FLIPSKY new 20s 100A tiny controller (vesc based)

[badgineer]

Hi Jaykup

Thermal performance at 80 phase amps, bench test, no air flow, 30V, 80-80 160kv motor. VESC firmware starts reducing current at 77.5C by default.

How did you test?
I am asking because I find your results... interesting.
I did expect the one with Aluminium PCB (With much better RDSon FETs) to do better than the normal 75100. But even the vesc6 is much better thermally? Is the 75100 really so bad? Can you check if anything was different / wrong with the 75100 test?

Thanks in advance
And please don't take this as a criticism / attack or something on those lines. On the contrary, I found you posts always informative and interesting, but now I'm really surprised by your results.

Br,

 

[jaykup]

Trust, but verify. I like it

I agree, I was really surprised by the results too. I will say that the 75100 aluminum PCB is a completely different design. It really shares the name only.

The aluminum PCB version has two PCBs, with the lower being the power stage. It has the 6 mosfets on the lower one, and the back side of the PCB is blank with thermal paste on it attached directly to the aluminum heatsink. That's a great way to cool these surface mount mosfets, similar to the Focbox Unity and Stormcore designs which are some of the best thermal performers in the category. Contrast this with the VESC 4.12 where the mosfets sandwich the PCB and it just feed heat into each other. The VESC 4.12 only manages 22 seconds in the same test.

Here is the dual motor version



Along with a video showing some more detail

https://www.youtube.com/watch?v=amiGHtlDCoM

So this was my test method:

• 80-80 160kv motor attached to a fixture on a bench
• 30v 10a CC/CV power supply to the ESC
• Connected to USB to the VESC tool
• Run a motor detection on 5.3 firmware
• Realtime logging on, heartbeat on
• Wait for ESC to get to 28-29C
• Motor at room temp
• run "foc_openloop 80 1000" in the terminal (80A phase, 1,000 eRPMs 142 physical RPMs)
• Export data to a CSV file, load in Excel and run a macro that formats everything and adds a column and summary.

Originally I thought the VESC thermal throttled at 77.5C, but it's slightly sooner than that. It's set to 85C by default, then starts backing off power in a gradual fashion 15% before that... closer to 72C, but this method looks at how long it takes to get to 77.5C. It's at least an apples to apples comparison.

The clock starts when the mosfet temperature is below 77.5C and the motor amps are greater than 70A.
The clock ends when the mosfet temperature is above 77.5C or the motor amps are less than 70A.

The formula can be seen in Column E in the attached spreadsheet of the raw data.
=IF(AND( [MosfetTemp] <77.5, [Phase Amps] >70),"GoodData","BadData")

I snapped a quick picture with the thermal camera when it was near it's full temp. It's interesting that the 12awg wires along with the 5.5mm bullets are the hotest point visible.



Lab tests are great and all, but it doesn't always tell the full story.

We swapped the old 75100V201 for this new one on my friends 20s ebike with a 7T 27.5" leaf motor. Here is a 22 mile log of him trying (and failing ) to keep up with me riding a skateboard last weekend at a State Park with some really nice steep hills. The new 75100 aluminum PCB version never went above 52c. It is in direct airflow, and the temp outside was 81F (27C) at the time of the ride.

Interactive metr log: https://metr.at/r/Dzn0x



Raw data of the thermal tests:

 

[Ianhill]

Its here theres a 75200 version is out or 72v 150amp 20s capable.

2 of these controllers on a kaabo wolf pro or alike would really wake it up, with a custom battery thats all about the amps and a jikong 200amp bms it would melt along nicely literally, it could challange a rion on short blasts but rion being lighter would have less heatsoak into the hubs over time so ferrofluid be a must.

I got a qd138 i been trying to light up i wonder if its upto that task 10kw should get some decent power out of it.

 

[Bengy22]

Just noticed the 75200 is live on the flipsky site for preorders, they ship out on the 31st so not much of a preorder. Spec's look pretty good though 150a cont 300a peak, dependent on external cooling factors. Hoping to see some reviews on it in the near future. I was wanting to get the aluminum 75100 to use with a bafang m620 but I was worried about the heat and running it at the full 120a phase so looks like this would be a good contender. Can you imagine pushing the full 150a or 300a peak though on one of the mid drive motors Someone's got to be crazy enough to do it right? Only thing I can find is people using phase runners with 96a and running sensorless, I want to see what these will do :lol:

 

[Ianhill]

I'm about to order the 75200 so next few weeks ill post how ive got on with it.

