THE VIDEO THREAD!

[Aneta]

3 minutes ago, Marty Backe said:

I'm not aware of anyone inspecting their motors. It's clear that the weak link is the MOSFETs.

We've been burning up MOSFET's for years with no evidence of motors being affected. I'm not concerned from a practical perspective.

Interesting. But the phase wires did melt/short, IIRC, on MCM5 on the same hill, in one of your videos? If they can melt inside the case, they can (and probably even easier, since they are inside a 2kW electric oven) melt inside motor, too.

[Marty Backe]

38 minutes ago, Aneta said:

Interesting. But the phase wires did melt/short, IIRC, on MCM5 on the same hill, in one of your videos? If they can melt inside the case, they can (and probably even easier, since they are inside a 2kW electric oven) melt inside motor, too.

You are incorrect. I'm not aware of any MCM5's on Planet Earth that have failed 

A few years ago I had an ACM that had one of the phase wires melt into the hall sensor cable, destroying the hall sensors (and thus the motor). BTW, this occurred on a different hill, before I discovered Overheat Hill.

For over two years now Gotway has been using heavier gauge motor wires. Motor wire melting has not been an issue since.

[Rehab1]

1 hour ago, who_the said:

Yes, it's true...

Great photo! I don’t think @Marty Backe could go much lower given his height. 

[EUC Custom Power-Pads]

9 minutes ago, Rehab1 said:

Great photo! I don’t think @Marty Backe could go much lower given his height. 

Though he can, I can still see 2-3 cm of air (in the tire).

[Mono]

2 hours ago, Aneta said:

There's a possibility that after replacing the controller the undiscovered/ignored damage in the motor can bite some day...

Judging from the frequency of reported burned boards vs broken motors the probability seems to be rather low. Then, the number one reason of (rare) motor failures seems to be related to the hall sensors. Do you have reasons to believe that motor damage is not only possible but somewhat probable?

Edited November 1, 2019 by Mono

[Nic]

I was under the impression all the overheat issues were solved, but this seems not to be the case. Is it really so difficult to properly cool the MOSFETs? Why don't they put a heatpipe in there to suck the heat away to some external cooling fins or a big hunk of metal structure? Maybe too much engineering and so few cases of failure to worry about?

[Lillian]

On 10/25/2019 at 3:42 AM, Moltimondi said:

 

Great video and demonstration of riding on wet pavements.

[Jrkline "Wheel Whisperer"]

1 hour ago, Nic said:

I was under the impression all the overheat issues were solved, but this seems not to be the case. Is it really so difficult to properly cool the MOSFETs? Why don't they put a heatpipe in there to suck the heat away to some external cooling fins or a big hunk of metal structure? Maybe too much engineering and so few cases of failure to worry about?

It isn't an overheat issue,it's an over amp issue.My wheel was modded to be supercooled by me due to my last board failure(double the area space of heat sink plus vents and ducted air fan) and my temp never reached 60C. on that hill and still failed. 

[Mono]

2 hours ago, Jrkline "Wheel Whisperer" said:

It isn't an overheat issue,it's an over amp issue.

How do the amps kill the structure if it is not by heat? And in reverse, how could an EUC overheat without large currents flowing?

Edited November 1, 2019 by Mono

[meepmeepmayer]

Did the motor cables short, or did the boards blow?

The MSX was supposed to be super-robust. How can this happen with just a bit of heavier riders and a quite short if steep hill?

[Jrkline "Wheel Whisperer"]

1 hour ago, Mono said:

How do the amps kill the structure if it is not by heat?

Heat is a product of the resistance to current flow(amps).Too much current flow through the Mosfet literally melts the metal leads of the mosfet(s).So yes,heat does "kill the structure",but the current creates the heat.So too much current = too much heat=melted mosfet=fried board.

[Jrkline "Wheel Whisperer"]

48 minutes ago, meepmeepmayer said:

Did the motor cables short, or did the boards blow?

The MSX was supposed to be super-robust. How can this happen with just a bit of heavier riders and a quite short if steep hill?

One or more of the mosfet leads melted which may or may not fry other board components.Although the MSX mosfets are more robust than previous versions,they still have a fail point which was exceeded.Had I or Frank been on smaller diameter wheels such as a Nikola,we might have missed the fail point and made it up the hill.

