Can any FOC or sinewave sensorless controller handle an astro 3220?

neptronix

neptronix

Administrator
Staff member
Joined
Jun 15, 2010
Messages
17,719
Location
Utah, USA
Just curious, as I know that we can do better than the old castle creations units today.

I want to use maybe 3kw continuous at most, 5kw peak and spin up to 550rpm unloaded, which would translate to something like 58k eRPM unloaded.

My eye is on an asi bac2000 at the moment, because I know it can do sensorless mode and fairly high erpm
 
Well, technically the SFOC5 by Incememed (Hall Sensors B Gone thread) could do it, though I have only used it on DD hubmotors so far, with limited success (the MXUS hubs kinda suck in physical design, plus I don't have accurate numbers for the motor specs to input into the SFOC's settings--when I got it working it was great until I broke the axle...twice so far, from torque)

From the manual:

Supply voltage
- Max 84 VDC (HV version 126 VDC 1)
- Min 18 VDC
Max phase current2 360 Apeak
Off-state supply current3 600 µA (max)
Max e-RPM 100k
Weight 800 g
Enclosure 140 x 87 x 47 mm
 
The website says 8 pole, but i looked up the specs on the ASI series ( phase runner ), and apparently ASI says 60,000 eRPM is the breaking point.

[youtube]EB3lNG5ykgs[/youtube]

[youtube]6ojifdokMKw[/youtube]

Seems that the phaserunner can run these biggies so i imagine the larger ASI controllers can handle it too.
 
amberwolf said:
Well, technically the SFOC5 by Incememed (Hall Sensors B Gone thread) could do it
Click to expand...

Where can one be purchased?
 
You'd have to ask incememed in the Hall Sensors B Gone thread to see if he has them available.
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680
mine is a test unit so I can help him make it better for situations like mine, with heavy cargo/bikes/hubmotors/etc.
 
I see. the application is for long distance touring though, so i don't think i want to place my bets on a beta version controller.
 
understandable

gotta say at this point even if it's "beta" it's way better than the many generic commercial ebike controllers i've used. :lol:

if i could fix my axle problems os i would know the torque couldn't break the motors, and we can resolve the startup-from-a-stop issues, and enable the negative-torque-braking, then i'd love to just run both motors on the trike with just the SFOC5, and rely completely on it for my work commute and regular riding around.

you can ask larsb about his expereinces with it too, he's using it for a revolt middrive, iirc.


the one thing i'd really like to change about it is the user-interface for programming and realtime info access; it's usable but not very easy. but need a programmer to write something to do those things; cant' do it myself.
 
BAC2000 sees my astro 3230 (yes) inductance / resistance as 10 uh / mohms respectively. It can spin that thing really, really fast.
 
can it spin really fast AND apply power to the road without sputtering? :)

3230.. you are very, very lucky to have one of those..
 
I sold one of my 18 fet vescs to a forum member here who is using it to power a delta wound 3220. Not sure if it is up and running but build thread is here https://endless-sphere.com/forums/viewtopic.php?f=28&t=72499

The 60k erpm is a fairly conservative limit from ASI. The cpu doesnt crap out and crash or anything. In any case, 4 pole pairs is only 40000 erpm at 10k shaft rpm.
 
flat tire said:
BAC2000 sees my astro 3230 (yes) inductance / resistance as 10 uh / mohms respectively. It can spin that thing really, really fast.
Click to expand...

3230? I thought there were only 2? I have one and last I heard, the other wasn't for sale.
 
I have been running a BAC2000 with a Astro 3220 for around 2 years on an Electrathon racing vehicle. I use a 66.6 volt nominal lithium pack. It all seems to work well together. The ASI website has enough information to set up Modbus communication, so I am using an Arduino to get data. I also run a max efficiency mode where the Arduino controls the BAC2000 current by Modbus based on RPM.

I am running sensorless. The regen works well. It took a little time to set up as the software is not very user friendly. If you have questions, I am happy to try and help.

Cliff
 
Thanks so much for this reply. What RPMs are you spinning the astro to? i'm wondering if the ASI series controllers can handle the full range of RPM.. IE up to 10k, perhaps beyond?

How's the sensored start?

Also, is it really 8 pole or 4 pole?

And in reality, what would you say the continuous wattage of the motor is?

That is really amazing what you are doing with the modbus.. i don't know such things were even possible with ASI controllers. I have some reading to do. I can handle dealing with complex software though as i am a computer programmer myself..

