How worry are you about your EUC's going up in flames?

[Funky]

1 hour ago, Vanturion said:

So I was thinking about this, how those caps are oriented (presumably 90 degree orientation required for clearance) and very much not ideal long cap solder legs which clearly didn't hold up here. Here's an generic square-wave motor controller from an e-bike for capacitor comparison purposes.

I think in normal electrical design engineering, these legs are kept very short for reasons of efficiency and probably even reliability. Now I'm no electrical engineer, but you I would think there's easy room for improvement to add an adapter with a copper conduction path which would allow these cap legs to be soldered as short as possible in this orientation. Something like this:

Seems like some low-hanging fruit from my layman's perspective. Think it's worth forwarding to the eWheel suggestion box for a better chance to get through to the Begode team?

That would be to expensive... "Wink-Wink"

[0000]

38 minutes ago, Funky said:

That would be to expensive... "Wink-Wink"

Yeah for sure, I've seen enough here to really hammer this point home. In this industry in particular, anywhere a dollar yuan can be saved, even and sometimes it seems especially at the cost of safety and reliability, it absolutely will be done. I'm convinced they must have some kind of official national corner-cutting incentive program not limited to any one company.

Anyway before I go off the rails into the "cultural differences" nonconstructive criticism land, I think I'll forward this to eWheels just to see what they say if they're not too busy to respond.

[0000]

Already heard back from eWheels which was cool.

The feedback was appreciated and it was acknowledged that having good motor controllers go bad because of a poor design choice resulting in a deficient conduction path blows (100% my words there, also pun intended). But. Based on what was said (or wasn't said rather), it's unclear one way or the other if there is even is an open dialogue or feedback loop for design change requests and improvements, even for the largest, or one of the largest, EUC vendors in the Americas.

Who knows, it's not really my business, well, beyond wanting to receive safer and more reliable products in the future. So maybe it is 5% my business, but I don't blame any vendor for not being explicit about whether they can influence overseas product design.

An elegant way to 3D print a small part with a proper conduction path doesn't come to mind, and I'm not really keen on removing my controller for a reliability mod unless I absolutely have to (which would realistically be after a failure). So probably going to leave things off here.

In any case, it sounds like eWheels is excited for the MTen4 controller with TO-247 HY5012 FETs, that was mentioned in the response. So there are improvements being made in the motor controller department. I just can't tell from the top-down screenshot if they've done anything about the less than ideal capacitor conductor path.

Edited September 1, 2022 by Vanturion
english mf, do you speak it?

[0000]

This has still been bouncing around in my head for some reason. Decided to search the topic and came up with this:

https://www.learningaboutelectronics.com/Articles/Why-capacitor-leads-should-be-kept-short

Inductance, knew there was a reason.

Quote

...capacitors lead must be kept short, less than 1.5mm in length, to effectively stop inductive effects, which can limit a capacitor's ability to pass high-frequency signals.

To be fair I don't know enough about the electrical design to say one way or another if that's a problem on EUCs though. Assuming the frequency of the current seen by the capacitors is produced directly from wheel RPM at a 3:1 relationship (3 phase motor), at an upper speed of say 40 mph and given a max tire diameter of say 20 inches, that translates to a little less than a frequency of 33 Hz. Not what I would call a high-frequency signal.

So basically I'm second-guessing the controller design engineers without the electrical design experience or knowledge to back it up because of some of their other choices in general over the years have not exactly been confidence inspiring. Seeing that picture, mechanically-speaking, leaving the capacitor legs didn't look right, but is it truly a reliability problem? I don't know.

If anyone with some electrical design knowledge wants to comment, I'd love an education

Edited September 5, 2022 by Vanturion
motor phases

[Tawpie]

Our wheels are definitely low frequency devices. The 'fast' electronics run just out of audio hearing range, so stray lead inductance isn't an issue at all for the caps. These caps are little energy storage buffers and their exact capabilities can vary quite a bit before they affect anything—the long leads are a non-issue.

