ASI Acclerated Systems controllers BAC2000 review

[ElectricGod]

My title is a bit confusing, but I figured that would get more search results.

I recently purchased an ASI BAC2000 motor controller. I've had it for several months now and can't get it to work without lots of help. You need a login to get access to their resources, documentation and knowledgebase. The first controller I received, I eventually sent back to ASI and they confirmed that it worked correctly. These controllers support bluetooth and RS232 communications. I couldn't connect to it at all. I have several USB RS232 cables, two PC's with RS232 and 2 phones...nothing worked. The second controller they sent the RS232 cable they use so I can at least communicate via RS232. They claim my phone is incompatible. It is running LineageOS 14.1, but it communicates with my two PowerVelocity controllers and a whole passel of other bluetooth devices so I doubt my phone is the problem. I think it's their android app. The Android BT settings sees the controller, but you have to use the app to connect to the controller and it will not list in the app. Anyway, the lack of ability of just basic communications was frustrating to say the least. Now that I have a controller that I can at least communicate via RS232, I can start setting it up. Then comes the next step...getting the controller running. They have a manual called "Get Your Bike Running with ASi Controllers". It's a basic "getting started" manual. I followed it religiously and failed to get any motor to work at all. I was told that the controller will run at a max of 72 volts, but actually it runs at 82 volts with some finagling. It turned out that the manual was wrong in several places. Last night, I got on line with ASI support and almost 2 hours later with their expert looking at my controller setup, we got my small block running. The throttle actually does stuff now...like run the motor. I'll get to the throttle later...which is weird. Anyway, after tech support set me straight with how to do the basic set up, then I actually got response from the motor. Still it didn't run right...it was jerky and slow and got hot really fast.

The first thing to do was get the motor working in sensorless mode like the getting started guide says. With much fiddling with settings in the controller we finally got the motor running in sensoreless mode. Initially the motor ran super slow, sucked down loads of current and was uber jerky and would stall out several times. after tweaking settings multiple times, we finally got the motor to run smoothly and efficiently.

Then came setting up sensored mode...more fiddling until that worked. The controller has auto detect functions for sensorless and sensored detection, but we failed to get it to work and had to resort to guessing at some of the settings until the motor ran.

Then came setting up the throttle..more setting tweaking and it still didn't work. It ended up being the brakes were on despite NOT having any brakes connected to the controller. 20+ minutes later, we had the controller brake settings turned off sufficiently that the controller would run the motor with the throttle. The throttle settings are vast and deep and many...oh wait...that's the entire controller and if you thought you knew something about motor controllers, well think again, the ASI controllers will make you feel like you know nothing at all. So back to the throttle. The default mode is called Torque mode. I'm used to throttle control where a little throttle, barely turns the motor and lots of throttle gets it spinning at full RPM's. Torque mode works nothing like this at all...in any way. You turn the throttle a tiny bit and the motor just spins up to full RPMs. Max throttle doesn't effect this at all...you just get full throttle all the time. Supposedly, I can't see what is really happening since the motor is not under load. Torque mode somehow adjusts torque, but tries to run the motor at full RPM. The idea is low throttle is low torque and hi throttle is high torque...you just can't see it without being under load. I was told that 100% of the time this is how they use these controllers. I asked about what I would call "normal" throttle control. We made a setting change in the throttle section and that was supposed to get me variable speed control instead, but it didn't change the way the motor ran at all. The support guy told me he would look into it further and find the correct option later. Throttle on this controller has so many options you will be over whelmed.

Since it was all running we said good night. I saved the configuration to the controller and to a file so that if I screwed anything up, I could restore a working config and get running quickly. I then made one small change. We had set the battery voltage to 76 volts. I changed it to 82 volts since I am running a 20S pack. Immediately the controller was not working again. I set it back to 76 volts and still nothing. GRRRR! I restored the backup config...nada...GRRRRRRRR!!! @$#%$#&$%& LOL...supposedly I have the brakes turned on again...I'll screw with those settings tonight and see if I can turn them off again. How that happened, your guess is as good as mine.

On a positive note, the BAC2000 controller is sooooo not a 2000 watt controller. These controllers use the AOT290 mosfet and they are good for 500 watts each. 12 fets and that's more like 4000 battery watts. The controller has a monitoring section of the config app that can access its internal sensing and monitoring functions. It tells you motor RPM, battery amps, phase amps, throttle voltage, speed and lots and lots and lots and lots of other things. I was running 200 phase amps at 82 volts or 16000 watts through the small block. The controller never got slightly warm, but the motor was hot. It only ran that way for 30 seconds. All I can say is OMG!!! crazy strong controller. I can't imagine what the 4000 or 8000 watt versions will do once they are released. That of course is assuming I can actually get the controllers running. Also, you will not need a watt meter or CA or any thing other than your phone and the android app to configure and monitor your EV. The controller knows more about your EV than you will ever need to know! I just wish I could get the controller to talk to my phone! It has several analog inputs that you can define for whatever purpose and it supports canBUS so whatever you want to connect that way is doable too. All this is in a less than $300 controller.

