MOTOR PORN !! EMRAX 228 teardown !!

Doctorbass

100 GW
Joined
Apr 8, 2007
Messages
7,495
Location
Quebec, Canada East
get two box of kleenex guys... :twisted:

Enjoy!

http://build-its-inprogress.blogspot.ca/2017/05/emrax-motor-teardown.html?m=1

DSC00249.JPG


Doc
 
I found this quote interesting:

This model (228LV) turns out to be particularly miserable to control. It's combination of ultra-low resistance (1.12 mΩ) and inductance (10 uH) means you would need to switch ~100 kHz at rated voltage, in order to have reasonable amounts of current ripple just from the PWM, which is out of reach of normal silicon MOSFETs or IGBT's of sufficient current and voltage rating. So it's a great motor, but essentially useless unless you can afford to build a GaN-FET or SiC-FET inverter for it.
 
Punx0r said:
I found this quote interesting:

This model (228LV) turns out to be particularly miserable to control. It's combination of ultra-low resistance (1.12 mΩ) and inductance (10 uH) means you would need to switch ~100 kHz at rated voltage, in order to have reasonable amounts of current ripple just from the PWM, which is out of reach of normal silicon MOSFETs or IGBT's of sufficient current and voltage rating. So it's a great motor, but essentially useless unless you can afford to build a GaN-FET or SiC-FET inverter for it.


Yes i have read that too, but that is certainly not for the higher voltage version.. i guess that at 100Kw burst and 100V the 1000A batt current become like 2000A phase.. so that's normal that at 98% eff that the winding resistance and inductance are very low.. these can rotate at 10500rpm!

Doc
 
The thing that surprised me the most about this motor is what appears to be the a very simple method used to retain the stator segments.
 
Glue alone and no mechanical retention for magnets is sketchy. Axials with non-pocketed rotors can/do toss magnets.

Heat path looks pretty terrible, I'm not surprised outside aircraft use they cook fairly easily.
 
Yes, these low weight high RPM motors all strike me as being under-performers in the thermal management department. Perfect application for a glider at 3000 m ASL, but for a car/bike project? It's certainly possible but not without some challenges.
 
I've been looking at these motors in a motorglider.

I'd prefer a higher voltage to get the current down but no one seems to make a high voltage low current ESC for this size range.
 
helno said:
I've been looking at these motors in a motorglider.

I'd prefer a higher voltage to get the current down but no one seems to make a high voltage low current ESC for this size range.

This is a bit like asking for higher drama for approximately no advantages in an EV power train build.

I personally also wouldn't trust my life on that motor, but I respect that everyone has their own comfort levels.
 
For an aircraft higher voltage makes a lot of sense.

I can build a high voltage battery that can easily handle the required current.

If I try to make the same power with ~100 volts then I need a battery that can handle 300+ amps and require a higher amp hour battery not because I need more amp hours but because I need to be able to safely handle the current. Very quickly I run out of weight budget.

As far as safety goes the glider is currently using a high strung two-stroke that is known for breaking cranks. Any electric power train should be able to beat it as far as reliability goes. This particular glider has been flown using a brushed Lynch motor but it is pushing that motor quite hard.
 
If you package it right, your battery, controller and motor all end up with a power harness length approaching zero.
 
That is true but aircraft have rather strict center of gravity limitations so it cannot be a single big lump.

http://i.imgur.com/cxW1aMJ.jpg

Everything has to fit were the existing engine, exhaust, ballast weight, and fuel tank were and maintain the CG. There is about 160-180 lbs available for the entire power system.

This plane is ripe for conversion since it was designed for a larger gasoline engine and has a substantial amount of ballast mounted on the engine to get the CG in the right place. It has a very lightweight engine paired with a heavy steel exhaust and a ballast block that is about the same weight as the engine itself.

The higher voltage EMRAX motors would be a very good fit but a controller that can handle the high voltage is huge since they are designed for 100kw motors rather than the 30 kw peak needed.
 
helno said:
That is true but aircraft have rather strict center of gravity limitations so it cannot be a single big lump.

http://i.imgur.com/cxW1aMJ.jpg

Everything has to fit were the existing engine, exhaust, ballast weight, and fuel tank were and maintain the CG. There is about 160-180 lbs available for the entire power system.

This plane is ripe for conversion since it was designed for a larger gasoline engine and has a substantial amount of ballast mounted on the engine to get the CG in the right place. It has a very lightweight engine paired with a heavy steel exhaust and a ballast block that is about the same weight as the engine itself.

The higher voltage EMRAX motors would be a very good fit but a controller that can handle the high voltage is huge since they are designed for 100kw motors rather than the 30 kw peak needed.

