US69 Posted March 18, 2017 Share Posted March 18, 2017 5 minutes ago, Marty Backe said: The way I read this was, the 84 volt wheels were being addressed by the comment, and there are some MSuper V3s/s+ wheels that have the old style. I assume all the 67v wheels have the potential problem. Yip...and she also states that NO ACM and Monster have this problem....while failures and GW France told another story... I think @houseofjob nailed it: Open and check :-) Link to comment Share on other sites More sharing options...
yourtoys7 Posted March 18, 2017 Share Posted March 18, 2017 Situation like this and lack of communication from company really slows me down in spending $2000. usd. ($2700. cad.) for Msuper V3S + this is not a small pocket change. For now I'm still enjoying my ACM 680. I just never really push it that hard, so for the most part I should..... should be ok Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 18, 2017 Author Share Posted March 18, 2017 No update yet, sorry if you thought otherwise I agree with the discussion here, especially @Xima Lhotz. Unless it's the motor connections, then why was there no warning (for any kind of failure)? Where is that temperature sensor and what's the point of having one if it does not monitor the actual critical part but something else? Might as well monitor the weather in Shenzen then... 1 hour ago, yourtoys7 said: I just never really push it that hard, so for the most part I should..... should be ok That's the million dollar question. Anyways, this is speculating for now. Sorry to keep you in uncertainty. - And Linnea's information can only be interpreted as... alternative facts (there seems to be proof of the opposite). If you don't check, you don't know. Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted March 18, 2017 Share Posted March 18, 2017 Do you think you could carefully open the side panel and take some photos to see whether there is any visible signs of failure? Maybe take a video of the wheel's behaviour before doing so to document the problem. I think we have to pull the zoom out a little and realize that we are all used to North American or European standards where we have safety certifications like UL/CSA/CE to help protect consumers. If you watch that Shenzhen video @esaj posted you can see how quickly products can come to market. The Chinese likely design and engineer to a "safe enough" standard that may work most of the time whereas our standards tend to be more over-engineered for harsh usage / consumer safety. For example, having live voltage at a charge port would probably never fly over here. It's just not safe, but some EUC makers figure put a cap over it, and it should be safe enough rather than adding some simple components to ensure it is safe. I hear that passing CSA/UL approvals can be a lengthy, time consuming process that can cost thousands of dollars and if not approved the manufacturer has to re-engineer their designs adding to delays and cost. How many UL approved hoverboards do you see on the market, and how much do they cost? For such a huge market with tons of cash to be had, very few companies seem to be investing in the time and money to become compliant. Sometimes it is more of a matter of getting the product out there and make money rather than let's test the heck out of it first before releasing. Dealing with the odd problem here or there after the fact can be cheaper. Actuarial science is a real eye opener. Just look at car manufacturers and how they decide whether a recall would be in their best interests. The cost of a few deaths and lawsuits versus the cost of recalling thousands of vehicles does come into play. Considering that our market might not be that large, and taking the failure rate from heavier than your average Asian riders who climb steep mountains you likely get a low number of issues. It probably makes more sense to Gotway to just quietly deal with individual problems than to issue warnings and try to patch all the wheels out there somehow when most wheels likely will be fine. Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 18, 2017 Author Share Posted March 18, 2017 Haha, I know how you feel, it's not like I don't want to pry open the wheel and finally know what's going on! But sorry, no opening before confirmation that the guarantee is unaffected, because otherwise I'd just send it back. I don't want to risk being stuck with a 1600€ purchase just because a few days of impatience. Especially if I don't know whether it's a fixable issue (motor connectors) or something that effectively ruins the entire wheel for me (bought it for the mountains, if it's only safe in flat terrain and on mountains you have Chinese roulette, I'd rather have my money back and buy a better wheel). I don't know why, but I have the feeling it's less bad than we might think. Also please don't assume Ian is slow. Not only is it weekend, the first thing I wrote was for him not to interrupt his weekend, and the "can I open it?" question is hidden at the end after a long problem description, so possibly kind of my fault if he misses it and keeps the harder work for workdays. -- Non-helpful generic quote: "Patience is not the ability to wait, but the ability to keep a good attitude while waiting." And as for cultural difference between Chinese "average case" vs western "worst case" engineering, while you're right, that helps none of us, and complaining won't hurt either so whine away Link to comment Share on other sites More sharing options...
