Go Back  R/C Tech Forums > General Forums > Radio and Electronics
power versus torque at rpm >

power versus torque at rpm

Like Tree4Likes

power versus torque at rpm

Reply

Old 12-22-2018, 11:23 PM
  #16  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

Bry195,

Your lack of fundamental knowledge of the hobby is glaringly obvious.

1. Motor architecture is set forth by the governing bodies. This was done as a cost control measure, and to prevent proprietary motor/esc combos. Additionally, there are classes of racing where the motor must be a specific number of turns of wire in the stator and rotors of specific dimensions. Typically these classes also prohibit the use of speed controls with dynamic timing. The manufacturers make motors to fit to these regulations.

2. An rc car is not anything comparable to a servo motor driving a robotic arm (or whatever motion control system with which you have experience). You can't calculate a load and select some magic motor. The amount of current a motor is going to draw is always going to change. Grip levels change as tires wear and the track groove changes. How perfectly is the car set up? Is the car mechanically perfect? Is the car tweaked from a crash? And the most important question, how close to the limit can the driver drive? An rc car isn't something controlled by a computer. A big part of how close to the limit you can drive comes from how the throttle responds. Motor wind, gearing, rotor size, motor timing, and the bevy of esc adjustments are plenty to make a car drive how you want.

So you want to build a dynamic load dyno? What do you think you are going to learn and do with that? You aren't going to find some magic timing setting that is absolutely perfect. Your dyno based setting will not be anything more than a starting point, just the same as the unloaded current draw recommended by some. The whole point of making adjustments is to lower your lap times, but if you can't drive at the limit, you lack the ability to test to see if the limit moved. When you can set lap records, and can run 95% consistency you have the ability to use lap times as your metric, which is the whole point of making adjustments. This is the way the fastest guys make adjustments. Additionally, instead of trying to apply whatever nearly useless data you will come up with from a dyno, it is much easier to use an esc with data logging. That way you can see what the car is doing when you are actually driving at the limit, and then make adjustments after that. Can you get a car to load and unload on the dyno the same way it does on track?

Post 5 race results where you set fastest lap and have 95%+ consistency. If you can't, you lack the ability to push current technology to the limit, which means you have no authority to try to move the limit. Don't worry, there are a few other engineers on here who think they can magically make up for their lack of driving ability by trying to reinvent the wheel.

Last edited by waitwhat; 12-23-2018 at 12:16 AM.
waitwhat is online now  
Reply With Quote
Old 12-23-2018, 12:41 PM
  #17  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by waitwhat View Post
Bry195,

Your lack of fundamental knowledge of the hobby is glaringly obvious.

1. Motor architecture is set forth by the governing bodies....

2. An rc car is not anything comparable to a servo motor driving a robotic arm (or whatever motion control system with which you have experience). You can't calculate a load and select some magic motor. The amount of current a motor is going to draw is always going to change....

So you want to build a dynamic load dyno...You aren't going to find some magic timing setting that is absolutely perfect.... Additionally, instead of trying to apply whatever nearly useless data you will come up with from a dyno, it is much easier to use an esc with data logging. That way you can see what the car is doing when you are actually driving at the limit, and then make adjustments after that. Can you get a car to load and unload on the dyno the same way it does on track?

Post 5 race results where you set fastest lap and have 95%+ consistency. If you can't, you lack the ability to push current technology to the limit, which means you have no authority to try to move the limit. Don't worry, there are a few other engineers on here who think they can magically make up for their lack of driving ability by trying to reinvent the wheel.
lets play a game where it seems like we are talking to each other but really we aren’t. The good news is that i have no intentions of debating concepts off in left field and I have been insprired by your comments to talk about something that makes sense. I am a person first... and an engineer second. If im overly/underlay descriptive you have the right to appeal to humanity and I will take pride in providing language that is understood by a wider audience. Its my intention after all.

1-take identical motors and mount 1 of them to an aluminum bulkhead and one of them is mounted to a piece of plastic and run them at 10 amps for 20 minutes. If you determine that the one mounted to the bulkhead stays cooler longer can you use that to your advantage? Can you make improvements other than just a fan or mounting to the bulkhead? How much can i improve the power my motor delivers by understanding the thermal limitations of my choices? What if the number the manufacturer provides (5.6 amps) has 30% safety margin built into it?

Positioning application priorities-High rotor inertia then power followed by thermal time constant
Continuous velocity application priorities- power, then thermal, then rotor inertia
reciprocating between starting and stopping-in between positioning and velocity-temperature first, rotor inertia second, power third.

2-Load dyno-well i suspect i can answer most of my questions from number 1 but Im pretty sure a few guys would want to tell me something about myself rather than have a reasonable debate. So what will a load dyno do? Well my dyno is and isnt what I expected. The flywheel has a huge inertia compared to what i would have expected for a power source that reciprocates (up and down) on a regular basis. Automotive dynos are measuring an acceleration only and from 0-100% should be 10 or 15 seconds. An electric motor is different. Gauss and rotor inertia may have a combustion engine equivalent but its not important because it only powers in one direction. I hope to develop a continuous duty value on the load dyno. An obviously make the peak duty better but thats not a big needle mover. As I hinted the other thing I would like to do is quantify rotor inertia to gear train ratios and how to maximize your driving style.

There is a transmitter input to the dyno and my radio can export csv files. Eventually I hope to feed the ESC telemetry back into the dyno via CSV.
the load of an inertia dyno is completely unrealistic for an electric motor...great for a gas motor but not for electric. Does it take 10 seconds for your car to accelerate on track? In fact for repeated acc and dec you need a gear ratio that gets the rotor inertia around 30:1 if you want to accel and deceleration optimally.