The ASI controllers are waaaaaaaay far more complex than any controller I've ever seen before. There's a file that tech support showed me. It's a list of all the functions in the controller. The file is 8000 lines long! However that makes up about 2-3000 options which is still a lot of stuff that can be set up. These controllers can wash your car, do your laundry and manage your childrens trust funds all while your just riding around town. For someone who is a noob to EV stuff, get something else, this is NOT the controller for you. For someone who is an advanced EV'r...like way advanced...well this may be an option for you, but there us a serious learning curve. These controllers seriously need a noob mode that hides its many thousands of options and makes them usable for the general EV'r.

All the controllers I would be interested in are good for 82 volts. They will tell you 72 volts, but that's BS. Anyway more to come...

Here's some pics of the BAC2000. The build quality is pretty nice, but getting at the mosfets if needed will need some effort and a special tool. If there's a build issue to this controller, it's the status LED. You can't see it without removing the cover. I added a small light pipe to get the LED visible outside the side cover. The mosfets are mounted to that thick sloped area on the bottom of the heat spreader.

This is the connectors on the controller. The pins project through the shell. The smaller connector is halls and motor temp sensor. The larger connector is brakes, enable, throttle and lots more stuff.

This is the Hall cable. I added an IP68 connector to the end of it since I use this connector on all my motors and controllers. Both cables just terminate in wire ends. You connect to them however you choose.

This is the correct phase colors. The manual doesn't say beyond U, V and W.

 

[fechter]

I think that's the same platform as the Grin Phaserunner. You might get some programming tips from them:
https://endless-sphere.com/forums/viewtopic.php?f=31&t=65031

 

[ElectricGod]

It ended up being that the brakes were set to on and I saved the config file before we fixed that item. I got it running today.

 

[ElectricGod]

I think that's the same platform as the Grin Phaserunner. You might get some programming tips from them:
https://endless-sphere.com/forums/viewtopic.php?f=31&t=65031

Thanks...I've read through a lot of this thread.

 

[ElectricGod]

I got on gotomeeting with Martin at ASI support and he showed me tons of things about setting up these controllers. I've now set up every spare motor (sensored and sensorless) I have on this controller and ran them all. Now that I know how to do it correctly and a few trouble shooting steps, it's fairly straight forward. For people looking for a controller that can quite literally do anything that any EV will EVER possibly want and do it well, then the ASI controllers are a great choice, but they are also complex. They are not noob friendly at all. You will never plug one into a motor, battery and throttle and have it spin up the motor. The chances of that working totally in the blind are minuscule at best. You REALLY need to set up the controller to your specific arrangement and then it's a really amazing controller. I thought I knew something about decent controllers, but until I learned this one, I was deluded by average controllers. These controllers diverge from the standard way of setting one up. The reason why they have a complex setup is because there is so much functionality crammed into the MCU. One of those things is that you don't need to know what is the correct hall and phase pairing. Just plug things in and let the autodetect steps do their work. The controller will figure the pairing out and then at worst all you need to do is tell the controller to spin the motor the other direction. Every other controller I have ever used required that you get the hall and phases paired correctly first and then it will run the motor. Not so with the ASI controllers. Flux weakening is really good too. I've seen quite a few controllers that do 30% and used a few that realistically gave you maybe another 15% more motor RPM. Not so with the ASI controllers. Put 30% in the correct field and you will get 30% more motor RPM. Put 60% in and you will get 60% more RPM. Martin told me that theoretically you could enter 200%. That's 2X the KV limit of the motor. There is little need for more battery voltage when you can use flux weakening to accomplish the same thing. I've read things like this before, but since the best I have seen is a small RPM increase, I really didn't believe it. Now that I have seen real flux weakening for myself, well I am a believer. I haven't done any load testing yet to see how well it works, but under no load, current draw on the pack does increase to get the higher RPM's. I can live with that.

I really can't wait for their 4000 and 8000 watt controllers to be released. I bought the BAC2000 just to review their products for future use, but I really wanted the more powerful ones. I'll be buying more of them in the future. They are complex, but they are also VERY capable.

 

[mrbill]

Hi ElectricGod:

Thanks for your review.

I have two of these controllers that I purchased through Grin, one on each of my bikes. Like you I found the initial setup challenging, but once I got the controller working (with some help from Grin), I found that it worked very smoothly and efficiently with my DD hub motors. Like you I observed that the controller never got hot even if the motor did. At most it got only slightly warm. I am using the BAC2000 as a "dumb" controller with the CA3 as the "smart" front-end.

How did you gain access to their knowledge base? I have a few outstanding issues that I am trying to solve:

1) Please post a followup if you find a way to switch throttle mode from torque to speed. When driving a mid-drive motor or a motor connected to the wheel through a "springy" drivetrain, the torque mode can cause an irritating resonance or shudder as the controller attempts to compensate for the varying load the motor sees.

