Jump to content


Full Members
  • Content Count

  • Joined

  • Last visited

Everything posted by trevmar

  1. Update: The second board I bought yesterday via Amazon has shipped via DHL 3-5 day. Within a week will be able to assess if it is a lemon or a Swallow :)
  2. Time for an update: EBAY is again advertising the 'Swallowbot' motherboards. For several weeks all the sellers had removed their auctions from Ebay. I received an email from my seller saying that a new stock of boards had arrived at the end of April and they are testing this batch of boards to make sure they work properly (more time-wasting...). I have ordered a second system from a different seller, and also paid for express shipping. hopefully I will have more info soon...
  3. After I mounted a near-UV laser on my 2418 CNC to burn black paint off PCBs, I observed 3 "furrows" being burnt across a 0.1mm gap. I set the Coppercam "hatches" spacing to 0.033mm and got the same pattern - 3 'furrows' across 0.1mm. Now obviously the laser focus is superb, but aside from that, I have seen more precision that you are getting, and I assume it is due to motor mass and inertia. I have just bought a low-vibration 52mm motor with less than 0.1mm runout, and when I get it mounted up I will post my observations here. Thanks for letting us know what you see. Incidentally, it is simple to replace the CNC stepper motors with 0.9 degree units, 400 steps per revolution, and just reprogram the parameters in the Woodpecker GRBL board. This gives double the resolution, something I have only needed on my Z axis. (ps: the laser I used was "GUUQA Blue Purple Laser Module 405nm 500mw Focusable for 3D Printer/CNC Engraving/Laser Engraving 12V" from Amazon) (pps: I odered the mount for the 52mm motor from Thingiverse: https://www.thingiverse.com/thing:2719169 )
  4. I designed a 'Ninebot' battery tester a year or more ago, and documented it on these forums. What I found was that the difference between the packs was the type of cell they used. The lower power uses a cheaper LG cell while the 340 uses the 'top-of-the-line' LG cell. I suspect that electric vehicles, etc, are using up the supply of premium cells and Ninebot can't get enough of them for all its products. Both cells put out the rated 15Amps peak current, and both cells were available in the USA (I put links in the discussion I posted), the only difference was rated capacity. I am not sure of the battery hardware is capable of testing for one or the other, but I doubt it.
  5. Remember that there is the Android utility from @MRN76 that will show the voltage for each cell in your battery before you have to open it up. Very handy! There is a discussion about this on the forums here, I am sure a search will find it :)
  6. Sounds like the electronics does not remain stable with temperature. Perhaps as the battery cools, the voltage being sensed rises (due to temperature dependence of voltage sensor). That would be a pretty easy coding/hardware error to make. Now as to why the new firmware version suffers and the old one doesn't, it is just another sign (to me) that the original guru(s) who write the basic firmware for the MiniPRO have long since found better jobs, and the current team is essentially incompetent. If I had coded that firmware I would not be asking for error-msg-screenshots, I would be replicating the condition you documented, and making sure I understood what could possibly be going wrong...
  7. Jojo, if Ninebot had shut Swallowbot down I am sure that somebody would have told us. I doubt that, as the Belarus dealer is still (apparently) fully operational.
  8. @MRN76 posted info on how to reverse the hall sensor it you want to operate a MiniPRO in reverse. Several weeks ago
  9. I am still being told by the Chinese Ebay seller of the "MiniPRO control board with Swallowbot firmware" that they are waiting for product to arrive from the factory, expected within 2 weeks now. No idea whether they are a fraud, or whether there are multiple things called "Swallowbot," but I will let you know if that channel is real as soon as anything arrives. It will only take 4 days to get here once they hand it to DHL. As far as I can see, it is a reputable seller, whose feedback continues to be nearly perfect. So I haven't given up all hope yet. If it does prove to be @WEagle95 work I don't mind sending him a donation, and @MRN76too, but it would be nice to understand what caused their change of mind...
  10. You let the battery go very flat. There have been many, many, discussions here about your options. Please browse around the forum to find the info you need to hone in on the problem.
  11. I just got the notification from the Ebay supplier shown below. All listings on Ebay have now disappeared except MRN76's original CPU offer.
  12. I am becoming convinced that my MiniPROs will never fly with Swallowbot The Ebay seller said that when my control board arrived from the manufacturer they found it was 'faulty' and are waiting for a replacement to arrive. I will post whenever there is any progress...
  13. The STM32F103 CPU used in the Segway MiniPRO uses RDP1 read-protection, which is easy to crack with a 'DFU voltage injection' attack. MRN76 can tell us if this is an 'official' reseller but I wouldn't be surprised if somebody has cracked the Swallowbot code and marketed it as a product. Or maybe they have some inferior code and are calling that 'Swallowbot'. Who knows? Anyway, summer is coming, and I want to get Swallowbot in my miniPROs. I have not been getting replies to my emails to Ninebot.run, and in any case don't like the risk of having a motherboard lost in transit to Moscow. This outlet is cheaper and Ebay is far less hassle to deal with. I have ordered one of these Ebay boards with shipping via FedEx, and I will report what I see when it arrives next week. At worst, I will have a spare control board I can put Ninebot code into! I will share whatever I find with MRN76 and WEagle. I notice there are other sellers too: https://www.ebay.com/itm/273788875366 and https://www.ebay.com/itm/303114821577?
  14. They suggest you 'hold off' on a repair for a number of weeks??? Did they offer you your money back??