3-none of this is new. Sometimes you can apply science it to all kinds of things that may appear to be a reinvention of the wheel. Where its coming from, its old news and where i hope to apply it...allot of RC people talk about or around it just because they dont do it everyday or wouldnt know where to look if they wanted to focus on it. Its almost like its there subconsciously.

Last edited by Bry195; 12-23-2018 at 03:47 PM.
Bry195 is offline  
Reply With Quote
Old 12-23-2018, 10:35 PM
  #18  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

I outlined some information about duty cycle and windings shorting across insulation to create a distinction between safe peak, intermittent, and continuous current. I guess it adds come context to how motors are limited by the engineering. Maybe i can add some juicer facts that might get you a feel for whether or not Im describing things that result in large or small results.

motion control is really about applying physics to electrical or fluid power motion. It can get complicated but 80% of everything you will ever apply comes from the fundamentals that anyone can understand in a very short amount of time if they want to.

When you use a system (motor and (drive) ESC) from a company with limited resources and thus limited documentation its usually not blindingly inferior to products from the big boys. If you spend the time testing you can bridge the gap and develop your own specifications. For example 1 motor mfg might publish an intermittent current rating 4 times higher than continuous while a smaller company would limit you to 2 or 3 times. You can determine in situ if you can do 4 times. safety margins are selected to protect you but also protect the mfg.

After you understand torque to accelerate your motor and drive train you might feel that you didnt really achieve the “feel” you were hoping for out of a particular system. The next thing to understand about acceleration is inertia mismatch. The resistance to change in speed. The rotor in your motor has an inertia. Everything that it is accelerating (wheels, drive shafts, belts, ..has an inertia. The gearing in between is used to create a mathematical relationship between the inertia of the motor and the inertia of the drive train.

Smooth acceleration come from a motor to drive train ratio above 30:1 but gets very laggy above 60:1. For example 60:1 might take 8 seconds to achieve full speed while 30:1 would be a couple seconds. 1:1 would be .5 seconds for the motor to accelerate the drive train.

Well it doesnt end because once you master torque and inertia mismatch there are a couple things that limit you. If your accelerations are shorter in time you are generating torque/efficiency and heat in a different way. To keep it simple a smaller mismatch will make the car punchy. It will also make the motor hotter because you are running through the efficiency islands quicker. You will be running the motor closer to peak but for a shorter amount of time. our MFGs are bringing us great products on very limited resources and im sure they are playing it safe. That safety comes from (as far as i can tell) 1 number. For example the 5.6 amps of timing discussed earlier. I’ll bet there is significant increases available by being able to accelerate 10% better or have 10% higher top speed. Rotating inertia and continuous/peak power are simple concepts.

Some people are artists and develop a “feel” that they can repeat time and time again. I’ll do some inertia and duty cycle testing in the near future and we can decide if there is anything of value in the science but at the worst it will help guys like me that arent artists be repeatable.

Bry195 is offline  
Reply With Quote
Old 12-24-2018, 01:19 AM
  #19  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

See, what you fail to realize, is that there is the limitation of the track layout. You can gear a motor taller an throw more timing at it, and up to a point it will go to a higher top speed, but you will never be able to use that top speed on track. You also lose acceleration, which means you will go slower. The larger load will also build more heat, with a diminishing return on power. My data logging let's me see exactly what my car is doing, and how the numbers change when I make an adjustment.

When I said can you get your dyno to load the car properly, I meant the loading that happens as the car yaws though a corner, something you can't test on a dyno (or in a wind tunnel coincidentally).

There is an approximate 4° timing window (according to the motor manufacturers engineers I've talked to directly) where the motor is going to give optimum performance. There is also a window for gearing, and as long as you are in that window, you will have every bit of power that you can. If your track layout changes, a small tweak in timing or gearing is all it takes. Rotor selection is based off what type of vehicle in which the motor will be used, in addition to the wind of the stator. A 1/12 scale pan car on 1s needs a different rotor from a 4wd buggy on 2s. Operating voltage and drivetrain mass have noticeable impacts on performance, but stator winding has a much larger impact on performance.

Are you talking about a spec motor or a mod motor? A spec motor has to be pushed close to the limit to get as much as you can out of it. With a mod motor you can run a really powerful motor timed and geared conservatively. It will make almost no heat, and have still have every bit of power you can use.

Even still, all of these things (aside from the stator windings) amount to marginal gains that will never be realized by someone who can't drive. If you can't drive a car to the limit if the motor is at 95%, moving the motor to 98% won't make you go any faster.

How about this. Build your dyno, do your testing and figure out where you want to set timing and gearing. Run your car and record your lap times. Then, set the timing to where it was set from the box, change your gearing to what the manual specifies, run again and record your lap times. I guarantee you will not see any noticeable change in lap time.

Unless you are going to design some fundamentally different motor (which wouldn't be legal for sanctioned racing), there is nothing you can do on a dyno you can't do on the track. There is a timing range on the motor, and you have a range of gearing options with a certain resolution of steps between ratios. I can try the same things you can try, on track, with lap times and real world telemetry as my metrics. I don't doubt your ability to collect and extrapolate data, I doubt you will see a noticeable difference in lap times because the gains really are that small.

Do you seriously think the engineers that design these motors don't know exactly where the sweet spot is on their motors regarding timing and gearing? Do you not realize there are team driver setup sheets that show esc settings, gearing, rotor selection, et cetera? The high performance segment of an industry puts emphasis on maximizing different aspects as opposed to the manufacturing automation segment.

