Jump to content


  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by esaj

  1. I just bought a < 10€ guitar stand from Thomann years ago. Old picture, beautifully covered KS16S in the front
  2. I'm fairly certain that the above is wrong, but if I'm wrong, somebody please do correct me, always happy to learn. In King Songs, the BMS does not "output" voltage towards the charger, there's a reverse polarity protection in the charging side which as a side-effect also prevents reading the battery voltage from the charge port with a multimeter (you might see a few volts as the mosfets leak a tiny bit of current "backwards", but not the actual voltage). On the other hand at least some older Gotways, if not also all the new ones, don't have reverse polarity protection, and you can read the voltage from the charge port. However, the charger does not "use" this voltage reading whether it's there or not, instead, the charger drops its own output voltage if the current limit (like 5A) is hit until the current stays at that limit. It cannot "directly" control the output current, but the control is done through altering the output voltage. This is the CC (Constant Current) -charging phase. Once current stays within this limit when hitting the maximum output voltage, the CV (Constant Voltage) -phase begins, where it just keeps the output voltage at the maximum (like 84V), and the current dies out slowly as the battery charges and the battery voltage goes up towards the charger voltage (the difference between the charger output voltage and battery voltage drops, since the resistance between the two is more or less constant, the current drops: I = U/R). I'd speculate (like others have before) that in this case the faulty charger output voltage has raised high enough to destroy the BMS charging-side protection mosfets. Never seen the KS BMS, but I'd guess they're using something like 100V max mosfets there, and a high enough voltage spike from the charger has fried the protection, allowing the cells to overcharge. They might have gotten severe damage at this point, but not enough to cause a fire. Once the badly damaged cells were charged using the normal charger, an internal short circuit inside a cell or such caused the temperature to raise beyond the critical "thermal runaway" -point (something like 125...150 Celsius), at which point the cell catches fire that keeps feeding itself. But that's all just guesswork.
  3. Rigol makes some relatively cheap oscilloscopes (300-400€ for 2/4 -channel 50MHz bandwidth, hardware unlockable to 100MHz + double memory etc): https://www.batronix.com/shop/oscilloscopes/DSO.html Right now they have a deal where you get all the options at the same price, except the bandwidth? I have the DS1054Z, haven't unlocked all the options (basically they should be bought separately, but there's a key generator online... ), but it's already out of warranty (3 years), so maybe one day if I need more bandwidth or such. To my knowledge, all the 4-channel DS1000's (1054, 1074, 1104) are actually the same hardware, except for the logic analyzer in the Plus-models, otherwise the difference is just the (software limited, unlockable) options. They're not super high quality precision instruments (such cost high 4-figures or 5-figures anyway ), but easily good enough for occasional hobbyist usage and they've got a lot of features for the price. The only downsides I've hit is that the internal voltage noise is relatively high if wanting to measure really low ripples (can't seem to measure reliably below about 0.8mV, which seems to be a lot for an oscilloscope) and the FFT is slow and not very precise, so a separate spectrum analyzer would be needed for such measurements.
