Buick Lacrosse 2012 e-assist moto-alternator

Rube said:
e-vektor said:
There are lots of dyno types; inertial (only mass, but not capable of performing steady tests), water braked, eddy current water cooled, friction dynos, etc.
Hi E-vektor, from your posts I read you have experience with dynometers. I'm interested in a DIY solution for an eddy current braked dyno. Have you observed successful amateur dyno in the power range of common electric motors say 1 to 50kw? In particular, I'm curious about the tools needed to log outputs. Apologies if this is off topic.

basically you need the retarder, a sensor for speed (hall sensor or encoder), a sensor for force/torque (load cell), and a power supply to control the brake.
The ideal way to control the brake is with a PID controller, so you can get steady conditions and also perform sweep / ramp tests.

In some old dynamometers you could see a potentiometer to provide a direct torque to brake.
Indeed, electric motors are a special case as they can sometimes be tested from high rpm until it stalls by increasing the brake's torque. This normally works with BLDC controllers, since motor torque increases as speed decreases, which leads to a balanced condition as the operator increases brake torque.

But I am not sure if it will work with vector control as torque is more flat in big areas, so speed would be unstable at all those areas (say from 50% speed to 0% speed)

I use the SP5 kit from my company which includes a built-in PID controller, to ease both steady and ramp tests.
 
I picked up one of these motors, is there a 10 pole resolvers for dummies video somewhere?
 
well I've got a function generator on the way, guess I will have to experiment. I get groups of 3 for poles, just not 10 poles.

looking at: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ResolverToQuadratureConverter
taking the samples at fixed points in the excitation cycle makes a lot of sense, as does using a board with 2+ adc.
 
it is difficult to do it with code due to the high speed pulses

there is an Analog Devices chip that does it for about $15, at high speed and without any error

http://www.analog.com/media/en/technical-documentation/data-sheets/AD2S1210.pdf

dcb said:
well I've got a function generator on the way, guess I will have to experiment. I get groups of 3 for poles, just not 10 poles.

looking at: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ResolverToQuadratureConverter
taking the samples at fixed points in the excitation cycle makes a lot of sense, as does using a board with 2+ adc.
 
first off thanks for all this vector, great stuff here. Very rare to see someone actually dyno.

I've seen the AD2S1210, but would still be interested in what a $10 dual channel adc stm board can do (and I prefer software problems to high pin count smd and pcb manufacturer), though a simple encoder is looking very tempting :)
 
I have done the first test with Motenergy ME1403 + Sevcon size4, the goal is to measure the TRUE max power of this motor, as some people say it can provide 60 HP and others say much lower....
By now we got only 12 HP at 5000 rpm (75V)

[youtube]5H20iD-CvtA[/youtube]
 
do you have a CSV file by any chance?

e-vektor said:
Motenergy ME115

I've not heard of a 115, it looks a bit like a water cooled me1304 though.

motenergy has a me1304 72v graph which looks like about 12hp (at 140 amps peak).
https://sep.yimg.com/ty/cdn/yhst-129399866319704/ME1304dynoplot.pdf

from here:
http://www.motenergy.com/mepmwacomo.html

says it can do 125A continuous air cooled, or 180 continuous 450 peak with 6 liters/minute liquid cooling. Battery current may be more.
 
Yes it is ME1403, sorry
It came configured for 160 Amp AC
By now I managed to increase current to 420 Amp, and stall torque to 76 nm, which is in line with ME spec. But the problem now is that current falls very fast as the motor accelerates, then power still is low. I think I have to setup the FOC settings
I will continue on Monday, now I am having a long weekend

I'll upload more videos, pictures and csv charts if a manage to do them

dcb said:
do you have a CSV file by any chance?

e-vektor said:
Motenergy ME115

I've not heard of a 115, it looks a bit like a water cooled me1304 though.

motenergy has a me1304 72v graph which looks like about 12hp (at 140 amps peak).
https://sep.yimg.com/ty/cdn/yhst-129399866319704/ME1304dynoplot.pdf

from here:
http://www.motenergy.com/mepmwacomo.html

says it can do 125A continuous air cooled, or 180 continuous 450 peak with 6 liters/minute liquid cooling. Battery current may be more.
 
I *think* with pmac it is more about field weakening (which can be done w/FOC), change the angle of the synchronized stator to push back on the magnets or something (which I think is why you can have more battery current than motor current in pmac, it is kind of a boost converter), and hope your controller doesn't fail at high rpm. The lacrosse induction motor doesn't have that issue.

