B.F.E.S.C -- has a nice ring. :mdr:

Sensored motor setups or great for CNC servo motors (and others that need high precision) but I see no need for RC applications were that kind of precision is pointless. As smooth as the MMM is I see no reason to toss a sensor in their, it is every bit as smooth as the novak ESC's and way more powerful.

So much for that idea then...

Smooth can motor per-chance? The fins on the Castle-neu are so small I doubt they do much really?...

I don't know...I bet they're fairly effective actually. Even a "ridged" can like the Castle one can more than double the surface area of the aluminium over a smooth can which will have an effect on operating temp. IMO, it's worth having the ridges. Plus, they look pretty cool too!:yes:

The only place a sensor would be of any help would be extremely slow and/or heavily loaded starts. Once the rotor gets enough rpm to generate suitable back-EMF pulses, the sensors are not needed.

And, as DT said, any increase in surface area is helpful. It's about exposing as much of the motor to air as possible.

:lol: One of our power supply guys here loves the term BFC. Of course, that's usually just something in the 10's of uF in a 270VDC system.

I like M^3. The cubed difficulty of the design the bigger they get.

Is this a potential option for a more expensive MMM that doesn't need a fan? :)

Cheers

What about a hybrid system like the Speed Passion GT ESC? That takes advantage of the sensored starts, but can still do the automatic timing. I never heard of a sensored motor having lower efficiency; if anything, I've noticed that they run cooler for the revs they have to spin (I know they don't put out THAT much power, but they're no slouches- especially once you get into the lower turns).

If someone could put out a sensored system that can deal out the amount of power your guys' system puts out, I think it'd be a hit (screw ROAR... I could care less if it's ROAR legal). I'm pretty happy with my Mamba Max and Medusa setup though.:mdr:

Methinks I need a CC BEC so I can run 3s happily all day.

Patrick, I have a little question for you that's a little off-topic. I'm just looking for an "official" opinion from an expert. :smile:

The common consensus here seems to shoot for a battery and motor kv setup to get around 30-35k rpm total. This most likely came about because Feigaos tend to heat up at speeds higher than that. However, lower rpm means less EMF pulses at really slow speeds, which increases the chances of cogging.

But, I kinda like running the Neus at higher rpms, say ~50k, and then gear down for the same overall speed. This type of setup seems to run cooler and just better. Doesn't that provide more EMF samples at lower speeds making for smoother starts and low-rpm running?

From an ESC/motor standpoint, which is better? Judging from the kv values of the bundled motors you plan to release, it seems that you are leaning towards the second option.

Yeah, sensorless systems are significantly more efficient. The reason is that the controller can vary timing based on RPM, load, etc. and uses the OPTIMUM timing at all loads and RPMs.

With the sensorless startup algorithm we have, there is no need for sensors unless you are doing something like rock crawling -- so why have the reliability issues of the additional wiring harness and PCB, AND the additional cost?

Don't fall for the BS Hype -- sensored is worse in almost every way.

Now for rock crawling, where the motor needs to have torque when not rotating, sensored has a slight advantage.

You missed my point about the Speed Passion GT ESC though. It combines both worlds. Apparently, if you run something like a Novak on this ESC, it runs far better due to the variations in timing that the ESC can select, like you say.

I don't understand how having sensors creates reliability issues? I have an original SS5800 motor that's still going strong. It's been run on anything from 6 cell Nimh to 2s lipo to 12 cell Nimh. No problems yet and the motor has to be a few years old. I've read of the odd problem with a Novak motor here and there, but in comparison to some other motor issues, it's not NEAR as prevalent. Sure, more parts means a higher chance of failure, but for the sheer amount of motors they have out, I've only heard of a few reports where the motor was messed up due to sensors.

I'm sorry, I didn't miss your point. I'm saying there is no reason to run sensors AT ALL. Adding sensors to a motor / speed control has no benefit, and proven drawbacks, so why do it?