OliverH Posted March 18, 2017 Share Posted March 18, 2017 22 hours ago, esaj said: Don't know about the current situation, but in the past we've had numerous reports of mosfets failures, typically in situations where high currents are demanded or high power is used for a longer while. Likely, the temperature at least plays a role here, but of course we can't be sure. Two ways come to mind how the temperature affects mosfet breakdown are the rising Rds(on) (resistance at full conduction, ie. the power loss over the mosfets goes up, causing it to heat more, causing it to resist more, causing it to loose more power and heat more, ie. a self-destructing chain-reaction) and the lowering Vgs(th). The latter is a more complicated situation, the Vgs(th) is the threshold voltage of the gate-to-source where the mosfet will begin to conduct. Usually this is something like 2...4V, but as the mosfet heats up, it goes down. The failure would then be caused by the lowered threshold voltage keeping the mosfet still conducting due to residual gate-charge (the gate acts similar to a small capacitor) while the other side of the half-bridge opens, causing a bridge shoot through. But it's pretty hard to prove without fairly precise measurements from the half-bridge when the breakdown occurs... Also, we don't know where the board-temperature is measured from. Older generation Gotways had the app-data in similar format as what the MPU-6050 gyro uses, and they also used exactly those gyros, so it's possible they read the temperature from the gyro (which is usually relatively far from the mosfets). Without having access to the source code and doing an examination/ code review we can't tell if the motor driver works perfect with the motor/ uses the correct configuration data. Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted March 18, 2017 Share Posted March 18, 2017 11 hours ago, meepmeepmayer said: Haha, I know how you feel, it's not like I don't want to pry open the wheel and finally know what's going on! But sorry, no opening before confirmation that the guarantee is unaffected, because otherwise I'd just send it back. I don't want to risk being stuck with a 1600€ purchase just because a few days of impatience. ... Non-helpful generic quote: "Patience is not the ability to wait, but the ability to keep a good attitude while waiting." And as for cultural difference between Chinese "average case" vs western "worst case" engineering, while you're right, that helps none of us, and complaining won't hurt either so whine away I guess I'm just an impatient realist. Watching too much YouTube and listening to satanic music will do that to a person. I figure it's like not wanting to open up the hood of a new car when it broke down by the side of the road because one might be afraid of voiding factory warranty. Or not opening up a new desktop computer to add a video card as it could void the warranty. I wonder if you would need to check with Ian before changing a flat tire. Sometimes understanding different viewpoints and realizing how things are done in different parts of the world can make people more tolerant of cultural and manufacturing differences. Setting our expectations and standards too high of products which are mainly targeted elsewhere can sometimes be unrealistic and set ourselves up for disappointment. I'm not saying that people should roll over (literally) and accept product shortcomings, but a little more knowledge of the big picture can help assess things better and put us in the correct mindset. BTW Satan compels you to not wait and to open your wheel now! EDIT: That last sentence was a joke just in case anyone misinterpreted it and didn't follow the music thread. Link to comment Share on other sites More sharing options...
