What are the differences between CAv3 and ERT controllers?

As the title says, I can't understand the differences between the bare CAv3 (which I believe is a display that shows all parameters read from the motor) and this ERT controller: https://electricrt.myshopify.com/products/ultimate-bbshd-upgrade-grin-phase-runner-foc-sinewave-controller-plug-and-play

Is it a external controller replacement for Bafang BBSHD controller, with a CAv3 integrated? But then why they also say "Base price does not include CAv3 although it is highly recommended". If it is "integrated" as they say "Full Cycle Analyst V3 Integration", how can not be included? I'm a bit confused here...

Also, what's the difference with this othe ERT controller for BBSHD? https://electricrt.myshopify.com/collections/controllers-1/products/bafang-bbshd-sinewave-external-controller-upgrade-72v-40a It looks just the first one is a "compact" version, while this one is bigger.

Then you should use XPD software to program the controller settings, and CAv3 software to program the display settings, right?

Cycle Analyst integration usually means there a plug hanging out of the controller to plug the CA into, but the CA isn't included. The second controller is a bigger more powerful one, that goes to 72v instead of just 60v.

Voltron said:

Cycle Analyst integration usually means there a plug hanging out of the controller to plug the CA into, but the CA isn't included. The second controller is a bigger more powerful one, that goes to 72v instead of just 60v.

Click to expand...

Ok, still the more compact one costs double the bigger one.

What I can't understand is: which is the controller and which is the display (CAv3) in their image? https://i.imgur.com/ee52JZh.jpg
Is this maybe the controller (shown in video): https://i.imgur.com/oKiExBC.jpg ?

I expect the controller to be a rectangle metal box with a PCB inside, but I just see a bunch of cables and CAv3 display in that picture.

Its a terrible picture of the controller...



But its more expensive due to the miniaturization and clear thermal potting etc, as opposed to big boxy regular controllers.

And CA mounted...




Now I'm confused.. I was just looking at the images... but it says the Phaserunner goes to 90v in the text?

racingame said:

As the title says, I can't understand the differences between the bare CAv3 (which I believe is a display that shows all parameters read from the motor)

Click to expand...

If you go to the http://ebikes.ca site and look at the CA pages, you can see all the details of what it is and does.

But it does nto read anything from the motor (except if it has a thermal sensor, and if you use the hall sensors in a DD hubmotor in the wheel to detect speed rather than an external wheelsensor).

It does not read anything from the controller either (except the voltage on the shunt inside if you use that vs a standalone (SA) shunt), nor does it program the controller in any way.

What it does do is take various sensor inputs and compare them to whatever settings you've got setup in it for them, and then control the throttle of the controller to have it perform the way you prefer it to be customized.

The smarter the controller, the less necessary the CA controls can be (and in fact you can have conflicts between the controller doing it's own internal rollbacks/etc and what the CA is trying to do, so the dumber the controller the easier it is to setup the CA to make the bike work exactly like you want it to).

But then why they also say "Base price does not include CAv3 although it is highly recommended". If it is "integrated" as they say "Full Cycle Analyst V3 Integration", how can not be included?

Click to expand...

That just means they include the connectors for the CA to plug directly in, so you don't have to add or modify the wiring of either one to make them work (because there are no standards for wiring or connectors in the ebike world).

You'd still need to setup the CA to work with it and make it operate the way you prefer. There's a fair bit of setup work to doing that, so if they provide a pre-setup CA you'd want to get that. If they don't, then you could get the CA anywhere and then set it up yourself following the Guide (UUG) by Teklektik, linked on the Grin site from the CA pages and here on ES in various places. There are threads about BBS* and Phaserunner that may have helpful info in doing the setup.

Then you should use XPD software to program the controller settings, and CAv3 software to program the display settings, right?

Click to expand...

XPD only works on a few of the generic controllers, that happen to have the right MCU chip in them, and the right firmware to talk to it.

So to program things like the Phaserunner (or any other advanced controller), you use the software specific to that model of controller, from that controller's manufacturer.


Note taht the Phaserunner is not more expensive just because it's smaller and cleaner setup and potted. It also has advanced control features that can make it better at controlling a specific motor than a generic type would be able to--but it also is more complicated to setup for any specific motor, so unless it comes pre-programmed for the motor you're going to use it for, it's a very DIY-type of thing to get it working well. See the Phaserunner for-sale thread here on ES for some examples.

Thank you Voltron for the photo depicting the controller.
But mostly, huge thank you to amberwolf for your very detailed explanation, I got a clearer view now.

Just a few other questions:
I saw CAv2 used to connect to the controller being just a "display" instrument, while the CAv3 directly connects to all the components, and also the controller.



