jayjay23 Posted October 11, 2015 Share Posted October 11, 2015 So I finally got everything up to do my write-up of the controller/motor replacement, so if anybody want's to know what this was about, I put some fairly detailed log here: http://www.myewheel.org/index.php5/Controller_replacement_30km/h Link to comment Share on other sites More sharing options...
Colestien Posted October 17, 2015 Share Posted October 17, 2015 Is the final results some thing like.... $350 for a generics, $300 for parts an their shipping = $650 for a wheel that performs like a $1000+ wheel? Or is it better than that? Link to comment Share on other sites More sharing options...
Colestien Posted October 17, 2015 Share Posted October 17, 2015 Sorry, I just reread what I posted. I just jumped to the bottom line. What I should have started with is. That Rocks! you have done something BIG, and with your documentation makes it understandable. Thank you for sharing it. Link to comment Share on other sites More sharing options...
jayjay23 Posted October 17, 2015 Author Share Posted October 17, 2015 The state as of today (as in the blog) with regards to performance and money is as follows (as I'm from germany I use the euros I really paid):299,- Euro Wheel from within germany218,42 Euro for motor and controllermotor 71,10controller 58,50bluetooth module 8,91shipping 79,91Goes at least 25 kph and has all 500W componentsBUTThe wheel today has the original battery of just 130 Wh, so I have had no time yet to do some more testing of the power limits of this system and after all I don't really mind what is the cost of the system, I like to work and understand the things, so for me it has a lot value of knowing how things work, knowing exactly what I have in my hands and finally how to repair it if something is not working.Because next stage will be the real expensive, the battery.I ordered 4 BMS PCBs 25,- Euros each (so I don't know how many I will finally use, 2 or 4) and 64 Samsung cells (with 2,6Ah, and 4,10 Euro each!). So I leave it to the reader what he thinks is necessary to be put into the wheel and what would be the outcome of a comparison. Link to comment Share on other sites More sharing options...
Colestien Posted October 17, 2015 Share Posted October 17, 2015 I'm agreeing with you on all levels. You have taken something, and have made it better. You have taken something and made it yours. The time you put into the documentation show it very well. Of course I'm not a rich man , and when I see numbers that I covert to USD (google) show me better for less, I have to smile. Good work. Link to comment Share on other sites More sharing options...
jayjay23 Posted October 31, 2015 Author Share Posted October 31, 2015 Battery update, so I finally finished my second battery pack (156Wh each) and would like to share a bit of the experiences I made. I switched from the original 132Wh pack to a first single 156Wh and noticed nothing but a increase in range from about 7km to something about 10km, which was somehow cool but of course not a big step. Today I put together the second pack and installed in on top of the existing, using longer screws and some tape to add about a good cm to the cover plate, still it's possible to reach the pedal and ride it very well. Yes it's not professional, I also ordered a case when I try to make up to 4 packs. Now driving with 2 packs I noticed a lot of difference in driving experience. When going with one pack the wheel was always quite unstable at higher speeds of maybe above 16km/h depending on the ground. I noticed this (and thought it was normal) when accelerating not just very very very gently the wheel struggled to keep up with my speed and had to do a notable late acceleration to keep itself under my body. Still there was some margin to fall, but it felt like the wheel struggled and I thought it was me that had to keep the balance better to avoid the instability. Finally after falling I only felt like going 14-20km/h during normal riding to stay more safe. With 2 packs now the sensation is somewhat different, the instability feeling has nearly completely gone, going faster (25km/h) feels much more safe and stable, so I guess that the small battery is just too small if you don't want to compensate the instability with exceptional skill. The overall power is also already quite great going our steepest hill around here I was so far always happy that the wheel got it at all, now it does it with 18km/h and feels like there is still room for more. I will not make a documentation about the battery packs as I think there already quite a few and I also think the soldering I did is not the right thing to do. For others to keep an eye on I would also like to state some mistakes I made on the way: - I thought the batteries are connected only to the PCB, so I ordered them U soldering connectors, but in fact there is snake form along the PCB that is made from cells with Z connectors (or directly welded) (now that is logical), the connectors that go up to the PCB are only for the balancing, nothing more. I now understand OleTC to make a spotwelder, for doing packs on a regular basis and/or in a good manner this is 'the' solution. - I bough the cheapest cells I could find with reasonable capacity (2,6Ah Samsung ICR18650 - 26F), but I missed some of the other important parameters. These cells only have a discharge rate of 2C 5,2A (which would be 312W, quite less), if you now read also that to get really long life out of your cells you should ideally never ever discharge them even for short moments above the rating it's probably not a good sign for long living packs unfortunately. The original 132Wh pack that came with the wheel also has Samsung cells (2,2Ah) but with a discharge rating of 10A (600W), so nearly double. So if I would do another pack with other cells I would go for higher C rates, especially I read somewhere here that some wheel draws up to 5000W peak, this is stress to the cells. Anyway going with 2 or later even 4 packs should make a decent result, but there is room for optimization :-) Link to comment Share on other sites More sharing options...