houseofjob Posted March 19, 2017 Share Posted March 19, 2017 So, my Monster 2400WH (first speedyfeetUK batch order, early January 2017 delivery) sports black plastic covered motor connectors, the same as in @Marty Backe's 2 Monster and ACMV2S securing cables videos below: And, in this Korean Naver Blog post (google-translated), an EUC service repairman (one of many; yes, Asian countries like Korea have such a service!), details the history of this particular Gotway motor connector issue (same has been happening in Korea), confirming that these black plastic motor connector soldering jobs are the culprits: The newer connectors are the metal bullet style (I believe; might be wrong) sporting white sleeves, the ones that both Linnea posted here, and the aforementioned Korean repairman posted on his blog (from the most recent ACM+): Additionally, the tech says that, on the problem cases, there wasn't enough solder to bond safely the 5mm motor cable to the connector terminal: .... and further notes that soldering jobs with cold solder at either end can result in the melting and severing of the bond under high currents demanded by the motor, alluding to a possible explanation of the issue at hand. He's also seen the occasional case where the bad soldering job causing the severing issue was at the motor cable-to-control board end. Apparently, with the new ACM+, they've soldered the crap out of both connector and control board ends. While the tech's solution is to just solder the wires directly together, I'm thinking I might do a whole new connector job on my Monster, soldering on these 8mm Gold Bullet Connectors that can handle 300A (I believe the Monster with 84V and 3500W max comes out to around 41A current max). If anyone more knowledgeable than I can refute this logic, please let me know. Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted March 19, 2017 Share Posted March 19, 2017 Those connectors look heavy duty, but how is the wire connected to it? 8 mm diameter sounds like it might be more for heavy gauge wire? Maybe look into a crimped connector where you can pinch the end of the wire and maybe solder at the same time. Perhaps one of those butt connectors with the built in solder and adhesive? I think they should incorporate a header or heavy duty pin connector that's locked (bottom side pin twisted or some sort of board bolt/nut combo) and soldered to which the motor wire with crimped/soldered plug can plug into directly bypassing the need for a midwire connector altogether. Basically ensure it is 100% disconnection impossible without intentionally wanting to pull a connector. If you want to use a non-crimped connector I wonder if silver soldering with a piezo butane torch might do the trick. Too bad the wires aren't a heavier gauge like Euc Extreme mentioned as that likely would eliminate the melting solder issues. It sounds like the wiring is carrying too much current for their small size. There's some good reading here: http://120studio.com/tech/soldering-cables.htm Link to comment Share on other sites More sharing options...
Rehab1 Posted March 19, 2017 Share Posted March 19, 2017 4 hours ago, Hunka Hunka Burning Love said: Too bad the wires aren't a heavier gauge like Euc Extreme mentioned as that likely would eliminate the melting solder issues. It sounds like the wiring is carrying too much current for their small size. The wire size on my ACM 1600 is only 14 gauge which is rated at 20 amps max for short runs! The wires are definitely carrying too much amperage for their size! Since my machine is 90 % disassembled I may upgrade to a 10 gauge wire. Unfortunately the circuit board may not accept the larger diameter wire. @EUC Extreme was this an issue during your modification and if so did you drill a larger hole in your circuit board to accept the wire? @esaj you have a lot of experience drilling holes in circuit boards. Any suggestions? Link to comment Share on other sites More sharing options...
Rehab1 Posted March 19, 2017 Share Posted March 19, 2017 This is something I was never aware of! I need to re-evaluate my soldering techniques. Link to comment Share on other sites More sharing options...
EUC Extreme Posted March 19, 2017 Share Posted March 19, 2017 Yes. I drilled a slightly larger holes. Before connecting. I soldered the cables, the cable thinner with a file right from the tip. Link to comment Share on other sites More sharing options...
Rehab1 Posted March 19, 2017 Share Posted March 19, 2017 3 hours ago, EUC Extreme said: Yes. I drilled a slightly larger holes. Before connecting. I soldered the cables, the cable thinner with a file right from the tip. Perfect! Thanks! Yes I would tin the end of the wires first. Also filing the ends of the wire to a point makes sense! Many people suggest using a Pin Vise for precision drilling into a circuit board. I ordered this from Amazon: https://www.amazon.com/tool-home-Precision-Model-Drill-Twist/dp/B010BV7190/ref=sr_1_1?s=industrial&ie=UTF8&qid=1489945558&sr=8-1&keywords=Pin+handle+drill Next question, have you experimented with MOSFET liquid coolers? Link to comment Share on other sites More sharing options...
