Test Equipment Anonymous (TEA) group therapy thread

[mnementh]

Heh. Slinki-T-Rex is actually quite appealing. Just flopping him back & forth & the clikkity sound he makes is viscerally pleasing.

EDIT : And now he is no longer mine. The boi saw him and just HAD to have him... and now I have orders requests from all the fam for one of their own. 

mnem
 

[nixiefreqq]

hey med. 

these both suck and blow.  if you want any of them let me know and i will stick whatever you might be able to use in a box.

ps......don't take any crap from people who don't drive on the correct side of the road.  (with the possible exception of spec.  he does seem to understand that heat rises and shit flows downhill).

pps  bd almost made me piss myself with that comment about misspelling "count".

[cyclin_al]

Also, shame on you Spec, for bringing that vid in here with our conversation aboot med's blower; the two are aboot as closely related as an ox is to the cart it pulls. 

I have half a mind to give you 40 lashes with a wet noodle.

med's motor is an induction/repulsion motor; it works by the electrically induced magnetism in a laminated metal core pushing against the field which created that magnetism. By nature, these motors will attempt to reach an rpm that is dictated by the number of poles in the rotor vs those in the stator and the fixed AC frequency applied; essentially a small-scale 2-phase AC industrial motor. It will tend to run at that speed over a very broad voltage range; as a result, "speed control" by varying voltage doesn't really work; we need to change speed via PWM, etc.

A modern high-KV (K-RPM/V) quadcopter motor is essentially a PWM-switched 3-phase permanent-magnet DC repulsion motor operating at high frequency; its speed is governed by the applied DC voltage and the frequency/duty cycle of its PWM commutation cycle.

As we began making motors capable of developing higher RPM, we hit up against the wall of primitive slow fixed-commutation ESCs. This wall was broken by the development of ESCs which actually read the reverse-EMF pulses created by a magnet passing over a pole, and alter the timing of the next commutation cycle accordingly. These ESCs then developed to the point where speed of the computer running them and speed of the FETs used for switching were the limiting factor.

Think the difference between fixed ignition timing on a old Briggs lawnmower engine vs dynamic ignition timing on larger industrial, motorcycle and automobile engines.

Then began the race for ever-faster ESCs and motors, with each generation of motor and ESC forcing further evolution of the other. These machines are forcing the development of BL motors in all sorts of other applications; it really has been an exciting time to be involved in them.

mnem
Except, of course, when the FAA had to stick their dicks in it.

You are now showing your ignorance of motor operation.
1/ You cannot change the speed of an AC motor by PWM, you change it by changing the frequency of the supply. The only exception to this is crude power reduction that causes increased slip and even pole skipping in the motor This is just about OK for fans  but not much else.
2/ The speed of a brushless DC (BLDC) motor is in NO WAY governed by the applied voltage. A BLDC is basically a a 3 phase AC motor with a permanent magnet rotor. As with all 3 phase AC motors the speed is determined by the frequency and number of poles. The K-V factor just tells you what the back EMF is at a given speed. Obviously you need a supply higher than this to drive any current (producing torque and power), but the speed is still set by the frequency.
3/ PWM Duty cycle does not set th speed, it controls the average power (current)
4/ The sensorless motor / controller designs were developed to reduce cost and weight, not because the sensors were not fast enough. They do have the potential advantage of being able to compensate for field distortion / torque reaction but not all controllers do this.

Yes, hence the etc part of PWM etc, which is what is used to control the 3-phase H-gate controllers commonly in use today.

Jeebus dude... I love a good argument as much as the next guy, but fuck... this is not a good argument.

[EDIT] You are trying to apply motor control principles designed for reliability and fixed-RPM efficiency to motors & ESCs intended for the exact opposite: to switch as much current as possible and to be able to change speeds as quickly as possible in order to effect attitude control of a small aircraft. Literally a apples and ocelots comparison.

mnem
 

Ah ha!  I think I figured out what is going on here.  Either I am going to play conciliator or facilitator to this argument....