Lastly, a properly utilized cooling fan is the most effective passive cooling device for a race rc car on track. I can run my car for an entire battery with a fan, and my motor comes off the track at ambient temperature. I've also pre-cooled my motor with dry ice, and have even considered a small onboard dry ice compartment to chill the air blown by my fan. Not really necessary based on my conditions, maybe if I was running touring car on asphalt in the 105° summer. Also, I've got an aluminum gearbox on my car (the stock gearbox is plastic) and one of the reasons is the heat dissipation. It also adds rigidity and ballast to the rear end of the car. Finally the machined cases line up much better than the plastic counterpart which translates to less drag. I can guarantee every aspect of an rc race car is there for the sole purpose of going fast. There are teams of engineers and drivers whose literal livelihood depends on it. Trust them, they know what they are doing.

​​​​​​
waitwhat is online now  
Reply With Quote
Old 12-24-2018, 08:51 AM
  #20  
Tech Elite
iTrader: (7)
 
Join Date: Jul 2004
Posts: 2,536
Trader Rating: 7 (100%+)
Default

Originally Posted by Bry195 View Post

Some people are artists and develop a “feel” that they can repeat time and time again. I’ll do some inertia and duty cycle testing in the near future and we can decide if there is anything of value in the science but at the worst it will help guys like me that arent artists be repeatable.
A lot of us are racers, and are interested in seeing tangible results. So, there's a high degree of skepticism when some of us see a ton of technical discussion without any final takeaway in the form of a product. When people question you, the condescending responses don't do much to help your cause.

just my $0.02
gigaplex and waitwhat like this.
071crazy is offline  
Reply With Quote
Old 12-24-2018, 02:23 PM
  #21  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by waitwhat View Post
See, what you fail to realize, is that there is the limitation of the track layout. You can gear a motor taller an throw more timing at it, and up to a point it will go to a higher top speed, but you will never be able to use that top speed on track. You also lose acceleration, which means you will go slower. The larger load will also build more heat, with a diminishing return on power. My data logging let's me see exactly what my car is doing, and how the numbers change when I make an adjustment.

When I said can you get your dyno to load the car properly, I meant the loading that happens as the car yaws though a corner, something you can't test on a dyno (or in a wind tunnel coincidentally).

There is an approximate 4° timing window (according to the motor manufacturers engineers I've talked to directly) where the motor is going to give optimum performance. There is also a window for gearing, and as long as you are in that window, you will have every bit of power that you can. If your track layout changes, a small tweak in timing or gearing is all it takes. Rotor selection is based off what type of vehicle in which the motor will be used, in addition to the wind of the stator. A 1/12 scale pan car on 1s needs a different rotor from a 4wd buggy on 2s. Operating voltage and drivetrain mass have noticeable impacts on performance, but stator winding has a much larger impact on performance.

Are you talking about a spec motor or a mod motor? A spec motor has to be pushed close to the limit to get as much as you can out of it. With a mod motor you can run a really powerful motor timed and geared conservatively. It will make almost no heat, and have still have every bit of power you can use.

Even still, all of these things (aside from the stator windings) amount to marginal gains that will never be realized by someone who can't drive. If you can't drive a car to the limit if the motor is at 95%, moving the motor to 98% won't make you go any faster.

How about this. Build your dyno, do your testing and figure out where you want to set timing and gearing. Run your car and record your lap times. Then, set the timing to where it was set from the box, change your gearing to what the manual specifies, run again and record your lap times. I guarantee you will not see any noticeable change in lap time.

Unless you are going to design some fundamentally different motor (which wouldn't be legal for sanctioned racing), there is nothing you can do on a dyno you can't do on the track. There is a timing range on the motor, and you have a range of gearing options with a certain resolution of steps between ratios. I can try the same things you can try, on track, with lap times and real world telemetry as my metrics. I don't doubt your ability to collect and extrapolate data, I doubt you will see a noticeable difference in lap times because the gains really are that small.

Do you seriously think the engineers that design these motors don't know exactly where the sweet spot is on their motors regarding timing and gearing? Do you not realize there are team driver setup sheets that show esc settings, gearing, rotor selection, et cetera? The high performance segment of an industry puts emphasis on maximizing different aspects as opposed to the manufacturing automation segment.

Lastly, a properly utilized cooling fan is the most effective passive cooling device for a race rc car on track. I can run my car for an entire battery with a fan, and my motor comes off the track at ambient temperature. I've also pre-cooled my motor with dry ice, and have even considered a small onboard dry ice compartment to chill the air blown by my fan. Not really necessary based on my conditions, maybe if I was running touring car on asphalt in the 105° summer. Also, I've got an aluminum gearbox on my car (the stock gearbox is plastic) and one of the reasons is the heat dissipation. It also adds rigidity and ballast to the rear end of the car. Finally the machined cases line up much better than the plastic counterpart which translates to less drag. I can guarantee every aspect of an rc race car is there for the sole purpose of going fast. There are teams of engineers and drivers whose literal livelihood depends on it. Trust them, they know what they are doing.

​​​​​​
I hope i wasnt too tough on you before. It seemed like you were talking to everyone else or yourself but not me. I apologize. I disagree or suspect you may be wrong on a few of this recent post but its all reasonable and the way you delivered the message is very much appreciated.