2) I am trying to disable the plugging component of engine braking, leaving only pure regeneration when engine braking. Unfortunately, there is no check box in BacDoor that one can set to turn plugging on or off. Grin has sent me a firmware file that does this for their Phaserunner that I may upload to this controller, but if you have access to a file (perhaps through the knowledge base) intended specifically for the BAC2000, I would be interested in trying that first. My fear is that a Phaserunner-specific firmware file may not be compatible or may cripple the BAC2000 in some way.

3) Do you have a BAC2000 firmware file that sets the default stock configuration? If I upload any custom firmware file I'd like to be able to go back to the factory settings if the process goes awry or if I discover that I prefer the stock configuration.

BTW, the firmware file loaded into BacDoor through the File->Bootloader menu is not the same as the parameter file, loaded through the Parameter->Load From File menu, where users can set speed, voltage, current, and other limits and thresholds.

 

[grindz145]

Sorry for your troubles there. I've set up the smaller version for a handful of different motors / configurations and I can tell you that the intended audience is an engineering team at an OEM and not an individual hobbyist. It's pretty awesome that they will entertain working with individuals though Thay are absolutely fantastic controllers.

 

[ElectricGod]

I don't believe the controllers that Grin has are the current revision. Mine includes bluetooth. Does yours?

I'll be using this controller 100% on inrunners and outrunners and probably never on a hub. That was one of my challenges with ASI support. They were used to setting it up on hubs only and then I come along and won't ever use a hub. I intend to use it on mid drives and direct drive via chain.

What is the wattage for your hubs?

I'll be using the BT functionality once I get that working. The phone app is pretty nice and using it as my CA will make the CA look like a toy.

This is the home screen...speed, battery status and amp draw. I think via the third page in the app which is where you do all the settings stuff you can change what the 1st page displays.

The second page, just swipe right to left to get to it, is lots more status information...anything you would ever need to know. It took 4 photos to capture the whole second page.

Then there is the 3rd page for motor setup and whatnot. This is the demo screen. I'm not connected to the controller so I can show all of it.

There is also this menu listing. You get to it via that 3 bar icon in the home screen.

You go here...and request a login.

http://accelerated-systems.com/support/documents/knowledge-base/

1. Talking to support...apparently there isn't an option to NOT do torque mode. However in the throttle settings, there is this option. I haven't tried it out yet to see how it responds differently to torque mode. You have to add the feature first to throttle and then you can set it, if it isn't already listed in throttle. It defaults to 1, but I want 0 to work.

2. Tell me more about this plugging component. I haven't messed with regen yet since I am only bench testing ATM. I'll mount it on my Currie scooter which I just completed and is currently running a PowerVelocity 12 fet controller that has AOT290's in it...same as this controller.

3. I can't find it right now, but I found an option in the BacDoor PC app that let me factory default the controller. That's what you are looking for. I assume you are saving your config to your PC. It's just an XML file and wouldn't be hard to edit and set the values back to defaults.

Bootloader option...wondered what that was for...now I know. I'm trying to get them to change a few firmware settings in the controller. For example...the filter caps and mosfets are good for 100 volts. That means you should be able to run this controller at 82 volts (20S) continuous with a max battery voltage of 90 volts and not EVER hurt anything. I have 4 other controllers that have 100 volt mosfets and caps and they all work exactly like this. Then in the the firmware, the rated system voltage is limited to 81 volts which is illogical. I intend to run at 82 volts. Their claims of 72 volts is just them being overly cautious. I also am insisting that they add voltage/speed tracking to the throttle functionality. This is just a firmware change in the controller. PM me with the firmware you have for the Phaserunner. It's just code...I bet I can mod it for throttle and other things if they wont. They don't have the firmware on the support site or in the KB section. They have an FTP site which is pretty basic. You can get the BacDoor apps there and that's about it. There are no firmware downloads there. You need a login to get in.

What version of BacDoor are you running? I have 1.6.0 and 1.6.4

Hi ElectricGod:

Thanks for your review.

I have two of these controllers that I purchased through Grin, one on each of my bikes. Like you I found the initial setup challenging, but once I got the controller working (with some help from Grin), I found that it worked very smoothly and efficiently with my DD hub motors. Like you I observed that the controller never got hot even if the motor did. At most it got only slightly warm. I am using the BAC2000 as a "dumb" controller with the CA3 as the "smart" front-end.

How did you gain access to their knowledge base? I have a few outstanding issues that I am trying to solve:

1) Please post a followup if you find a way to switch throttle mode from torque to speed. When driving a mid-drive motor or a motor connected to the wheel through a "springy" drivetrain, the torque mode can cause an irritating resonance or shudder as the controller attempts to compensate for the varying load the motor sees.