Reliability issues are because of additional small wires that can get pinched, broken, fatigued, etc, as well as an additional PCB in the motor that can get damaged or fail. More failure points in the system, lower reliability.

To each their own I suppose. I figured you missed it since you mentioned nothing of that particular ESC.

Why is there no benefit to sensored? Wouldn't that give the ESC an advantage for performance if you could come up with software to take advantage of it? I would assume the more information the ESC knows for sure, the better.

No, because the sensor data is arbitrary to timing -- this is another reason why sensors are a bad idea -- the controller has no idea how much advance the sensors have. Also, usually there is some phase skew from sensors as well -- so that upsets timing as well. As the saying goes, Garbage in, Garbage out...

Adding sensor data to sensorless data would actually make timing worse, and reduce efficiency...

I have a sensored 1/10th scale system from another vendor & I can say for sure. There is no more cogging in my MMM/Nue set up than in that other vendors system. On top of that the MMM is WAY more programmable than that other system as well. Before I ran the MMM, I was worried about cogging but now that I have run the MMM, the subject of cogging does not even come into my head (unless you guys post something about it).

Besides, a lot of what is commonly referred to as cogging is actually not cogging and lack of power or various other things.

I will disagree with you, Patrick, about sensored motors, and there is no "Hype" here. For given application there is a given motor.
Brushed motors can deliver 1000% torque at stand still
Sensored Brushless motors 1/3 of that
Sensoreless not even 20-30% of that sensored systems
You know why and there is no BS here.

Extra wires and sensors are weak links as well as brushes, as long as mentioned them.

Timing with sensors can me more advanced and flexible if you write software accordingly and there is no need to set sensors exactly at "0", you can adjust it anytime as wide as you wish. And back check with motors efficiency, you can even run in dual mode and use sensors for back-check, startup, pulling, holding and braking. If you creative you can even generate brushed mode. Just need to set sensors perfectly spaced to motor poles. If you use encoder instead of hall-sensors, or even IC you can be completely independent of motor type/pole count. I am not going too deep in details, but I think you know the rest.

All the rest about sensored systems in only limited by flaws in software.

I actually was saving this sensored idea from suicideneil for desert, but I guess it would be impossible to convince you otherwise... :(

To BrianG:

Brian, I wasn't expecting question like that from you. Efficiency curve defines given motor RPM range. Number of Poles, Motor diameter and rotor length defines torque and speed factors. (Rotor weight, air gap, core material, windings, delta/star...) From motor parameters you can plot it quite easy.
For the same Power motor with
More poles -> Lower RPM Higher Torque
Less poles -> Higher RPM Less Torque.
Smaller diameter -> Higher RPM, Lower Torque
Bigger diameter -> Lower RPM, Higher Torque
Longer rotor -> Lower RPM, Higher Torque
Shorter rotor -> Higher RPM, Lower Torque
I do not even know what to comment on "Less" EMF pulses...
I will call it Monday and I am in bad mood and picky, sorry for harshness.

Artur, I already "know" what the answer is, but thought it would carry weight if it came from a designer of the product. I'm just thinking forward to when CC starts carrying their MMM and motor combos.

I see a lot of people running kinda low rpm setups and gearing almost too high, and also the occasional startup cogging.

Sounds like someone has a case of the Mondays! (Office Space reference in case you didn't get it :wink:)

From a typical R/C use standpoint, sensorless commutation works well enough. As Patrick pointed out Rock Crawlers could certainly benefit from a sensor, but I can't imagine any way that it would improve performance anything else out there in the ground market (besides tall rollout 4 cell nimh carpet racers).


Boy what I wouldn't kill for a sensored outrunner system. Maybe in one year.

I'm going to disagree with you here. Sensored motors are marketed with a lot of HYPE -- The makes of sensored systems have been BSing the public for a couple years now, telling them that sensorless is inferior, in the hopes that people will buy into their hype. And some people have bought into the hype.

