Reversing MiniPro Steering sensor

[RooEUC]

10 hours ago, Vik's said:

Our oppo at russian forums did a wonderful job at connecting jtag interface to ninebot (both One and minipro), to change the serial number making effectively a P model out of E+

Link to original content: http://electrotransport.ru/ussr/index.php?msg=1042396

No modifications is done to the firmware itself, only the "user information" is being written to the EUC's memory - milage, serialnumber etc.

 

On top of that he made it possible to change fw on 9b1 though the original app:

 

 

Thanks for this. It's a lot more involved than I'm ready to get yet but I find it very interesting.

I wish we had an elite level coder among us who could edit the MiniPro firmware and get 22kph. I bet there are a few of us here who would happily spread the cost between us if we could find a 'coder for hire' for such a job.

Edited May 2, 2017 by RooMiniPro

[FreeRide]

I would certainly contribute to any method to get between 20-25 km/h out of the Mini Pro.

Anyone else tried this swap or reversing of the magnets?  

Another way to reverse the steering would be a very simple electrical circuit this would ensure that the center and calibration stayed the same and would be slightly less work to insert it inline between the sensor board than it would be to change the magnets which seems to change the calibration either due to slight differences in the magnets, the symmetry of the enclosure, the precision in their placement, etc.  

It seems that if only the reversing of the steering is needed for a good speed mod, the electronics approach might be best until they crack the firmware and loader.  I tried going backwards just to see if there was any issue with the position of the bar and I didn't find any.  It seemed to me that if the steering was properly reversed it would be perfect.  Of course I didn't test the other safety features such as alarms and tilt-back if riding backward.  I believe though that tilt-back is in fact more dangerous than the problem it is trying to solve in many cases.

I just might have to buy an extra one to have it for experimenting as mastering the EUC is turning out to be challenging for me, and some things like using binoculars and cameras are easier from the MiniPro since it is so easy to stand still and made minute movements.

Edited May 12, 2017 by FreeRide
typos

[Pezman]

You know all of this would be so much better if the Mini Pro just didn't tilt back at all. If it just got up to 17.5kmh and stayed at that speed rather than tilting back it would be great. The tilt back slows you down so much and then you need to get back up to speed again only to get slowed down again.

[RooEUC]

3 hours ago, Pezman said:

You know all of this would be so much better if the Mini Pro just didn't tilt back at all. If it just got up to 17.5kmh and stayed at that speed rather than tilting back it would be great. The tilt back slows you down so much and then you need to get back up to speed again only to get slowed down again.

All self-balancing vehicles like these must have tilt-back. It's a fundamental part of how the self-balancing technology works. There is no other way to slow the vehicle down once the top speed is reached. Without tilt-back to reduce speed, the motors would continue to speed up until they reach their maximum output, and with no power left over to keep self-balancing, the rider would be dropped forward and fall off. 

These machines keep us upright by increasing speed in order to stay under us. Imagine running on a big metal ball the size of an elephant. As you stand towards the front it will begin to roll forward. You keep your body and the centre of gravity closer to the front to keep rolling forward.  In order to stay on the ball you have to walk backwards, keeping that COG towards the front.  As the ball increases in speed you have to walk faster. Eventually the speed will become too fast for you to run and you must either slow the ball down by moving yourself back far enough on the ball that the centre of gravity is behind the ball (sort of what tilt-back does) and you exert a force that wants to roll the ball backwards to slow it down, or you must jump off and crash.

 

Edited May 12, 2017 by RooMiniPro

[Pezman]

Wow, good to know. 

 

[kasenutty]

I'm with all you guys, tiltback must be stopped. I think it should be changed to forward tilt like my EUC. Wanna go fast? Fly away, mofo

[RooEUC]

 @kasenutty, you find a way to make a joke in almost every post.  How did you get to be so cheerful?  Do you smoke that much weed?

[kasenutty]

Yes, I do. Naw. Hmm. I dunno. I smoke about 2-3 grams of BHO shatter per week. 