I havent failed to realize there are track limitations. Once you quantify the track limitations you work the car specs to fit those limits. system sizing is based on torque, inertia, and rpm. The data comes from the application. In this case the track. If you add up all the peak torques and average them then add up all the rpms and average them what you will find is if the track is very rpm dependent or very torque dependent. If you average 1nm at 20000 rpm then you set the timing to center the bell curve on that average. (Tiiming)

So you generate a point to center the curve (makes the car efficient for that track) and set your timing. We know more power means more of whatever and in this case more efficient racing for the track that the averages were developed from. but this is just making the system efficient and your average point is inside the curve but not on the peak of the efficiency curve.

if you can run this track non-stop for 30 minutes and the temperature never exceeds 120 at 5.6 amps you are running below the safety margin of the motor. Being safe is good but being fast means you want to more rpm or more toque and less safety. (This also applies in reverse if you are too hot). Choose a new lap time you want to run and back calculate the new accel and speed rates to achieve that average time. If you keep doing this you will eventually hit 170f. Every time you will be moving the center of the curve to maintain efficiency and obviously your power is going up. Once you stabilize at 170 you can say that 1.2nm at 23000 rpm with the weight of your car is 100% duty for the combination.

you have a dyno so you can create the torque curve. Based on what you learned from the first track you can apply the duty cycle average you applied to every track and change going forward. I dont disagree that people have developed a feel for FDRs and timing and so on that is pretty good. What im saying is that if someone as smart as the guys who developed those motors shows them how to use it they will be significantly faster and consistent.

the only thing a motor sees is torque and speed. No reason I cant see yaw reflected in one of those numbers or a derivative of it.

Im talking about taking 2 like motors and making one better than the other but i run stock motors. Your explanation on mod motors makes sense.

if i take 2 identical cars with identical drive train inertia and I select a 17.5 neodymium rotor and a 17.5 iron core rotor and run them side by side the iron core will bring the inertia mismatch closer to 1:1 than the Neo will because iron is more dense. The iron core will be more responsive than the neo but that may be the oposite of what you need.

windings and voltage are all important but we should probably just talk about comparing 2 stock motors.

I see a window of 4 degrees for a given load and rpm that shifts current dramatically but if i flatten the curve and extend it to the right i could ad 30 degrees of timing and then tune that bulk move by plus or minus 2 degrees to make it efficient.

You would be amazed by what factory engineers dont know. Its not a knock on them but some are better than others and the best go to the money. RC doesnt have the money. There really isnt that many motor designers out there. Probably less than 1000 globally. even if we assume they all know there sweet spot the best way to sell your product if you know it is to document it. I havent seen a torque curve from a motor mfg in a long time. I promise you that if i run a motor to its limits but under the torque curve it wont be damaged and it will be the fastest it can be. I can also promise you that if i run it over the limits 1 time it will be broke. If they know there product why dont they provide real documentation? I would love to review anything I have talked about with one of the motor designers from any of the suppliers.

air is not good at cooling. Its not good at cooling quickly. Peak current is quick heat. The flange on the end of your motor will dump allot more heat to another piece of metal much quicker than a fan. You need a fan to dump bulk heat. You need the motor flange to soften the peaks and give the air time to work.

I trust everyone even after they make a mistake or have a misunderstanding. There are 50 ways to skin a cat if you study the cat or have skinned enough of them. Ive made mistakes and Ive designed motor and drive systems for more than 20 years. Ive never used it as a reason to not weigh the problems of a lesser experienced engineers design problems. If a motor design engineer tells you that all you have to worry about is one variable.....or gives you a clear straight answer he is probably not a motor design engineer.
Bry195 is offline  
Reply With Quote
Old 12-24-2018, 02:41 PM
  #22  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by 071crazy View Post
A lot of us are racers, and are interested in seeing tangible results. So, there's a high degree of skepticism when some of us see a ton of technical discussion without any final takeaway in the form of a product. When people question you, the condescending responses don't do much to help your cause.

just my $0.02

much appreciated and I tried to make a point rather than insult in response. If someone is standing face to face with you and the first thing they say is how unqualified you are you have to make a choice about how you will respond. i put allot of thought into what i responded with. The fault is really mine. Nobody questioned me. That would have been fine but debating with enginees can develop some habits that the rest of the world just has to see to understand. I could have claimed that i was speaking to completely unqualified people and broke it down inch by inch. That’s not me (most of the time). But I also have a hard time ignoring people because most of my favorite people are just as screwed up as me and they could become friends.
Bry195 is offline  
Reply With Quote
Old 12-24-2018, 03:50 PM
  #23  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

How do you plan to quantify an on track load? Literally every single time a car hits the track the exact load you can put through the car is different. By treating a load as a constant instead of a variable you're already aiming for a target that isn't there. Even F1 teams can't perfectly correlate data to make their simulations and dyno runs perfect. They could literally program an AI to control a virtual car absolutely perfectly to 100%. They don't because they work in the fine area in what happens when the human driving the car drives it to 100.5%?They still test the car because there is the human variable, and just like rc racing, it doesn't matter what you can do anywhere else but on track.

If you are slow driver you will have a wildly inconsistent set of data and if you tune everything to that you are leaving more performance on the table. If you analyze data from a pro driver and tune your data set to that then congratulations, you now have a small data sample size to set a motor for the way someone else drives.

Every motor is different, every battery is different, every car is different every driver is different. You will never be able to make an exact representation of what the car is doing. You will never eliminate your lack of driving ability from the equation. Your time would be much better spent learning to be a better driver. That is the only thing that will drop your lap times to get you on pace.

You assume my motor comes off the track at 120° without the fan. Congratulations, you are wrong. My motor comes off the track at 160° after a full battery. My timing has never been set by an arbitrary number from running unloaded, so congratulations on being wrong again. My timing is set by performance on track, and the plateau temp of the motor. I know a small timing or gearing change is all it takes to keep my motor in that sweet spot. I run a fan to prevent heat fade, and because my fan is hooked straight to the battery it is powerful enough to keep the motor within 20° of ambient temperature.