2) I am trying to disable the plugging component of engine braking, leaving only pure regeneration when engine braking. Unfortunately, there is no check box in BacDoor that one can set to turn plugging on or off. Grin has sent me a firmware file that does this for their Phaserunner that I may upload to this controller, but if you have access to a file (perhaps through the knowledge base) intended specifically for the BAC2000, I would be interested in trying that first. My fear is that a Phaserunner-specific firmware file may not be compatible or may cripple the BAC2000 in some way.

3) Do you have a BAC2000 firmware file that sets the default stock configuration? If I upload any custom firmware file I'd like to be able to go back to the factory settings if the process goes awry or if I discover that I prefer the stock configuration.

BTW, the firmware file loaded into BacDoor through the File->Bootloader menu is not the same as the parameter file, loaded through the Parameter->Load From File menu, where users can set speed, voltage, current, and other limits and thresholds.

 

[mrbill]

I've seen quite a few controllers that do 30% and used a few that realistically gave you maybe another 15% more motor RPM. Not so with the ASI controllers. Put 30% in the correct field and you will get 30% more motor RPM. Put 60% in and you will get 60% more RPM. Martin told me that theoretically you could enter 200%. That's 2X the KV limit of the motor. There is little need for more battery voltage when you can use flux weakening to accomplish the same thing.

I was pleasantly surprised to discover this, too.

It has been suggested that adding field weakening results in loss of efficiency, but my own simple experiment with two DD motors has found this not to be so, at least with the BAC2000.

The two motors I used were the Nine Continents M3006RC with a Kv around 9, and the Edge1500 with Kv around 11.5. After setting up 30% Maximum Field Weakening Current (Peripheral Configuration -> Advanced Motor in BacDoor) and confirming that the maximum free spin RPM of both motors was now roughly the same, I observed under load that the M3006RC drew roughly the same amount of current as the Edge1500 at all available speeds, including those speeds above which the M3006RC could not normally be driven without field weakening. Available regeneration current was also the same for both motors. As RPM exceeded even the maximum unloaded RPM of the motor with field weakening, the motor/controller behaved naturally as a regeneration brake without exhibiting any artifacts to alert the user that field weakening was being used.

It is true that the unloaded maximum RPM current draw was higher on the M3006RC, but when loaded both motors behaved similarly.

I am working with the assumption that although the controller is injecting additional current to achieve higher motor RPM, the motor naturally produces a back-EMF that cancels out this injected current. The net effect is that battery current does not increase at RPM above which the motor would be incapable of running without field weakening, hence efficiency is not degraded.

 

[ElectricGod]

I've seen quite a few controllers that do 30% and used a few that realistically gave you maybe another 15% more motor RPM. Not so with the ASI controllers. Put 30% in the correct field and you will get 30% more motor RPM. Put 60% in and you will get 60% more RPM. Martin told me that theoretically you could enter 200%. That's 2X the KV limit of the motor. There is little need for more battery voltage when you can use flux weakening to accomplish the same thing.

I was pleasantly surprised to discover this, too.

It has been suggested that adding field weakening results in loss of efficiency, but my own simple experiment with two DD motors has found this not to be so, at least with the BAC2000.

The two motors I used were the Nine Continents M3006RC with a Kv around 9, and the Edge1500 with Kv around 11.5. After setting up 30% Maximum Field Weakening Current (Peripheral Configuration -> Advanced Motor in BacDoor) and confirming that the maximum free spin RPM of both motors was now roughly the same, I observed under load that the M3006RC drew roughly the same amount of current as the Edge1500 at all available speeds, including those speeds above which the M3006RC could not normally be driven without field weakening. Available regeneration current was also the same for both motors. As RPM exceeded even the maximum unloaded RPM of the motor with field weakening, the motor/controller behaved naturally as a regeneration brake without exhibiting any artifacts to alert the user that field weakening was being used.

It is true that the unloaded maximum RPM current draw was higher on the M3006RC, but when loaded both motors behaved similarly.

I am working with the assumption that although the controller is injecting additional current to achieve higher motor RPM, the motor naturally produces a back-EMF that cancels out this injected current. The net effect is that battery current does not increase at RPM above which the motor would be incapable of running without field weakening, hence efficiency is not degraded.

Were you looking in BacDoor at the phase amps (motor current)? with and without field weakening? This number does rise unloaded for me. I don't know...yet about loaded and I'll be watching my watt meter too. Phase current and battery current can be wildly different. My watt meter was reading 11 amps and the phase current was nearly 200 amps.

When regen engages, you are already off the throttle so field weakening wont be relevant anyway. Also field weakening only comes into play above the Kv limit of the motor. You won't see it's effects below this limit. Field weakening works on the idea of running the phases longer after they have passed top dead center. This creates the boost in motor RPM at the cost of a little efficiency. The extra current use isn't due to pumping more current into the motor. If that was the case, then you could just turn up the phase current to get the motor to exceed it's Kv limit. FOC uses more current because it is running the phases longer and after they have reached their maximum efficiency. I'm uncertain about the bEMF effects.