DebboR Posted October 31, 2015 Share Posted October 31, 2015 Batteries with lower C rating usually have a internal resistance that is a lot higher than cells with a high C rating. Therefore, drawing a big current from the cells will result in a bigger voltage drop. This could be why the wheel seems to 'lack' peak power (the voltage across the motor windings drop, causing it to generate less force). Using two battery packs in parallel causes the current to be split between them, resulting in the voltage drop being halved and thus gaining peak performance (in theory ). I'm also planning to upgrade my generic wheel to 500W with the 30km/h controller, and eventually upgrading the cells in my battery pack if necessary. What are your thoughts on the quality of the motor and controller, and have you been brave enough to test its top speed? Link to comment Share on other sites More sharing options...
jayjay23 Posted October 31, 2015 Author Share Posted October 31, 2015 Yes, I have thought about that too and I thought that the internal resistance of two packs is like two resistors in parallel, whose resistance is half, letting more power through. The question that popped up for me is, is there a low resistance value that will cause too much current to flow? About the quality, hmm, well I have only on sample that I can judge, but there is indeed a quirk that I didn't looked into it any further. The bluetooth stuff is working with a older gotway app (the newer one crashes on my phone), but I only used it change the ride mode once. The outside quality of the components look good, it's up to you whether you put glue everywhere, I've not done it. Now the quirk that I have is that there is a small interruption / vibrating in the rotation occuring every full wheel rotation. To me it doesn't seem to be mechanically so I thought that the controllers software has some problem there which I could not fix easily, but maybe the problem is somewhere else. I have never opened the motor as I don't have the tool for pulling the ball bearing from the axle, so I don't know the inside. The fact that the vibration occurs at every rotation and feels absolutely non mechanically I believe it's the controller software. Electronic problems in the motor can not occur per full wheel rotation as to my understanding as the motor has steps that are much smaller and so do the hall sensors. Anyway during driving the vibration is not very dominant, I actually decided to ignore it. If you hold the wheel in your hands over the ground and accelerate the force is very notable though. The fastest speed I have gone is about 26km/h, so I really don't know how the ground behavior is, when lifted off the ground it goes between 28 and 32 then just switches off, but this does not mean much as it cant do regulation (with the two packs maybe 90% full it goes 30,8). Sometimes, with empty battery going uphill I had the feeling that the controller uses slight padel lifting or breaking to keep me from going fast, but maybe this was just a sensation from the lack of power. In case it just switches off I could only imagine trying it on a small path with meadow left and right using all protections I have from snowboarding and skating together :-) (peants with foam (for falling on your ass when snowboarding, it really hurts if you don't have them and land on ice), back protector and helmet, from skating: knee, elbow and hand protectors). But currently I have no feeling of trying this out :-), btw when I fell I slide on a wet street and as it was wet I finally had no injuries at all (was very lucky this time). Link to comment Share on other sites More sharing options...
