Mobipus Controllers

Status
Not open for further replies.
Mobipus do have a full range, the website doesn't appear to be super up to date. SamD might be the best person to elaborate on the other models.

Generally I think the 72200 and 72600 cover the range of requirements pretty well - The 72200 can make a crazy fast ebike, the 600 can make a crazy fast motorcycle. More than 200a battery into an 'ebike' sized hub is kinda pointless anyway, you're going to be on the wrong side of saturation and just generating exponentially more heat. Plus even with a fast wind, I'm not able to keep the front wheel on the ground as it is with still another ~30a to go.

Worth point out that the 'fancy graphics' are definitely not a major selling point of these controllers, sure it's cool - but what you're really purchasing is a well engineered controller. I don't think you're going to be able to blow up one of these very easily at all. I'd certainly take this over the APT any day of the week. Certainly having seen Shenta going through all the hidden, advanced settings - this is at a whole other level compared to the rest of the rubbish we have available (Sabvoton, Adaptto included in that!)
 
Maybe I got this wrong? I thought the 72200 was 200 A peak? But it is actually 200 battery A? So peak phase amp would be in the 350A region?
Don't get me wrong, I see the graphical display as a pure bonus. A cherry on top. And what a cherry it is. Also the user interface looks great and well designed, so dialing in might not be too hard. What really got me excited except from the hardware was the ability to share controller settings. Load a config file and be done with it. No guesswork or fail and error like some other hi specced controller I know of. If one got the same motor as you and has gotten the controller working great, sharing the config will be highly appreciated by fellow e-bikers.
 
macribs said:
Maybe I got this wrong? I thought the 72200 was 200 A peak? But it is actually 200 battery A? So peak phase amp would be in the 350A region?

That's correct, battery amps in the model number unlike bs Kelly ratings. Phase amps around 350. I believe Sam has also seen them perform beyond the ratings in bursts, frankly I don't need to push it as bike is scary fast anyway with my wheelbase. I'm running about 140 battery, 320 phase on my normal commute and it's plenty. Boost mode I've got set to 180/350 for surprising sport bikes. Had the DH up to around 110kph and still had a bit more in it, but I have no interest in pursuing that on the street.

One of these pushing the new mxus 5k with ferrofluid and hubsinks would be more than enough for all but the most loony among us on a <50kg chassis and would be bloody hard to overheat.
 
Do you know how to get your hands on the mobipus 12250? That would be nice step up from the 72200 and not crazy like the 72600. I'm looking for continuous 20kw and peaks of 30kw for a cr80 conversion. Also has anyone pushed a golden motor 10kw liquid cooled motor to see what it can handle?
 
devmanner said:
Do you know how to get your hands on the mobipus 12250? That would be nice step up from the 72200 and not crazy like the 72600. I'm looking for continuous 20kw and peaks of 30kw for a cr80 conversion. Also has anyone pushed a golden motor 10kw liquid cooled motor to see what it can handle?
I have heard that they are not going to be making the 120V models. I think for safety and hardware reasons. If you choose a high-enough RPM motor, you don't need the added volts anyhow. Design it from the start for a 20S battery, stay on the "sane" side of the voltage spectrum, and choose the rest of your components accordingly. 600A is enough to push ~43KW into a motor. :twisted:
 
@Ohbse thanks for clearing that up fo rme, I was a bit confused there. Always great to learn things perform better then you initial believed.
I think a Mobipus will be the 2017 controller for me.
 
Hi KingJamez will drop you a message. Not sure what happened but its not a problem.
 
parabellum said:
devmanner said:
Also has anyone pushed a golden motor 10kw liquid cooled motor to see what it can handle?
I think Luke "liveforphysics" did, ask him.

I asked him but he hasn't gotten back to me on that question. Plus his was the air cooled I think. I will shell out the money for a mobipus 72600 if I know the motor can give the performance I want.
 
I just purchased a 72200 from SamD and he's shipping it to me in the states. I'll be using it on a QS205V3 3.5T with FF/Hubsinks and 20S 20AH LiPo. I'm going to start working on a 8" tablet mount for the bike so I can take advantage of the cool display.

-Jim
 
kingjamez said:
I just purchased a 72200 from SamD and he's shipping it to me in the states. I'll be using it on a QS205V3 3.5T with FF/Hubsinks and 20S 20AH LiPo. I'm going to start working on a 8" tablet mount for the bike so I can take advantage of the cool display.