 

[mrbill]

I don't believe the controllers that Grin has are the current revision. Mine includes bluetooth. Does yours?

Grin hasn't said anything about Bluetooth being featured in the models they sell, so I doubt they include it. That said, I haven't tried to see if my controller is sending out a BT signal.

I'll be using this controller 100% on inrunners and outrunners and probably never on a hub. That was one of my challenges with ASI support. They were used to setting it up on hubs only and then I come along and won't ever use a hub. I intend to use it on mid drives and direct drive via chain.

1. Talking to support...apparently there isn't an option to NOT do torque mode. However in the throttle settings, there is this option. I haven't tried it out yet to see how it responds differently to torque mode. You have to add the feature first to throttle and then you can set it, if it isn't already listed in throttle. It defaults to 1, but I want 0 to work.

I'm very interested to learn how your project goes.

I first started out trying to get these controllers to operate well with my then-current crank-drive and mid-drive systems. In both cases I ran into problems related to the torque feedback causing a resonance or shuddering through the drivetrain. The problem is that my bike frame acts like a spring with a low-frequency resonance. That frequency appears to be close to the same as the feedback of the torque throttle in the BAC2000.

I tried changing settings for current regulator (Ki, Kp, and bandwidth), speed regulator (Ki, Kp, and mode), and PLL (bandwidth and damping), all to no avail.

For mostly unrelated reasons I have since switched my bikes over to DD hubs and find that these controllers actually give me greater overall efficiency than my mid-drive over most terrain, provided I take full advantage of regeneration on the downhills.

What is the wattage for your hubs?

The Nine Continents hub is rated for up to 1000 watts, and the Edge1500 is rated to 1500 watts. The BAC2000 is overkill for these, but I wanted to leave myself room to experiment with bigger DD hubs and to increase reliability by not stressing a smaller controller. That said, I have pushed up to about 3500 battery watts into each of these motors when testing.

I'll be using the BT functionality once I get that working. The phone app is pretty nice and using it as my CA will make the CA look like a toy.

Toy or not, the CA3 has the advantage of being hard-wired and not subject to smartphone app reliability. It also does exactly what I want and has a development team that is responsive to bug fixes and feature requests. I also use my smartphone on the bike, but I use it for GPS tracking, directions, and other non-control functions.

You go here...and request a login.

http://accelerated-systems.com/support/documents/knowledge-base/

Done. I've requested a login. I hope that my being an end-user, having bought two of their controllers through one of their customers, doesn't disqualify me from being able to get an account.

2. Tell me more about this plugging component. I haven't messed with regen yet since I am only bench testing ATM. I'll mount it on my Currie scooter which I just completed and is currently running a PowerVelocity 12 fet controller that has AOT290's in it...same as this controller.

Engine braking on the BAC2000 appears to have two components. At higher speed regeneration is sufficient to provide the necessary braking force. But, at lower speed the controller attempts to run the motor in reverse to achieve the necessary braking force. With the typical e-bike hub motor in a 26" wheel the transition between regeneration and plugging is gradual and occurs below about 30 kph. I can tell when some of the braking force is accomplished with plugging when the motor temperature rise accelerates. Sometimes my highest motor temperature is observed at the bottom of a long, steep downhill where the controller was plugging much of the time. Plugging makes the motor hot! When commanding high braking force I can sometimes feel the motor braking force increase suddenly, and then I know it is has transitioned to plugging. Another way I can tell is that the current displayed on the CA3 increases from negative to less negative or swings positive while the braking force remains constant. With regeneration one cannot skid a tire, but plugging can skid a tire.

With the current firmware there is no way to turn off plugging. I manage it now by being aware of when it occurs or is likely to occur. When I sense it, I ease up on the throttle (used to command engine braking) and use more friction brake so that I don't overheat the motor and unnecessarily consume valuable battery energy.

I can see that plugging would be useful for applications where all braking is done with the motor or when it is desirable to use the controller to achieve tight speed control over varying terrain. But, I design my bikes for efficiency, and I'd rather use my friction brake at times when the controller needs to consume battery energy to accomplish braking.

3. I can't find it right now, but I found an option in the BacDoor PC app that let me factory default the controller. That's what you are looking for. I assume you are saving your config to your PC. It's just an XML file and wouldn't be hard to edit and set the values back to defaults.

The XML file is the parameter file. This file can be read directly in a text editor. Yes, I save this file for each bike/motor combination, and that all works fine. I'm interested in the firmware file, or the latest default firmware file for this controller. This is an EHX file loaded through the File->Bootloader menu, I think. (I haven't yet tried it.)

The EHX file is also text, but it appears as gibberish in a text editor.

