn't want a Sherman with suspension, except for maybe the price and Begode-ness, but if the only way to get it is from AliExpress I see how a lot of people would be turned off. I'm interested to see if AR is able to maintain stock as they do other "regular" wheels.

Even more peculiar is the RS Resolute, which nobody except AR seems to carry and it has supposedly been discontinued, but they have stock right now. Plus, Begode for some reason has not taken the Monster v3 off of their site header image yet, despite having had the opportunity to do so for a while now. I'm sure AR would carry it if it was still available, it's a great wheel. 

[EUC Sharkman]

I wonder if Begode is using the Master to see how to make this succcessor. Things like if they will offer different battery pack options like with the Master, or if they just go with whatever option was most popular with the Master. Or whether or not they do a 134v C38, C30, or both. And then whatever price they decide to put this wheel at. There could also be issues discovered with the Master that they would want to correct.

 

The C30 version of this wheel is coming into AR next month. If a successor were to come out, when would that be? Within a year, or more?

[EMA]

7 hours ago, meepmeepmayer said:

First sentences:

"Another new product from Begode in stock. And it seems that this is just an intermediate solution that will remain in history."

Looks like many think this wheel is already  halfway to being replaced by a successor?

i think it will take a while before seeing something new, 100v is a super solid tech right now, let's see how the master goes...

EX20s is quite good from my perspective, if it's fit your need you will not be disappointed imho

of course during time Begode lineup will go gradually to 134v system, but they usually start with smaller wheels like RS 
 

[BleepBloopBlop]

The main thing is there is no need for high discharge cells when you have 8 cells in parallel. The high capacity cells offer all the current discharge needed and isn't a weak point in this 3600wh packs.

Even if Begode transitions up to a 134V system you would still have 6 cells in parallel. In my opinion 75A output current from the packs from 50E cells will be sufficient. (6P x 15A (non continuous rating) = 75A output current.

It is easy enough to switch the cells to 40T but at a 20% reduction in capacity I don't see that being needed. 

Edited May 4, 2022 by BleepBloopBlop

[InfiniteWheelie]

I would take the high discharge cells any day of the week. As long as you have near 3 kWh, range will likely never be an issue. However cutouts at high speed or extreme lean can still be a problem, and high discharge cells add more buffer. I believe they'd also last longer, being less stressed by the power drawn.

Edited May 4, 2022 by InfiniteWheelie

[Tawpie]

11 hours ago, BleepBloopBlop said:

It is easy enough to switch the cells to 40T but at a 20% reduction in capacity I don't see that being needed.

Still, it makes me wonder.

What is the trigger for an overlean cutout? Insufficient current or insufficient voltage? By this question I mean: is the controller watching the battery voltage to determine when enough is enough, or is it watching the phase current monitors (or the PWM)? If, and it's a big if, the cutout is triggered by the battery voltage getting too low under load, then I can see a headroom advantage from the high current cells. Less internal resistance, less voltage sag under load.

Of course, if a cutout is commanded due to max phase current then sag doesn't apply if the parallel 50Es can still supply current that meets the firmware's "too much current" set point. You would have the option to increase the firmware limit when high power cells are installed.

If it's due to max PWM duty cycle, you've exceeded the motor's ability to convert energy to thrust and the 50Es are probably sufficient.

Without seeing a schematic or firmware flow chart, my bet is on voltage as the trigger. It's cheaper to implement. But it well could be more sophisticated than that. One might be able to test to get a hint... get to very low battery but before low battery tiltback and try the ankle flick trick. If it cuts out, the cutout is probably initiated by the voltage monitors (it could be due to lack of current, but lack of current is directly related to lack of voltage and lack of current isn't "too much current").

Edited May 5, 2022 by Tawpie

[Nick McCutcheon]

44 minutes ago, Tawpie said:

Still, it makes me wonder.

What is the trigger for an overlean cutout? Insufficient current or insufficient voltage? By this question I mean: is the controller watching the battery voltage to determine when enough is enough, or is it watching the phase current monitors (or the PWM)? If, and it's a big if, the cutout is triggered by the battery voltage getting too low under load, then I can see a headroom advantage from the high current cells. Less internal resistance, less voltage sag under load.

Of course, if a cutout is commanded due to max phase current then sag doesn't apply if the parallel 50Es can still supply current that meets the firmware's "too much current" set point. You would have the option to increase the firmware limit when high power cells are installed.

If it's due to max PWM duty cycle, you've exceeded the motor's ability to convert energy to thrust and the 50Es are probably sufficient.