  4. Where is the data captured from if you don't have a Charge Doctor, from the wheel telemetry? That's a lot of noise... I've never looked very closely at the wheel voltage measurements, I'd guess that at least in most cases it's just a voltage divider feeding directly into an ADC-pin of the MCU, and if done "poorly" (no filtering, lots of SMPS noise in the system), there can be a lot of noise in the signal. They (likely) just divide the voltage, so the battery voltage range is "mapped" from 0-67.2V to 0...less than 3.3V (to leave headroom in case of overvoltage, so it won't blow the ADC), say 0...3.0V. 12bit ADC can give 4096 different values, but if roughly 10% of the top end is left for handling possible overvoltage and spikes, that leaves around 3600 values for the entire range. 67.2V / 3600 = about 19mV per LSB, which isn't that bad really, should be easily enough resolution for battery voltage measurement. The noise amplitude seems to go up to around +-0.5V from the average, so a total of about 1000mV or about +-50 on the ADC-reading, so likely it's not just "normal" noise of a few LSBs that tends to end up in the measurements unless special care is taken in layout, shutting down peripherals etc. Maybe there's no filtering at the divider at all? If you have an oscilloscope and some courage, you could try to measure the ADC-pin noise when it's not charging and during charging. One possible source might be the charger itself, I don't think that the cheap chargers have much in terms of filtering the output noise, so that might be another place to take a peek at (with an oscilloscope).
  5. There's still ambiguity about the terms, but what I think of as "cut-out/cut-off" was the early wheels that would actually seem to cut power to the motor based on reaching certain speed, regardless of battery state, which is just crazy, or the BMS overcurrent protection killing all power under heavier load. At least to my knowledge, no (brand) wheel produced these days or within the last 2-3 years does those.
  6. Found this, standard sizes here: https://www.woodproducts.fi/content/standard-sizes-thicknesses-widths-and-lengths
  7. Two-by-four is a commonly known term here, but nowadays at least hardware stores seem to indicate the dimensions in millimeters, for example 45x95 or 48x98 are more or less standard sizes, both are close to 2x4", the slimmer one's planed. 48x98mm comes down to about 1.89" x 3.86". Also "even" numbers are common, like 50x100mm, which is actually even closer to 2x4 (1.97 x 3.94"). My guess is that the sizes are based on "traditional" imperial sizes as they seem come very near the whole inch-values.
  8. Muistelisin jonkun ( @EUC Extreme ?) laittaneen kuvia/videota joissa käytettiin hydrauliprässiä akselin vaihtoon joskus aikoja sitten, eli ei ihan helppo homma. Tai ehkä on, jos löytyy sellainen nurkista
  9. @hobby16 has a list of connectors and their polarities in his site ( http://hobby16.neowp.fr/2016/11/20/charger-customization/ ), I guess it's ok to copy-paste them here. Do note that these are from several years ago, and newer wheels might have different pin-out or connectors, so check beforehand to be sure: Most frequently found connectors e-Wheel Socket Wiring Voltage Topo* Remark Most e-wheels GX16-3 1: V+ ; 3: 0V 67.2V 16S connector for Solowheel, Airwheel, Gotway, Firewheel, King Song… Gotway MSuper3 84V/1600Wh GX16-4 1: V+ ; 2: 0V 84V 20S other MSuper3 with battery <1600Wh have GX16-3 connectors IPS (Holtz, Zero…) GX16-4 2: V+ ; 4: 0V 67.2V 16S warning, same connector as above but different wiring! InMotion V8 ** GX12-3 1: V+ ; 2: 0V 84V 20S same connector & wiring as hoverboards InMotion V3,V5 Lenovo Int:V+ ; Ext: 0V 84V 20S Lenovo square socket, originally for Lenovo laptop power supply Ninebot One*** Lemo 63V 15S see photo for wiring*** Ninebot Minipro, Xiaomi ** GX12-4 1: V+ ; 4: 0V 63V 15
  10. I missed all the "fun" of this shit storm... Big thanks to other more active moderators for keeping things running, I guess I'm more or less retired now, haven't been that active for the last... year?
  11. If it's indeed the same device, then it's a good deal. But the datasheet for MP157 does say that: "Avoid the minimum DC voltage below 70V. Low DC input voltage will bring the problem of thermal shutdown." So I don't think they're actually the same device. MP9488 could probably be used without issue in place of MP157 with rectified AC (it can handle the high input voltages), but likely not the other way around, when the input is lower DC voltage. But, of course if it works without issue down to minimum voltages needed in charging, then why not. Can you try to run it for several hours straight at low voltage and see whether it overheats / goes to thermal shutdown? Say, 50V DC?