Bootloader option...wondered what that was for...now I know. I'm trying to get them to change a few firmware settings in the controller. For example...the filter caps and mosfets are good for 100 volts. That means you should be able to run this controller at 82 volts (20S) continuous with a max battery voltage of 90 volts and not EVER hurt anything. I have 4 other controllers that have 100 volt mosfets and caps and they all work exactly like this. Then in the the firmware, the rated system voltage is limited to 81 volts which is illogical. I intend to run at 82 volts. Their claims of 72 volts is just them being overly cautious. I also am insisting that they add voltage/speed tracking to the throttle functionality. This is just a firmware change in the controller. PM me with the firmware you have for the Phaserunner. It's just code...I bet I can mod it for throttle and other things if they wont. They don't have the firmware on the support site or in the KB section. They have an FTP site which is pretty basic. You can get the BacDoor apps there and that's about it. There are no firmware downloads there. You need a login to get in.

ASI are probably keeping a tight control on firmware files. And, given your description of the controller above, it looks like they conservatively rate their controllers.

What version of BacDoor are you running? I have 1.6.0 and 1.6.4

My BacDoor version is 1.5.4.0. I'm not surprised there is a more recent version available.

The identifying sticker on one of my controllers reads:

10-000614
[bar code]
1525-00014

 

[mrbill]

Field weakening works on the idea of running the phases longer after they have passed top dead center. This creates the boost in motor RPM at the cost of a little efficiency. The extra current use isn't due to pumping more current into the motor. If that was the case, then you could just turn up the phase current to get the motor to exceed it's Kv limit. FOC uses more current because it is running the phases longer and after they have reached their maximum efficiency. I'm uncertain about the bEMF effects.

Thanks for correcting my understanding.

That said, I did not observe a significant loss of system efficiency (as measured by comparing battery watts) running the M2006RC with 30% field weakening compared to the Edge1500 without field weakening. If there is a loss, it must be small. Behavior of the two motors was nearly indistinguishable. I should add that I was not driving the motor in the speed range where field weakening would be active, only coasting in that speed range.

 

[mrbill]

When regen engages, you are already off the throttle so field weakening wont be relevant anyway.

I'm having trouble with this statement.

My bike running the M3006RC runs up to about 52 kph before natural regeneration kicked in.

When I set up this motor on the BAC2000 with 30% maximum field weakening, I was able to coast downhill up to about 67 kph before natural regeneration occurred. No throttle use involved.

The BAC2000 must be doing something in connection with field weakening even when the wheel is not being driven by the controller.

 

[ElectricGod]

When regen engages, you are already off the throttle so field weakening wont be relevant anyway.

I'm having trouble with this statement.

My bike running the M3006RC runs up to about 52 kph before natural regeneration kicked in.

When I set up this motor on the BAC2000 with 30% maximum field weakening, I was able to coast downhill up to about 67 kph before natural regeneration occurred. No throttle use involved.

The BAC2000 must be doing something in connection with field weakening even when the wheel is not being driven by the controller.

Interesting...I wonder if something is set wrong in the controllers. Regen is pointless when you are on the throttle. It's the exact opposite of throttle. The whole point of regen is to recover energy that would otherwise be lost to braking. Why would you ever hit the throttle and brakes at the same time? That makes no sense. Regen is effectively braking since that's when it gets used...to slow down and replace braking with a method of regaining energy. Regen should never kick in when you are under throttle. Most controllers have a couple of ways to implement regen. The brake switches are commonly used to activate regen. On my PV and Kelly controllers, they have a feature for engaging regen when you let off the throttle. I use my brake switches for activating the brake lights which need 12 volts. This is not compatible with motor controllers which use 5 volts to activate the brakes. The "regen at throttle off" works well for me to engage regen.

I know the ASI controllers have various forms of limiters in them, but I believe all of them can be used or not. I have the speed limiters and other things turned off. I want the controller to run my motor at whatever maximum speed I can get out of it. It's total BS IMHO to limit a controller at all for speed. That's the realm of commercially made bikes. I don't know if this will work for you or not, but in the motor name plate section is an option for gear ratio. This is used to determine actual vehicle speed. Try setting it such that the controller can never calculate any speed because it is calculating way too low. Since you are using a CA, you don't care about what the controller thinks the speed is.

 

[ElectricGod]

The higher speed stuff (plugging) isn't regen, that's speed limiting. Turn off the speed limiting or set it to something like 100mph so it never engages or change the gear ratio or tire size in the controller so the controller can't get close to where it comes into play. The reason why it's making your motors hot is because it's doing work, not doing regen. It's running the motor backwards like you said or else it's shorting the phases together. Either way, the fields and magnets are working against each other. Regen does not operate this way. The motor spinning is creating electrical pulses. You can collect them or not and you can collect them aggressively or just a little bit. The motor resists rotation a lot or a little, but never to the point of locking up the wheel. That's shorting the motor windings or running it backwards. During regen, the motor is never doing as much work as it is when it's generating torque. It ought to cool down during long phases of regen like coasting down a hill, not get hotter.

 

[mrbill]

When regen engages, you are already off the throttle so field weakening wont be relevant anyway.