You both appear to agree that frequency is being used to control the motor speed.  You are both getting caught up in how it is done; in the end the output from the micro-controller is changing frequency to control speed.  However, the software programmer is using the PWM function to do it.  However, they are not necessarily controlling the pulse width to control motor speed.

In a normal application, the PWM function is configured with a frequency.  From there, the control algorithm varies the pulse width to achieve the control function.

In this case, the control algorithm is varying the frequency of the PWM function.  Is the pulse width even being modulated?  I do not know for sure, but I guess probably not.  In the end, the frequency of the micro-controller output is being modulated.

Should it even be called PWM in this case?  Probably not, but if you read the code, it is going to be most definitely using the PWM function.

Court adjourned?

... and back to work; I wasted lunch time on this!  I can read the thread while working, just can't post while working.  That keeps most of my posts in the catch-up category...

[cyclin_al]

pps  bd almost made me piss myself with that comment about misspelling "count".

Same here, once the moment passed and I finally got the joke.

[Cerebus]

So if I understand: The Count is counting to see if Cunt Lame Duckula is out for the count. Did I get that right?

[Robert763]

Other than to say the laws of physics and electromagnetism don't care if it's on dodgy quadcoptor or not, I'm done on the exploding motor.

[mnementh]

@cyclin_al: PWM is a bit of a PITA term with regards modern high-performance quadcopters, TBH. 

PLEASE NOTE: The following is a gross over-simplification of a very complex ecology that is still in its infancy. Please have mercy and try to overlook oodles of huge nits. PLEASE.

The basis of modern radio control is a PWM signal that represents the relative position of the joysticks on the radio. These multiple signals are then combined into CPPM analog signal, and then again into packetized digital radio. In its simplest form, the signal received from the RX radio is that PWM reconstructed at the other end. Modern RC turns that all into digital signal over SBUS or other proprietary formats.

The earliest brushed-motors ESCs were actually little more than a linear DC voltage regulator controlled by the varying voltage from an optoisolator across a capacitor/resistor pullup circuit. That was controlled directly by the PWM signal from the RX radio which doesn't know the difference between controlling an ESC or a servo. A flight controller inserted into this mess allows 4 joystick signals to control however many motors are on the aircraft; adding in feedback from an accelerometer/gyro sensor chip brings attitude control and self-leveling capability into the mix.

That PWM signal to the ESCs was soon found to be too slow, so flight controller designers started upping the PWM frequency to be able to respond faster. That evolved into more complicated non-PWM signals being used to control the ESCs.

At the same time, the processor on the ESC controls the H-Gate directly using the libraries intended to synthesize PWM for conventional motor control; these capabilities are then modified to change both frequency and duty cycle to control RPM and current output to motors. Eventually we exceeded the capability of most cheap processors to synthesize this signal fast enough, and had to move on to processors with... integrated hardware PWM generation. 

So yeah... in RC you'll hear the term PWM bounce around quite a lot. It is very confusing, annoying even, as exactly what it means at any moment is entirely context-dependent.

mnem
*back out into my shed-misery*

[Cerebus]

Ah ha!  I think I figured out what is going on here.  Either I am going to play conciliator or facilitator to this argument....

You both appear to agree that frequency is being used to control the motor speed.  You are both getting caught up in how it is done; in the end the output from the micro-controller is changing frequency to control speed.  However, the software programmer is using the PWM function to do it.  However, they are not necessarily controlling the pulse width to control motor speed.

In a normal application, the PWM function is configured with a frequency.  From there, the control algorithm varies the pulse width to achieve the control function.

In this case, the control algorithm is varying the frequency of the PWM function.  Is the pulse width even being modulated?  I do not know for sure, but I guess probably not.  In the end, the frequency of the micro-controller output is being modulated.

Should it even be called PWM in this case?  Probably not, but if you read the code, it is going to be most definitely using the PWM function.

Court adjourned?

... and back to work; I wasted lunch time on this!  I can read the thread while working, just can't post while working.  That keeps most of my posts in the catch-up category...