You greatly underestimate the engineering that goes into rc race cars. Manufacturers definitely run computer simulations on their designs, and do continuous development work. They get paid to go fast, they know better than you, trust me. There is a reason it is called a setup window.
waitwhat is online now  
Reply With Quote
Old 12-24-2018, 10:11 PM
  #24  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by waitwhat View Post
How do you plan to quantify an on track load? Literally every single time a car hits the track the exact load you can put through the car is different. By treating a load as a constant instead of a variable you're already aiming for a target that isn't there. Even F1 teams can't perfectly correlate data to make their simulations and dyno runs perfect. They could literally program an AI to control a virtual car absolutely perfectly to 100%. They don't because they work in the fine area in what happens when the human driving the car drives it to 100.5%?They still test the car because there is the human variable, and just like rc racing, it doesn't matter what you can do anywhere else but on track.

If you are slow driver you will have a wildly inconsistent set of data and if you tune everything to that you are leaving more performance on the table. If you analyze data from a pro driver and tune your data set to that then congratulations, you now have a small data sample size to set a motor for the way someone else drives.

Every motor is different, every battery is different, every car is different every driver is different. You will never be able to make an exact representation of what the car is doing. You will never eliminate your lack of driving ability from the equation. Your time would be much better spent learning to be a better driver. That is the only thing that will drop your lap times to get you on pace.

You assume my motor comes off the track at 120° without the fan. Congratulations, you are wrong. My motor comes off the track at 160° after a full battery. My timing has never been set by an arbitrary number from running unloaded, so congratulations on being wrong again. My timing is set by performance on track, and the plateau temp of the motor. I know a small timing or gearing change is all it takes to keep my motor in that sweet spot. I run a fan to prevent heat fade, and because my fan is hooked straight to the battery it is powerful enough to keep the motor within 20° of ambient temperature.

You greatly underestimate the engineering that goes into rc race cars. Manufacturers definitely run computer simulations on their designs, and do continuous development work. They get paid to go fast, they know better than you, trust me. There is a reason it is called a setup window.
Im sorry if you assumed I was speaking to you throughout the topic. My apology was to you and the rest was hypothetical numbers. When I disagreed and agreed it was nothing personal. we are talking which is good and I’ll try harder to leave room for you to disagree and refute.

every single application is different and the wants and needs that go into controlling it. every application can be described mathematically or anecdotally. that really is what the mid level system designers move on to after they are done applying the latest and greatest cooling techniques or saving 10 cents on copper. they start working out the math for applications of the motor and refining the design for those applications.

I want to tell you about AI but im not trying to insult you. AI is responsive to its senses. it wont respond to anything it hasnt been taught or doesnt sense. You are correct but your comment isnt all or nothing. i mentioned torque, velocity and position control awhile back. these are the principals that AI use in an application. this is what 99% of motion control is about. But its not anymore. Response times are boarderline predictive now and the amount of code that can prepare responses in advance is turning AI on its ear. I worked for bosch and we developed a method to rewrite a set of responses automatically. essentially rewrite a program if we were in water or on land and everything in between. Darpa is funding development of this philosophy applied across motion and analytics...Stochastic control if your interested.

I dont know what to say about how i should spend my time other than you are right. im a good new racer and I agree that practice is the most important thing I can improve. What i understand about how to apply motors consumes very little of my resources because its what I do. In fact if you are looking for a weak spot in my comments you might see im not highly motivated to do allot of work to prove any of it beyond what I have committed to. The reason being that i have done just what I have commented on 10’s of thousands of times and taught to at least a 1000’s of times and disagreements are a dime a dozen. Im always open to a reasonable investigation or conversation but im old and prefer to pick my battles. I can apply what i know and share the results and someone will say hey that car is accelerates and decelerates 10% better than anyone else around me and I’ll be ok with that while I work on my driving skills but I will put very little effort into that 10% because its not my priority.

the amount of work that anyone has to put into understanding the physics really isnt that much and while they are working on driving better they very well could decide to simultaneously develop a better runnning car and understanding of that car. Maybe a top 10 performer would improve by allot or a little its hard to tell without being exposed to those people. What I do know is that it takes allot more information than 5.7 amps at 42 degrees timing and set your fdr to a temperature. What i am willing to provide is torque curves with peak and continuous values that allow people to apply control systems more consistently. Which will ultimately tell at least me (if im censored) why the information isnt published and nobody is asking for it. its either the audience or the manufacturers. either its too much for the casual user or it means that a manufacturer may feel singled out and everyone wants to feel like a winner. its also possible that nobody investigated it so nobody thought they were missing out on anything.

if a mfg or participant wanted to dive in before i start providing the info they could look at motion control sizing software (free) from allen bradley, siemens, yaskawa, fanuc or my favorite Bosch. I think they call it indrasize now. you can figure out your torque at an rpm and use that to input the application requirement. it will spit out information that tells you what rotor inertia you need. where the torque curve needs to be centered and how much harder you can work the motor based on the intermittent (averages) required for the track to end at 160F or whatever number you prefer.
Bry195 is offline  
Reply With Quote
Old 12-24-2018, 10:34 PM
  #25  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

I trust that manufacturers and racers do and dont know better than i because assuming anything less than that is not respectful to that person. you’ll have to assume some level of knowledge or experience assigned to me Whether you understand what im saying or i speak reasonably or ...When i make this evaluation i like to look at what it costs me and what i may potentially get in return for that fee. If i lose nothing and have an unclear vision of what i would get i wait until i understand the fundamentals of the concept and then i ask questions or relate experience that has the same context. im not saying you have no clue but you also havent lost anything nor will you unless you are not interested and continue to read. i dont want you or anyone else to stop reading but my use of your teeing up descriptions is probably getting frustrating to you and i cant really see it being that fun for people that just want to know. i dont mind addressing generalizations because people need to start somewhere and I appreciate that you are not irate. If you want to understand Im your guy and i am learning at the same time. I may not be your hero but if you get to know anyone including individual motor developers have they proven anything more to you than I? I’m trying to be helpful.
Bry195 is offline  
Reply With Quote
Old 12-25-2018, 03:49 AM
  #26  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

Let me try to explain it another way.

