NEW - ESC vs ESC

[sarpilot]

You guys are confusing the heck out of me but staying tuned in in the hopes of learning some new stuff...


Sent from my DROID3 using Tapatalk 2

[Iceberg86300]

Originally Posted by kimmik

50% throttle means motor current is 200% of input current, and motor voltage is 50% of input voltage.

80A and 44V input at 50% throttle is thus 160A motor current at 22V motor voltage.

You will never see this situation in a 600 unless your gearing is say 5 teeth too large, in which case you will indeed blow the esc or motor in no time.

160A motor current at 50% duty cycle translates to 80A rms ripple current through the input. The two capacitors cop the majority of this current, which is why one of them fused so quickly. Can you imagine 40A flowing in each 26awg lead.

A more realistic scenario, e.g. 80A at 90% throttle:

89A motor current
40V motor voltage
3.6kW
30A rms ripple current (rms is root mean square)
15A est. per capacitor
40W total freewheeling and i2r loss.

That's why this esc in correct use doesn't fail in this mode.

Even if you have freewheel, that esc will still fail in the same mode as the ripple current is same. the esc was tested at 2x current rating and 50% throttle, no esc manufacturer ever designed for that.

To make a esc survive that scenario, you need 1. A big low esc buffer cap bank, and 2. Afw to reduce get thermal 3. Lots and lots of cooling air.

Trying to wrap my head around this.I am a mechanical engineer, not EE.

Mainly, why does the current go so high at 50% throttle?

ub3p:
Where and how are you getting current readings from?

AFW, is this just not having brakes, or not forcing the motor to slow down when throttle and or load decrease, or something totally different?

Also, could you throw a hyperion on there?

And cool dunno setup!

Steve
Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[Mercuriell]

Originally Posted by Iceberg86300

Trying to wrap my head around this.I am a mechanical engineer, not EE.

Mainly, why does the current go so high at 50% throttle?

Steve
Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

It's a hard one for us mechanical. non-blackbox types to visualise

I don't think of it as current going in, magicked up to double after the ESC, but rather as the load drawn by the motor due to the stress of the dyno. In this case, of Tony's torture test, the motor is drawing a peak current of 160A, the PWM of the ESC is chopping power to the motor to only supply at 50% of the time so the power going into the ESC, ignoring losses, and available for 100% of the time, only needs to be at 50% of the motor current, to supply the power for the same in intput voltage.

This example gives an ideal of the stress, electrical and thermal our motors can experience - in this example that's 50 x 160A ie 8kW of power - thats around 400W of heating - makes you understand why the motors and pinions get warm!!

For those with limited understanding of brushless ESC (that includes me) this site gives some theory of the circuitry and a nice pictogram of ESC function !

[kimmik]

AFW is very different from brake.

Brake is rectification of the motor AC to recharge the battery. This drains kinetic energy and brakes the motor.

Freewheel is when the motor is allowed to spin during no power application. But during this time, in passive fw, the inductive current flows through the body diode because the freewheeling fets are off.

In active fw, we select out the fets that are involved in fw, and turn them on during this time. So the the voltage drop decreases from ~1V of diode to I*R, which is a lot lower.

This is my attempt at conceptualising pfw vs afw. I am keen to hear tony uber's formal explanation.

[FireNWater]

.
So I'm still confused. . . is this a valid test of a realistic scenario I'm likely to see in my model, or someone just showing us how quickly they can smoke an ESC?
.

[Mercuriell]

Interesting graph Kim we really need some foreplay to explain normal ESC function! So is freewheel basically dealing with inductive currents during FET off status by either shunting through the body diode (whatever that is lol) in passive FW and through the FETS in active FW? How does this benefit the ESC if is still having to "shunt" induced current ?

FireNWater: I think the CC illustration unlikely real-life scenario, as others have posted, no one goona hook up a Scorp 4530LE and pull 80A through at 50% throttle, BUT I bet a FW ESC like the Jive could have managed it. This is a voyage of discovery and learnng for all of us, exploring different aspects of ESC, rather than a didactic 'This One's Best'

[Proud Polac]

This thread is really making me consider majoring in EE over Computer Science. IDK though maybe both? Keep the tests coming I really want to see the Jive and YGE torture tested

[sarpilot]

Man these dudes are WAY over my head w knowledge.... I just wanna know what won't grenade......

[heraldo]

I hope tony doesn't plan on torture testing the kosmic in the same way? Would hate to see that burn.


Sent from my iPhone using Tapatalk

[Iceberg86300]

Originally Posted by kimmik

AFW is very different from brake.

Brake is rectification of the motor AC to recharge the battery. This drains kinetic energy and brakes the motor.

Freewheel is when the motor is allowed to spin during no power application. But during this time, in passive fw, the inductive current flows through the body diode because the freewheeling fets are off.

In active fw, we select out the fets that are involved in fw, and turn them on during this time. So the the voltage drop decreases from ~1V of diode to I*R, which is a lot lower.

This is my attempt at conceptualising pfw vs afw. I am keen to hear tony uber's formal explanation.

The AFW seems to line up with motor breaking turned off in the esc.

Seems like a bad idea to be sending power back into the lipo. But I guess the micro controllers can handle that with todays tech.