Sounds like a cool project, helno! You should post a thread in the electric aircraft section. We run a Rinehart PM150DZ in Voltron Evo, and the battery sits at 700 V top of charge. High voltages in batteries can be managed, and in my opinion it's no more challenging than a high current, low voltage option.
 
helno said:
If I try to make the same power with ~100 volts then I need a battery that can handle 300+ amps and require a higher amp hour battery not because I need more amp hours but because I need to be able to safely handle the current. Very quickly I run out of weight budget.

In theory at least this shouldn't be an issue. Cells arranged into a battery are constant power, as you rearrange cells from series to parallel you trade off voltage for current by power (watts) and capacity (watt-hours) stays the same.

In sounds like an interesting project and perhaps if you put up some numbers of where you're hitting a roadblock someone may be able to help - or at least appreciate the problem you're facing properly :wink: There are few people on this forum that work at OEM level in the EV industry so sometimes you can get some good nuggets of wisdom that would normally be expensive and hard to come by :)
 
I created a post in the electric airplane section so I don't futher sidetrack the discussion here.

https://endless-sphere.com/forums/viewtopic.php?f=38&t=89000
 
liveforphysics said:
Glue alone and no mechanical retention for magnets is sketchy. Axials with non-pocketed rotors can/do toss magnets.

Heat path looks pretty terrible, I'm not surprised outside aircraft use they cook fairly easily.

Hi Luke, I'm looking to put the 228 LV into a bike build. I'm guessing by your response and knowledge of electric motors that running the Liquid cooled version would be a wise choice in a motorbike. I was keen to avoid the extra complication water cooling brings into the design and build but I'm fairly committed to using this motor now... Have you got any specifics where this motor has cooked? I would be interested to hear of any applications where Emrax motor overheating has occurred as this could help me determine if my design/build with this motor is at risk of overheating.

Punx0r said:
I found this quote interesting:

This model (228LV) turns out to be particularly miserable to control. It's combination of ultra-low resistance (1.12 mΩ) and inductance (10 uH) means you would need to switch ~100 kHz at rated voltage, in order to have reasonable amounts of current ripple just from the PWM, which is out of reach of normal silicon MOSFETs or IGBT's of sufficient current and voltage rating. So it's a great motor, but essentially useless unless you can afford to build a GaN-FET or SiC-FET inverter for it.

I have contacted Emsiso requesting them to address above statement and there EmDrive 500 controller which is recommended by Emrax for the LV motor. I'll report back once I receive a response.

Cheers, Toby
 
It's funny how this motor is generally always the 1st choice for various college projects, but those who use it never select it a second time for an application...

If you've got unlimited time and funding, you can learn all the same lessons in control and heat path challenges for yourself to eventually end up with a neat looking paperweight.

That mass of iron and copper and magnets contains no magic, but does pack many creative ways to fail. Maybe the liquid cooled version solves some thermal issues, still depends on glue for magnets, difficult to control current, and the outside of the can spins which adds new failures and risks. It's not the worst motor, just ends up with a final powertrain system approaching the worst.
 
Ohio State appear to have got their Emrax working to good effect the second time around. It's a popular choice for college teams because its cheap and the catalogue quoted peak power / mass is attractive.
But it's not an easy motor to apply to a bike project, so a whole bunch of extra work needs to be done to make it work.
 
Maybe Toby will find a way. If the em-Drive500 controller was recommended by Emrax most likely they should pair up nice.
As for cooling as the Emrax is an outrunner liquid cooling should be both efficient and relatively easy. A radiator and hi-flow pump will shed a lot of heat. Not to mention if heat still is a problem the now well tested Ferro Fluid and heat sink combo should help too.

I don't know what price the controller Toby are looking into has, but I saw a few expensive ones powering up Emrax 228 on youtube a while back. One australian controller, can't remember the name and a german unitech.

To reach max power for the emrax will come at a premium price, as I think the motor itself is 3K + and those controllers are twice that. Then you need a big battery to juice up that motor.

But hey if you build it Toby, we will watch and follow your thread. That is one light weight motor and controller combo, should make for an insane e-motorcycle.You will be in the Energica Eva class. That Italian beauty also got a 100 kw PM motor.

tyyxkcm.jpg
 
macribs - afaik FF will only work on hubs. FF bridges the air gap between the stator and the magnets. Emrax is an axial motor not radial like a hub so I don't think it would work the same. Although you probably could put FF in the emrax it would probably spin off because the magnets are mounted vertically instead of horizontally and are on the side of the stator rather than around it.
 
Hi guys,

I received a response from emsiso regarding possible control issues of the emrax LV. Although response was brief I was informed that the 228 LV emrax motor paired with the emsiso emdrive 500 controller was being run in the E- raw bike built by essence motorcycles with no reported problems.
 
spinningmagnets said:
The blog that hosted the pics of the teardown when this thread was started, took the pics down. Here's what I could find:



This is amazing how influent the E-S can be on many E-V company! :p

Doc
 
You know I love porn Doc... Good job... But the blog was down.