The general scheme of things in BLDC/PMSM controllers is that at their heart they have a current control loop, this has a demand input of motor current (torque is a function of current), and a feedback input of winding currents, the output is a duty cycle setting for the PWM half bridges. This control loop runs at the PWM frequency. The resultant waveform after the PWM is integrated/filtered by the motor inductance is a (typically sinusoid) that follows the commanded current input. The current waveform is at a much lower frequency than the PWM signal.

The control loops look a bit like this:


(image from one of the slides accompanying Dave Wilson's excellent video series on motor control)

There's feedback of rotor position (which can either come from Hall effect sensors, a shaft encoder, or be guestimated by algorithm from back EMF voltage); from rotor position versus time, velocity can be derived.  The frequency of the commanded current (typically a sinusoid) is what determines the motors rotational speed. Obviously the outer loop for position control is only used for servo applications and is missing on 'just need a wizzy motor' applications. The velocity loop is, obviously, controlling angular velocity.

So what controls the motor's speed is the frequency of the commanded current signal. If the amplitude of the commanded current signal is not big enough the motor will stall or at least slip. Any voltages involved are side effects of the stator inductance and stator induced back EMF from the rotor flux - there must be enough instantaneous voltage available to overcome back EMF and produce the required current but that's taken care of by the current control loop.

Current is king here as current dictates the stator flux and the interaction of stator flux and rotor flux is what dictates the torque, torque versus load (including friction) determines speed. If the motor physically tries to run faster than the stator flux allows, the stator flux produces negative torque and slows the motor.

This stuff is not easy and I quite understand that people can get confused, I did until I'd read/watched everything in my stack of study materials at least twice.

[mnementh]

Now delete the current-monitoring portion and substitute a gyro/accelerometer as the feedback loop to the flight controller. The commanded position/measured position is dynamic at speeds in excess of 100MPH in some cases, and degrees/sec attitude changes so fast that we had to move to chips connected with a faster bus to keep up. Oh, and the ultimate attitude/postional feedback loop is still visual to the pilot. 

mnem
 

[bd139]

I would just like to apologise to everyone for posting such a contentious explosion 

I’d like to share the following as a less contentious one to refresh people:

https://youtu.be/pXI1PDTp5gk

[Specmaster]

OMG, mnem strikes again. 

[Mr. Scram]

Oh boy, I almost this week's episode of “knowing better and talking louder”. 

[Cerebus]

Now delete the current-monitoring portion and substitute a gyro/accelerometer as the feedback loop to the flight controller. The commanded position/measured position is dynamic at speeds in excess of 100MPH in some cases, and degrees/sec attitude changes so fast that we had to move to chips connected with a faster bus to keep up. Oh, and the ultimate attitude/postional feedback loop is still visual to the pilot. 

mnem
 

Mnem, this is just the motor control part, just, in your terms, the ESC. The commanded motor angular velocity (ultimately thrust) comes from your flight control system and goes into the ESC. Delete the current monitoring bit and your BLDC/PMSM motor won't go around and won't be controlled. A gyro/accelerometer reading vehicle velocity/attitude can't tell the motor what electrical angle it is at.

I can't tell if you understand that the motors we are talking about are fundamentally AC motors; whoever decided to label them BLDC motors did everybody a disservice. You differentiated brushed DC motors above so I tend to believe you do, but then you argue with Robert and insist that voltage sets the speed of the motor (it does for a brushed DC motor, it doesn't in any meaningful way, for a Brushless DC motor, which is much better described as a Permanent Magnet Synchronous Machine). So I can't tell whether you're all arguing at cross purposes or whether you (Mnem) have some misconceptions about how BLDCs/PMSMs work.

Everything I know about the subject says that Robert clearly understands what he's talking about and that you're either a bit confused or are doing a terrible job of explaining what you do understand.

[Cerebus]

Oh boy, I almost this week's episode of “knowing better and talking louder”. 

Hey, at least it's not "knowing better and making up stuff to prove it" which seems to be the modus operandi of some people on this forum (who will remain nameless for fear of being deluged in another steaming pile of 'expertise').