Try this: Take a sensored motor and measure how much phase imbalance the sensors have -- you'll be astonished and shocked at how BAD the sensors are. Typical sensor phase imbalance is around 10 degrees from sensor to sensor, and that's just PHASE imbalance. Even on high-end military motors there is at least 5 degrees of phase imbalance -- just because Hall sensors trigger points aren't precise. And timing itself is completely ambiguous with a sensored system - - and because it is ambiguous, it can't add any meaningful data to the rotor position estimate.
Not only that, but sensors also trigger incorrectly all the time during run (by winding flux) -- and the sensor controllers blindly follow the incorrect sensor outputs...

Granted, sensored systems can produce more torque at stall than sensorless systems. But in RC applications we can generate more torque at startup than is required using only sensorless algorithms.

You state: But that really isn't true. Once the sw PLL is locked (within a few commutation cycles) we can control timing to within .1 degrees -- how can an ambiguous sensor system with jitter and phase imbalance add to that? Adding sensors into the equations just makes the position data LESS precise.

Believe me, we build hybrid controllers (that start sensored and run sensorless) for both the military and industrial applications, and have been building those types of controllers for years.

We've done all the math, the simulations, and modeled the systems extensively. Once the sensorless software is locked, sensors add nothing but noise to the system.



Sensors are useful for generating a lot of torque at stall (> 20% PWM duty cycle lengths), but with the motors we use in RC, we can't even PWM at 20% during stall -- they would draw thousands of amps. So even at a 20% max duty at stall we can generate PLENTY of torque at startup -- Seen the videos of people backflipping Monster Trucks with the MMM from a standing start? :smile:

So there's my rant about sensors. With respect to timing and control, Sensorless = precision, Sensored = ambiguity.

Ok, so I've kinda got the "Mondays" myself. 'Cause I have to get up at 5:30am to fly to California. And I don't want to go to California.
And it's been raining here a lot, and my backyard is flooding and mulch is pouring into my pool and I keep having to spend hours cleaning it out.
And I got an E-Revo and I haven't even had time to install a Monster in it yet.

And I keep editing this post 'cause I don't want people to think I'm angry or being aggressive about this... I just want to present the facts...

I must be the only person who had a pleasant Monday...:lol:

my monday was fine. the only thing that is getting me down is Pdelcast has an e-revo and I dont (haven't even heard of traxxas shipping yet). :mdr:

luckily my job allows me 4 days off at a time :mdr: ..... anyways all this extremely deep electric talk was just basically bathroom reading until i saw that patrick has an e-revo .... :yipi: ...:surprised: ....i want one:no:

I don't know all the fancy shmancy terms and techno-blurb, but I have a lot of experience with a lot of controllers. Sensored USED to be better. It was not hype, it was fact. A year or more ago, sensorless simply was not as smooth, especially at low speeds. The sensored-smooth advantage is all but gone now and I won't likely buy another sensored system myself. There was a time though, that I wouldn't even consider sensorless because of the rough initial acceleration.

A lot more than a year ago, if ever IMO. I had a mamba max system over a year ago, and it was the smoothest thing I ever drove even back then, on any batteries :)

Don't feel bad, I think I have a pool in my basement! The water is coming up through the floor almost faster than I can pump it out. And just when I make decent headway, the stupid weather decides to drop another 1-2" of rain in 5 minutes!

And since my workshop is in my basement, it's not fun. I only have standing water is the low spots, but still tough to work on things hopping from high spot to high spot. Grrr.

I guess when it rains, it really does pour...

Just to rub it in guys, we are experiencing a mini-heat wave today; boiling hot and sunny!

My MMM has been 100% smooth in bench testing, save for the brake issue which I am working on, so I guess sensors arent really needed in this application anymore with the better software of sensorless. That said, I still would like Novak (boo! evil!) to release a 6s sensored esc, with realistic specs, and proper motors. I guess I will be able a realistic comparison now since I can run both escs on 4s lipo and 14 cells, with the hv6.5 motor providing the motive power. Cant say fairer than that really...