But burning through a lead… that usually means the lead was nicked during assembly and was a bit thinner than it was intended to be. The spot where it thinned out is higher resistance than the rest of the lead, so it gets hotter. @Dosingpsychedelics's lead got hot enough to melt and become a fuse. This isn't a fabulous sign, that a nicked lead is enough to cause it to burn through. It signals that the part is pushing a lot of current around and in general, cheap audio grade electrolytic capacitors don't like that much. They can do it (think bass drivers), but it's stressful.

[0000]

2 hours ago, Tawpie said:

Our wheels are definitely low frequency devices. The 'fast' electronics run just out of audio hearing range, so stray lead inductance isn't an issue at all for the caps. These caps are little energy storage buffers and their exact capabilities can vary quite a bit before they affect anything—the long leads are a non-issue.

But burning through a lead… that usually means the lead was nicked during assembly and was a bit thinner than it was intended to be. The spot where it thinned out is higher resistance than the rest of the lead, so it gets hotter. @Dosingpsychedelics's lead got hot enough to melt and become a fuse. This isn't a fabulous sign, that a nicked lead is enough to cause it to burn through. It signals that the part is pushing a lot of current around and in general, cheap audio grade electrolytic capacitors don't like that much. They can do it (think bass drivers), but it's stressful.

Thanks Tawpie for the lesson. So not an inherent design flaw like I thought per say, but the design of using long capacitor legs leaves an opening for assembly defects in the current path. Makes sense.

With my old square-wave Infineon e-bike controller, I ended up blowing one of these caps before at only 45A battery and 90 phase amp limits. That was the picture I showed earlier actually. It still went even without the long legs, so yeah stressed parts for sure.

I think I ended up getting name brand caps rather than the cheapo randoms that came with the controller and didn't have any capacitor related issues since until I upgraded motor controllers. Something to think about, maybe it wouldn't be a bad idea to be proactive about replacing the caps with known-brand ones. IIRC, I had about 2500 miles on that ebike controller before one popped too.

Edited September 4, 2022 by Vanturion

[Tawpie]

12 minutes ago, Vanturion said:

I think I ended up getting name brand caps and didn't have another issue with them rather than the cheapo randoms that came with the controller. Something to think about, maybe it wouldn't be a bad idea to be proactive about replacing the caps with known-brand ones. IIRC, I had about 2500 miles on that ebike controller before one popped too.

Higher quality definitely helps... and if you read the fine print on the "good" ones, there's a little statement about lifetime. It's true, these caps age and go bad. They're filled with liquid electrolyte that eventually dries out and doesn't work as well, which means they can't store as much energy and gradually fail to do their job. Nobody pays attention to it though, for the most part if you're not stressing them, all is good enough.

For best long term results we'd see polypropylene caps in our driver boards, but they're HUGE. And very costly. One of a myriad of tradeoffs made in the design, and every manufacturer has to make the same decisions. The V13 touts the fact that they've chosen to go with many many caps in that design—depending on what they're doing it may ease the burden on an individual part but one never knows until one sees the schematic. Moar is not always better. Often times it is a good move, but not always (the "can 9 women make a baby in 1 month" question comes to mind).

But we humans are simple creatures, and since more is usually what we desire, it helps sell.

Edited September 4, 2022 by Tawpie

[0000]

Wow, you're pretty knowledgeable about this stuff. Thanks for sharing, it's really interesting. I'm always keen to know more about the inner-workings of controllers that make these great hobbies possible. On a totally different subject, 2 more questions came to mind...

2 hours ago, Tawpie said:

"can 9 women make a baby in 1 month"

Do you know if this test is ongoing and are they accepting volunteers?

[Robse]

On 8/19/2022 at 4:06 AM, Glock43x said:

After reading a few threads on here (esp the one regarding storage), it seems like many are worry about it. Assuming batteries are not damaged or wet and not charging, is it possible for it to just randomly spontaneously combust?  I live in an apartment and  I own 2 16S's and 1 16X. I'm a bit paranoid now lol. Kept thinking.....go to work and come back to a burned apartment building. lol.