I'm having trouble with this statement.

My bike running the M3006RC runs up to about 52 kph before natural regeneration kicked in.

When I set up this motor on the BAC2000 with 30% maximum field weakening, I was able to coast downhill up to about 67 kph before natural regeneration occurred. No throttle use involved.

The BAC2000 must be doing something in connection with field weakening even when the wheel is not being driven by the controller.

Interesting...I wonder if something is set wrong in the controllers. Regen is pointless when you are on the throttle. It's the exact opposite of throttle. The whole point of regen is to recover energy that would otherwise be lost to braking. Why would you ever hit the throttle and brakes at the same time? That makes no sense. Regen is effectively braking since that's when it gets used...to slow down and replace braking with a method of regaining energy. Regen should never kick in when you are under throttle. Most controllers have a couple of ways to implement regen. The brake switches are commonly used to activate regen. On my PV and Kelly controllers, they have a feature for engaging regen when you let off the throttle. I use my brake switches for activating the brake lights which need 12 volts. This is not compatible with motor controllers which use 5 volts to activate the brakes. The "regen at throttle off" works well for me to engage regen.

My statement above refers to coasting at high speed above the speed at which natural regeneration due to motor back-EMF occurs. The throttle is at zero. The controller is (presumably) doing nothing, EXCEPT that if I have programmed the controller to apply some field weakening, the threshold speed at which natural regeneration occurs increases in proportion to the percent of field weakening I have programmed. The motor is acting just like a higher Kv motor when coasting.

I know the ASI controllers have various forms of limiters in them, but I believe all of them can be used or not. I have the speed limiters and other things turned off. I want the controller to run my motor at whatever maximum speed I can get out of it. It's total BS IMHO to limit a controller at all for speed. That's the realm of commercially made bikes. I don't know if this will work for you or not, but in the motor name plate section is an option for gear ratio. This is used to determine actual vehicle speed. Try setting it such that the controller can never calculate any speed because it is calculating way too low. Since you are using a CA, you don't care about what the controller thinks the speed is.

I've set the BAC2000 not to limit speed in any way. I do speed limiting on the CA3 (that I can vary using pre-programmed AuxD steps and Grin's 2-button control), when I like to set a moderate speed for descents so that I maximize the regeneration: fast enough to get regeneration and no plugging but slow enough that I'm not losing too much energy to air friction. Somewhere between 25 and 40 kph works best.

 

[mrbill]

The higher speed stuff (plugging) isn't regen, that's speed limiting. Turn off the speed limiting or set it to something like 100mph so it never engages or change the gear ratio or tire size in the controller so the controller can't get close to where it comes into play. The reason why it's making your motors hot is because it's doing work, not doing regen. It's running the motor backwards like you said or else it's shorting the phases together. Either way, the fields and magnets are working against each other. Regen does not operate this way. The motor spinning is creating electrical pulses. You can collect them or not and you can collect them aggressively or just a little bit. The motor resists rotation a lot or a little, but never to the point of locking up the wheel. That's shorting the motor windings or running it backwards. During regen, the motor is never doing as much work as it is when it's generating torque. It ought to cool down during long phases of regen like coasting down a hill, not get hotter.

I do no speed limiting on the BAC2000. Plugging occurs when a user-commanded engine braking force is higher than can be achieved by regeneration alone.

The BAC2000 is capable of variable regeneration, and the CA3 sends the BAC2000 the proper throttle voltage signals so that the latter knows when the throttle voltage is controlling propulsion or engine braking. (Propulsion between 1.0 and 3.5 volts; braking between 0.8 and 0.1 volts.) The throttle on my handlebars does double duty: a normal throttle when driving, and a variable regeneration lever when the CA3 e-brake circuit is closed. The system works like a charm.

This afternoon I loaded Grin's "no plugging" firmware for the Phaserunner onto my BAC2000, and my quick test suggests that it does indeed disable plugging. The behavior is a bit disconcerting as the regen brake acts like an anti-lock brake in anti-lock mode when plugging would otherwise occur. The motor exhibits a mild rapid vibration. Since battery current remains negative at all times, I believe the firmware change is doing what I expected, although I would prefer smoother behavior. What I don't know yet for sure is whether the controller's maximum current limits have been now capped at those of the smaller Phaserunner controller.

 

[ElectricGod]

The firmware file is meant to be an executable that runs in the MCU. It may look like gibberish text, but that's your text editor attempting to read a binary file. There is a software development tool (IDE) that will correctly read this file and interpret it as something human readable. This is why I want the firmware file you have for the phaserunner. If I can figure out whatever IDE they are using, then I can edit any firmware file. It's a readily available IDE...just got to find out the specifics. If I knew what MCU was in the controller, that would get me to the correct IDE. I bet the phaserunner used the same MCU. Talk to the grintech folks and see if they can tell you.

