Compact Field Oriented Controller, ASI + Grin, limited run

[Tats]

Thanks, Setting the Threshold to 0.0A gave me what I was looking for.

The EM3EV website says the 33G 14s7p pack can run 35A continuous and 43A burst and 12A charge. Hence my settings.

On the ride today tho the peak Amps was 38.10A and max regen of -8.98A. The regen matches my settings but the max amps is low, that's why I think the smart BMS may be limited. I'll dial the CA back a bit on this as I don't need that much.

I use the bike on PAS mode with a 2.00 V/Sec and a 12 pole PAS I just sometimes use the throttle to pass someone or get across a junction a bit quicker. I'll look at a torque PAS soon tho as Justin has some more options available now.

[magooXI@aol.com]

Just received a Phaserunner controller, so exciting, maybe a bit over my head, I have been reading this thread and on the Grin site for months for education and information purposes.

I have a bike on the stand with a new QS 30H DD motor mounted, AFAIK motor is 1500w with 23 pole pairs, Phaserunner mounted on heatsink, 52v 11 AH Litium battery with 50 amp bms, V2 CA DPS, Half twist throttle.

All wired up ( throttle direct to controller) and spinning motor nicely but in reverse, Phase wires are Y-Y, G-G, B-B, switching G and B results in stuttering/shuddering.

Connect USB-TTL cable to Mac OS El Capitan with Phase Runner Suite and controller connects fine, Auto tune Static test makes motor buzz but then says motor not detected make sure all 3 phase wires are connected to motor windings.

I have tested connections may times but cannot get past this stage.

Motor works well with grinfineon and work well in reverse with phase runner so it seems connected.

I thought I’d give it a shot here before I bring it in for set up.

I will be using a CA V3 but didn’t want to scavenge it off another bike, I will if it helps to make this setup run.

Oh and 6 times blinking is constant.

I have a CA3 / Phaserunner / BBSHD setup. I ran the test initially which spun the wheel properly, but after the 3rd test I can hear the motor during the test.. but the internal gears are not spinning. I swapped motors from my other bike thinking this was the problem.

[kevinscargobike]

Could it just be the reverse checkbox? (I didn’t double check, but I think I recall that being a thing)

[davideserin]

Just received a Phaserunner controller, so exciting, maybe a bit over my head, I have been reading this thread and on the Grin site for months for education and information purposes.

I have a bike on the stand with a new QS 30H DD motor mounted, AFAIK motor is 1500w with 23 pole pairs, Phaserunner mounted on heatsink, 52v 11 AH Litium battery with 50 amp bms, V2 CA DPS, Half twist throttle.

All wired up ( throttle direct to controller) and spinning motor nicely but in reverse, Phase wires are Y-Y, G-G, B-B, switching G and B results in stuttering/shuddering.

Connect USB-TTL cable to Mac OS El Capitan with Phase Runner Suite and controller connects fine, Auto tune Static test makes motor buzz but then says motor not detected make sure all 3 phase wires are connected to motor windings.

I have tested connections may times but cannot get past this stage.

Motor works well with grinfineon and work well in reverse with phase runner so it seems connected.

I thought I’d give it a shot here before I bring it in for set up.

I will be using a CA V3 but didn’t want to scavenge it off another bike, I will if it helps to make this setup run.

Oh and 6 times blinking is constant.

I have a CA3 / Phaserunner / BBSHD setup. I ran the test initially which spun the wheel properly, but after the 3rd test I can hear the motor during the test.. but the internal gears are not spinning. I swapped motors from my other bike thinking this was the problem.

This has been a nightmare. Same result. I am desperate at this point to get this working.. Can someone help me?

[magooXI@aol.com]

Could it just be the reverse checkbox? (I didn’t double check, but I think I recall that being a thing)

Figured it out. The CA wasn't loaded properly with the correct firmware.

Now I'm trying to figure out an ON / OFF switch for 84V / BBSHD / PR / CA3. This thing sparks like crazy every time I disconnect it. Dielectric grease has made the spark even more intense.

[magooXI@aol.com]

My error message came back, it had only gone because the battery voltage went below 84v. The error message is a result of the battery voltage being higher than the Regen voltage start/end range, regardless of whether you're using it or not.