EUC Extreme Posted March 19, 2017 Share Posted March 19, 2017 I have not tried. I prefer passive cooling. Because it always works if it is sufficient. All fans and others. In the end, they all break down Link to comment Share on other sites More sharing options...
esaj Posted March 19, 2017 Share Posted March 19, 2017 6 hours ago, Rehab1 said: The wire size on my ACM 1600 is only 14 gauge which is rated at 20 amps max for short runs! The wires are definitely carrying too much amperage for their size! Since my machine is 90 % disassembled I may upgrade to a 10 gauge wire. Unfortunately the circuit board may not accept the larger diameter wire. @EUC Extreme was this an issue during your modification and if so did you drill a larger hole in your circuit board to accept the wire? @esaj you have a lot of experience drilling holes in circuit boards. Any suggestions? Typical PCBs are FR4, an epoxy laminate with glass-fiber reinforcement "mesh", a good metal drill should cut it just like that. I mostly do the drilling on the CNC-machine, but on occasion I've made holes bigger using hand drill too. Nothing really "special" about that, only suggestion I'd have is to check beforehand that there aren't any unrelated traces running right next to the hole you're going to make bigger, so you don't cut any extra traces by accident, although that might be hard/impossible if there's a more or less solid ground plane -layer sandwiched inside the board (assuming multiplayer-board), then something like shining a bright light through the board won't work. But, I don't think they'd place any small signal-traces or such right next to the motor phase-wire holes anyway, as the phase wires themselves likely give out a lot of interference... Link to comment Share on other sites More sharing options...
Marty Backe Posted March 19, 2017 Share Posted March 19, 2017 1 hour ago, EUC Extreme said: I have not tried. I prefer passive cooling. Because it always works if it is sufficient. All fans and others. In the end, they all break down That's why I was disappointed to learn that the Monster has two fans. I hope they don't break. I was riding my Monster in the mountains yesterday (low 25c's temperature) and the internal temperature climbed as high as 65c. I have a feeling the Monster is not going to do so well when I start riding it in the mid-30's. My ACM, with no fans, rarely gets in the 60's even when I'm riding at it's 33c outside. Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 19, 2017 Author Share Posted March 19, 2017 Gotway: "Cable Management... what is that?" Customer: "Well if the cables don't get hot or anything..." -- The board looks ok. The only problem description I can give is... there's a mess of molten cables in the wheel. Need your help here to find out what exactly happened, and some other questions. Finally, pictures: #1 Overview... looks more or less reasonable so far #2 Burn marks on the shell and the respective spot of the electronics #3 Board looks good, no unwanted sights or smells #4 Closer look at the bad spot #5 Motor connectors, and the bad spot top right #6 I don't even know how to describe this... #7 (this forum needs a confused emote) #8 Looks like everything just... melted together? Concerning the last 3 images, the big fat black cable comes right from the motor. At it's end (where the issue is), the 3 fat colored cables and a lot of thinner colored cables (which go to the top of the mainboard) come out and it looks like one of the bigger motor cables melted everything else? Hard to describe for me, I don't know what the other cables do. Questions: 1. Is the yellow battery connector (see pic 1 and 2) the only connection to the battery? Is there any other connector that can be disconnected? Ian wrote if I disconnect the battery, the motor should turn easily again, but it does not. Did I miss any battery connector? Does the short (assuming the melted mess constitutes one) prevent the motor from turning freely? Do you believe the motor is ok? 2. What are the thinner colored cables for? Does it matter? 3. Can you tell if the motor connectors are the good or bad (old) type without removing the black heatshrink around them? (might as well check now) -- Well, if you can help, thanks. I'm a bit perplexed now. Will make more pictures of stuff if anybody wants... Extra detective points go to @Rehab1 because he kind of predicted the issue: the cables themselves are too weak (that's my interpretation for now) Enjoy Link to comment Share on other sites More sharing options...