Have you sat through dave wilsons series?
https://training.ti.com/teaching-old-motors-new-tricks-part-1-introduction-motor-control-pi-controllers-pid-controllers-and
 
The point is that while rpm is lower than nominal speed the current should be constant (if FOC is well setup) and when rpm is higher as the back emf is higher then the field weakening should start and current has to decrease.
Something like this:

1682147d214d1e617551359320d2ee56.jpg


Nominal speed with this motor should be about 3000 rpm at 69v
But I know very few about FW in PMAC. With induction it is automatic, indeed the lacrosse motor has very low nominal rpm, about 1100 rpm in star configuration

Why should sevcon fail? Do you mean it can break?

dcb said:
I *think* with pmac it is more about field weakening (which can be done w/FOC), change the angle of the synchronized stator to push back on the magnets or something (which I think is why you can have more battery current than motor current in pmac, it is kind of a boost converter), and hope your controller doesn't fail at high rpm. The lacrosse induction motor doesn't have that issue.

Have you sat through dave wilsons series?
https://training.ti.com/teaching-old-motors-new-tricks-part-1-introduction-motor-control-pi-controllers-pid-controllers-and
 
Has anyone made or seen any effort to use the OEM inverter, or at least some of it's components, for the eAssist Gen1 motor?

It is in a somewhat unusual package in that the DC to DC converter and inverter are in the same case and the common controller board for the two is ~60% of it with the other being the IGBT section of the power stage. There are two other boards in the case. One is the power side of the DC to DC and the other I think you would call the filter section for the inverter. You can see most of this in the pictures below. What is missing is the power side of the DC to DC. It was basically destroyed in exploring this unit.

The inverter has some pretty nice components on it. The IGBT module and the two current sensors would be easy to remove I think. The IGBT is a real nice looking Infineon FS400R07A1E3 and the current sensors are LEM HC5F400-S.

I picked up 6 of these units, mostly for the 1kWh battery in the package, and 6 of the motors. Didn't pay much for them and they are easy to get so I don't feel bad destroying most of one in the name of DIY science.

Dropbox link to the pictures because only one of them actually shows up for me...

https://www.dropbox.com/sh/ygvy0u7b080fvc1/AACgHFFrCH1PK6s5H8k-TTQna?dl=0
 

Attachments

  • DSC_1108_01.jpg
    DSC_1108_01.jpg
    100.4 KB · Views: 3,331
  • DSC_1109_01.jpg
    DSC_1109_01.jpg
    154 KB · Views: 3,331
  • DSC_1110_01.jpg
    DSC_1110_01.jpg
    204.3 KB · Views: 3,331
  • DSC_1111_01.jpg
    DSC_1111_01.jpg
    244.5 KB · Views: 3,331
  • DSC_1112_01.jpg
    DSC_1112_01.jpg
    211.8 KB · Views: 3,331
  • DSC_1113_01.jpg
    DSC_1113_01.jpg
    209.8 KB · Views: 3,331
  • DSC_1114_01.jpg
    DSC_1114_01.jpg
    209.8 KB · Views: 3,761
  • DSC_1115_01.jpg
    DSC_1115_01.jpg
    153.4 KB · Views: 3,331
superb pictures!
400A (IGBT) seems good for this motor, although in the Delta setup it can manage more (but with less efficiency)

BTW: I uploaded some videos of e-vehicle testing on a rolling road dyno
e-kart:
[youtube]VThKHwb1Tv4[/youtube]

e-ATV:
[youtube]ksymUROLMeo[/youtube]

themelon said:
Has anyone made or seen any effort to use the OEM inverter, or at least some of it's components, for the eAssist Gen1 motor?

It is in a somewhat unusual package in that the DC to DC converter and inverter are in the same case and the common controller board for the two is ~60% of it with the other being the IGBT section of the power stage. There are two other boards in the case. One is the power side of the DC to DC and the other I think you would call the filter section for the inverter. You can see most of this in the pictures below. What is missing is the power side of the DC to DC. It was basically destroyed in exploring this unit.

The inverter has some pretty nice components on it. The IGBT module and the two current sensors would be easy to remove I think. The IGBT is a real nice looking Infineon FS400R07A1E3 and the current sensors are LEM HC5F400-S.

I picked up 6 of these units, mostly for the 1kWh battery in the package, and 6 of the motors. Didn't pay much for them and they are easy to get so I don't feel bad destroying most of one in the name of DIY science.