 

[Hummina Shadeeba]

How to know if it’s worth getting such a high amperage esc, or is it a waste that will magnetically saturate the motor?

I plan to wind this to about 60kv

https://alienpowersystem.com/shop/brushless-motors/80mm/aps-83100sw-sensored-outrunner-bldc-motor-180kv-7kw-water-cooled/

 

[scianiac]

Yesterday I happen to have been looking at what it takes to saturate those 63mm outrunners. I found this on an eskate forum.

The stator saturates pretty fast unfortunately.

I know that on paper amps = torque, but if i go from 6354’s running 70A to 6374 running the same amps, the 6354 saturates at around 45-50 amps and the rest is wasted heat.

The 6374 however saturates at around 80-85A so pressing full throttle will launch me like a damn tesla car.

Running both at 10A should in theory produce the same torque, potentially slightly less from the bigger motor as the losses due to the size of the rotor are bigger

So from that you may be able to roughly guess what a 63100 will saturate at, assuming that person's experiences are correct.

 

[Hummina Shadeeba]

Isn’t there a generalization about staturating that could be made based on the stator size and kv?

 

[Ianhill]

One things for sure i wont be saturating a qs138 its rated 3kw but it takes 10 times that to reach saturation, so what ever this controller throws at the motor it will survive for a decent amount of time but weighs alot to do so.

 

[BalorNG]

Saturation is about current, current, current. Wattage is more or less meaningless in the context, why keep brining it up?

 

[BalorNG]

Yesterday I happen to have been looking at what it takes to saturate those 63mm outrunners. I found this on an eskate forum.

The stator saturates pretty fast unfortunately.

I know that on paper amps = torque, but if i go from 6354’s running 70A to 6374 running the same amps, the 6354 saturates at around 45-50 amps and the rest is wasted heat.

The 6374 however saturates at around 80-85A so pressing full throttle will launch me like a damn tesla car.

Running both at 10A should in theory produce the same torque, potentially slightly less from the bigger motor as the losses due to the size of the rotor are bigger

So from that you may be able to roughly guess what a 63100 will saturate at, assuming that person's experiences are correct.

That's pretty interesting info, but at what KV?
I think saturation *torque* is most useful metric, and heat production at this torque.

 

[scianiac]

Saturation is about current, current, current. Wattage is more or less meaningless in the context, why keep brining it up?

Wattage is an sometimes helpful information here because of how it's linked to Kv, after all we're really talking about current and motor turns so having wattage helps us make estimations when missing other data points. Sure it would be nice if motor mfgs would just tell us saturation current at a known Kv/turn count but we're kinda left piecing together random bits of information.

That's pretty interesting info, but at what KV?
I think saturation *torque* is most useful metric, and heat production at this torque.

Unfortunately he didn't say but considering he was talking about eskate motors we're probably talking around 150-200KV

 

[Hummina Shadeeba]

Anyone have a decent guess on what current saturation would begin with this common motor wound to 35kv?

https://alienpowersystem.com/shop/brushless-motors/80mm/aps-83100sw-sensored-outrunner-bldc-motor-180kv-7kw-water-cooled/

Other than using a dyno how could it be figured?
Surely based on the stator dimensions a close guess could be made.

I got the 75100 which I think uses the same firmware as the 75200. Wonder if it would be worth it to use the 75200

 

[Ianhill]

Saturation is about current, current, current. Wattage is more or less meaningless in the context, why keep brining it up?

I dont know if you have noticed but its brought up due to the manufacturer giving rated continous outputs as that current current current means nothing without time time time.

For example the motor i mentioned is rated at 72v so its assumed one can work it out that 32kw saturation is no more than 445 battery amps.

As actual phase current runs alot higher and get confused with battery current by newbs so for the understand of the basics a continuous power draw is used the rest is gravy for us tech nerds.

 

[Hummina Shadeeba]

as that current current current means nothing without time time time.

Time isn’t relevant to magnetic saturation


http://www.qs-motor.com/product/qs138-90h-4000w-mid-drive-bldc-motor-with-428-sprocket-and-sine-wave-controller-for-electric-motorcycle/

That motor’s stats show a max phase current of 660. Maybe that’s the max current before saturation. I assume the max bus current ,which is much lower (180), is related to the esc limits.