Warnings[8]: VdcHighFLDBK

I also resolved my issue with the G01 (SWXH)) motor. I had mistaken it for another model on Bafang's spec sheets. So with the correct pole count of 44, it tests fine.

It worked fine with the basic settings. I've set it to 40 phase amps and 15 battery amps.

How did you clear the Warnings[8]: VdcHighFLDBK?

[Killer-B]

Could it just be the reverse checkbox? (I didn’t double check, but I think I recall that being a thing)

Figured it out. The CA wasn't loaded properly with the correct firmware.

Now I'm trying to figure out an ON / OFF switch for 84V / BBSHD / PR / CA3. This thing sparks like crazy every time I disconnect it. Dielectric grease has made the spark even more intense.

Have you looked at the switches from ebikes.ca, I have one of them wired into the white and yellow wires coming outta the heatshrink on the throttle connector on the PR.. Works like a charm!

[Marc S.]

Hi Grin Tech,

Since the maximum regulated current available to power external accessory devices from Cycle Analyst v3 is limited, I was wondering if my Phaserunner v1 can power the hall throttle, 3x hall brake sensors, PAS, etc (drawing between 25-30mA combined); if I use a 74V (20s) battery.

What is the max current of the 5V line, the Phaserunner can safely provide at up to 90V?

Thanks!

Best regards

Marc

[justin_le]

What is the max current of the 5V line, the Phaserunner can safely provide at up to 90V?

I will try to get a proper empirical answer to this soon, but suffice to say it's well above what any normal set of ebike accessories (PAS sensors, brake cutoffs, throttles etc) would draw. The 5V output is coming from a DC-DC buck regulator, so it doesn't have the same reduced output at higher voltages that the CA has with its linear regulator.

[justin_le]

Do post the results of this. I'm not sure when I'll get time to do the promised wind tunnel thermal dissipation tests so any data points here will be appreciated.

It turns out the time came to pass in December 2018!

The setup was a fun. You can see the phaserunner zip tied to a diagonal tube representing the downtube on a bicycle. For the load motor, I have two large DD hubs with 45mm wide stators that are bolted to each other. The first motor is hooked up the phaserunner, while the 2nd motor has the output phase wires shorted together in order to provide significant resistance, and a resistance that will increase with the drive RPM. That way as the phaserunner goes into thermal rollback, the torque will decreases and then the load motor RPM will similarly decrease to a value that produces a matching torque.

In my first tests I was using a standard 9C style hub motors with 27mm stators, and that was totally inadequate. It got over 150 degrees in no time, and even with the windings fully shorted couldn't provide sufficient resistance to fully load the phaserunner to get into thermal rollback at the higher wind speeds.

Anyways, with the two 45mm motors back to back with Statorade added it worked great. The results show about 40 amps of continuous phase current when crawling at 5kph. with it increasing to 50 amps at 20 kph and reaching 55A at 45 kph.

Phaerunner Thermal Rollback.jpg (41.87 KiB) Viewed 1628 times

This is with the phaserunner mounted on top of the tube, where the heatsink it is somewhat in the wind shadow of the tubing itself. I was curious to see if there was much difference if the Phaserunner was hanging under the downtube, both with it inverted and the heatisink facing into the wind, or more conventionally done with the heatsink against the frame tubing and having thermal contact to the metal.

Phaserunner Hanging.jpg (146.88 KiB) Viewed 1628 times

Phaserunner Hanging Inverted.jpg (146.8 KiB) Viewed 1628 times

I also did the test with a large finned heatsink attached to the Phaserunner.

Heatsunk Phaserunner.jpg (123.16 KiB) Viewed 1628 times

The results show that the varied orientations of the stock Phaserunner didn't change the effective thermal rollback current, so the lesson here is that you can strap it to the bike however you like, and so long as it's in the air stream it'll give similar continuous phase currents.

Phaserunner Varied Mount Thermal Rollback.jpg (57.4 KiB) Viewed 1628 times

The additional heatsink not surprisingly had a large effect, allowing 60 amps continuous at 10 kph and approaching 70 amps at high speeds. I couldn't actually complete the 40 kph test since the combination of the wind tunnel fan and the Phaserunner power to drive the load motors kept blowing the 20A breaker feeding the lab room.