  12. Nice, I looked at some of the MPS-regulators earlier, but all the suitable models were out of stock and no information when they might become available again (but that was 3-4 months ago ). That's a huge price reduction vs. the Linear Tech-devices. EDIT: Oh wait, that's an AC/DC -regulator, the datasheet suggest rectified 85-265V AC input, what's the lowest DC voltage it still works with? The device would need to work on a 16S or 15S with depleted batteries, so the charger output might be <50V, thus why I was looking for input ranges from about 40...45V (empty 15/16S) or less up to 105V or more (full 24S). I've never seen anything other than 85V AC mentioned as the lowest input on the AC/DC-models, and never tried anything but DC/DC-switchers, if the AC/DC's can work down to 45V DC or so, then there's a lot of options. Although if the minimum RMS AC voltage is 85V, it would mean min. 120V DC, but apparently you're already using it on lower voltages. A quick glance on Electronics Stack Exchange advises against using less than the minimum voltage though, apparently the device can overheat due to undervoltage lockout -problems even if it seems to work fine for a while. This was the MPS-model I was looking at earlier, out of stock and lead-time of 19 weeks: https://www.mouser.fi/ProductDetail/Monolithic-Power-Systems-MPS/MP9488GS-P Works from 7.5V up to 450V DC input.
  13. There are PCB-mounted XT60's/90's available, for example: https://www.tme.eu/fi/en/katalog/dc-power-connectors_112990/?visible_params=2%2C6%2C613%2C422%2C7%2C9%2C1247%2C1322%2C18%2C2555%2C5%2C416%2C77%2C11%2C412%2C2671%2C413%2C419%2C2546%2C424%2C177%2C1427%2C1428%2C68%2C13%2C2467%2C32%2C205%2C1424%2C516%2C646%2C426%2C909%2C82%2C511%2C21%2C1182%2C1335%2C536%2C328%2C329%2C138%2C1323%2C247%2C527%2C49%2C69%2C35%2C1382%2C117%2C10%2C20%2C635%2C1605%2C418&mapped_params=2%3A1260%3B416%3A1641369%2C1641370%2C1641371%3B1322%3A1436472%3B (Filtered by AMASS, XT30+60+90+150, mechanical mounting on PCB) I haven't really worked on this apart from doing basic calculations for the step-down circuitry. Maybe one day, but lately I've been too busy to even think much about this... If the monitor is separately powered, it's relatively simple and cheap to make, but "proper" step-down for the entire voltage range (still about 40-45V to about 105V to account for all 15-24S battery systems at full and empty voltages) seems to add too much to the cost, unless someone has found a good & cheap SMPS-controller? Of course, if limiting to max. 84V systems (20S batteries), there a lot more options.
  14. I had a metal cap with the Firewheel, no issues there, looks like it's similar (rubber inside the cap to prevent short circuits).
  15. Just search for "gx16 cap" in Aliexpress or Amazon or whatever, they're quite cheap, couple of examples: https://www.aliexpress.com/item/32830722831.html https://www.aliexpress.com/item/32820098498.html https://www.amazon.com/GX16-Aviation-dust-cap-Waterproof/dp/B06XFQBXCP
  16. My study's been out of commission for a while, electrics were redone, now sporting 2 x separate 230V/10A feeds dedicated for this room... earlier it was fed through a single feed shared with the room next to it, and turned out I was overburdening the wiring at times Since the power usage has been several kilowatts at times (at worst, something like two desktop computers, 3 monitors, several power supplies and table lighting, CNC-machine and/or 3D-printer, 750W hot air station + in the adjacent room, a freezer and two wheels charging, all running at the same time), surprised that the fuse never blew. Still might take a while to get back up to speed, I had to pack everything up and tear down the tables for the rewiring, slowly building things back up, but it's still a mess: This is actually the first time since 15th of last month that I've even turned on my computers (so almost a full month without a computer while being at home, that must be a record for me ) , also explaining why I haven't been around on the forums much, only occasionally checking the notifications with the phone.