 

[erk1024]

1 hour ago, Jrkline "Wheel Whisperer" said:

Had I or Frank been on smaller diameter wheels such as a Nikola

I think we also have to distinguish between the Nikola and the Nikola+ because the Plus has the control board mounted on the side, and the heatsink is not exposed outside the shell to take advantage of airflow. (I don't know if the 84v version is or not). 

In any case, hope you guys get your wheels fixed soon! 

Edited November 1, 2019 by erk1024

[Aneta]

7 hours ago, Mono said:

Judging from the frequency of reported burned boards vs broken motors the probability seems to be rather low. Then, the number one reason of (rare) motor failures seems to be related to the hall sensors. Do you have reasons to believe that motor damage is not only possible but somewhat probable?

Here's a simulation showing that on a 30% hill the motor will generate almost 2kW of heat and overheat over 150C in just a minute or so:

https://www.ebikes.ca/tools/simulator.html?batt=cust_72_0.2_14&cont=cust_100_150_0.03_V&wheel=18i&mass=140&hp=0&blue=Lbs&autothrot=true&throt=50.4&frame=cust_1_0.02&grade=30

Also, I've burned a 1200W motor on my e-scooter, with no damage to its 25A controller. The soldering between phase wires and windings melted. The windings became dark brown from extreme heat, too. So if the EUC controller has burnt from an overload, I'd definitely inspect the motor, too.

[Jrkline "Wheel Whisperer"]

48 minutes ago, erk1024 said:

I think we also have to distinguish between the Nikola and the Nikola+ because the Plus has the control board mounted on the side, and the heatsink is not exposed outside the shell to take advantage of airflow. (I don't know if the 84v version is or not). 

In any case, hope you guys get your wheels fixed soon! 

Airflow is of no consequence.The only solution would be to design a board with components with greater ampacity ratings.I can't fault Gotway.I push the upper tolerances when I climb steep hills and so does everyone who rides like that.Jason is already shipping new board!

[Nic]

5 minutes ago, Jrkline "Wheel Whisperer" said:

Airflow is of no consequence.The only solution would be to design a board with components with greater ampacity ratings.I can't fault Gotway.I push the upper tolerances when I climb steep hills and so does everyone who rides like that.Jason is already shipping new board!

I would blame Gotway ... the wheel should cut out before any failure occurs. They obviously haven't accounted for abnormal situations.

[ZenRyder]

34 minutes ago, Jrkline "Wheel Whisperer" said:

Airflow is of no consequence.The only solution would be to design a board with components with greater ampacity ratings.I can't fault Gotway.I push the upper tolerances when I climb steep hills and so does everyone who rides like that.Jason is already shipping new board!

The question is would the 100v MSX perform better? The higher voltage should result in less amps to put out the same amount (watts) of power. Lower amps also results in better efficiencies and less current loss due to those inefficiencies. Any 200 lb rider ready to take on Overheat hill with a 100v MSX?

[Aneta]

1 hour ago, ZenRyder said:

The question is would the 100v MSX perform better? The higher voltage should result in less amps to put out the same amount (watts) of power. Lower amps also results in better efficiencies and less current loss due to those inefficiencies. Any 200 lb rider ready to take on Overheat hill with a 100v MSX?

For a given % grade of incline and speed (read: air drag), the current must be the same on any particular motor regardless of system voltage, because thrust (read: torque) is determined by current, not voltage. Is this correct?

Where 100V may have an advantage over 84V is a safety margin in maximum torque at given speed, because the back EMF is set by speed, so voltage applied to windings will be higher, and thus max phase current will be higher. So, where 84V may overlean when bumping in a rock on steep incline, 100V will plow throw using its additional torque.

[ZenRyder]

2 minutes ago, Aneta said:

For a given % grade of incline and speed (read: air drag), the current must be the same on any particular motor regardless of system voltage, because thrust (read: torque) is determined by current, not voltage. Is this correct?

Um, I don't think so. Given that all other things being the same (weight, incline, speed, drag, and motor) are the same. If the voltage (electrical pressure in a water analogy) is 100v instead of 84v and the amount of energy/power to move the wheel forward via (thrust/torque/watts) is the same, then amps (electrical flow) going through the wires/mosfets/motor are definitely less. It should run cooler and more efficiently.

This is one reason why you run a pool motor at 220v instead of 120v, because the higher voltage doesn't use as many amps and it's more efficient.