Hopefully this info is useful for people doing planning. What I realized doing this is that even though the Phaserunner heats up much faster than a hub motor due to its much lower thermal mass, it actually has a higher steady state current capability than all but the very largest DD hub motors. Most common hub motors in the higher power class (MAC, Crystalyte 'H', 9C, Bafang BPM / CST etc) will eventually overheat with phase currents in the 30-40A realm. With these setups longer term hill climbing power limit bottleneck would be coming from the hub motor and not the phaserunner motor controller.

The phaserunner controller will generally go into thermal rollback first, but then the motor will keep getting hotter and hotter until it takes over starts driving the thermal rollback via a CA3 device, which will reduce the phase currents even further than what the PR was limiting it to.

[amberwolf]

The phaserunner controller will generally go into thermal rollback first, but then the motor will keep getting hotter and hotter until it takes over starts driving the thermal rollback via a CA3 device, which will reduce the phase currents even further than what the PR was limiting it to.

Too bad the PR doesn't have a secondary thermistor input (and firmware segment to monitor it, too).

Then it could deal with it all on it's own, no external device needed.

[StinkyGoalieGuy]

Is the 90V max of the Phaserunner a hard cutoff?

If I use LiFe cells with 27 in series, once they settle to their nominal voltage of 3.3v each, that would be 89.1v. But hot off the charger, that would be 3.6v per cell giving 97.2v. My cells seem to settle down to 3.3v after about 20 minutes. Can the Phaserunner handle 97.2 volts for that short period of time?

[davideserin]

Is the 90V max of the Phaserunner a hard cutoff?

If I use LiFe cells with 27 in series, once they settle to their nominal voltage of 3.3v each, that would be 89.1v. But hot off the charger, that would be 3.6v per cell giving 97.2v. My cells seem to settle down to 3.3v after about 20 minutes. Can the Phaserunner handle 97.2 volts for that short period of time?

The absolute max I was able to run on mine was 90.8 before it would go into error and cutoff.

You need to set the Regen value higher also to avoid that throwing up errors/cutting off when the voltage surpasses it.

I was running lithium ion 22S and charged to 90.8 instead of 92.4v which is good for the battery. But it’s worth considering that the Regen won’t work until the battery depletes a bit, so this made me change down to 21S. Funnily enough it suggests 21S as the max in the manual.

I’m running a Mxus 3k turbo at 96A phase and 70A battery limit and it’s working great.


Sent from my iPhone using Tapatalk

[cycborg]

It's been on the Grin website since January, but it hasn't been mentioned here or elsewhere on ES that I can find. There's a new lower-power version of this called the Baserunner which was designed for the Hailong battery base.

Specs are max 55 A phase current and 60 V battery, vs. 93 A and 90 V respectively for the Phaserunner, but at less than 2/3 the price and 2/3 the physical volume, it seems like an excellent choice for applications that don't involve a lot of cargo or hills or NEED MOAR POWER.

Is anyone here using these yet? Anyone from Grin want to chime in?

News Blurb
Product page

[neptronix]

Looks like you may have to be the guinea pig

Quick Q about field weakening.. is there any way to selectively enable and disable it?

[tomjasz]

Have one for a new battery build but a bit out of my knowledge base. Watching with keen eyes...

[cycborg]

Quick Q about field weakening.. is there any way to selectively enable and disable it?

It's trivial to do this in the config program. If you're talking about handlebar control, you can't do it via CR or a simple switch. You can change any settings via MODBUS if you code up an Arduino or something.

[davideserin]

Quick Q about field weakening.. is there any way to selectively enable and disable it?

It's trivial to do this in the config program. If you're talking about handlebar control, you can't do it via CR or a simple switch. You can change any settings via MODBUS if you code up an Arduino or something.

I’d like to try this, are there any instructions anywhere?


Sent from my iPhone using Tapatalk

[cycborg]

I’d like to try this, are there any instructions anywhere?

Not sure what "this" means, but if you want to implement field weakening, it's on the Advanced tab in the config program, and if you search this thread for "weakening" there's been a fair amount of discussion about it, including detailed testing by Justin.