What's the main difference in this new approach? From what I could understand, you are not just able to display the inputs taken from the various components, but you can also change the parameters to directly control them from the CAv3 now. The CAv3 is basically a proxy, so since all sensors inputs have to go through it, it's able to read all values directly instead that asking to the controller and can also either validate them or even change them before those data are sent to the actual controller, is that right?

The thing is that, as you said, this sounds redundant to me, since the CAv3 kinda acts like a controller on his own, so there are basically both the controller and the CAv3 telling the motor parts what to do. I guess that, being the phaserunner designed to work in pair with CAv3, there shouldn't be any kind of interfere in doing their job, rather a synergy in achieving a better result overall tho.

What association there is between GrinTech Phaserunner and ASI controllers (http://accelerated-systems.com/products/ebikes/)? Are they based on the same PCB or what exactly? Since I see also their PDF/software on the Phaserunner site. Just two Canadian companies doing the same thing: compact controllers? Also, why ASI specifies the Wattage for each controller, but there is none for the Phaserunner instead?

In short practical terms what's the difference between the CA3-DP and CA3-DPS?

Any guide if I would like to set up a Bafang BBSHD motor by myself both for the Phaserunner controller config part and the CAv3 "display" config part?

racingame said:

I saw CAv2 used to connect to the controller being just a "display" instrument, while the CAv3 directly connects to all the components, and also the controller.

Click to expand...

THey both connect the same way, just that the v3 has several sensor inputs, and more control routines/settings to deal with them to more precisely customize the control of your system to exactly how you want it to operate, and an analog throttle control output vs a cutoff-throttle-output.

From what I could understand, you are not just able to display the inputs taken from the various components, but you can also change the parameters to directly control them from the CAv3 now. The CAv3 is basically a proxy, so since all sensors inputs have to go through it, it's able to read all values directly instead that asking to the controller and can also either validate them or even change them before those data are sent to the actual controller, is that right?

Click to expand...

Neither one has any communication with the controller, period. Neither can ask the controller for anything at all.

All they do is take the sensor inputs and display various standard info (watts, volts, amps, etc) and some customizable stuff depending on the version. V3 has more inputs than v2 (which is just the shunt, IIRC, for volts and amps and any derivable info).

Then they process that info based on your settings and send it out to the throttle input of the controller in the case of the v3, and IIRC v2 just grounds out the throttle signal to disable the controller from doing anything when a limit is exceeded.

NO data is sent to the controller, none of the sensors that go into the CA have any of their info going to the controller. Only the throttle output voltage from the v3 goes to the cntroller.

FWIW, controllers that do have any kind of communication all have their own ways of doing it, no standards, so it'd be tough to make one device to talk to all of them (especially since they're all proprietary so you'd have to figure out each new model as it comes out every day/week/month/etc and then update the CA to make it work with those. Not practical, even if it was possible.

The thing is that, as you said, this sounds redundant to me, since the CAv3 kinda acts like a controller on his own, so there are basically both the controller and the CAv3 telling the motor parts what to do.

Click to expand...

Not really. The CA tells the controller when (v2, v3) and how much (v3) to apply throttle. The controller (if it's got any smarts of it's own) decides what to do about that, based on any other inputs it has.

The cotnroller, dumb or smart, does all the motor control, power handling, etc. If it's a smart controller (like the PR) it can have a lot of stuff that is customizable, to make it ride the way you want it to.

If it's a dumb on elike a tpiypcal generic eibke controller there's not much if anything you can customize, it just does whatever it was built to do when you use the throttle, and if you don't like that you eitehr get used to it and/or learn to handle the thorttle well enough to make it do what you want, or you toss it out and buy a different one until you find one that works the way you like.

BUT: that's where the CA comes in; the v3 you set it up teh way you'd like it to run, then it uses the throttle input to the controller to run it the way you want to. YOu can think of it as adding a "brain" to a dumb controller.

What association there is between GrinTech Phaserunner and ASI controllers

Click to expand...

the PR is a custom-packaged ASI, somewhat reengineered by Grin.

In short practical terms what's the difference between the CA3-DP and CA3-DPS?

Click to expand...

The S just means you have a speed sensor on the wheel instead of using the hall sensor from a DD hubmotor in a wheel.

If you don't have a DD hubmotor in a wheel you require the S version or the CA can't detect bike speed, only motor speed (which is not useful for most info you want to gather/control with).

The SA version is the same as the DP except it uses an external shunt you put between your battery and controller; it's for controllers that don't have a CA plug (most don't).

Any guide if I would like to set up a Bafang BBSHD motor by myself both for the Phaserunner controller config part and the CAv3 "display" config part?

Click to expand...

There's sevearl threads about builds like that, if you look for BBS* and Phaserunner in the title. Could be spelled as one or two words.