 

[scianiac]

Anyone have a decent guess on what current saturation would begin with this common motor wound to 35kv?

https://alienpowersystem.com/shop/brushless-motors/80mm/aps-83100sw-sensored-outrunner-bldc-motor-180kv-7kw-water-cooled/

Other than using a dyno how could it be figured?
Surely based on the stator dimensions a close guess could be made.

I got the 75100 which I think uses the same firmware as the 75200. Wonder if it would be worth it to use the 75200

After it's wound to 35Kv probably not very much, if we assume that quote is correct 100mm length vs 74mm puts you up to 108A maybe but once you lower the Kv from probably 190Kv to 35Kv your saturation current will be much lower, like 20A? My math and motor theory may be off there but there are good reasons why winding these outrunners for such low Kv is not optimal. Outrunners like this with lower Kv generally are much larger in diameter with much higher pole counts for a reason.

Saturation is about current, current, current. Wattage is more or less meaningless in the context, why keep brining it up?

I dont know if you have noticed but its brought up due to the manufacturer giving rated continuous outputs as that current current current means nothing without time time time.

For example the motor i mentioned is rated at 72v so its assumed one can work it out that 32kw saturation is no more than 445 battery amps.

As actual phase current runs alot higher and get confused with battery current by newbs so for the understand of the basics a continuous power draw is used the rest is gravy for us tech nerds.

Saturation current and continuous power are both important to factor here but they are different, you guys are just talking about different things. Selecting a ESC based on motor saturation current is totally valid and then you can be more sure that the motor will run into thermal issues before the ESC will. And you can select the motor based on it's continuous wattage rating.

 

[Hummina Shadeeba]

Selecting a ESC based on motor saturation current is totally valid and then you can be more sure that the motor will run into thermal issues before the ESC will.

I was thinking it would be the opposite and if the motor is saturating it’s going to be getting hotter before esc.. especially if the esc is intended for the higher amps like the 75200 instead of the 75100

 

[scianiac]

Selecting a ESC based on motor saturation current is totally valid and then you can be more sure that the motor will run into thermal issues before the ESC will.

I was thinking it would be the opposite and if the motor is saturating it’s going to be getting hotter before esc.. especially if the esc is intended for the higher amps like the 75200 instead of the 75100

Isn't that what I said? The motor will run into thermal issues before the ESC, if your near saturation current it will get too hot before the ESC if you've selected the ESC to have a higher current rating than that required to saturate.

 

[BalorNG]

Anyone have a decent guess on what current saturation would begin with this common motor wound to 35kv?

https://alienpowersystem.com/shop/brushless-motors/80mm/aps-83100sw-sensored-outrunner-bldc-motor-180kv-7kw-water-cooled/

Other than using a dyno how could it be figured?
Surely based on the stator dimensions a close guess could be made.

I got the 75100 which I think uses the same firmware as the 75200. Wonder if it would be worth it to use the 75200

After it's wound to 35Kv probably not very much, if we assume that quote is correct 100mm length vs 74mm puts you up to 108A maybe but once you lower the Kv from probably 190Kv to 35Kv your saturation current will be much lower, like 20A? My math and motor theory may be off there but there are good reasons why winding these outrunners for such low Kv is not optimal. Outrunners like this with lower Kv generally are much larger in diameter with much higher pole counts for a reason.

Saturation is about current, current, current. Wattage is more or less meaningless in the context, why keep brining it up?

I dont know if you have noticed but its brought up due to the manufacturer giving rated continuous outputs as that current current current means nothing without time time time.

For example the motor i mentioned is rated at 72v so its assumed one can work it out that 32kw saturation is no more than 445 battery amps.

As actual phase current runs alot higher and get confused with battery current by newbs so for the understand of the basics a continuous power draw is used the rest is gravy for us tech nerds.

Saturation current and continuous power are both important to factor here but they are different, you guys are just talking about different things. Selecting a ESC based on motor saturation current is totally valid and then you can be more sure that the motor will run into thermal issues before the ESC will. And you can select the motor based on it's continuous wattage rating.