I am so worried that i ended up building a "external storeage" box for my RS19.  Maybe one more for the 16X and Ninebot.  After all; in my opinion lithium batteries can not be trusted.  

 

 

[Funky]

29 minutes ago, Robse said:

I am so worried that i ended up building a "external storeage" box for my RS19.  Maybe one more for the 16X and Ninebot.  After all; in my opinion lithium batteries can not be trusted.  

 

 

I built also something like that. (Doh in metal cabinet, and wheel drives itself out.) Finished it last week.  I also will post a thread some day.

[mrelwood]

 

11 hours ago, Tawpie said:

"can 9 women make a baby in 1 month"

Yes... on average. 

[0000]

11 hours ago, Robse said:

I am so worried that i ended up building a "external storeage" box for my RS19.  Maybe one more for the 16X and Ninebot.  After all; in my opinion lithium batteries can not be trusted.  

Pretty slick, great the fireproof material was tested first too rather than winging it. We all have different tolerances for risk and mitigation strategies, definitely can't fault a family man for taking the steps to achieve piece of mind. Ideally I'd do something similar, for now I get by with checking voltage often, always attended charging, never charging at temperature extremes, and never storing the packs near full (below 50% if it can be helped).

7 hours ago, mrelwood said:

Yes... on average. 

Ah, math jokes ^{reluctant upvote}

[MrP-MrF]

No begode. So no worried...

[0000]

Lol, well, the truth is all the manufacturers can and do have issues, sometimes very serious issues. Just perusing new wheel posts here, that point becomes abundantly clear. I recently poked fun at Begode with a parody song, but it was mostly tongue-in-cheek. You could've replaced their name with any of the other manufacturers and still capture the element of truth.

In western markets, we've come to expect the best, 100% reliability, 110% performance, and will flash cash and expect quick results. But there is a consistent gap whether by outright failures or other shortcomings with what is expected and, seemingly often, what is delivered, particularly for first batch new wheel owners. On some level you have to applaud the folks willing to put up with buying what sometimes equates to prototypes and basically offer themselves up as "test dummies" as I've seen it put. Without people willing to accept this level of risk, perhaps we wouldn't get the same progression of new wheels. One way to look at it.

One thing I'm fairly certain of is without this kind of overseas manufacturing, low-regulatory grayish market, we probably wouldn't get this type of transportation technology in an affordable consumer product at all for a lot of reasons, none of which too many probably want to dwell on.

It's easy to poke fun from the POV of not having experienced any severe consequences from one of these failures whether by fire or by injury, I've obviously done it, but on the other hand I don't think it hurts anything to repeat on here variations of the same conversations about charging safety, wearing gear, and fire risk because sometimes shit does happen and stuff on the forums and chat gets buried or de-emphasized over time. That said, I think @Robse's got the right idea. If you don't have access to a fireplace or industrial oven for indoor charging and wheel storage no matter the brand, building a fireproof containment area is praise-worthy.

In general, besides for reasons of personal protection, keeping the fire and injury incident count low for as long as possible keeps EUCs away from the line of sight of overbearing bureaucrats and regulators. Even having a plan of action of how to remove (launch) your wheel (with giant CO-triggered slingshot through the window) should the batteries start popping off isn't an outlandish idea (or is it?).

[Ben Hatfield]

felt better since I swapped out the original Gotway/Begode LGM50T packs for the higher output, Litech packs from eWheels.  They provided the swap packs for the MSP I purchased from them, but I did buy new packs for my EX.N, purchased from Free Motion.  

[Circuitmage]

On 8/21/2022 at 4:01 PM, Dosingpsychedelics said:

Definitely losing confidence in quality of euc’s .replaced dc converter at 2 months of buying it new ….now the capacitors blew And scorched the main board at around 4 months …sent ewheels an email just waiting to hear back from them but yes I’m becoming worried that eventually my wheel will burst into flames .

 

So, another thing about that picture, is that I don't see any conformal coating, covers, or tie-down on the caps or board. Another issue with having the caps like that (with relatively thin leads) is vibration and environmental wear. 

The KS boards I have seen have conformal coating on them, which is inspiring.