[Robert763]

You better hope I understand a bit about this motor stuff, there are full size aircraft flying around out there with motor controls I designed and you might be on one someday 

[Saskia]

praying everytime I board a plane

[Mr. Scram]

Hey, at least it's not "knowing better and making up stuff to prove it" which seems to be the modus operandi of some people on this forum (who will remain nameless for fear of being deluged in another steaming pile of 'expertise').

The "louder" seems reserved for a very limited group, found participating in these "discussions" surprisingly often and purporting a body of expertise equivalent to the physical heft. Sometimes a man's ego is his own worst enemy.

[med6753]

You better hope I understand a bit about this motor stuff, there are full size aircraft flying around out there with motor controls I designed and you might be on one someday 

Hopefully not including FB310 Fan Bases.

BTW.....if you all want to hear how loud this damn thing is tune into Discord tomorrow and I will turn on my Mic and give you a sample of the noise pollution. 

[mnementh]

Now delete the current-monitoring portion and substitute a gyro/accelerometer as the feedback loop to the flight controller. The commanded position/measured position is dynamic at speeds in excess of 100MPH in some cases, and degrees/sec attitude changes so fast that we had to move to chips connected with a faster bus to keep up. Oh, and the ultimate attitude/postional feedback loop is still visual to the pilot. 

mnem
 

Mnem, this is just the motor control part, just, in your terms, the ESC. The commanded motor angular velocity (ultimately thrust) comes from your flight control system and goes into the ESC. Delete the current monitoring bit and your BLDC/PMSM motor won't go around and won't be controlled. A gyro/accelerometer reading vehicle velocity/attitude can't tell the motor what electrical angle it is at.

I can't tell if you understand that the motors we are talking about are fundamentally AC motors; whoever decided to label them BLDC motors did everybody a disservice. You differentiated brushed DC motors above so I tend to believe you do, but then you argue with Robert and insist that voltage sets the speed of the motor (it does for a brushed DC motor, it doesn't in any meaningful way, for a Brushless DC motor, which is much better described as a Permanent Magnet Synchronous Machine). So I can't tell whether you're all arguing at cross purposes or whether you (Mnem) have some misconceptions about how BLDCs/PMSMs work.

Everything I know about the subject says that Robert clearly understands what he's talking about and that you're either a bit confused or are doing a terrible job of explaining what you do understand.

No, I understand the difference. I am bad at explaining it... let me see...

Okay. On a quadcopter, we literally don't care about limiting the torque of the motors. Quite the opposite; we want the motors to always accelerate and decelerate as quickly as possible. Therefore, the ESCs do not monitor current. Current limiting is entirely a function of how much current and voltage a given battery can drop across the combined ESCs vs the KV of the motor vs the load of the prop on the motor.

Okay, lots of little electrical losses; shunt drop in the PDB, iR of the FETs, etc... but functionally, if you overprop a motor and you have enough battery capacity to do it, you're going to smoke a motor, and when that happens, usually smoked FETs in one or more ESCs.

Conversely, we are running these ESCs right on the ragged edge of their performance envelope; deliberately. In many cases, they can handle the RPM of a loaded motor, but the motors can and will run faster unloaded than the circuitry and software can keep up with. If you run a hot motor with no load, it will near-instantly (much too fast for any amount of cooling to matter) exceed either the capacity of the CPU to detect or the capacity of the laminated stator to translate the reverse-EMF these ESCs use to determine timing for the next commutation cycle. If that happens, you have HI & LO side FETs on at the same time. This condition will short the FETs in question in a millisecond whether there is a motor attached or not; but as the motor is connected, THEN you have a huge current dump across the windings.

There is no sine wave generation here; that is why I characterized these as a DC motor. The signal generated is nearly a square wave (well, a really dirty one at high RPM); this is of course rounded up quite a bit by the inductance of the motor because it is heavily loaded by large current dropped across few windings of heavy gauge wire, stator inductance and powerful permanent magnets that are moving, etc. This is on purpose, to keep the FETs out of linear operation as much of the time as possible. We have ESCs the size of a quarter that handle 40-50A continuous without heatsinks, because ultra-low iR FETs on huge masses of copper and always in switching mode.