Dropbox link to the pictures because only one of them actually shows up for me...

https://www.dropbox.com/sh/ygvy0u7b080fvc1/AACgHFFrCH1PK6s5H8k-TTQna?dl=0
 
e-vektor said:
superb pictures!
400A (IGBT) seems good for this motor, although in the Delta setup it can manage more (but with less efficiency)

BTW: I uploaded some videos of e-vehicle testing on a rolling road dyno
e-kart: https://youtu.be/VThKHwb1Tv4
e-ATV https://www.youtube.com/watch?v=ksymUROLMeo


Thanks! I figured it would be worth a few extra minutes to break out the D5600 DSLR instead of using grainy cell phone pics.

From reading your thread Delta definitely seems the way to go. I actually have 2 projects in mind currently for these motors. Both of them are dual motor setups. Although with the power you are getting out of the motor in Delta the ATV project might change to a single motor. The other is more of a remote controlled utility chassis, for lack of a better term, and will always be low speed so it will likely stay dual motor, one per wheel on whatever axle I decide is powered. It will need to be able to move 1-2 tons so it may actually end up being hydraulically motivated with the eAssist motors driving the pumps.

The ATV, based on a Polaris Sportsman style, I had initially planned one motor front and one rear using a 2 speed gearbox to directly drive the differential. But again, seeing the power you are getting one may just do the job as I do not need high speed. It will largely depend on efficiency losses introduced by using the existing transmission/pilot shaft to differential setup and if I can convince the OEM electronic controls for things like 4WD engagement, hill decent control, etc to work without an OEM ECM. Both of them are largely mechanical in operation and simply on/off to make them available so I don't see that as being a big hurdle. Perhaps hydraulic motivation would be applicable here as well but that would make any sort of regen almost impossible, or at least very complicated, from what I know of hydraulics. And there is potential for quite a bit of regen going downhill, especially if I make Active Decent Control all electronic controlled.

Right now I am planning on either 24s or 36s using modules from Chevy Volts, probably at least 2p on top of the internal 3p that they already have. So I will have either ~89v or 133v nominal to play with and easily 5-700A peak discharge.

I'll put up my own thread once I get closer to actually starting...
 
many ATVs are 40 HP or below, for instance LT-Z 400 was 36 HP, Kymco 500 (AWD) was 36, only the 600+ cc engines normally had 50 HP or more. For me in an ATV the goal is to have at least the same thrust than weight (without driver) even if it makes the final speed to be a little low. For instance the banshee chasis weights about 170 kg, and I got a thrust of 188 kg in the star configuration, and probabl 175 kg in the delta one. This ensures that the ATV could climb high slopes, athough maybe not as high as the original ATV. For instance a small Kymco MXU 250 has 180 kg thrust, but LT-Z 400 has more than 210 kg at 1st gear


themelon said:
e-vektor said:
superb pictures!
400A (IGBT) seems good for this motor, although in the Delta setup it can manage more (but with less efficiency)

BTW: I uploaded some videos of e-vehicle testing on a rolling road dyno
e-kart: https://youtu.be/VThKHwb1Tv4
e-ATV https://www.youtube.com/watch?v=ksymUROLMeo


Thanks! I figured it would be worth a few extra minutes to break out the D5600 DSLR instead of using grainy cell phone pics.

From reading your thread Delta definitely seems the way to go. I actually have 2 projects in mind currently for these motors. Both of them are dual motor setups. Although with the power you are getting out of the motor in Delta the ATV project might change to a single motor. The other is more of a remote controlled utility chassis, for lack of a better term, and will always be low speed so it will likely stay dual motor, one per wheel on whatever axle I decide is powered. It will need to be able to move 1-2 tons so it may actually end up being hydraulically motivated with the eAssist motors driving the pumps.

The ATV, based on a Polaris Sportsman style, I had initially planned one motor front and one rear using a 2 speed gearbox to directly drive the differential. But again, seeing the power you are getting one may just do the job as I do not need high speed. It will largely depend on efficiency losses introduced by using the existing transmission/pilot shaft to differential setup and if I can convince the OEM electronic controls for things like 4WD engagement, hill decent control, etc to work without an OEM ECM. Both of them are largely mechanical in operation and simply on/off to make them available so I don't see that as being a big hurdle. Perhaps hydraulic motivation would be applicable here as well but that would make any sort of regen almost impossible, or at least very complicated, from what I know of hydraulics. And there is potential for quite a bit of regen going downhill, especially if I make Active Decent Control all electronic controlled.