If you're talking about MODBUS, there's a lot less to go on there, so you'll need to do more of a deep dive, but it's been discussed a bit in this thread, so search for that as well. Also, this thread has some detail.

[neptronix]

Thanks a bunch for that thread.

[drdrs]

I put together a few examples that communicate with an ASI controller using their MODBUS implementation on a computer using a python installation along with some example log outputs, byte listings, and descriptions. In the future, I'll also put up some arduino code that performs the same thing.

https://github.com/dsoto/modbus-examples

Once I climbed the learning curve, I've been able to build a few Arduino dashboards using these commands.

[justin_le]

Hey everyone, we're getting very close to having a V1.1 release of the Phaserunner software suite ready to release. This can be downloaded for people to beta test from the following links:

Windows Versions:
https://www.dropbox.com/s/jojzx9fx1xodi ... 0.exe?dl=0
MacOS Version:
https://www.dropbox.com/s/ecpawbjsvubgr ... 0.dmg?dl=0

There have been a number of tweaks on the back-end that we are hoping will fix the random communication errors or other glitches that some people would experience during the autotune process, and we've also updated the error dialogs when they do show up to be much more informative. So if you are someone who has a setup that has been fussy with the V1.05 software we'd be very interested to know if all the issues go away with the 1.1 beta release above.

In addition to that, we've also changed some things around on the layout and added a few features. First and foremost is something that I've alluded to on our GMAC product page which we're calling virtual electronic freewheeling.

Virtual Freewheeling.jpg (82.06 KiB) Viewed 731 times

This causes the Phaserunner controller to inject a small amount of phase current through the motor winding even when the throttle is off. As soon as the motor stops turning (based on the stall timeout) then the current drops to zero properly so you don't have to worry about it wasting power when the bike is parked.

For systems that have an always engaged motor (direct drive or a geared motor with a locked clutch), what you can do is tweak this value so that it's just about cancels out the drag torque of the motor itself. Then when you let go of the throttle to pedal unassisted, there will still be a small ~10-30 watts of power going through the motor in order to cancel the drag, and the experience for the rider is exactly the same as if the motor was actually freewheeling. The downside is that this draws a bit of power from the battery (unlike an actual freewheel), but on the upside you'll almost always get more energy back from regenerative braking than you expend on this virtual freewheeling behavior.

That's our main intention of the feature, but there is another use with mid-drive or geared motor setups in order to keep the motor always engaged with the drivetrain. That way there is no windup delay and jarring kick once the motor spins and catches up with the clutch, which has been the source of many premature mechanical features.

One tweak on the layout side is that we've made the corresponding Nm of torque visible when setting the phase current limits. This way it's easy to know what are appropriate current limits based on what a manufacturer suggests for the maximum torque for the gears, or for a given torque arm or dropout spec.

Nm Display.jpg (54.19 KiB) Viewed 731 times

For instance, our spinout testing showed that with aluminum dropouts and a tabbed washer system you probably don't want more than 40 Nm on the axle ( see viewtopic.php?f=2&t=93562&start=100#p1374044 ). With this feature you can simply drag the max phase current setting down to the point where you see it correspond to 40 Nm of torque without having to do the computation based on the KV.

Anyways we'd appreciate any test feedback from both Windows and MacOS users currently using a Phaserunner or Baserunner. There are a few more small updates (like missing tooltips) that aren't yet in this 1.1 beta release but most of the core function enhancements are there to check out!

[davideserin]

This is a nice update, thanks. I’ll have a test ASAP.


Sent from my iPhone using Tapatalk

[justin_le]

This is a nice update, thanks. I’ll have a test ASAP.

Apparently there was a chance of a memory leak with the beta build that I posted the other day so use these links here instead

Windows:
https://www.dropbox.com/s/bh92ddhvsh2yt ... 0.exe?dl=0

MacOS:
https://www.dropbox.com/s/fzyk4iq9ydfu2 ... 0.dmg?dl=0

We're really interested both to hear from people testing virtual electronic freewheel feature, and also for people who've had communication issues with their computer using the 1.05 release to see if those are resolved in this one.

[rowbiker]

"... learning to be a dad ..."

First of all, *Congratulations!*, and second of all, link to details??