So even if I have the brake turned off when I "downshift" the motor/esc are acting like a generator? And when freewheeling is engaged it acts like a flyback loop with current looping around the motor and esc, with the motor acting as the resistor (and the rotor of course) to dissipate power until it the lower input is achieved?

My mechatronics minor is starting to come back to me

The power thing turned out to be a doh moment. I was thinking Pout=Pin (wire, heat, etc are being disregarded here). Just from reading your posts I'm going to assume you agree with conservation of energy. The doh for me came when I realized I was just thinking about current, but you are totally correct here.

I'll steal your example and elaborate a bit. The esc controls voltage going to the motor , which needs to match the Kv chart, and it doesn't care what current is going through it. So if you attach a load (dyno, rotor head, etc) that esc will kill itself trying to match that load by increasing the current, as you have already set the voltage using the throttle. current to the motor is the only thing that can satisfy the conservation laws, again, because the voltage is already set.

Sorry for the long msg! and any errors as I'm freaking tired.

Steve
Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[KU-HELI7]

Hello,

AFW is correctly called "active rectification" or "synchronous rectification"

http://en.wikipedia.org/wiki/Active_rectification

Regards, Hemut

[ub3r]

Ive uploaded a video to YouTube explaining how Active Free-Wheeling works.
I apologise in advance if im not clear enough, ive never been great at explaining things.

Ill be uploading the video file to HF shortly. Ill also write up a detailed explanation.

[ame]http://www.youtube.com/watch?v=uqzOQGiwGnE&list=UU7lc_DFDwD3wDHd-ixH1gfA&index=1&feature=plcp[/ame]

[Mercuriell]

Wow, thats's a very clear explanation Tony

Cant' understand why any manufacturer would not use FW and for the price difference of a FET or a silicone diode why wouldnt all use AFW? Are there any downsides? Be fascinating to see the 'scope of none, passive and active FW

[heraldo]

Makes sense to me, and I'm still in highschool!
Very good explanation tony and much appreciated.
PS: you have impeccable taste in music


Sent from my iPhone using Tapatalk

[ub3r]

Originally Posted by Iceberg86300

Where and how are you getting current readings from?

Im using my clamp meter for the current readings.

Originally Posted by Iceberg86300

AFW, is this just not having brakes, or not forcing the motor to slow down when throttle and or load decrease, or something totally different?

Totally different.

Your motor is an inductor, and it has freewheeling currents during the PWM-OFF cycle. During the PWM-OFF, current normally flows via the body diodes in your FETs. AFW allows the free-wheeling currents to flow via the appropriate FET and save us the diode losses.

Originally Posted by Iceberg86300

Also, could you throw a hyperion on there?

Ill try dyno as many as i can, just need time.

Originally Posted by Mercuriell

Wow, thats's a very clear explanation Tony

Cant' understand why any manufacturer would not use FW and for the price difference of a FET or a silicone diode why wouldnt all use AFW? Are there any downsides? Be fascinating to see the 'scope of none, passive and active FW

John, all power FETs have inbuilt body diodes. And all ESCs have free-wheeling. They either have Passive free-wheeling (using the FET body diodes, like the castle and scorpion ESCs), or Active free-wheeling using the FETs themselves (Kontronik and YGE). AFW > PFW.

If there are no diodes, the circuit would fail because you will have huge voltage spikes occurring to try and leak the inductive (free-wheeling) currents.

In Passive FW, you'll see the ph voltage drop below ground level (during PWM-OFF) as the free-wheeling diode conducts.
In AFC, the lower FET turns on and clamps the voltage to ground level.

There is no price difference between them at all. Its all in the control.

[ub3r]

Originally Posted by heraldo

Makes sense to me, and I'm still in highschool!
Very good explanation tony and much appreciated.
PS: you have impeccable taste in music

Glad you like the music.

[KU-HELI7]

@ub3r
Your explanation of AFW is 100% correct. But because you know in detail the advantage of AFW and of course the disadvantage of PFW, I'm just wondering why you are doing a test with 50% throttle and that high load on an ESC without AFW.
Such a test will always end up in burning the ESC (without AFW). It's only a question of (quite short) time.
So for me, the real title of this thread is: "How fast can I burn an ESC having no AFW"

Regards, Helmut

[ub3r]

Originally Posted by KU-HELI7

@ub3r
Your explanation of AFW is 100% correct. But because you know in detail the advantage of AFW and of course the disadvantage of PFW, I'm just wondering why you are doing a test with 50% throttle and that high load on an ESC without AFW.
Such a test will always end up in burning the ESC (without AFW). It's only a question of (quite short) time.
So for me, the real title of this thread is: "How fast can I burn an ESC having no AFW"

Regards, Helmut

I'm glad you agree with my explanation.

Well the test shows us the advantage of AFW over PFW.

[sarpilot]

Thx for the great video explanation of the different types of freewheeling systems! Boy that HeliJive is starting to look better all the time…


Sent from my DROID3 using Tapatalk 2

[KU-HELI7]

Originally Posted by ub3r

Well the test shows us the advantage of AFW over PFW.

Yes, but demonstrating this advantage by burning ESCs without AFW is IMHO quite unnecessary since AFW is nothing new on the market and the advantages of AFW are well known for years.

Regards, Helmut