  17. The 170Wh battery of the cheaper model won't carry you for 20km, maybe 15km if you're light weight, and you can't ride as fast towards the end. As a rule of thumb, you can divide the watthours by 15 or 20 to get the actual range in kilometers, it varies depending on your weight, riding style, uphills etc though. And in case you were wondering, buying the cheap model and then upgrading the battery packs to the bigger option later on will likely be more expensive in total than buying the big battery-version directly. The batteries are usually the most expensive part of the entire wheel, except for the very small batteries, like that 170Wh, where the most expensive part is likely the motor. I wouldn't look at anything that's less than 680Wh (for 16S/67.2V) or around 800WH or so (for 20S/84V), especially for the more powerful motors, the small batteries (less packs in parallel) cannot even give out as much power as the motor could use and you'll hit range anxiety sooner or later.
  18. You measure the charger unloaded (or specifically very lightly loaded just by the 10 megaohm or similar input impedance of the meter) to see that the maximum voltage is correct. Under load, it starts to drop the output voltage so that only a specific maximum amount of current (constant current mode) flows to the packs. Once the voltage has risen high enough to reach the maximum, the constant voltage mode starts, where the current will drop as the battery voltage gets closer to the charger voltage, dropping to 0 when the battery has reached the same voltage as the charger (which happens much, much later than when the LED changes from red to green in most chargers). The BMS-comment refers to what the charger voltage should ideally be. The balancing circuitry in the wheels is based just on slowly "bleeding" charge off the cell through a resistor once a certain voltage threshold has been reached (which could be up to 4.2V), so for the balancing to occur, the battery pack must be charged until it reaches high enough voltage for all the cells to "catch up" and charge to the same maximum voltage. For example, if your 20S-battery pack was charged to a voltage of 83V, you still don't know whether all the cells have charged equally (83V / 20 cells = 4.15V per cell), or if 19 cells were at the full 4.2V and the one was at 3.2V, or a mixture of these. The previous is a bit bad example, if there really was one cell that 1V behind all the others, likely that one cell would already have been so degraded that it should be replaced, but even much smaller differences cause some of the cells discharge faster than the others, causing extra stress on them and they will die sooner. But ideally the charger voltage should be high enough to fully charge and balance all the cells if left charging for long enough, which basically should be the amount of cells * 4.2V. Should, but, most wheels have some form of reverse polarity protection in their BMSs (apparently at least Gotways don't, some other older wheels like Firewheels don't). The idea is to protect the battery from a charger with reversed polarity (wrong type of charger) or external short circuit through actual diodes or mosfets. If there are BMSs that use diodes for this protection, they will have a "side effect" that the diodes cause a drop of their forward voltage (around 0.3-0.7V depending on type of diode even at low current) before the cells. Thus if you have a diode dropping 0.7V before the actual cells, and your charger outputs 84V, the batteries won't charge but to around 83.3V. I originally thought that the Firewheel charger was set to 67.8V (for 67.2V wheel) because there are protection diodes dropping around 0.6V in the BMS, but in reality it was just a badly adjusted / drifted charger. For your Gotway you can simply check that the charger's at 84V, since there are no protections. But, in case there were protections and you wouldn't actually know whether there are diodes or mosfets there, it'd probably be better still just to adjust the charger to the typical 4.2V per cell voltage, or at least not much above it. The only reason for higher voltage, while still avoiding overcharging which can cause a fire or explosion at worst, would be to ensure that the cells reach their full voltage when leaving it on charger for balancing to occur. But there you'd be better knowing the protection circuitry, or even better, being able to measure the individual cell voltages, it would be so much easier if the wheels had BMSs that report the individual cell-voltages (or even better, active balancing, but likely that won't be seen).
  19. esaj