What I want to see is a Novak HV motor run on an MMM. If sensored technology is really THAT bad, AND and people are consistantly getting sub 150* temps on the motors and I am getting an easy 45-50 mph with my hv4.5/8ight, I can only guess that using an MMM/Novak HV setup would be pretty sweet. Not only that, but a HV6.5 or 7.5 on 6s would also be pretty sick.

You may be in luck my friend.

The MMM runs my hv6.5 perfectly, no cogging at all or even the slightest stutter. I plugged in one of my 6s A123 packs, which is equal to about 5s lipo in voltage, for testing with. I wound it out to max rpms, and all I can say is fook may, its abit scary at that speed.....

I would like to thank Patrick for his outstanding attitude and willingness to come on such forums and others to listen to customer feedback and field questions (tho innane and ignorant some may be.)

That being said, I still stick to my wish list. A 8S capable controller with outstanding hardware and software. Good customer service is obvious (and well handled by CC.) A relatively small footprint, simple mounting, reliable heat dissipation w/o the use of fans (as Pat dC said, K.I.S.S.) and an amp capability of 120-160A or so. Preferably splashproof. 6mm or 5mm bullets would be great. A switching BEC is gravy. Oh, also anti-spark tech built in. (Anyone that's plugged a 40V battery into an ESC w/o pre-charged caps know what I mean.)

The MGMs are outstanding hardware, but the software is primitive compared to CC's. Its not horrible, but CC is so far above everyone else software-wise it's no comparison. As said before, a CC version of the MGM 16024 would be the end all of controllers. I realize this would be a very niche controller of a already small market, but wishes can be anything.


Patrick, I do have one question on the controller you are toying around with.. what would be the purpose of of going higher than 8S? IE, is what you are targeting an 1/8th scale or 1/5th scale controller?

For 1/8th scale, from my bit of experience, going above 8S gives diminishing returns. Ideally you are shrinking the battery capacity as you are upping voltage. 10 or 12S of even a 3.2Ah pack is pretty heavy. Small lipo batts (say below 3.2Ah) start losing energy density ( ie mAh/g) so any eff benefits lose ground to greater weight to get the required runtime. If its the same FETs to be used, then great, more V headroom. If its pricer FETs, can the specs/price come down? Tho I'm sure you've already considered this...

OTOH, are you attempting a 1/5th scale/Prius capable Big @$$ M.F. MMM ESC (BAMFMMMESC)? I just wonder if it would be a bit overkill for 1/8th scalers who want to run HV (a minority of a minority) in power and price. I can't see an 1/8th scale that would ever need or even use a 12S 3200 mah pack.

For marketing purposes you may want to cut down the name a bit tho, maybe just "Godzilla Max" (he was a lizard too.) BAMFMMMESC does seem kinda catchy tho. :)

Enjoy it! Weird weather in the US lately. Intense heat on the East Coast, snow on the West Coast (north, but still abnormal for June), and crazy rain/floods around my area that are damaging properties all over the place.

its from all those nitro people

I bow to Mr Patrick del Castillo (Where is the bow smilie). Reading posts like that makes me wish I worked at Castle.

Monday-Monday - it is Friday :)

Patrick you are:

• talking about spindle motors...
• miss the fact that we would like to have hybrid system (or based your opinion)
• sensors imbalance is fixed (constant for given motor) and can be offset (I am not even going into phase shift with RPM business)
• optical sensors are rock-solid
• motor windings are even worse in terms of matching, not even considering that load on the motor shaft is not constant in car-applications
• once motor spinning use sensoreless feedback, it is free no hardware to add
• braking, pulling and holding that is what you can do great with precise rotor position feedback
• for $$$ you can add encoder and it can be single wire, or even wireless

Couple thoughts, sensoreless ESC starts at stand sill only if you have backslash or you being in motion before or at least kept memory of position (Or very clever software :) ). 20% PWM at stall is not high current if you matched with magnet pole and motor is rated for the load.

Hall sensors are cheap and sensitive to all kind of things, including magnets or screwdrivers in their proximity, but can be shielded or replaced with opto...