Otherwise, there's the UUG by Teklektik, intended to setup *any* system from scratch, which needs to be followed from page 1 to the end of setup (if you skip stuff you think you're not using, there will be things not setup teh way you expect as its' often interrelated).

racingame said:

What's the main difference in this new approach? From what I could understand, you are not just able to display the inputs taken from the various components, but...

Click to expand...

First, either model can be hooked to any controller either by a special CA_DP connector on the controller or lacking that, by means of an external shunt - there is no need to open or internally modify the controller.

The CA2 works by 'limiting' the user throttle that goes directly to the controller. Normally the CA2 does nothing but monitor and display electrical and speed/distance information. Only when one of these monitored parameters goes out of a configured range (speed, current, voltage) does the CA limit the throttle applied by the operator via 'throttle override' to diminish controller power to alleviate voltage sag, excessive current, or high speed.

The CA3 operates very differently. It is a true fly-by-wire system where the operator throttle, PAS unit, temp sensor, speed, and ebrake signals go to the CA - not the controller - and the CA alone runs the controller by means of the throttle connection. The controller is specifically operated in dumb mode with respect to all these features so there is no conflict. If the controller can provide 'torque throttle' then that can work normally, but otherwise the CA runs the bike - not the operator. This means that even the simplest controller can get autocruise, basic or torque-sensing PAS, motor thermal rollback, low voltage cutoff, and speed limiting.

In addition to these features, the CA3 provides throttle ramping, matches throttle and controller input/output voltages so there are no throttle dead-zones and can remap throttle operation to provide current or power throttles which are significantly smoother and more controllable (very similar to 'torque throttle'). If used with Grin controllers, the CA3 provides throttle-controlled proportional regen and automatic regen braking to maintain speed descending hills. The CA3 also provides connections for up to two manual handlebar controls to allow quick manual 0-100% scaling of any of the configured maximums (speed, throttle range, PAS level, current, power). This lets you do on-the-fly adjustments to make the bike perform exactly as you want at any particular moment.

In short, the CA2 can act to override what the operator does while the CA3 adds a variety of more complex behaviors that the controller does not have. (In point of fact, the CA2 does have a basic 'Current Throttle mode' that requires some custom wiring, but that's not really a menu-selectable out-of-box function).

So - they may look the same, but these are very different products.

From what you say telektik, then it would be useless having something as a "smart" controller like PR, since the CAv3 pilots it transforming it into a dummy puppet. What I'd like to understand is: which is the hardest part? Configuring the controller or the CAv3?

From what you said also, the CAv3 is still just a display that reads data from all the sensors and sends nothing to the controller, except the throttle voltage. It checks if parameters are within user-specified ranges, but if are not, what it can do? I understand that if you got a shit controller, it can make it work much better trying to ease his tasks and even allow it to do something that otherwise wouldn't be able to do emulating extra functions that the original controller doesn't support by itself. But in the very specific case of mounting it in combo with a PR, a crazy smart controller, shouldn't be better to let the latter one to handle all the stuff and really use the CAv3 just a simple display to monitor stuff?

- When I finish to configure the controller, what benefits do I see? (considering I haven't even plugged a CAv3)
- When I finish to configure the CAv3 system (I don't even know how to call it at this point, since it's a display, but does way more than just retrieve data) and connect it to the ebike and controller, what benefits do I see as extra?

Going through the PR manual: http://www.ebikes.ca/documents/Phaserunner_Manual_Rev1.0.pdf
and telektik CAv3 guide is enough to having everything fully set up?

Why if the PR is based on ASI BAC 800W, it can deliver even 3KW and more without any issues?

racingame said:

...the CAv3 is still just a display that reads data from all the sensors and sends nothing to the controller, except the throttle voltage. It checks if parameters are within user-specified ranges, but if are not, what it can do?

Click to expand...

You are confusing the resulting functionality with the means of control. I gave you a long list of things that the CA can provide to any controller - some of which are pretty complicated. The fact that the functionality can be achieved only by controlling the throttle is of no consequence.

racingame said:

...But in the very specific case of mounting it in combo with a PR, a crazy smart controller,...

Click to expand...

In the world of controllers, the PR is not that smart - it's pretty much a one-trick pony - it's really good at running a motor - but that's it. Grin took the ASI controller and revised it specifically to work nicely with the CA3. This isn't an exercise in 'what could best be done in the controller' to make a complex and feature-rich controller, it's about which features to put in different products to appeal to the broadest market and to provide a phased upgrade path so folks can leverage previous purchases with low risk and modest added cost. It's about the ability to add a CA3 to a cheap controller or upgrade the controller and keep the CA and all its sensors, switches, and configuration.

You have a lot of questions but this is one of those situations where you need to envision the features you want and sit down with the manuals and see exactly how they can be provided - the devil is in the details. There are lots of ways to build very nice bikes that make their riders very happy - all without PRs and CA3s. You need to discover a mix of parts that makes you happy.