    126-Volt Nikola

    I'd at least like to think that the engineers designing the wheels do know what they're doing, but likely it's more a matter of far less stringent safety and quality requirements/culture in general and trying to save on manufacturing costs to keep the wheel prices low(er). Hopefully when pushing the envelope towards something not seen so far they do take their time and test things more carefully. Or just use @Marty Backe as a crash test dummy before real people
  20. esaj

    126-Volt Nikola

    Yeah, it's a small miracle that nobody has (to my knowledge) died so far on a high speed crash... for the motors, it's really hard to say how much can they actually handle. In the e-bike world, people have "overvolted" the stock motors in the past for higher speeds, some could take it, others not. There's quite a lot of metal and mass on the motors, and the metal side covers should conduct heat pretty well, plus there's more or less constant airflow around the covers while riding, so it might be that it works just like it is with higher voltage and still using maximum currents... or not. The motor arcing voltages are likely pretty high, but how high, no idea. Several hundred volts? Less? More? There's more to it than the motor alone. Higher voltage mosfets tend to have higher internal resistance, which likely means more losses (and heat) in the mainboard to get rid of, which may require better cooling. Higher voltage "kick-backs" from the motor. Bigger capacitors for the higher voltages. Beefier step-downs to handle dropping the voltage for the mainboard electronics (12V or so for gate drivers, 5V for USB and such, down to 3.3V for the MCU, IMU and some other parts). Larger BMSs for more cells in series. Maintenance becomes more dangerous the higher the voltages go.
  21. esaj

    126-Volt Nikola

    Another thing that came to mind is (if they're using the same motor), how much power can the motor handle continuously? With higher voltage, you use less current to produce the same power vs. a lower voltage. If the motor can handle the same duty cycles with higher voltage as it did with lower voltage, that means more total output power. If not, and they have to limit the power (average current, lower duty cycle) to keep the motor heating in check, it means that they're trading higher top speed for less torque (less current) for the same output power.
  22. esaj

    126-Volt Nikola

    Slightly off-topic, but wasn't it revealed some time ago that GW's never had any "80%" alarm, just the typical speed- & battery-percentage based alarm that starts to go down once the battery is depleted enough? Can't find the topic right now, but there was a picture with different models, voltages and speed-limits... Edit: Here it is: Official values directly from Gotway: Don't know where the "80%" -thing originated.
  23. esaj

    126-Volt Nikola

    AFAIK, for the exact same motor, and leaving out details like power losses at different currents due to internal losses (they increase in square, that is current to the second power times resistance), yes, I guess you could say that x% higher voltage gives x% higher speed, but in practice, the increase in speed won't likely be linear, but still faster at higher voltage. Of course at some point, the voltage will be high enough to cause a short circuit in the coils, as the wires adjacent to each other will "strike through" the lacquering or whatever it is the coil wiring is covered with and the entire thing shorts and likely melts... well, maybe not melt, the motor will just come to a sudden halt and you break your forehead on the pavement.
  24. esaj

    126-Volt Nikola

    It's a good question, and I have no definite answer... basically both could be thought of as capacity, but the other gives the charge capacity and the other the power capacity, or something along those lines? People are more accustomed to seeing amp hours, because things like tablets and cell-phones use single-cell batteries, so the voltage is always the same, so you can just as well measure amp hours. Same for cars (at least as long as they use nominal 12V lead-acid batteries, for electric cars you of course look at kilowatthours). But when you've got devices using different battery configurations between models (different voltages), the amphours are misleading, because a lower voltage wheel will have a higher nominal amp-hour capacity, and using watt-hours gives you the "reality".
  25. Yeah, saw that. I was actually talking in Discord today about electronics in general with dongie (don't know if that's his forum handle or who he actually is really ), and delved a bit into the dead ACM board Rehab sent me a couple of years back. It just now dawned on me that the weird 140+V @ 75V input (apparently) voltage-doubling node on the step-down is probably a voltage doubler so they can use a cheaper AC/DC -converter there (minimum input 85V) to make it work with the 16/20S packs while saving a bunch vs. using a proper DC/DC -converter... well, if it works...
  • Create New...