There are a finite number of gear ratios, and you have to select a ratio. There are a finite number of parts you could change to alter the rotational mass of the car and you have to have all of those parts in place to have a working car; furthermore there is a weight minimum, so at some point you have to add ballast to keep the car at legal weight as you remove more rotating mass. There are a finite number of different rotors to adjust the rotor mass, and you have to select a rotor. The timing on the motor will have an optimal settings based on all of these (and many other) factors.

I regularly run on 3 different size tracks. If I am at the small track I gear down a tooth or two and bump the timing up a degree or two. This gives the car a snappier power band good for the tight track. If I go to the largest track I would gear up a little bit and turn the timing down a little. This gives the car more legs up top, with a little less down low snap. These are easy adjustments to make, and they are noticeable on track. 1/10th scale off-road cars almost exclusively run 48p gears, but I run 64p for a finer adjustment resolution. There is an optimal gear ratio and timing setting where the increase in timing or gearing beyond that point produces more heat and no noticeable increase in on track performance. I've found that corner, and it equates to my fanless motor temp plateauing at ≠160°F after a full run. If my fan fails while racing I don't want to risk cooking a motor, so my fan is there to eliminate heat fade, rather than allowimg me to pull a taller gear or run more timing. My timing and gearing already reside at the point of diminishing returns.

I have an outlaw 17.5 that I run from time to time to test different ideas. It has a stator that has extra thin material for the laminations, and a mod rotor that has a lot more magnetic mass than a spec rotor. The motor can pull incredibly tall gears and the top speed is insane. All that top end doesn't do any good on track. Very soon I will have a new rotor to test to see if I can make the motor an rpm monster, and then shift the gearing down significantly, similar school of thought as a mod motor setup.

Next time you go to the track ask some of the fastest guys there to take your car for a spin. When you see them go way faster than you with your own car, maybe it will drive home the fact that your motor maybe being 3-5% from the perfect setting is not holding you back at all, rather that your lack of driving ability is holding back your car. Racing is about the challenge to extract every bit of potential from the car. If you are getting motored down the straight, it is because they slowed down less than you did to get around the corner before the straight, not because your timing is a little off.

Search "tekin how to clinic" on YouTube. It's an hour long, and should give you an insight into how in depth the manufacturers understand their products.

The graph below represents the ideal gearing (green box) for the optimal timing adjustment (red X). There are many factors that determine these two settings and testing on track will absolutely let you arrive at the optimal settings much sooner than your theoretical dyno ever will. There is no advantage you can find because every car is bounded by the same physical limitations. Some people are incredibly good at finding the absolute maximum of these limitations, and I can tell you they are the ones getting paid to design this stuff and to help their driver's go as fast as possible.
​​​​​
Bry195 likes this.
waitwhat is online now  
Reply With Quote
Old 12-25-2018, 10:57 PM
  #27  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by waitwhat View Post
Let me try to explain it another way.

There are a finite number of gear ratios, and you have to select a ratio.

(Me)You just outlined motion control. You have 10 variables in the track and 10 in the car. Each variable in the car gets refined and matched to the requirement as an individual and as a whole.

I regularly run on 3 different size tracks. If I am at the small track I gear down a tooth or two and bump the timing up a degree or two.

(Me)This is the inertia mismatch ratio and centering the (curve) power requirements to available power. There is allot more refinement in this than is apparent.

I have an outlaw 17.5 that I run from time to time to test different ideas. It has a stator that has extra thin material for the laminations, and a mod rotor that has a lot more magnetic mass than a spec rotor. The motor can pull incredibly tall gears and the top speed is insane. All that top end doesn't do any good on track. Very soon I will have a new rotor to test to see if I can make the motor an rpm monster, and then shift the gearing down significantly, similar school of thought as a mod motor setup.

(me) iI love power. If there is enough of it but not too much of it you can probably be more dynamic than a small light weight system that is custom for the application.

Next time you go to the track ask some of the fastest guys there to take your car for a spin. When you see them go way faster than you with your own car, maybe it will drive home the fact that your motor maybe being 3-5% from the perfect setting is not holding you back at all, rather that your lack of driving ability is holding back your car. Racing is about the challenge to extract every bit of potential from the car. If you are getting motored down the straight, it is because they slowed down less than you did to get around the corner before the straight, not because your timing is a little off.

(me) Im an average guy with 25 years of control and 2 of driving. A guy with 25 years of driving and 2 of control would be my wonder twin.

Search "tekin how to clinic" on YouTube. It's an hour long, and should give you an insight into how in depth the manufacturers understand their products.

(Me) thank you i will.

The graph below represents the ideal gearing (green box) for the optimal timing adjustment (red X). It looks close to what I will generate for the motors i get to test. It looks like a torque to timing curve of sorts. If it is then I propose to improve it by adding a way for you to run the motor over continuous. If it is a torque curve a continuous, intermittent, and peak curve will allow you to run closer to 400% peaks and know how many times you can do it and how many times you need to do it.
I take the end bell to 0 and then generate the dynamic commutation in the ESC. I havent decided whether the timing is on an axis or part of the curve but certainly the temperature portion of the curve is integrated into something like your trapezoid.

​​​​​
Lots of great info.

A college kid gets out of school and lands a job as an engineer to design motors. He lives in the theoretical world of amps and volts and thermal mass blah blah blah. The guy before him is now working closer with marketing and/or customers. For 5 years he develops himself by applying what he learned. Motor design only becomes fun when you become an application engineer. Its not right or wrong but you understand the application very well and could probably take a motor that was designed by that 5 year guy and beat him with it. anything more i say isnt intended to take anything away from how much you understand but is intended to respectfully unlock a passion for you or others like you. Guys that may not even completely understand what you wrote.