  15. Great... Since leanback can be eliminated I will follow the procedure for doing so in the users manual. Oh, wait, there is nothing about how to control leanback in the user manual, in fact, there isn't really a users manual On the hilly roads near my home, leanback is totally unpredictable. It apparently depends on the ambient temperature, the MiniPRO temperature, the battery charge, and many things not under a riders immediate control. @WEagle95 tells us that a rider is given 'de-merit points' by the firmware which also are used to determine when the machine decides to leanback. As for the MiniPRO being a toy, well I bought mine at a time when it was being offered as a "Personal Transporter without the hazards and limitations of hoverboards". It has a 100Kg load rating (pretty big kid!). To this day I see nothing in Segway advertising about the MiniPRO being for kids only. Sure, Ninebot makes a model which is for kids, but it is extremely cut down from the MiniPRO which we are discussing here...
  16. Leanback is a safety issue, especially on the very steep roads (but within MiniPRO specifications) around my home. I messed up installing the CPU that @MRN76 so kindly supplied, partly because I had soldered these CPUs so many times before that I didn't take enough care to protect the motor driver transistors which immediately blew because I had a couple of shorted CPU pins My mistake. But summer is nearly here, and I am thinking I really need to fix that pushback. Luckily @MRN76 has been keeping in contact, even though the news has not been good (I hope you are healing after that fall, you didn't give us an update when you got out of hospital). For the new members, here is a link that @WEagle95 post last June- describing what he had achieved at that point: https://mimod.ru/en_US/swallowbot-firmware/
  17. Alexander and Alexei have written an app that will allow you to detect a faulty cell in your battery. Run the app and record the output when fully charged, and when discharged >50% and it will be able to give you a pretty good idea if any cells are looking weak
  18. Exactly! Even though I have a ledge at the end of my driveway which always causes problems it is just 2cm high. But there is a slight down-slope from the bitumen getting to it, and a slight up-slope on the other side. In my case the problem is repeatable. We can take bigger bumps (pine cones and tree roots are common around here, for example) but it is this innocuous one which causes trouble... Interestingly, the entrance to our local park is shaped exactly the same, and it also throws me in exactly the same way... But other driveway ramps are not a problem...
  19. I have been suggesting the veering right has all the hallmarks of a rogue CPU interrupt, but I was programming a DC-removal-digital-filter last night and it struck me that the firmware processing the miniPRO accelerometer data may overload at the precise ledge where this problem occurs, and that numeric overflow may be causing the veering (the accelerometer data has to have gravity filtered out of the miniPRO-induced accelerometer fluctuations). Further, any difference in the impact (for example, when we changed tyres) would mitigate the potential for overload. It will be interesting to take a closer look at the problem when the Russian Hack firmware is eventually delivered to us all. If momentary overflow is indeed the cause, I would expect the failure to be not changed much in the hacked firmware... One way or another we need to get this loss-of-control fixed..
  20. Yes, I have both models. There is no difference in the motors part number, but they do paint it in a different part of the motor frame whcih, on first glance, makes them look different. As far as I can see, there is no difference between the 260 and 320 beyond the batteries and some cosmetic changes. Exactly the same firmware works on both platforms. However, the difference in battery quality is huge. The extra capacity of the 320 makes so much difference, especially when you are half-way through your ride :) I am hoping that the hacked firmware will give the 260s a little more range, as the standard firmware goes into a 'power saving' mode once it drops to 60% or so...
  21. Uncommon? It happens on all four of my family's miniPROs, better than one out of 4 times when we run over the ledge at the end of our driveway. I consider that 'replication', and certainly not 'uncommon'. The miniPRO takes larger obstacles OK, and smaller ones, but this particular size (and/or shape) it doesn't like... There is an identical road-edge at the inlet to our local park. It threw me there, too. And the 1" pebble on the well-groomed dirt pathway (in the park) twisted my ankle for several months. I couldn't walk for a month. No, this is a very common problem. I only hope that the MRN76/Angel firmware modifications make it go away...
  22. Did you note that in the MiniPRO specifications that they say the maximum obstacle height is 2cm. Yet it must easily take larger tree roots, and has to be able to recover from larger bumps without throwing the rider. I assume Segway know that this swerving is a real problem that they can't fix, and anybody who complains is just pointed at the legal disclaimer saying that hitting a 2cm bump is riding it beyond its capabilities. By the way. Is the original design team still available to Segway/Ninebot to debug the problem? I heard that the original firmware writers had left the company. Indeed, at one point there was a rumor that the primary firmware developer was a high school student who had written all the key stuff in his holidays. That story is in line with my experience as to how these breakthrough products tend to come together...
  23. All our miniPRO (4 of them) fail in the same way at this ledge, but the heavier the rider the worse the problem. But here is the main reason why I believe it is a firmware issue: When we put knobbly tyres on the miniPRO the problem goes away, it can negotiate the 3cm ledge with knobblies, but not with the stock tyres. Yet the load on the miniPRO motors is higher due to the larger radius of the knobblies, so we are not dealing with a motor-strength issue.. When I put the 85/65-6.5 (soft) tyres on the miniPRO the problem largely disappears, but it still fails occasionally. It may be something to do with impact sensors, but I suspect timing is the culprit. Two unexpected signals are hitting the CPU at once, and it has not been programmed to cope with that occurrence. I have been programming microcomputer firmware since the 2650, 6805 and 8048, and have seen this type of intermittent failure happen many times before.
  24. Scary, but very common in Interrupt-driven firmware designs. I have my driveway bump which allows repeatability, and it is definitely not a limitation of the platform, but an overload warning the software calculates which is incorrect... I am keen to get my hands on a copy of MRN76 & Alex's hacked firmware. Which may well no longer have the problem, if we are lucky (it may be in the 'pushback' or 'battery saver' code somewhere)
  • Create New...