The motors most of us think of when talking speed control are essentially stationary motors, being controlled for efficiency and matching speed to some fixed requirement. Too much torque breaks things, so they have current monitoring in the loop and sine-wave output for smooth acceleration.

ESCs in quadcopters are the exact opposite... we want things to change absolutely as quickly as possible... it is that constant state of flux... the juggling act... that makes it possible for 4 little fistfuls of angry pixies to cut holes in the air fast enough to stay aloft.

Like I said... these things have evolved differently for a reason. These ESCs are VERY primitive; there's a limit to how much you can fit on a PCB the size of a nickel and still expect it to handle 30A continuous and more.

You say "properly designed"... yes, speed controls properly designed for reliability and longevity work as you suggest. Quadcopter ESCs have been designed for high-speed switching, high-current capability and to be as small and light as absolutely possible, because literally every gram counts when you're hoisting an aircraft aloft by brute force on battery power.

The two design scopes pretty much diverge at the word "speed control"... 

mnem
 

[Saskia]

@med will you be joining the voice channel ? That would be kinda nice, my English is getting rusty

[mnementh]

You better hope I understand a bit about this motor stuff, there are full size aircraft flying around out there with motor controls I designed and you might be on one someday 

Hopefully not including FB310 Fan Bases.

BTW.....if you all want to hear how loud this damn thing is tune into Discord tomorrow and I will turn on my Mic and give you a sample of the noise pollution. 

I hope to be able to... but I'll probably be up to my ass in shed-sheetmetal. 

mnem
If I can find a ladder that can handle my fat ass...

[med6753]

@med will you be joining the voice channel ? That would be kinda nice, my English is getting rusty

For you My Dear I'll join the voice channel but you may not be able to hear me over the racket. 

[med6753]

You better hope I understand a bit about this motor stuff, there are full size aircraft flying around out there with motor controls I designed and you might be on one someday 

Hopefully not including FB310 Fan Bases.

BTW.....if you all want to hear how loud this damn thing is tune into Discord tomorrow and I will turn on my Mic and give you a sample of the noise pollution. 

I hope to be able to... but I'll probably be up to my ass in shed-sheetmetal. 

mnem
If I can find a ladder that can handle my fat ass...

Remember, DST ended for us last weekend so Discord starts an hour earlier (15:00 EST) 

[Cerebus]

The "louder" seems reserved for a very limited group, found participating in these "discussions" surprisingly often and purporting a body of expertise equivalent to the physical heft. Sometimes a man's ego is his own worst enemy.

The impression I get of the principal suspect I'm talking about is of an being old, small and wizened, I have no clue as to their actual physical characteristics. On the Internet no one knows you're a dog or indeed a very small man. I've never associated [imagined] physical characteristics with the degree of pugnacious of people online. When I have had some genuine clue to the stature of people online I've sometimes noticed the "little man effect", where unusually small men try to compensate with aggressiveness. Both the men that I've known in real life who are highly likely to start a bar room brawls were both obviously short (and Scots, but experience of the Scots suggests that attribute had nothing to do with their attitude).

[Specmaster]

I'll be on discord voice chat tomorrow and I look forward to hearing the loudness of the fan as well. As for this quadcopter, PWM, ESC and all this BLDC/PMSM and whatever, I don't have the foggiest idea of what is going on there, so I'm keeping quiet on the topic, know your limitations I say, but I read and try to understand in order to broaden my experience a bit.

As for the small man syndrome Cerebus speaks of, I too have noticed that it does tend to be the smaller packages that are most aggressive in all forms of life, be it man, cat, dog, bird etc but there are always the exception to that. I hope that he was not referring to me with that statement because I can assure you all that I'm not a small man, being 6ft 4" tall and built like the proverbial brick sh1t house with such large hands that I struggle to find any gloves to fit me