-Jim

Nice one. Exact same motor as me, I can perhaps save you some time with the settings.
 
I have mounted a hydraulic pressure transducer on my left brake lever to act as a kinda of virtual rear brake, unfortunately this wasn't working particularly well, the reported voltage on the analogue brake input was lower than I expected, maxing out at around 2.1v on a 5v sensor. While all of the threshholds are configurable, when I set it up for the voltage swing I was seeing it would only work intermittently and drop out at higher brake force.

It turns out the unexpectedly low voltage isn't actually low, it just displays at 1/2 of reality in the real time stats and logging graphs. Having configured all the thresholds for double the numbers it works beautifully!

Now with mild slip regen when off throttle and then my left lever acting as incredibly smooth and intuitive regen braking right down to practically stopped I really don't have to ever use the front brake. The sensation is not at all dissimilar to large hydraulic brakes on the rear, just much more confidence inspiring than bicycle parts.

At speed it's easily hitting my maximum regen limit of 50a back into the battery and can be precisely controlled to threshold of lockup at any speed you like, even down at walking pace.
 
kingjamez said:
I just purchased a 72200 from SamD and he's shipping it to me in the states. I'll be using it on a QS205V3 3.5T with FF/Hubsinks and 20S 20AH LiPo. I'm going to start working on a 8" tablet mount for the bike so I can take advantage of the cool display.

-Jim

I have the same motor run by a MaxE and pretty much love everything about it accept throttle response at take off. I'll be interested in hearing how yours does. The Adappto has a delay of some sort that can not be adjusted out unlike the near instant response of the Lyen controller I had used before on different motors. I want that "gas motorcycle" throttle type response for pulling the front end up at low speeds when needed.

Ohbse, do you have any comments on throttle response since you also have the same motor?

Thanks,

Tom
 
Ohbse said:
I have mounted a hydraulic pressure transducer on my left brake lever to act as a kinda of virtual rear brake, unfortunately this wasn't working particularly well, the reported voltage on the analogue brake input was lower than I expected, maxing out at around 2.1v on a 5v sensor. While all of the threshholds are configurable, when I set it up for the voltage swing I was seeing it would only work intermittently and drop out at higher brake force.

It turns out the unexpectedly low voltage isn't actually low, it just displays at 1/2 of reality in the real time stats and logging graphs. Having configured all the thresholds for double the numbers it works beautifully!

Now with mild slip regen when off throttle and then my left lever acting as incredibly smooth and intuitive regen braking right down to practically stopped I really don't have to ever use the front brake. The sensation is not at all dissimilar to large hydraulic brakes on the rear, just much more confidence inspiring than bicycle parts.

At speed it's easily hitting my maximum regen limit of 50a back into the battery and can be precisely controlled to threshold of lockup at any speed you like, even down at walking pace.

Care to share details on your pressure transducer? It sounds like you aren't running a brake on the rear at all? Very interesting. I'd love to do that as there isn't very much clearance on my motor between the 203mm disc / frame/ phase wires .

Does the Mobipus have a regen vs. battery voltage setting? If so does the amount of available regen phase in as the voltage drops or does it only start at a certain pack voltage?

-Jim
 
The lack of starting torque youre describing is just what actual control feels like lol. The lyen controllers essentially run unlimited phase amps, this does give them great 'punch'. You can totally replicate that with the mobipus, the throttle is very configurable! There are limits though, as the 3.5t wind is very fast it requires many phase amps. As the mobipus is limited to 350ish this might fail to rise on a build with a longer wheel base or heftier rider thand me. Realistically we bought the wrong winding, 4t would be much superior.
 
I agree on the 3.5T not being the best one. I wanted the 4T and had the deal signed and sealed with QS but when I found the 3.5T laced with a matching front wheel for almost $500 less than QS, I jumped ship.

Sounds like this might be pretty equal to the MaxE in terms of total performance then. I can dial in 196 amps and 360'ish phase amps on the MaxE.

Sounds like I just need to try a QS 273!

Tom
 
litespeed said:
I agree on the 3.5T not being the best one. I wanted the 4T and had the deal signed and sealed with QS but when I found the 3.5T laced with a matching front wheel for almost $500 less than QS, I jumped ship.

Sounds like this might be pretty equal to the MaxE in terms of total performance then. I can dial in 196 amps and 360'ish phase amps on the MaxE.