 

I'm not always happy, but I mostly only messageboard when I'm in a good mood. I've also been on messageboards a long time. You have to be all nice at first for a while before you start calling people names and insulting their mothers. I think I have about 5-6 weeks left to go before I go crazy

[jojo33]

3 hours ago, kasenutty said:

Yes, I do. Naw. Hmm. I dunno. I smoke about 2-3 grams of BHO shatter per week. 

 

I'm not always happy, but I mostly only messageboard when I'm in a good mood. I've also been on messageboards a long time. You have to be all nice at first for a while before you start calling people names and insulting their mothers. I think I have about 5-6 weeks left to go before I go crazy

It's like french administrative agents : they are odious really they talk to you nastily because they dont f**** last night...when one is smiliing to you It's ok you can smile too. .. In France we said "he's pleasant like a prison'doors. .."

Forum is the same...If you are in chat with many frustrating people or yourself if you are very Hardly frustrated ...begin the war. ...

 

 

 

Edited May 12, 2017 by jojo33

[RooEUC]

I've thought about this reverse mod and for the work involved I don't think that the result is worth it.  No one has done it and commented that it's worth keeping in the reverse state.  I took my EUC out for a ride today in strong winds and then the MiniPro right after.  The MiniPro was much more stable and safe in the wind and gave me bonuses like being able to stop without getting off and without pain on my legs but the speed is too low.  We have to figure out a way to increase the speed on this thing.  I know with my weight the MiniPro is more than capable of running at 23kph and still have plenty of power left as a safety margin.  I would chip in money towards paying a genius to modify the firmware if it can be done.

[FreeRide]

51 minutes ago, RooMiniPro said:

I've thought about this reverse mod and for the work involved I don't think that the result is worth it.  No one has done it and commented that it's worth keeping in the reverse state.  I took my EUC out for a ride today in strong winds and then the MiniPro right after.  The MiniPro was much more stable and safe in the wind and gave me bonuses like being able to stop without getting off and without pain on my legs but the speed is too low.  We have to figure out a way to increase the speed on this thing.  I know with my weight the MiniPro is more than capable of running at 23kph and still have plenty of power left as a safety margin.  I would chip in money towards paying a genius to modify the firmware if it can be done.

Running backwards is all about increasing the speed, the amount of increase definitely seems worth it.  

There have been very few reports, and the one guy that did it with the magnets later found out he had made a mistake and the housing for the magnets was not fully secure.  He was also pretty vague about why he didn't like it, but then he discovered part of the problem was his mistake.  I have not yet seen him report that he tried it again, and maybe he just got nerves.

I have not yet seen someone reverse the steering yet with electronics either.  I think I'm buying a second machine just to try this mod (meaning backward) on it before it is too late.  Have not decided yet if I will just try the magnet change or will try other options.  It could be this glitch in reverse exists because there is a secret to switch the steering for those developers that like top ride faster.  Also in case anyone does break the firmware, it would be nice to have an extra machine to experiment with.  

If it went faster I'd be much happier, and then I might just carry an extra battery in a knapsack for a field swap when needed.

[RooEUC]

3 hours ago, FreeRide said:

If it went faster I'd be much happier, and then I might just carry an extra battery in a knapsack for a field swap when needed.

Yes I agree, it would be a little dream machine if it could do 25kph.  I am considering doing the magnet reversal, just to see better how this works and if it's a viable option or not.

[donaldduck8]

14 hours ago, FreeRide said:

He was also pretty vague about why he didn't like it

 

Here in Germany it is only permitted with 18KmH with handlebar, registration and insurance to drive.
With this change, however, it is only possible to drive with the kneebar, in addition, the insurance expires upon manipulation of the speed (> 18KmH).
The police pulls ninebots, which only go with kneebar immediately from the traffic.

14 hours ago, FreeRide said:

I have not yet seen him report that he tried it again, and maybe he just got nerves.

I have it back because I drive with the handlebars much safer and I do not want to do anything illegal!

In other countries, of course, the legal issue may be different.

11 hours ago, RooMiniPro said:

I am considering doing the magnet reversal, just to see better how this works and if it's a viable option or not.

Switching the two magnets is really easy.

 

Greetings Donald

[FreeRide]

Thanks for the update!  Ride Safe!