Magnets:. Glue alone is not acceptable to me. It will age and fail... Or quality in Production will create great risk for failure. Good Production pieces can not fail (Mill Spec).

As for the tiring arguments around making power with Volts and not Amps...
Volts

Yes there is the copper weight issues with big current.
Yes there is the I^2*R issue at every connector and bolt
More importantly is wear and tear on Batteries.

When taking Luke advice it is important to know what he is up to. Luke is an extreme guy... He wants extreme things... Like... A 20Ah battery that can discharge at 2000A.

That's cool... Plenty of need for that... But I see a different need in our sector.

Low cost
High cycle count
Light weight

If I can string cells in series and pull 1C off of them instead of 10C with bursting of 30C... This treats the chemistry better and will result in longer life. I would like my cells to cycle thousands of times... And I have never seen cells last that many cycles while discharging at super high rates.

Not here to argue.
Let's just see what the lifecycle ends up being on Luke's 28S packs. They get pushed pretty hard... Upward of thermal limits. My guess is they will show premature wear.

It is nice to be able to charge fast. That is of super high value.

I can say this:. As lithium batteries get used and abused their internal resistance goes up over time. IF you depend on that IR being low then performance will degrade with time... As users bump against thermal limits.

On the other hand... If you make power with voltage and draw only 1C off your cells... The internal resistance can go up and it will have much less effect (exponentially less).

Drag racing and Motorcycles are one slice of a very big pie. They have strict boundaries and clear goals. These limitations and goals do not nessesary align with the limits and goals of other platforms.

Then there is simple scaling.

420V 1000A Tesla
Do you think you will see 105V 4000A Tesla
No
You won't
Ever

Ohm's law.
To break it is to fail to scale.

-methods
 
This motor is all about the peak efficiency to me.
If you are running close to the 98% peak efficiency, you aren't making a lot of heat in the first place, so the poor thermal path is less of an issue.

Is the thermal path still poor when you run the liquid cooled versions?

I wonder if the company is amenable to changing their design of how the magnets are held. I agree that glued magnets is ridiculous considering the heat issues and the speed at which it rotates..

If i was finally doing my dream project today - a Toyota MR2 with high density 18650 packs from a Tesla - i'd be using one of these motors though. It's short length and high power ratio would allow you to fit the controller and motor into the lower part of the engine bay, allowing you to easily rest a big bank of batteries on the top half of the engine bay.

On a car that originally had a big heavy iron block, the battery, motor, and controller weight in the engine bay could be made just a little bit heavier than the original powertrain, and you could end up with a quick car that handles much like the original.. all thanks to the super low weight of the motor.
 
methods said:
As for the tiring arguments around making power with Volts and not Amps...

Yes there is the copper weight issues with big current.
Yes there is the I^2*R issue at every connector and bolt

If I can string cells in series and pull 1C off of them instead of 10C with bursting of 30C... This treats the chemistry better and will result in longer life. I would like my cells to cycle thousands of times... And I have never seen cells last that many cycles while discharging at super high rates.

I can say this:. As lithium batteries get used and abused their internal resistance goes up over time. IF you depend on that IR being low then performance will degrade with time... As users bump against thermal limits.

On the other hand... If you make power with voltage and draw only 1C off your cells... The internal resistance can go up and it will have much less effect (exponentially less).

Then there is simple scaling.

420V 1000A Tesla
Do you think you will see 105V 4000A Tesla
No
You won't
Ever

Ohm's law.
To break it is to fail to scale.

-methods

I'm super surprised to see such a fundamental misstep in your train of thought methods.. regardless of the series/parallel configuration of any given quantity of cells, for any given power discharged from the pack the C rate is going to be the same. 100v/10ah/1000w = 1c, 10v/100ah/1000w =1c.

While you're correct that cell interconnections add losses, there's plenty of ways to increase total losses with more cells in series VS less in parallel. Good design applies to both scenarios equally. Higher voltages require more substantial insulation/air gap/safety margin, all of which is mass or volume that could be dedicated to more active material in the form of cells or copper connecting cells.

Regarding the Tesla example, they (like any competent engineering team) design the best products they can within the boundaries of competing motivators like cost, complexity etc. In many cases costs and complexity are reduced by using commodity components. For a variety of reasons, there's a lot of historic R&D for voltages around that level (3 phase AC mostly) and that leads to certain 'sweet spots' for delivering any given power level at various voltages. There's no inherent reason I can think of that would mean the tesla pack couldn't be configured to deliver half the voltage with twice the AH. You would require increased conductor area for the pack > inverter connections, however you would require less BMS slaves, less complex monitoring, less insulation. The reason they didn't do this is it falls in to a no-mans land where the silicon doesn't really exist off the shelf.
 
Back
Top