You are correct but have a few beliefs that may or may not be true but cannot hold any higher value (to the general public) than the (without proffitt) science. Both of us are just having a conversation and beginning to settle on a language. I believe i understand what you are saying and i translate it into things that I have or can ponder. i still agree that to be a faster racer i need to race more and it is a higher priority than motor design. But i design motors, apply motors, teach others, and solve motion application problems that are almost always the first time anyone has attempted it. Dont take this the wrong way. You want me to be a better racer and set aside the engineering because there probably isnt enough impreovement that can be made. With all my heart i want to tell you in the best way that you should spend a little more time with the engineering. Have you ever wondered why an engineer often doesnt give a yes or no answer? Because engineers design on a curve and the biggest sin is being an engineer that is wrong and cant justify the decision. i can’t definitively tell you the science will do better than your ability to apply.

Because i have been applying in so many odd ways for so long i look at the knowns and the unknowns and then I decide for myself if its worth investigating and I believe it is. Modern engineering is all about eking the last ounce of dollar and performance trades to quanitify the finite so that there is less infinite to work on later. Anecdotally I can take your average torque and speed and inertia at one track and apply it to any track you havent ran before just based on the time requirements or aspirations you have. If you set your car up for a short track and a long track and i get torque speed and time data as well as confirmation that you like what you have I can expand and contract the relationship you established on the first track for a brand new long or short track with everything in between. there are factors that are unmeasurable (lets call that mathematical error) but that is usually quantifiable because it gets rolled up in your results. If i get the first one wrong it’s because i took the error out and wanted to minimize your error. However i wouldnt be very good at applications if I insisted everyone is a knucklehead and they need to drive the way i think they should drive but I can quantify the error in most cases and build it into the theoretical. This is what a hypothetical 15 year engineer could get lucky enough to learn. its scary to unravel engineering but it becomes a fact when its succesful over and over again but i still wont sin by making a promise because the hypothetical “you” hasnt been measured. i havent looked at an application and said that there is something more that can be done and been wrong. That probably doesnt help you “trust my gut” but the solution you want me to be happy with is as much trust as mine is. they are both equal.

you have a minimum weight. some weight is better than others. weight that dissipates heat or is unsprung. weight that accelerates rather than resists a change in speed. high frequency applications versus low frequency applications. the math isnt crazy. its a couple formulas (simple formulas) to create averages for what you have and the work you need to do.

i was thinking about another example that may or may not hit home. Hp is tqxrpm/5252. so lets say we have a motor designed to do 1hp of work all day long. center to center on the mounting flange is probably 6 inches or so. if you know that you only need 1hp or tqxrpm/5252 for half of its life can you use a 1/2 hp motor? if i stop thinking in terms of power I can build a system that is half the size and does all things better than the 1hp. there is simple extremely finite math that represents the track (or work) and the device performing the work can be selected or enhanced to do that work in a way that the unmeasurable are extremely minimal. i have been doing a sanity check on the un measurable here. Ive modified my focus and I will modify it more.

I have helped guys in F1 and just about anything you can think of. What you get from those guys is a willingness to investigate. there are hero’s in the world but the gap really isnt that big. I wouldnt be surprised if the domestic mfgs didnt have any motor developers. for the last 15 years torque/Frame size ratio of motors has gone down at least 10-15 percent per year. the reason for that is the tech is at that age (20-30) that everything is getting refined to the nth degree. I have participated in the progression and when i look at RC I see spots of brilliance but I primarily see 20 year old tech that is enough but could do more and if i can do something I will.

Last edited by Bry195; 12-26-2018 at 08:44 AM.
Bry195 is offline  
Reply With Quote
Old 12-26-2018, 05:20 PM
  #28  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

Here is the first part of a three part write up on brushless motor timing. I know it isn't new information to you, but more to show you people in the rc industry do understand exactly how a brushless motor works.

There are always design constraints. In your early example of using a 1/2hp motor instead of the 1hp motor, there is a reason the 1hp motor was specified. It may be possible to design a system that works with the 1/2hp motor, but the 1/2hp system would be physically too large to fit where the 1hp system has to fit. How much more complex do you have to make the 1/2hp system to be able to divide the work to be able to not over work the motor? Is that added degree of complexity worth the increased number of things that could fail within the system?

Perhaps you should read the ROAR rule book to see the exact design constraints of a ROAR legal motor.

It goes:

Overall maximum diameter is 36.02mm measured at whatever point yields the
maximum dimension, excluding solder tabs, screw heads or lead wires. Maximum
length is 53.00mm measured from the mounting face of the motor to the furthest most
point of the end bell, not including solder tabs, lead wires or original manufacturer’s
logo or name. Motor mounting holes must be on 1.00- inch (25.40mm) centers.
8.5.4.2 Stack/Stator:
For Stock, Super Stock and Spec Brushless Motors: The stator construction must be
continuous laminations having the same overall shape, one after the other without
anything in between. The laminations must be of one homogeneous material without
cut-outs, holes or hollow sections other than the three slots for the round copper coil
wires and three slots for the screws used to hold the entire can together. The overall
stator length parallel to the motor shaft shall be a maximum 21.0 mm. The thickness
of the laminations shall be 0.35+/- 0.05mm. A ‘go-no-go’ gauge 14.500 +0.000/-0.005
mm diameter shall pass into the stator, clearing the stator plus its windings and the
electrical collection ring at the end of the stator.
8.5.4.3 Winding:
8.5.4.3.1 ROAR Stock motor: Only three slot “Y” wound stators are permitted. No delta wound
or slot less stators are allowed. Only circular (round) pure copper magnet wire
permitted. The three slotted stator must be wound with 17.5 turns of 2 strands of a
maximum diameter of 20AWG or .813 mm per slot. The resistance for each slot of
the stator shall be tested and a minimum resistance figure will be determined upon
submittal.
​​​​​​Maximum retail price of $149.99