ProEV said:
I have been running a BAC2000 with a Astro 3220 for around 2 years on an Electrathon racing vehicle. I use a 66.6 volt nominal lithium pack. It all seems to work well together. The ASI website has enough information to set up Modbus communication, so I am using an Arduino to get data. I also run a max efficiency mode where the Arduino controls the BAC2000 current by Modbus based on RPM.

I am running sensorless. The regen works well. It took a little time to set up as the software is not very user friendly. If you have questions, I am happy to try and help.

Cliff
Click to expand...
 
Have you considered a VESC variant? The VESC 6 or VESC 75/300 can handle fairly high ERPM. I think the 75/300 would be better for a low inductance motor because you could set the PWM frequency higher.

This video shows the 75/300 running a low inductance high ERPM motor: https://www.youtube.com/watch?v=fbK2dcoYS7g
But I see that you've already commented on that video.
 
Astros are usually 8 poles / 4 pole pairs. They have that info on their site. Mine is 8 poles / 4 pole pairs.

Continuous wattage thru an astro enters the territory of unbelievable but they need active cooling. Water cooling the case is supposed to be best.

Yes the BAC2000 will spin in excess of 10K on 4 pole pairs, but I don't want to spin this particular motor to its limits to see if it does 15K+.
 
vesc 6 is said to do up to 150k eRPM, so it could even drive the difficult yet cheap turnigy rotomax which is appealing, as that motor has 0.2mm lams. of course you could only extract 90% of it's power as it's rated to spin to 178k eRPM, but i am okay with only getting 8500w peak out of a 5.5lb motor.. :lol:

I am not sure what lams the astro has. I'm assuming 0.2mm, considering the 95-95% peak efficiency claims. Have been trying to get ahold of someone who can answer basic questions for a few days now.

The 75/300 is certainly an option but it does look like a bit of a hack job. There are a few efforts to build 12FET VESCs, which IMHO would be ideal if they pump out 100A peak. I am holding out for someone to design an ebike oriented controller around the VESC idea. Until then, the ASI units seem to be the ticket. just today, powervelocity replied in a thread stating they were working on a VESC clone.
 
flat tire said:
Continuous wattage thru an astro enters the territory of unbelievable but they need active cooling. Water cooling the case is supposed to be best.
Click to expand...

Are there any astro motor watercooling bits out there? or is it a matter of rolling your own out of pre-existing parts?
If i go with an astro, i think even just an external heatsink that air can flow through would help.
 
flat tire said:
Astros are usually 8 poles / 4 pole pairs. They have that info on their site. Mine is 8 poles / 4 pole pairs.

Continuous wattage thru an astro enters the territory of unbelievable but they need active cooling. Water cooling the case is supposed to be best.

Yes the BAC2000 will spin in excess of 10K on 4 pole pairs, but I don't want to spin this particular motor to its limits to see if it does 15K+.
Click to expand...

Bump again, what 3230 do you have? What wind? Did you get it from Holmes?
 
We try and gear for 12,000 but I think 11,500 is the highest we have run. Electrathon tactics rather than any mechanical or electrical limit.

The sensorless start has a brief delay which costs us at the start of a race. We would like to add sensors and go sensor start to sensorless but have not installed any halls sensors yet.

We see 100 amps from a dead start but immediately drops to 60 amps. 60 * 66 = 3960 watts, so that may be because we have set the limit at 4K in the software where you define your motor. 30 amp in regen which is sent to 50% of maximum.

The motor gets hot to the touch but nowhere near it's limits but Electrathon rules limit energy to 1000 whrs.

neptronix said:
Thanks so much for this reply. What RPMs are you spinning the astro to? i'm wondering if the ASI series controllers can handle the full range of RPM.. IE up to 10k, perhaps beyond?

How's the sensored start?

Also, is it really 8 pole or 4 pole?

And in reality, what would you say the continuous wattage of the motor is?

That is really amazing what you are doing with the modbus.. i don't know such things were even possible with ASI controllers. I have some reading to do. I can handle dealing with complex software though as i am a computer programmer myself..
Click to expand...
 
flat tire said:
3T with ported housing from the davinci dude
Click to expand...

Interesting. I wasn't aware that there ended up being more than the first two. Well at least we know there are three! Thanks for responding.
 
You must log in or register to reply here.

Similar threads

MitchJi
Justin's FOC Sinewave Controller Status Plus Some Info
Replies
14
Views
5,008
bowlofsalad
bowlofsalad
Back
Top