DebboR Posted October 31, 2015 Share Posted October 31, 2015 I don't think you have to worry about a too low battery resistance. Even if you got up to an 'ideal' battery with no internal resistance, this just means the voltage on the battery terminals would be constant (at the nominal 60-odd volts) regardless of the current draw. The current would then only be relying on the impedance of the motor, instead of on the sum of the motor impedance and the battery resistance. The motor should be just fine with this. About the vibration issue: if you turn the wheel off and spin the wheel by hand, do you still notice this irregularity? If you do, it should be something mechanical. If there's no problem when spinning it manually, I would also think that it should be a firmware issue (although this would surprise me). I think you're right about concluding that this shouldn't be an electrical problem, as the motor performs multiple 'electrical rotations' in one single mechanical rotation. Have you reattached the buzzer? Might give you a warning when the wheel is about to cut off ;-) Link to comment Share on other sites More sharing options...
jayjay23 Posted October 31, 2015 Author Share Posted October 31, 2015 Probably you are right about that the motor is designed for zero additional resistance. Turning it by hand gives no irregularities. Yes when I came to the firmware conclusion I also thought, that this should be a really strange error. No, I have no buzzer, maybe I can use one from the LiPo testers? Actually I'm unsure about what should be the properties/parameters of a suitable buzzer. Having a buzzer would be nice to be more secure, the controller description said there should be 3 warning levels if I remember correctly (22km/h, 25 and 28?). Link to comment Share on other sites More sharing options...
DebboR Posted October 31, 2015 Share Posted October 31, 2015 Maybe the problem is still mechanical, but only apparent when put under load? Is it still smooth when you push down on your wheel and move it across the floor? Afaik almost every buzzer is interchangeable (they're just little pieces of piezo metal inside a plastic canister). The ones from a lipo tester should be fine. If you still have your old controller, you could desolder and reuse that one? Also, in some old desktop pc's, there used to be a seperate buzzer hooked up to a cable to provide the BIOS beeps. Link to comment Share on other sites More sharing options...
jayjay23 Posted November 1, 2015 Author Share Posted November 1, 2015 Yes cool the buzzer from the LiPo tester work fine! It's just that it is not so loud, when going 20km/h the wind sound starts to raise and you have to turn your head in order to still hear it. Though I put it into the case loose, so I have no clue which is the direction, could be easily optimized too. The quirk is not dependent on load, I did not checked with load but without, if the wheel is lifted up, powered on and you let it go, it has that move too, actually going higher speeds while lifted up your hands start to shake, but going on ground with the body weight gives no notable sensation to me anymore. One feature that this board has, which I have read somewhere on the internet in the specifications of a wheel is auto reset, so when fallen you can turn it to correct upwards posiiton again an it beeps two times letting you know balancing is switched on again. There are four types of beeps, one short, two short, three short (each with a small pause between the group) and short beeps with no pause. Going my favorite steep hill up I noticed that beeping sets in already in lower speeds, e.g. around 13-14km/h instead of 20-22, so it seems that the limits are based on power, maybe it just tells you which percentage of the battery power it currently let's through (via PWM modulation), which would be quite cool as on empty battery the warning would occur also in time. Link to comment Share on other sites More sharing options...
OleTC Posted November 1, 2015 Share Posted November 1, 2015 21 hours ago, jayjay23 said: - I thought the batteries are connected only to the PCB, so I ordered them U soldering connectors, but in fact there is snake form along the PCB that is made from cells with Z connectors (or directly welded) (now that is logical), the connectors that go up to the PCB are only for the balancing, nothing more. I now understand OleTC to make a spotwelder, for doing packs on a regular basis and/or in a good manner this is 'the' solution. So it is. I observed the old packet and saw the welded tabs, connecting 8 cells in a snakeshape. Those are the important connections, together with the tabs, that connect the 2 'snakes' at one end of the BMS. Naturally also the first and last tabs, forming the 0 and +60V of the serialized cells. All other tabs are just for the balancing and won't carry high currents. Unless something is very wrong. They also attach the cells mechanically to the board and should still be of good quality. I see it as very important to get these welds right, together with very good insulation between the 2 rows of cells on the BMS-board. I made the latter out of heatresistant auto cloth-tape and a strip of cardboard, to be sure, that not even a hard bump will be able to short the cell-ends of the 2 rows. A permanent short here, would likely cause a thermal runaway situation. If anyone want to build a batterypack, it is recommended to open and study a good quality one, to understand the details of doing this. In addition to what has been mentioned, I also used heatresistant cardboard gaskets that are sticky on each cell, to insulate the nickeltabs from the plastic underneath. And heat shrink tubing for the Whole pack to keep all parts tight together and insulated from possible outside shorting items. All these items can be found on the Internet auction portals. Building the first batterypack is actually a bit challenging. Link to comment Share on other sites More sharing options...