Without seeing a schematic or firmware flow chart, my bet is on voltage as the trigger. It's cheaper to implement. But it well could be more sophisticated than that. One might be able to test to get a hint... get to very low battery but before low battery tiltback and try the ankle flick trick. If it cuts out, the cutout is probably initiated by the voltage monitors (it could be due to lack of current, but lack of current is directly related to lack of voltage and lack of current isn't "too much current").

@supercurio would be able to explain it better than me (please correct me if I've got this wrong), but as you go faster, there's less of a gap between what the battery can provide and what the motor is taking. Once the battery is providing all it can to the motor, any attempt to go faster or load the system any more will result in a cutout. I too thought there was a trigger, and it took some convincing to get me to think otherwise, but it turns out it just feels that way due to how fast this barrier is crossed.

In more complicated terms, as motor speed increases, so does its back emf (voltage induced back into the system by the motor as a result of how the motor functions).  The closer the back emf gets to the battery voltage, the less of a potential difference there is between the two and thus the less current can be supplied. If a certain amount of current is needed to remain at speed, and the back emf gets high enough, the battery is physically not able to provide this current and a "cutout" (it's actually just a really fast overlean) occurs.

Higher discharge cells may help a bit due to their lower voltage sag (thus widening the gap between max battery power available and power taken by the motor), but in top-speed cutout scenarios, cell type can't reduce the back emf produced by the motor as it speeds up. It would be interesting to quantify how cell type affects cutout speed! 

@Chriull has a great thread where he determines how to see exactly when an overlean will occur by analyzing a KS-16(S? C?)'s cutout data from wheellog.

[meepmeepmayer]

Not sure if I understand the question correctly, but: the wheel switching off after an overlean is triggered by the tilt angle. If that is over 45° (or what it is), the self-balancing has officially failed and there is no point in continuing.

Same thing when the wheel tilts too much to the side and switches off.

-

And the overlean just happens.

current * excess voltage = power available for self-balancing

You can run out of free voltage from the voltage drop on acceleration. This drop depends on the cell type (high discharge or low discharge) and the battery configuration (lower drop on highly parallel config).

You can run out of current from the discharge limit of your cells, the battery configuration (more current available on high-p), and possible firmware current limits.

I think you expect way too much sophistication from our crappy wheels. No fancy triggers. You just overlean one, and it figures that out the obvious way ("Help, I've fallen and can't get up!") and switches off.

-

The EX20S being 8p means using higher discharge cells doesn't do much (if anything practically relevant) so I would prefer the biggest battery capacity possible.

Edited May 5, 2022 by meepmeepmayer

[EMA]

cut out are "not related" to battery voltage, it's not the battery that makes you go down but other limits

 

when you cut at high speed you have reached the end of your motor rpm

if you cut at low speed maybe it's the motor that not have the power or the board that reach firmware limit and prevent blew up

[Jean eRide.ie Community]

35psi on tire

150psi on suspension 

That's what's working well for me at the moment, just 15km so far on that setting, 15km before that.

Later I may increase both if needed. 

Suspension didn't bottom out yet on potholes, going down a curb or me jumping on the spot (it stayed a bout 1 cm away, can see it with the rubber rings).

The 35psi tire, I'll need to record some video to test if it is enough or if compresses too much when going down a curb.

[Chriull]

2 hours ago, Tawpie said:

Still, it makes me wonder.

What is the trigger for an overlean cutout?

As @meepmeepmayer wrote there is no cutout, as @Nick McCutcheon wrote it just could feel like one.

2 hours ago, Tawpie said:

Insufficient current or insufficient voltage? By this question I mean: is the controller watching the battery voltage to determine when enough is enough, or is it watching the phase current monitors (or the PWM)?

It's an overall system limit, mainly by the voltage difference between battery and the motor back emf, as written before.

2 hours ago, Tawpie said:

If, and it's a big if, the cutout is triggered by the battery voltage getting too low under load, then I can see a headroom advantage from the high current cells. Less internal resistance, less voltage sag under load.

A small headroom advantage. From the two points defining the limit - Stall torque and max no load speed (~lift cut off speed) "only" stall torque increases. So the connecting limit line becomes a bit steeper.

So the higher the speed, the lower the gained headroom.

The cells voltage sag is already regarded in the steepness of such a limit line and has no additionally influence.

Only the decreasing "internal" voltage by discharge additionally influences the limit.

2 hours ago, Tawpie said:

Of course, if a cutout is commanded due to max phase current then sag doesn't apply if the parallel 50Es can still supply current that meets the firmware's "too much current" set point. You would have the option to increase the firmware limit when high power cells are installed.