All the rest is irrelevant, because there is no good sensored controller on the RC market to do the comparison.

____
MGM hardware with CC software - I am hearing that allot. Patrick have you checked what is different between your creation and theirs?

• No, I'm not... talking about motors in general - - applicable whether using spindle motors, linear motors, axial or radial flux, etc.
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) miss the fact that we would like to have hybrid system (or based your opinion)

  Nope, didn't miss that either. Didn't say I wouldn't make a hybrid system either... just that the halls are useless in most situations. :yes:
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) sensors imbalance is fixed (constant for given motor) and can be offset (I am not even going into phase shift with RPM business)

  So -- Why would I want to take my high accuracy sensorless position and use it to correct an inaccurate sensor? Why? To what end? To lower my position accuracy? To lower efficiency?
  
  That's like saying -- "Lets measure this distance with a tape measure, and then I'll count how many steps the distance is. Then we'll correct the number of steps using the measured distance, because we know the number of steps in inaccurate." -- Doesn't make any sense to correct an inaccurate measurement using an accurate measurement in the first place. There is no additional information that the hall sensors add to the system. Only noise -- which I don't want to add to the system.
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) optical sensors are rock-solid

  SLOW, same imbalance problems as halls...
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) motor windings are even worse in terms of matching, not even considering that load on the motor shaft is not constant in car-applications

  Phase imbalance is compensated for in sensorless -- so it doesn't matter. Not compensated for with sensors... another sensor drawback...
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) once motor spinning use sensoreless feedback, it is free no hardware to add

  That doesn't even make any sense. Sensorless isn't free -- it requires a lot of hardware... :neutral: Adding sensor support on top of that adds cost.
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) braking, pulling and holding that is what you can do great with precise rotor position feedback

  Holding torque is better sensored. But not really important in this hobby except MAYBE for rock crawling. I don't understand where braking would be different - - same positioning algorithm as running the motor, just different FETs being turned on.
  
  Quote:

  Originally Posted by GriffinRU (Post 181801) for $$$ you can add encoder and it can be single wire, or even wireless

Throw more money at something that doesn't work well in the first place? No thanks... :)

Quote:

Originally Posted by GriffinRU (Post 181801) 20% PWM at stall is not high current if you matched with magnet pole and motor is rated for the load.

Incorrect -- remember there is no back-EMF at stall, and most (hobby type and high performance) motors don't have enough inductance to enter that into the equation -- So the only thing sure to limit current at stall is phase resistance. There is a non-linear (1/x) relationship to current draw vrs RPM at a given PWM level.
Exactly -- so why spend money trying to fix something that doesn't work well to start with?

Don't think I don't hear you -- I'm being purposefully obtuse to make my point:whistle:... -- there are situations where a hybrid would be nice to have. I just don't think that 95% of people need (or would benefit from) a hybrid system, and the reliability and efficiency issues outweigh the usefulness.

How about a high pole count sensored inrunner or outrunner for us Crawler guys? Optimize slot/pole ratio for about 2700kv and 3 cell lipo. If you don't do it I will as my number 2 project.

Speaking of crawlers, wouldn't a stepper motor be a better idea altogether? Granted, the wiring would be a PITA depending on the number of steps...

Patrick, as I said earlier for given application.

If optics are slow than you are really fast my friend!!!!!!

Quote:
Originally Posted by GriffinRU http://www.rc-monster.com/forum/imag...s/viewpost.gif
20% PWM at stall is not high current if you matched with magnet pole and motor is rated for the load.

Incorrect -- think about it. (Hint -- it has to do with back-emf)

No comments, Maybe I am not clear, but how you get the motor moving if you cannot draw the current...

P.S. It is pointless discussion, you like what you have, I am trying to let you know how it can be done right for given application.
You current sensoreless ESC is great for RC but I am looking for rock-crawlers as application for brushless as well.
And out-runners are great motors and with sensors can be direct coupled to wheels...

It would overheat in no time.