[RooEUC]

I did the magnet reversal and took my MiniPro out for a reverse test ride and here are my findings:

The mod works to an extent.  I took it up to 20kph but this was as fast as I felt safe going due to the issues with steering which I will get to.  At 20kph there was no tilt-back so I believe the 25kph top speed to be legitimate.  I can see the increased usefulness when being able to go at this speed and I would ride mine more if I could use it this way all the time. 

However, the steering needs some work.  When stationary, the MiniPro is always slowly turning left.  If left alone it will do a full rotation on its own.  So when driving it, I have to always be giving some right steering input to go straight.  Not only that, but left steering input is more sensitive and turns faster than right, so it is hard to get used to how much input to give for each turn.  The steering just generally feels weird and unpredictable.  It does not feel safe.  This is with auto steering sensitivity turned off and set to the lowest setting.

The knee bar is slightly farther back but still comfortable enough to use.  Everything else feels normal.  

Edited May 23, 2017 by RooMiniPro

[kasenutty]

You're a nut, Roo. Thanks for posting your findings

[ozzy]

39 minutes ago, RooMiniPro said:

I did the magnet reversal and took my MiniPro out for a reverse test ride and here are my findings:

The mod works to an extent.  I took it up to 20kph but this was as fast as I felt safe going due to the issues with steering which I will get to.  At 20kph there was no tilt-back so I believe the 25kph top speed to be legitimate.  I can see the increased usefulness when being able to go at this speed and I would ride mine more if I could use it this way all the time. 

However, the steering needs some work.  When stationary, the MiniPro is always slowly turning left.  If left alone it will do a full rotation on its own.  So when driving it, I have to always be giving some right steering input to go straight.  Not only that, but left steering input is more sensitive and turns faster than right, so it is hard to get used to how much input to give for each turn.  The steering just generally feels weird and unpredictable.  It does not feel safe.  This is with auto steering sensitivity turned off and set to the lowest setting.

The knee bar is slightly farther back but still comfortable enough to use.  Everything else feels normal.  

good job .Do you have any video?

[RooEUC]

52 minutes ago, ozzy said:

good job .Do you have any video?

No.  There's nothing to see anyway. You would not really notice a difference between me doing 17.5 and 20. The only interesting part would be my face as I'm thinking "holy @#%€ the steering is so f@$#%ed I'm going to die. [Bump] Wooahhhhh"

[RooEUC]

Today I took the MiniPro apart again to try to resolve the steering issue.  I used the tip on the German forum and loosened the hall sensor board, rotated it slightly and tightened it back in place.  After this the MiniPro stayed almost still, just barely rotating to the left ever so slightly.  Still though, the steering does not feel right at speed.  So I have put everything back they way it should go.  Unless someone comes up with another way to increase the speed I will just stick with it in stock form.  What I did find from my tests was that a constant tiltback-free cruising speed of 20kph makes a world of difference over 17 (17 is the real life speed I can cruise at without dipping into tilt-back at all) and I think most of us would be really happy with this.  If tilt-back kicked in at 21kph we could cruise at 20 all day long.  I ride my EUC at 20-21 (tilt-back begins at about 22) and it feels fast enough to get around without feeling so fast that there's fear of being mangled in a cut-out situation. 

The next question I have is: Can the hall sensor polarity be reversed?  If so, the magnets would stay in their stock position and the wires would just be re-soldered and hopefully the steering would remain perfectly straight.  But I believe that hall sensors like this fail when the polarity is reversed or they have reverse polarity protection but do not function when reversed.  I need some confirmation on this though.

If that's not possible then the next step would be a firmware hack.  And that's something I don't know how to do.   

Edited May 24, 2017 by RooMiniPro

[Keith]

On 01/05/2017 at 10:42 PM, kasenutty said:

 

Roo's Car

Obviously it's all done with smoke and mirrors!

[donaldduck8]

23 hours ago, RooMiniPro said:

  I used the tip on the German forum and loosened the hall sensor board, rotated it slightly and tightened it back in place.  After this the MiniPro stayed almost still, just barely rotating to the left ever so slightly.