And continuous *wattage* is all about heat shedding and maximum possible power is function of maximum possible RPM at maximum possible current.
You cannot get more *torque* out of the motor no matter the winding, and heat losses at this torque due to Ohmic heating are also constant no matter the winding (more or less, and so far only the motor is concerned - you'll lose more in phase wires if they are not thick enough, and ESC of course), provided you have same copper fill/number and size of magnets/etc.

Funny enough, maximum RPM is ALSO more or less the same regardless of winding, just would be achieved at different voltages with different winds, and the RPM limiters is:

a. Your ESC first and foremost (switching frequency, how much voltage it can take)
b. Bearings
c. Centrifugal forces tearing the rotor/motor apart
d. Iron losses shooting up to unreasonable values

It is extremely hard to achieve maximum possible power output as continuous power on any motor, even middrive, that would require an elaborate active liquid cooling solution.

 

[Ianhill]

What BalorNG said.

Saturation point is highly limited affair due to thermal limitions of shedding heat, most designs struggle to maintain a fraction of peak power achievable over time

Designs have differing weak spots but even the best to date needs a rest period inbetween peaking without advance water cooling circuitry to shed the created heat.

Erpm etc we should be on top of.

I'm only mere mortal myself got myself a £225 paper weight of a controller from china when i find out it has enough erpm but for some reason or another it limited to 12pole motor or more wont play nice with a 5pole qs138 that i wanted to spin up for a while now.

 

[Hummina Shadeeba]

What BalorNG said.

Saturation point is highly limited affair due to thermal limitions of shedding heat

Isnt heat irrelevant to magnetic saturation?

 

[Hummina Shadeeba]

Anyone have a decent guess on what current saturation would begin with this common motor wound to 35kv? ..

After it's wound to 35Kv probably not very much, if we assume that quote is correct 100mm length vs 74mm puts you up to 108A maybe but once you lower the Kv from probably 190Kv to 35Kv your saturation current will be much lower, like 20A? My math and motor theory may be off there but there are good reasons why winding these outrunners for such low Kv is not optimal. Outrunners like this with lower Kv generally are much larger in diameter with much higher pole counts for a reason.

With 86v I’m forced to lower the motor kv. the motor will do 3200rpm at full speed on paper.


the higher the voltage, and therefore lower kv, the less amps needed for same torque.

 

[BalorNG]

Anyone have a decent guess on what current saturation would begin with this common motor wound to 35kv? ..

After it's wound to 35Kv probably not very much, if we assume that quote is correct 100mm length vs 74mm puts you up to 108A maybe but once you lower the Kv from probably 190Kv to 35Kv your saturation current will be much lower, like 20A? My math and motor theory may be off there but there are good reasons why winding these outrunners for such low Kv is not optimal. Outrunners like this with lower Kv generally are much larger in diameter with much higher pole counts for a reason.

With 86v I’m forced to lower the motor kv. the motor will do 3200rpm at full speed on paper.


the higher the voltage, and therefore lower kv, the less amps needed for same torque.

Indeed. But you still cannot get *more* torque out of the motor, and HEAT LOSSES at this torque will remain basically the same +- a few watts really. I have a couple of motors with different winds, I've run the numbers.

As far as I understand, ebikes.ca simulator favors low kv motors because a considerable portion of losses modelled there happends outside of the motor - long, thin phase wires, pretty inefficient ESCs, etc.
A middrive with properly short, thick phase wires getting into an efficient ESC is a different matter.

Personally, I favor high KV, 8s setups because of small and efficient RC chargers first and foremost, plus chances of getting an electric shock is negligible at this voltage...

 

[BalorNG]

What BalorNG said.

Saturation point is highly limited affair due to thermal limitions of shedding heat

Isnt heat irrelevant to magnetic saturation?

Technically, you'll demagnetise your magnets much faster at higher temps with high flux density... I'm not an expert in this because I my priority is efficiency at pretty low power first and foremost, which is an other challenge entirely.
However, my point was that getting *continous* maximum torque (at the saturation point) out of any motor except one that has an engineering marvel of a liquid cooling for any length of time is night impossible due to enormous heat generation for given motor mass/area.

 

[Hummina Shadeeba]

I don’t think anyone here is saying u can get more torque out of low kv. It’s all the same heat to torque, but when u lower the kv u are more able to get to that max torque. No higher current esc needed

The magnets are high temp n42sh I believe. I’m not worried about them and just trying to figure when the stator saturates.