Firmware current limit exists only for (very) low speeds as the possible currents would annihilate the components quite instantly.

Overleans at normal/higher speeds just happen by reaching the system (battery/motor) limits - there is no interaction with the firmware. One can just get reported how near one is to the limit by watching the reported pwm duty cycle.

2 hours ago, Tawpie said:

If it's due to max PWM duty cycle, you've exceeded the motor's ability to convert energy to thrust and the 50Es are probably sufficient.

Pwm duty cycle is one measure that shows once one has reached the limit. Different to humans duty cycle cannot exceed 100%.

In most higher speed overleans one should not have any really notable difference between both battery types.

For details one can take a look at https://forum.electricunicycle.org/topic/7855-anatomy-of-an-overlean/ and the there mentioned links.

 

[Tawpie]

Thanks everyone!

I was trying to point out that in addition to the motor physics, there is a firmware layer that may actually be more important when deciding how useful cell selection will be.

Master doesn't appear to mind being forced into a wall... I'm quite certain 40T cells buy headroom with Master. I'm not so sure with S22 or other wheels that have firmware limits before forward tilt angle.

Something I don't know is the real world effect the lower capacity of the 40T has on range. The physics pretty much declare that it takes a certain amount of work to move an object from here to there, and so far there hasn't been a workaround. WattHours is a measure of work, and if you have fewer watt-hours available, you would expect that you simply can't get the same amount of work done... and won't be able to move your object as far. The unknown is due to this being a self balancing system that requires additional work to maintain balance, this work is over and above that required to get you from here to there. So I can image a nearly empty 50E system that can't do the balancing work when an equally empty 40T system still could because its voltage sag doesn't reach the firmware limits. I should be able to ride the 40T further than expected, because it can still keep me balanced—but this is firmware dependent.

Anyway, I do really think the high drain batteries are a "good" thing in general for all wheels, and headroom is only part of the reason. I also expect there to be a range penalty, but am hopeful that it's not the full 20%!

Edited May 5, 2022 by Tawpie

[techyiam]

21 minutes ago, Tawpie said:

WattHours is a measure of work, and if you have fewer wHs available, you would expect that you simply can't get the same amount of work done

Those discharge curves at various %C tells more of the real story. At higher discharge rates, the 40T charge capacity will start to look better.

[Tawpie]

4 hours ago, techyiam said:

Those discharge curves at various %C tells more of the real story. At higher discharge rates, the 40T charge capacity will start to look better.

The area under the curve will give you the total energy available. The high discharge cells can produce more current near the end of their discharge curve, and I think that's what allows the wheel to maintain balance 'longer'. Not much longer, but some longer!

Edited May 5, 2022 by Tawpie

[techyiam]

6 minutes ago, Tawpie said:

The area under the curve will give you the total energy available.

Discharge curves are plots of "Voltage vs Capacity Curves".

[Tawpie]

15 minutes ago, techyiam said:

Discharge curves are plots of "Voltage vs Capacity Curves".

Precisely. Capacity is measured in milliamp hours, voltage in volts. Multiply and you get power in milliwatt hours. Sum the area and you have total milliwatt hours of energy.

Edited May 5, 2022 by Tawpie

[Paul A]

Does this help?

___________

 

https://www.sciencedirect.com/topics/engineering/battery-capacity#:~:text=Battery capacity is measured in,battery will last 125 hours.

Battery Life Analysis

Shahin Farahani, in ZigBee Wireless Networks and Transceivers, 2008

6.1 Battery Discharge Characteristics

Battery capacity is measured in milliamps × hours (mAH). For example, if a battery has 250 mAH capacity and provides 2 mA average current to a load, in theory, the battery will last 125 hours. In reality, however, the way the battery is discharged has an impact on the actual battery life.

Discharging a battery at the manufacturer-recommended rate normally helps the battery deliver close to its nominal capacity. But the result cannot simply be extrapolated linearly to other discharge profiles.

 

In many short-range wireless sensor networks, although the average current consumption of a device is low, the instantaneous current can be high.

 

For example, in Figure 6.1, a transceiver with sleep current of 1 μA and peak active current of 20 mA wakes up every 2 seconds and stays active for 5 ms. The average current consumption is only 50 μA, but the peak current of 20 mA can have an adverse effect on the battery's actual capacity, especially if there is not enough time between the periods of high current discharge to let the battery rest and recover.

 

Figure 6.1. An Example of Current Profile of a Device in a ZigBee Network

This can be explained by the relaxation phenomena (or recovery effect) [1]. When a battery is discharged at a high and sustained rate, the battery reaches its end of life even if there are still active materials left in the battery.