You have to rotate the board a little further to the right place, you have to enlarge the holes with a file something

 

 

Greetings Donald

[RooEUC]

35 minutes ago, donaldduck8 said:

You have to rotate the board a little further to the right place, you have to enlarge the holes with a file something

 

 

Greetings Donald

I don't want to permanently modify the board. Even if I do, the steering does not behave the way I like it to.

[FreeRide]

12 hours ago, RooMiniPro said:

I don't want to permanently modify the board. Even if I do, the steering does not behave the way I like it to.

This is important, since the faster you go the more critical it is that the steering have the right response.

Still waiting for my parts to ship, but I'm still going to try this first with an attempted at a calibrated precise adjustment to the magnetic forces, and if that does not work I think I'm motivated to try an electronic modification even though I've not build electronic circuits for more than a decade other than more than a couple components, but this should be a relatively simple circuit, but until I open up the unit and make measurements I'm reserving judgement.  I don't see why a careful adjustment on the mechanical side would not work as intended, but again until I make measurements of the signal in actual powered on state i'm not really wanting to comment too much.

Edited May 26, 2017 by FreeRide

[RooEUC]

44 minutes ago, FreeRide said:

This is important, since the faster you go the more critical it is that the steering have the right response.

Still waiting for my parts to ship, but I'm still going to try this first with an attempted at a calibrated precise adjustment to the magnetic forces, and if that does not work I think I'm motivated to try an electronic modification even though I've not build electronic circuits for more than a decade other than more than a couple components, but this should be a relatively simple circuit, but until I open up the unit and make measurements I'm reserving judgement.  I don't see why a careful adjustment on the mechanical side would not work as intended, but again until I make measurements of the signal in actual powered on state i'm not really wanting to comment too much.

To modify it by adding an electronic circuit I think you would have to essentially reverse the signals coming out of the hall sensor. I'm not sure how to do that.  If not that, do you have another idea of the objective that the add-on circuit would fulfil? I hope you can come up with something.  If I can help, let me know. 

The issue with making mechanical modifications is that it may be harder to go back to standard operation if you don't like the result. It's also possible to completely ruin the steering.  As I found out, the slightest misplacement of the magnets or the hall sensor affects the steering significantly.

I wonder if we can increase the speed by approaching this from another angle entirely; by interfering with how the device detects the speed it's doing. Does anyone have a thorough understanding of how that works? 

[FreeRide]

13 minutes ago, RooMiniPro said:

To modify it by adding an electronic circuit I think you would have to essentially reverse the signals coming out of the hall sensor. I'm not sure how to do that.  If not that, do you have another idea of the objective that the add-on circuit would fulfil? I hope you can come up with something.  If I can help, let me know. 

The issue with making mechanical modifications is that it may be harder to go back to standard operation if you don't like the result. It's also possible to completely ruin the steering.  As I found out, the slightest misplacement of the magnets or the hall sensor affects the steering significantly.

I wonder if we can increase the speed by approaching this from another angle entirely; by interfering with how the device detects the speed it's doing. Does anyone have a thorough understanding of how that works? 

Modifying the signal electronically should be very simple stuff, the specific type of Hall Effect sensor used is of a linear type and essentially one needs to invert the signal which is biased at 2.5V according to measurements made on the german site.  A simple Inverting OpAmp with biased output should do the trick.  That's 4 resisters and one small chip in a typical circuit with a couple small caps to prevent and noise/feedback issues.  It should be very low power so hopefully could be powered right from the connector that connects the sensor. However until I measure the circuit myself, this is all theory.

In order to avoid any issues with going back to standard operation if need be I've taken precautions.  1st I bought a second machine, 2nd I ordered parts so I don't have to modify the original parts, 3rd I intend to only make the changes if I can measure the operation of the circuit all powered up and record the current operation.  So in the end I'm less worried about the mod and more about just general issues of disassembly and assembly especially if I'm working with it under powered conditions.  I also believe the mechanical approach should actually work if done carefully and/or with measured calibration.  In the end though I'm thinking of it as a fun project and I'll learn something either way and I'll be happy... just much happier if the mini-pro runs 20-25km/hr reliably and safely.  Then I change the tires, then I work on the battery capacity issue.