OleTC Posted November 1, 2015 Share Posted November 1, 2015 16 hours ago, jayjay23 said: Now the quirk that I have is that there is a small interruption / vibrating in the rotation occuring every full wheel rotation. To me it doesn't seem to be mechanically so I thought that the controllers software has some problem there which I could not fix easily, but maybe the problem is somewhere else. I have never opened the motor as I don't have the tool for pulling the ball bearing from the axle, so I don't know the inside. The fact that the vibration occurs at every rotation and feels absolutely non mechanically I believe it's the controller software. Electronic problems in the motor can not occur per full wheel rotation as to my understanding as the motor has steps that are much smaller and so do the hall sensors. Anyway during driving the vibration is not very dominant, I actually decided to ignore it. If you hold the wheel in your hands over the ground and accelerate the force is very notable though. Regarding the vibration, it is not as simple as this, no? Tyre balancing Link to comment Share on other sites More sharing options...
Jason McNeil Posted November 1, 2015 Share Posted November 1, 2015 On 10/31/2015, 3:43:27, DebboR said: Batteries with lower C rating usually have a internal resistance that is a lot higher than cells with a high C rating. Therefore, drawing a big current from the cells will result in a bigger voltage drop. This could be why the wheel seems to 'lack' peak power (the voltage across the motor windings drop, causing it to generate less force). Using two battery packs in parallel causes the current to be split between them, resulting in the voltage drop being halved and thus gaining peak performance (in theory ). I don't think there's any manufacturer who makes eWheels today understands this vital point about Internal Resistance! I've been on a crusade for getting better cells installed in eWheels prone to cut-outs owing to voltage drop, but it's been a real struggle... The crazy part is that in the past year, there is no longer a significant trade-off between capacity & power, a good modern cell can give both. Nor are they more expensive either, the Samsung 25R for instance can be bought at a wholesale level for $2/cell! If you're looking for cheap cells for a 16s pack, go with the Samsung 25Rs, IR = 18mΩ Higher capacity, but also more costly, than either Samsung 30Q or the LG HG2 IR = 18-20mΩ In +64 cell packs for maximum capacity, then either the LG MJ1, MH1 or Samsung 35E are good bets Don't waste your money on old cell technology, there's absolutely no rationale or justification to buying them! Continually amused by those newbies who ask 'where can they buy the cheapest eWheels,' are those weeks in hospital really worth a couple hundred saved for one prone to cut-outs or poor performance. Here's a power graph of a Ninebot E under acceleration (full battery), arrow on the right is when the voltage drops below the cut-off threshold. Link to comment Share on other sites More sharing options...
SlowMo Posted November 2, 2015 Share Posted November 2, 2015 I think it is the same controller that they use for Gotway MCM 14" so better be safe and do not dare to exceed the 30kph limit. Link to comment Share on other sites More sharing options...