Now if the 84v MSX had to lean further than a 100v to get enough torque for the hill, then the over-lean could be the cause of exceeding the ratings. Above I said, "if all things being the same". But in fact, you would probably have to lean further on the 84v to get enough torque to keep moving up the hill. Whereas the 100v would be able to provide more without going as high on the current. 

10 minutes ago, Aneta said:

Where 100V may have an advantage over 84V is a safety margin in maximum torque at given speed, because the back EMF is set by speed, so voltage applied to windings will be higher, and thus max phase current will be higher. So, where 84V may overlean when bumping in a rock on steep incline, 100V will plow throw using its additional torque.

Agreed, in theory the additional voltage can give more torque and watts of power with the same amount of amperage. This is why power tools have moved from 12v into the 18v range. More torque/power/efficiency from a similar sized battery. They even have some now using 24v.

[Aneta]

22 minutes ago, ZenRyder said:

Um, I don't think so. Given that all other things being the same (weight, incline, speed, drag, and motor) are the same. If the voltage (electrical pressure in a water analogy) is 100v instead of 84v and the amount of energy/power to move the wheel forward via (thrust/torque/watts) is the same, then amps (electrical flow) going through the wires/mosfets/motor are definitely less. It should run cooler and more efficiently.

This is one reason why you run a pool motor at 220v instead of 120v, because the higher voltage doesn't use as many amps and it's more efficient.

Now if the 84v MSX had to lean further than a 100v to get enough torque for the hill, then the over-lean could be the cause of exceeding the ratings. Above I said, "if all things being the same". But in fact, you would probably have to lean further on the 84v to get enough torque to keep moving up the hill. Whereas the 100v would be able to provide more without going as high on the current. 

Agreed, in theory the additional voltage can give more torque and watts of power with the same amount of amperage. This is why power tools have moved from 12v into the 18v range. More torque/power/efficiency from a similar sized battery. They even have some now using 24v.

Isn't it the phase current, not battery current, that burns the MOSFETs (and browns the windings, melts the solder and cooks Halls inside motor)? For 100V vs 84V for the same mechanical load, the battery current will be lower, but the phase current will be the same. Compare:

84V - battery current 40A, phase 106A

https://www.ebikes.ca/tools/simulator.html?batt=cust_84_0.2_14&cont=cust_100_150_0.03_V&wheel=18i&mass=140&hp=0&blue=Lbs&autothrot=true&throt=41.2&frame=cust_1_0.02&grade=30

100.8V - battery 32A, phase 106A

https://www.ebikes.ca/tools/simulator.html?batt=cust_100.8_0.2_14&cont=cust_100_150_0.03_V&wheel=18i&mass=140&hp=0&blue=Lbs&autothrot=true&throt=33.2&frame=cust_1_0.02&grade=30

Heat generated in both cases is 2kW.

Edited November 1, 2019 by Aneta

[xorbe]

46 minutes ago, ZenRyder said:

This is one reason why you run a pool motor at 220v instead of 120v, because the higher voltage doesn't use as many amps and it's more efficient.

What you're overlooking is that the pool motor is hard wired to a constant 220, but the EUC motor is getting a variable amount of energy through the controller.  With the same motor magnets/windings, 84 vs 100 will be identical until you eclipse what 84 can reach with the 100 setup.

There will be less current from the battery for the same motor load.  I'm a little fuzzy on the mosfets, as I believe they are high freq pwm pulsing.  I think at the end of the day, they are more beholden to power delivered to the motor load, not what's happening on the battery source side (because basically they have an in/out efficiency factor).

Edited November 1, 2019 by xorbe

[Mike Sacristan]

Monika chose the MSX today because she just wanted to be free and ride.
I took the opportunity to ride the 16X and talk a bit about the CX tyre.
Here is a teaser for now while the full version is processing.

 

Edited November 1, 2019 by Mike Sacristan
spelling

[ZenRyder]

22 minutes ago, Aneta said:

Isn't it the phase current, not battery current, that burns the MOSFETs (and browns the windings, melts the solder and cooks Halls inside motor)? For 100V vs 84V for the same mechanical load, the battery current will be lower, but the phase current will be the same. Compare:

84V - battery current 40A, phase 106A

https://www.ebikes.ca/tools/simulator.html?batt=cust_84_0.2_14&cont=cust_100_150_0.03_V&wheel=18i&mass=140&hp=0&blue=Lbs&autothrot=true&throt=41.2&frame=cust_1_0.02&grade=30

100.8V - battery 32A, phase 106A

https://www.ebikes.ca/tools/simulator.html?batt=cust_100.8_0.2_14&cont=cust_100_150_0.03_V&wheel=18i&mass=140&hp=0&blue=Lbs&autothrot=true&throt=33.2&frame=cust_1_0.02&grade=30

Heat generated in both cases is 2kW.