Sounds like I just need to try a QS 273!

Tom

With that cost difference I don't blame you. The 72200 is directly comparable to the Max-E in terms of peak power, perhaps a 10-15% higher - however its true continuous rating is much, much higher. Cooling is much less of an issue due to the design. I've managed to get mine to 36 degrees C after a ~1.5km long high speed run (~105km/h). I probably would have exploded a Max-e in the same conditions.

The other key difference is in configuration - the Mobipus is perhaps an order of magnitude more configurable and complex. Not to say it's hard to get running, it's just the range of tuning options is miles different. To give you an idea, at least for the parameter version I'm running there are 330+ user configurable settings. In the 'factory' mode, e.g Mobipus support there are many, many more than that.

These controllers can also run sensorless! I didn't realise until discussing it with Shenta. He was describing a 5% battery efficiency boost, essentially because it overcomes inaccuracy in the hall sensor placement. Only killer is it requires accurate measurements of the motor including inductance, which I've been unable to find for the qs v3 205 3.5t.

I was one of the vocal early adopters of the Adaptto, I have purchased several - it's conceptually still a fantastic product. But the technology has stagnated and is fundamentally the same as when first released now some years ago. I certainly won't be going back (at least not to their current product range) after having used the Mobipus.

kingjamez said:
Care to share details on your pressure transducer? It sounds like you aren't running a brake on the rear at all? Very interesting. I'd love to do that as there isn't very much clearance on my motor between the 203mm disc / frame/ phase wires .

Does the Mobipus have a regen vs. battery voltage setting? If so does the amount of available regen phase in as the voltage drops or does it only start at a certain pack voltage?

I am running a cheap moped left hand brake lever sourced from Aliexpress. I then purchased a 300psi pressure transducer, also Aliexpress. I drilled and tapped the outlet of the master cylinder and plumbed the transducer straight into the end. I then wired this as you would wire any other three wire throttle, as it's a 5v sensor this works out well. I put a very small amount of brake fluid in the reservoir and essentially just self bled it and pretty rapidly got a good feel on the lever. The resulting output is a dead linear increase from 0.32v to about 4.2v with increased effort on the lever but with essentially no motion. This works out to be quite confidence inspiring in practice, I have found setups based on travel rather than pressure felt unnatural. You're correct - I'm not running a physical rear brake at all.

Pressuretransducer.jpg

Yes - you can configure variable currents at different pack voltage levels. There are four configurable thresholds, you can then define limits at each. This is not really an issue in practice with Regen, as unless your battery is particularly shitty it won't be worried about absorbing relatively big pulses for very short durations - I've seen -62a or 5kw back into mine, however this by definition only lasts for a second or so because by then I'm just about stopped.
 
Ohbse said:
Only killer is it requires accurate measurements of the motor including inductance, which I've been unable to find for the qs v3 205 3.5t
That's actually a little surprising - I mean, if it doesn't have these parameters, I don't think it can do true field-oriented control. Doesn't it have a discovery mode to measure them? Maybe it does but it's locked down for end users?

I'm not as familiar with Adaptto or Sabvoton, but since I use a PhaseRunner I know the ASI controllers have this capability. If you know someone with a PhaseRunner or BAC2000, they can do the measurements for you! :)
 
litespeed said:
kingjamez said:
I just purchased a 72200 from SamD and he's shipping it to me in the states. I'll be using it on a QS205V3 3.5T with FF/Hubsinks and 20S 20AH LiPo. I'm going to start working on a 8" tablet mount for the bike so I can take advantage of the cool display.

-Jim

I have the same motor run by a MaxE and pretty much love everything about it accept throttle response at take off. I'll be interested in hearing how yours does. The Adappto has a delay of some sort that can not be adjusted out unlike the near instant response of the Lyen controller I had used before on different motors. I want that "gas motorcycle" throttle type response for pulling the front end up at low speeds when needed.

Ohbse, do you have any comments on throttle response since you also have the same motor?

Thanks,

Tom

I agree on that litespeed, My previous experience with the Lyen/C-lyte controlelr type was exactly what you described.. withc mean INSTANt torque response .. just like a ULTRA FRAGILE/ SENSITIVE / QUICK throttle... and the Kelly and Adaptto feel like having a delay compare to these.