Right now I am planning on either 24s or 36s using modules from Chevy Volts, probably at least 2p on top of the internal 3p that they already have. So I will have either ~89v or 133v nominal to play with and easily 5-700A peak discharge.

I'll put up my own thread once I get closer to actually starting...
 
e-vektor said:
many ATVs are 40 HP or below, for instance LT-Z 400 was 36 HP, Kymco 500 (AWD) was 36, only the 600+ cc engines normally had 50 HP or more. For me in an ATV the goal is to have at least the same thrust than weight (without driver) even if it makes the final speed to be a little low. For instance the banshee chasis weights about 170 kg, and I got a thrust of 188 kg in the star configuration, and probabl 175 kg in the delta one. This ensures that the ATV could climb high slopes, athough maybe not as high as the original ATV. For instance a small Kymco MXU 250 has 180 kg thrust, but LT-Z 400 has more than 210 kg at 1st gear

My 1999 Sportsman 500 had about 25hp and would go anywhere I ever tried to go. It weighed about 650lbs or so.

My current 2012 Sportsman 550 x2 has about 45hp and will go anywhere with reckless abandon, including 75mph which is totally unnecessary and unstable! With all the aftermarket gear I have on it the weight is about 900lbs and it's a 2 seater with a dump bed.

I'm planning to find a 2010-2013 ish single seat. It is amazing how many there are on the market in the fall here for under 3000usd. I want that year range because I know the driveline pretty well and it has double wishbone front suspension instead of strut style that the older ones had and that they went back to in '14 or '15.

Actually the rear differential in the units outside of that range was an integrated part of the engine/transmission unit. The ones I'm looking for have separate front and rear differentials using pilot shafts out of the gearbox. That was the other main reason I was thinking about.
 
Would you be willing to share your VCL/settings? Possibly sell them? I'd love to make and market a kit using the Curtis and the e-assist motor.
 
Sure, but they don't include internal FOC settings, right? User still needs to run the characterization process

coleasterling said:
Would you be willing to share your VCL/settings? Possibly sell them? I'd love to make and market a kit using the Curtis and the e-assist motor.
 
does anybody know how to replicate / store / export the characterization?

coleasterling said:
That's a good question. I've never messed with a custom VCL or characterization.
 
Here is a params file from the STAR configuration (max 20 HP), but as explained before there are no inductance, resistance, etc parameters inside, only user parameters (torque mode, throttle voltages, etc),
The only special characterization values are
* Slip Gain, which was already adjusted to 16% with the dynamometer for max torque.
* And regarding field weakening, I discovered that for this motor the best output was when the F.W. was "natural" (OFF) and not driven by the controller

http://www.sportdevices.com/uploads/pics/PARAM_curtis_GM_20 HP_quad.xls

(note that it is not a true XLS file, but CSV renamed to XLS, by the Curtis SW)



e-vektor said:
does anybody know how to replicate / store / export the characterization?

coleasterling said:
That's a good question. I've never messed with a custom VCL or characterization.
 
So Im new to the electric motor setups and I've been wanting to do a electric boost system project for some time. I just bought a Gen 2 EAssist motor/Gen. I want to put it on my Miata as a boost system and the have a ReGen system. Most of the stuff you guys are talking about is way over my head, but I want to learn more about how to make this work. Thanks e-vektor, keep up the good work
 
96Miata said:
So Im new to the electric motor setups and I've been wanting to do a electric boost system project for some time. I just bought a Gen 2 EAssist motor/Gen. I want to put it on my Miata as a boost system and the have a ReGen system. Most of the stuff you guys are talking about is way over my head, but I want to learn more about how to make this work. Thanks e-vektor, keep up the good work

I would suggest starting a new topic on this, referencing this thread as a starting point, since it is a very different application of the 2nd gen of this motor, not the first being discussed here. I know from the electronics side the Gen2 is vastly different from the Gen1 eAssist setup. I have not looked at the motor itself though.

What you are talking about doing is somewhat similar to a long term project I am working on though. I am working out a way to integrate a Chevy Spark EV trans-axle into the drive train of my 2004 GMC 2500HD Duramax pickup. My current plan, for the mechanical side at least, is to connect it to the front axle output side of the 4WD transfer case. The Spark motor is not very big but it puts out 400 foot pounds as Chevy has it setup. It is a permanent magnet motor though, not inductive like the eAssist unit.
 
Back
Top