If the current is increased by the circuitry on the phase side, then I agree that could limit the advantages of the lower amps on the battery side. 

But even on the simulator graphs, if you change vertical black dashed line (I think it's the speed in kph) on both graphs to 2.8 kph, then the 84v overheats in 2 minutes, but the 100v overheats in 8 minutes. Now maybe I'm not using the simulation tool correctly. But doesn't that prove that the 100v would work better than the 84v in this particular situation? 

Well, I'm not sure because I think using 30% as the grade number is not very close. I doubt the actual grade of Overheat Hill was over 20%. Roads in the US are generally never more than 10% and I measured a fire road that looked steeper than the grade in the video and it was 17%. I was originally thinking the 309 lbs was probably high too, but when you factor in the weight of the rider, gear, AND the 50 lb. wheel, maybe that is pretty accurate. But when I changed the numbers to have less grade 25%, the output didn't change the overheat time in some cases, but it did in others. So, my confidence in the tool is not very great. 

 

[Aneta]

15 minutes ago, ZenRyder said:

If the current is increased by the circuitry on the phase side, then I agree that could limit the advantages of the lower amps on the battery side. 

But even on the simulator graphs, if you change vertical black dashed line (I think it's the speed in kph) on both graphs to 2.8 kph, then the 84v overheats in 2 minutes, but the 100v overheats in 8 minutes. Now maybe I'm not using the simulation tool correctly. But doesn't that prove that the 100v would work better than the 84v in this particular situation? 

Well, I'm not sure because I think using 30% as the grade number is not very close. I doubt the actual grade of Overheat Hill was over 20%. Roads in the US are generally never more than 10% and I measured a fire road that looked steeper than the grade in the video and it was 17%. I was originally thinking the 309 lbs was probably high too, but when you factor in the weight of the rider, gear, AND the 50 lb. wheel, maybe that is pretty accurate. But when I changed the numbers to have less grade 25%, the output didn't change the overheat time in some cases, but it did in others. So, my confidence in the tool is not very great. 

 

The Motor Simulator gets very inaccurate at very low speeds, it must be a calculation implementation issue, not a real issue. For example, at 2.8kph, thrust is shown as 85lbs in the 84V version, and 45lbs in the 100V version. Which can't both be true, since nothing else (weight, coefficients of rolling and air resistance, wheel diameter) has changed! The load must be the same, but Simulator shows it different. I did notice that the Simulator is unreliable at very low speeds. But starting from ~10% of motor's no-load speed, the Motor Simulator is very reliable and has been verified by real life examples many times (moreover, the motor data actually comes from their test bench as described on the same page below the graphs).

@Marty Backe, what's the % grade at the point of TOH (The Overheat Hill) where both MSX's fried?

The steepest paved roads in San-Francisco are in 37-41% range (http://www.datapointed.net/2010/02/more-steeps-of-san-francisco/), so 30% for a dirt road/trail is not unheard of.

[Marty Backe]

58 minutes ago, Aneta said:

The Motor Simulator gets very inaccurate at very low speeds, it must be a calculation implementation issue, not a real issue. For example, at 2.8kph, thrust is shown as 85lbs in the 84V version, and 45lbs in the 100V version. Which can't both be true, since nothing else (weight, coefficients of rolling and air resistance, wheel diameter) has changed! The load must be the same, but Simulator shows it different. I did notice that the Simulator is unreliable at very low speeds. But starting from ~10% of motor's no-load speed, the Motor Simulator is very reliable and has been verified by real life examples many times (moreover, the motor data actually comes from their test bench as described on the same page below the graphs).

@Marty Backe, what's the % grade at the point of TOH (The Overheat Hill) where both MSX's fried?

The steepest paved roads in San-Francisco are in 37-41% range (http://www.datapointed.net/2010/02/more-steeps-of-san-francisco/), so 30% for a dirt road/trail is not unheard of.

I don't know. All I can give you is the Google Pin for the bottom of the hill. You can look at it in 3D. It's not super-steep, but it's very long.

https://goo.gl/maps/6oLB3vYjAXKH3N718