 

[fechter]

This afternoon I loaded Grin's "no plugging" firmware for the Phaserunner onto my BAC2000, and my quick test suggests that it does indeed disable plugging. The behavior is a bit disconcerting as the regen brake acts like an anti-lock brake in anti-lock mode when plugging would otherwise occur. The motor exhibits a mild rapid vibration. Since battery current remains negative at all times, I believe the firmware change is doing what I expected, although I would prefer smoother behavior. What I don't know yet for sure is whether the controller's maximum current limits have been now capped at those of the smaller Phaserunner controller.

Could that be an oscillation in the current feedback loop? There might be a parameter to smooth that out. There will always be some vibration at low speed due to cogging in the motor. There may also be a mechanical resonance at a particular speed (I get this with one of my motors).

 

[madin88]

thanks for sharing your findings. i might give one a try for my next project.

I wonder if it is possible to set up more than one power mode?
like:
mode 1: 20A batt, 50A phase, 25kmh, slight regen
mode 2: 40A batt, 100A phase, 45kmh, strong regen
and so on..

 

[ElectricGod]

thanks for sharing your findings. i might give one a try for my next project.

I wonder if it is possible to set up more than one power mode?
like:
mode 1: 20A batt, 50A phase, 25kmh, slight regen
mode 2: 40A batt, 100A phase, 45kmh, strong regen
and so on..

I'm not sure, but the controller has several inputs that can be set however you want to use them. A switch on one of those inputs probably could trigger your two modes.

 

[mrbill]

This afternoon I loaded Grin's "no plugging" firmware for the Phaserunner onto my BAC2000, and my quick test suggests that it does indeed disable plugging. The behavior is a bit disconcerting as the regen brake acts like an anti-lock brake in anti-lock mode when plugging would otherwise occur. The motor exhibits a mild rapid vibration. Since battery current remains negative at all times, I believe the firmware change is doing what I expected, although I would prefer smoother behavior. What I don't know yet for sure is whether the controller's maximum current limits have been now capped at those of the smaller Phaserunner controller.

Could that be an oscillation in the current feedback loop? There might be a parameter to smooth that out. There will always be some vibration at low speed due to cogging in the motor. There may also be a mechanical resonance at a particular speed (I get this with one of my motors).

I found the parameter, Current Regulator bandwidth, that I could adjust. The default was 1500 radians. I won't pretend to understand this parameter, only that it affects both the current regulator integrative and proportional feedback values that I cannot change directly in BacDoor, although I can see the calculated values after I change the bandwidth term. After much trial and error I reduced this to 30 radians. Now I experience only the slightest vibration at very slow (e.g. walking) speed.

I tried 25 radians, but on one of my test runs the controller threw a fault (i.e. shut down) during hard motor braking on a bumpy road. Sudden loss of motor braking could be dangerous, so I raised it to 30 radians. If I experience faults at 30 radians I'll gradually increase the bandwidth. Anything under 50 radians would be acceptable, but lower gives smooth braking at low speed.

Thanks for the suggestion.

 

[mrbill]

Bootloader option...wondered what that was for...now I know. I'm trying to get them to change a few firmware settings in the controller. For example...the filter caps and mosfets are good for 100 volts. That means you should be able to run this controller at 82 volts (20S) continuous with a max battery voltage of 90 volts and not EVER hurt anything. I have 4 other controllers that have 100 volt mosfets and caps and they all work exactly like this. Then in the the firmware, the rated system voltage is limited to 81 volts which is illogical. I intend to run at 82 volts. Their claims of 72 volts is just them being overly cautious.

There may be a way to allow the controller to operate in your desired voltage range without updating firmware. I discovered how quite by accident while I was trying to test the "no plugging" firmware at 24 volts nominal (7s LiPo).

Many of the Protection and Fault Thresholds for the Battery are based on the values entered into the Motor Nameplate Ratings -> (Rated system voltage and Rated motor power). From these two values is also calculated (using Ohm's Law) the "Rated system current", the basis for the Battery current limit and Regeneration current limit under Peripheral Selection -> Battery.

To get my system running at 24 volts I merely entered an acceptable value into Rated system voltage, e.g. "48". Then I adjusted the Battery thresholds to validate operation between, say, 40% of 48 volts and 60% of 48 volts.

If you are confident that the controller will not be damaged by presenting the battery terminals with 90 volts, then you should be able to do something similar by adjusting your threshold ranges based on an acceptable system voltage of 72 volts.

 

[ElectricGod]

I got the latest firmware today which was already loaded onto my controller.

 

[ElectricGod]

I also added a light pipe I scavenged from an old router to get that status LED visible outside the case. The LED is on the small daughter board. This is one area that I find irritating about ASI so far. They wont even acknowledge that something is a good idea. I presented several things...better manuals that are complete and accurate, actual speed throttle control, set the firmware to accept 90 volts max and this light pipe and I seriously doubt they will ever implement any of them. I've even offered to make the firmware changes myself and they are against that too. OKIE DOKIE...whatever.