 

However, if the discharge rate is not continuous and there are cutoffs or very low-current periods, the transport rate of active materials catches up with the depletion of the materials, giving the battery a chance to recover the capacity lost at the high discharge rate.

 

The actual capacity of a battery for a specific use-case scenario can be determined experimentally. Battery manufacturers might be able to provide an estimate for the actual battery capacity if the application's current profile (e.g., Figure 6.1) is provided.

The ratio of the battery's actual capacity to its nominal capacity is called the battery efficiency factor.

One way to avoid high current discharge periods, if possible, is to use a sufficiently large capacitor to supply the current to the transceiver when the node is active.

While the device is in sleep, the battery charges the capacitor, and when the device becomes active and requires a high discharge rate, this capacitor will provide current to the device.

In this way, the battery would not experience high discharge rate periods and the battery efficiency can be improved.

Another characteristic of a battery is its self-discharge rate. A battery, even when not in use, loses its capacity over time due to internal leakage.

Battery manufacturers quantify this leakage as self-discharge per month. For example, a 300 mAH battery with self-discharge per month of 0.5% loses 1.5 mAH of its capacity after one month.

The battery shelf life is defined as the longest time a battery can be stored before its capacity falls below 80% of its nominal.

[litewave]

I love these nerd posts. I sure miss @esaj and wish he'd come back.

Edited May 5, 2022 by litewave

[techyiam]

9 hours ago, Tawpie said:

The area under the curve will give you the total energy available.

Doesn't the 40T look better at higher discharge rates?

[Mike Sacristan]

I have done 14000km on my 2664Wh Nikola. All Samsung 50e cells in a 24s6p config.
It takes quite a bit of effort for me to push to 60A.
Samsung 50e has a 9.8A continuous discharge rating. They can go higher but 15A continuous would make them very hot.
It wouldn't surprise me if the Master with that HT motor spinning at high speed will be pushing those cells beyond the 9.8A spec very often.

My early EXN HS with LG cells will typically hit 60A if I am riding around 65kmh. If I hit a bump at those speeds I will usually spike a couple of amps and get a beep from the wheel. The Samsung 50e and LG M50T perform almost identically even though the LG have a lower max continuous discharge rating. The Samsung can handle higher discharge so in that sense they are "safer" but that's it.

Voltage sag can be quite insane on the EXN HS... I can dip one third of the battery or more at high speed and go from 90V to 81V in some scenarios. More precisely... at 64kmh slight incline the other day. The EX20S does have more packs in parallel though so it would take quite a lot of pushing on that wheel and that would only be for a brief moment anyway... like a high speed acceleration from 60 to 70kmh. But sag is a thing at these wheels at high speed.

I have seen high amp overleans on MSP HS at 56A and well below the 5 beeps. On a 24s4p the LG M50T cells simply could not provide the current needed for the donkey acceleration from 45kmh to 55kmh and overlean occurred at 51kmh and 56A... in less than two seconds. 

 

 

[goatman]

higher resistance cells lose mah to heat at higher amp draws

50e or m50lt may have more capacity than a 40T or p42a but are higher resistance

i wasnt surprised to read that the p42a S18 gets the same range as the original higher capacity packs but the volt sag is non existent with the p42a 

[GoGeorgeGo]

 

Did an awesome 12v modification to my EX20s today which came out great. So nice having it be so upgradeable without looking janky like so many other mods ive seen. No big seats, no duct tape. Added a USB port, a beefy headlight, and some nice bright tail lights

[Nick McCutcheon]

2 hours ago, GoGeorgeGo said:

 

Did an awesome 12v modification to my EX20s today which came out great. So nice having it be so upgradeable without looking janky like so many other mods ive seen. No big seats, no duct tape. Added a USB port, a beefy headlight, and some nice bright tail lights

Sweet!! Have you had a chance to test the headlight? I bought a smaller version of that light for the Sherman and it was uhh not great to say the least. Very low white accuracy (strong yellow and blue tinge) and despite appearances not as bright as expected either.

Did you just split off of the 100V line to power the buck converter? 

[chanman]

22 hours ago, techyiam said:

Doesn't the 40T look better at higher discharge rates?

Yes, but 15A per cell is a nutty current to use to represent average riding on these bigger capacity wheels. Maybe for a millisecond if you hit a pothole at 40mph.
5A is probably more typical is and much less clear cut. Also depends on where you cut off the voltage to stop riding quite a bit, extending that a bit lower gets a decent bit more mileage out of the high capacity cells, but better not get into a situation that needs a big spike in current demand if you're doing that.
 

Edited May 7, 2022 by chanman