Please, even if you get one of these Mobipus controller, please review it in term of Throttle response! I am very curious to see how these feel


Doc
 
Hello all at Endless Sphere,

This is Shenta, I am the architect for this controller, I haven't come up on ES for a while until seen this post recently,
when I first design this controller, I had a simple philosophy in mind, if it doesn't make people say "Wow", it doesn't worth the time and energy to make it.
After a few years of quest and try and error, we came up with something we though is a product good enough and we are welling to buy it our self.

I believe the users are the best critics for any product, so I will leave the comments to the users,
but I would like to give some answers for some of the questions.


Quokka: Can anyone tell me how water resistant it is?
I think the following video explain it all, even we won't guarrantee for full sub merge, but it is water tight.
https://youtu.be/y5DTO2g86bY

spinningmagnets, This is a "golf cart" controller
it is a "crazy" controller

romelec: What is the continuous and peak power capability of the 72200
Continuous power is 200 battery amps until thermal roll back, if you have good heat exchanger, 200 battery amps is its working current.

Ohbse: unexpectedly low voltage isn't actually low
We have also regen throttle voltage mapping, if you want to use Regen Throttle function, I could help you to set it up.

litespeed: throttle response at take off
The Throttle, Oh Throttle...
We have voltage map setting for the throttle, max min voltage, sensor failure protection, throttle power vs motor current mapping, soft start setting,
gas bike engine torque delay emulate setting...etc, and a lot more, we put a LOT of time and effort on the throttle feed back.

kingjamez: regen vs. battery voltage setting?
Yes, we are VERY concern about end user safety, we do have full voltage protection against regen, there's a voltage window for regen.

cycborg: Doesn't it have a discovery mode to measure them?
We do have discovery mode, but we found out it is rather useless against real world good equipment (LCR Meter) testing, for those controller offers discovery
mode, if you compare with proper equipment, there's only 1/10 times could get the correct numbers, another 9/10 times, the parameters are wrong,
it has to do with the accuracy of the internal voltage / current sensors, now our 72200 sensors are calibrated to +- 600amps, which is 1200amps of reading range,
with accuracty +-0.25%, for puny controller, when do 100 phase amp and with accuracy of 0.1%, you could get the motor impedance very accurate, but with 5 amps
swing around, it is very difficult to get correct value. However, we are working on a new algorithm, base on flux, we are currently working on it, we may able to
do a motor adoptive tuning on the fly in the near future.

cycborg: I don't think it can do true field-oriented control
I am very sure we have true FOC, actually we use STM32 as the core MCU, we used ST SDK stack on the SVM and Parker/Clark module, there's no point to reinvent
the wheel, we didn't use the rest of ST library, there's just too much bugs and the design was for speed control, not torque.

Doctorbass: please review it in term of Throttle response!
There's pretty good review on Facebook by Sketch Coleman,
you could definitely find his opinion and comments there:
https://www.facebook.com/groups/endless.sphere/?multi_permalinks=1325288857540179&comment_id=1325306374205094&notif_t=like&notif_id=1493800504099997

Thank you all very much for your support, and what I want is for everyone to enjoy the thrill of torque and fun of EV.
If anyone want to get a hand on Mobipus controller, or need more info, please check out SamD.

Cheer
Shenta
 
Any plans on building a really good BMS? That is the next thing that is really needed. Something that can put out 200 to 300 (or 600 amps even!) amps and let you charge at 50 to 100 amps. The Adaptto BMS is pretty great but not a stand alone unit. Build one of those in a small form factor that can take and put out some serious amps and you just made a "new wheel!"

Any free time to start a new project?

Tom
 
litespeed

Yap, I was going to have an open source BMS project, in memory of one friend died in lithium battery fire, but the project have been stall for about 1 year now, hardware is finished, needs someone to code it, any volunteers?
 
Thanks Shenta for all these responses.

I agree that it would be great to get a competitor for the Adaptto SYSTEM

And i say system because i think that having an all integrated system that include controller, display, bms,charge management with boost converter would be very nice!

I own 3 Adaptto MAX-E and i personally think this is the best ebike powertrain control system i have ever had!

The controller you designed seem very promissing in term of performance, but it still require people to get seperate display, seperate charger, seperate BMS... if you could work on an integrated system like Adaptto did it would make ALOT of people happy here! and i would be glad to try one and review it.

Your controller seem to be in the similar power as the Sevcon are offering but seem much more easy to program and dont require 900$ Canbus interface and program. That is great!

Doc
 
Status
Not open for further replies.
Back
Top