Hunka Hunka Burning Love Posted March 19, 2017 Share Posted March 19, 2017 Looks like the yellow and green motor wires (thinner ones) coming from the motor melted and shorted together. MOSFETs might be okay? That's a new one. Guess I lost my $20 bucks betting on MOSFETs. No thermal insulation tubing either. I think I now know why they added that. Definitely need thicker gauge wiring to help carry current and keep temps low. I think we can safely say that wires do seem to heat up. Anyone know what is that plastic melted thing near the cable into the motor? Is that a loose pedal magnet? It's weird to me that there are two different gauges of wire on either side of the black motor connectors. You would think both sides should be the same heavy gauge wiring. The black motor connector on the blue wiring also appears to be melting (it appears shinier than the others). One other possibility might be that the current was so great that the wire insulation was not able to shield the electricity between the wires. An arc from wire to wire formed and the resulting short heated and melted the wires up like a spot welder. There's signs of something molten (solder?) in the melted spot where there shouldn't be solder which is weird. Link to comment Share on other sites More sharing options...
US69 Posted March 19, 2017 Share Posted March 19, 2017 So much to Linnea Lins: That is a mosfet problem! So we clearly see that the high amount of amperage melts down the motor connection cables...That looks like you have to change them, from the black connectors as far as possible in wheel input direction...and if the melt is near the wheel input perhpas the complete cables from inside the wheel up to connectors must be changed!! I would definitely ask Ian.....perhaps he will do it hisself, or change the wheel...or whatever..... Did i say somewhere before that even if the black connectors are cold crimped now, a higher resistance could even produce more heat? Looks like a real design problem, with to thin cables for that amperage!!! Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 19, 2017 Author Share Posted March 19, 2017 Don't know if you can see it, the last few cm of the black cable (before it splits into the multiple colored cables) looks a bit messy... not sure if that is also heat damage. Link to comment Share on other sites More sharing options...
US69 Posted March 19, 2017 Share Posted March 19, 2017 7 minutes ago, meepmeepmayer said: Don't know if you can see it, the last few cm of the black cable (before it splits into the multiple colored cables) looks a bit messy... not sure if that is also heat damage. Thats what i meant in my answer...perhaps the complete cable from inside the motor has to be changed...leave that to Ian, if you are not electrician pro! And yes..the white fiber heat insulation cables are missing... Just in that moment here, my whish to buy/have a ACM 1600wh is going to get down :-( Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 19, 2017 Author Share Posted March 19, 2017 I'm definitely going to write Ian (have to get this thing repaired anyways), just wanted to ask here first to help me interpret. Looks like the worst case - the very motor cables are underdesigned and can melt. Link to comment Share on other sites More sharing options...
US69 Posted March 19, 2017 Share Posted March 19, 2017 to your questions/ - motor should be ok! - thin wires are very important!!! this are the sensor infos from the motor to the board!!! - the melted cable shorted the motor, thats why it blocks! - after you photographed everything you can try to seperate the cables. - if everything seperated...the motor should go free again,,,if not...there is still a short somewhere... - motor connectors are the old, but you can only see inside if the connec is crimped or soldered.. - seams the soldered connec has worked as "sollbruchstelle", and after this did not broke...temperature rises up an up and melted everything - conclusion: This can happen To ANY ACM...my 2 cents Link to comment Share on other sites More sharing options...
US69 Posted March 19, 2017 Share Posted March 19, 2017 Holly Cow! It melted the silicon from the magnet....this white thing with the magnet in it...it belongs on the white sidepad! Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 19, 2017 Author Share Posted March 19, 2017 Yea I can understand waiting to buy an ACM now (don't the msupers V3 have the same motor?)... on the other hand, first time this happened, right? Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.