jayjay23 Posted November 2, 2015 Author Share Posted November 2, 2015 @OleTC: Yes you are right the two snake' head and tail go up to the board (connection for main current). Hey the idea of the washers is great, the connectors on the plus pole are just the 2/10mm of the outer cell isolation, which is quite to damage, away from the minus casing. So I can put my work here as an example of how NOT to do it: But anyway I will use all the material I have already payed for and I have to repeat that I really like the wheel now with two packs, it's also absolutely forgiving if you miss a pumb and gives you the feeling that it is under control and not letting you fall each moment you go a little faster or don't be 100% concentrated. I made one pack with the lipo testers, but I didn't had suitable connectors so I used DIN9, but they are really ugly big, so my second pack has none (but will get more suitable ones later), when I got the cases for putting all 4 packs to the wheel I will think about adding all the lipo testers also (but with a switch). For isolation between the two cell rows I used the shrink tube I cut from the original pack, it's not too thick and makes a stable impression. As I didn't ordered the big shrink tubes I just put fiber tape around the packs. Ohh and one tip I have also for everybody making packs is, to think about where the wires should go and where it is thus best that they leave the pack. I initially made both wires the same length but that is totally unsuitable for letting the wire leave the pack on one side. If a pack is put into a wheel naturally everything I've seen so far is that the wires have to leave on one corner of the pack, it's also way more easy to integrate the wires into the isolation if the leave on the corner, so most likely you need a shorter and a longer wire for this. Link to comment Share on other sites More sharing options...
jayjay23 Posted November 2, 2015 Author Share Posted November 2, 2015 @OleTC: Regarding the vibration, it is not as simple as this, no? No, I don't think so, it doesn't sound mechanically and it feels actually like there is 1-3 steps of the motor that have less/inversed acceleration, it feels like when breaking down to zero, on low speeds you can feel the rippling of the motor steps, it feels like this, just with higher speeds. I made a video of it (unfortunately it only shows how it looks and sounds, but the sound in the video could easily interpreted as something mechanically) but I was not able to upload it here yet (complained about size although it's less then 3,1MB). OK, now it worked (is there a way, except youtube and other platforms, to integrate the video to be directly playable?) MOV_0094_small.avi Link to comment Share on other sites More sharing options...
electric_vehicle_lover Posted November 3, 2015 Share Posted November 3, 2015 Thanks for sharing all this information. My generic EU is on its way to my home and I will want to upgrade it ;-) Also, the question you have about the wheel, that's another reason why we should make our own firmware ;-) Link to comment Share on other sites More sharing options...
jayjay23 Posted November 5, 2015 Author Share Posted November 5, 2015 On 11/2/2015, 8:46:44, SlowMo said: I think it is the same controller that they use for Gotway MCM 14" so better be safe and do not dare to exceed the 30kph limit. What is the behavior of the Gotway MCM 14? Just shutting down at 30? Link to comment Share on other sites More sharing options...
SlowMo Posted November 5, 2015 Share Posted November 5, 2015 The MCM2s is reported to just shut down with no tilt back. I only gives audible warning beeps. Link to comment Share on other sites More sharing options...
OliverH Posted November 5, 2015 Share Posted November 5, 2015 I heard the same of a friend regarding his MSuper 850 Wh, shutting off if you keep on accelerating not listening to warnings. Link to comment Share on other sites More sharing options...
jayjay23 Posted November 6, 2015 Author Share Posted November 6, 2015 On 11/2/2015, 12:38:17, Jason McNeil said: If you're looking for cheap cells for a 16s pack, go with the Samsung 25Rs, IR = 18mΩ Higher capacity, but also more costly, than either Samsung 30Q or the LG HG2 IR = 18-20mΩ In +64 cell packs for maximum capacity, then either the LG MJ1, MH1 or Samsung 35E are good bets Thanks for all your measurements, I find that really interesting and yes you are probably right from a manufacturers point of view, but unfortunately I will not get those cells for 2$ :-), well anyway thanks for the list, I will consider this topic better when I build the next packs. 21 hours ago, SlowMo said: The MCM2s is reported to just shut down with no tilt back. I only gives audible warning beeps. It's really strange as I have some sort of another perspective on the specs. If somebody tells me the thing goes 30 I assume it's save to go 30, not that this is 0.01km/h from being injured really badly. Well I will be carefully, now with the speaker (which I unfortunately hear only sometimes) I try not to go too fast. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.