​​​​​​If you can design a motor that fits within these design constraints, and it produces more than a marginal gain, I'd certainly like to try it.
waitwhat is online now  
Reply With Quote
Old 12-26-2018, 10:20 PM
  #29  
Tech Adept
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 105
Default

Originally Posted by waitwhat View Post
Here is the first part of a three part write up on brushless motor timing. I know it isn't new information to you, but more to show you people in the rc industry do understand exactly how a brushless motor works.

There are always design constraints. In your early example of using a 1/2hp motor instead of the 1hp motor, there is a reason the 1hp motor was specified. It may be possible to design a system that works with the 1/2hp motor, but the 1/2hp system would be physically too large to fit where the 1hp system has to fit. How much more complex do you have to make the 1/2hp system to be able to divide the work to be able to not over work the motor? Is that added degree of complexity worth the increased number of things that could fail within the system?

Perhaps you should read the ROAR rule book to see the exact design constraints of a ROAR legal motor.


​​​​​​If you can design a motor that fits within these design constraints, and it produces more than a marginal gain, I'd certainly like to try it.
I appreciate the encouragement but my hands are full. Your author wrote something extremely beautiful because its understandable. He is an engineer and a human. That is a rare breed. My point is to provide some simple data that can be read like a map. Similar to your trapezoid that includes timing and a way to quantify heat generation for the purposes of extracting more out of the motor.

Once people get comfortable with working closer to the red (and gain confidence in the math) they will get ideas on how to push the red further out. My hope is that creates an arms war of technology in RC. I’ll chill out now.

Author said...I believe the reason high timing usually causes motors to run hot is that they do not actually spend most of their time near the top of their RPM range, that only happens at some point on the straight line of a track, say. But the rest of the time (most of the lap!), in the infield, the RPMs are lower, and the commutation is done further from the neutral plane, so a lot of heat is generated.

He believes right. Its easy enough to know where the motor spends most of its time. You align rpm, torque and timing to the track (this is what I have been rattling on about).... and stage 2 (not yet) I’ll see if i can simplify inertia mismatch. Widening the torque curve is great but not if you dont use that width. Dynamic commutation expands and contract the curve as needed. Set the end bell at 0 and run the dyno up. Set it at 10 then 20 then 30...then set your esc to mimic the curve and run it again (its beautiful but unusable in Blinky). The motor will get hot but slower. IIRC I had a ludicrous peak of something like 400 watts (continuous/intemittant) on a reedy splus 17.5. The motor gets less and less efficient but when you race Blinky and you know you average rpm you know the timing and torque curve for that timing because you found it on the dyno. You can run a curve for efficiency or peak torque on the dyno which will tell you the plus and minus to set the end bell to.

Great..you know what the motor can do efficiently and at peak torque across the range of RPM’s. So what if the load is half as much on a track? You will run cooler. What if the load is twice as much? You will run hotter (not from timing issue but because there is still an underlying efficiency unrelated to timing). You just learned to set the RPM and timing but how do we address this other underlying efficiency (below timing)? High frequency versus low frequency torque requirements (per the track) requires you average the torque like you did with the rpm. You can go back to timing but that doesnt change the epower to mpower efficiencies but it will allow you to ditch some heat if you go to your efficiency timing a couple degrees away. That’s not the real answer. High frequency torque heat is stored (conducted) in the bulkhead until air can take care of it. Find out how much it can hold. Low frequency is converted slowly by air. If your bulkhead isnt hot you have allot of torque left in the motor. Its like an intercooler for a turbo. I’ll overlay an intermittent and continuous torque curve so you can use your fdr to stay under what your bulkhead can conduct and convect.

you will end up knowing speed, torque, timing, heat and how much you can do with gearing for a short or long track. You can play with the inertia ratios in your choice of gearing to minimze high frequency or low frequency heat generation and ill figure out something on the inertia mismatch issue but there i suspect mfg rotor inertia are pretty close. Reality is you pick the motor based on a range of mismatch and then use the gearing to get the torque at an rpm but we will have to do this backwards.

if i show you how to measure motor performance for the way you want to drive you can pick a motor for torque or rpm based on a simple averages and a curve rather than this guy said that. What you will find is that it torque versus rpm just needs to be in a range because you will know how to apply either. But if you have 2 motors that generate the exact same torque by setting the timing to 5.6 amps you will no longer need that number. The better motor is the one that gets rid of heat the way you need it removed (low versus high). Then come back here and see if you agree with my educated guess about what 5.6 amps is. Its not the same from mfg to mfg but they are all 17.5 and subject to the same timing issues. i think 5.6 is safe torque + rpm efficiency + the motors ability to dissipate an average of low and high frequency heat but I would love to know definitively.

Last edited by Bry195; 12-26-2018 at 10:35 PM.
Bry195 is offline  
Reply With Quote
Old 12-27-2018, 12:10 AM
  #30  
Tech Apprentice
iTrader: (1)
 
Join Date: Dec 2018
Posts: 81
Trader Rating: 1 (100%+)
Default

The 5.6A (not what every manufacturer recommends btw) thing is just something that is repeatable based on the test equipment available to the guys that care to use a motor analyzer. It is a safe starting point based on what that manufacturer (Trinity I think) knows based on their motor architecture and the tolerances to which their motors are constructed. They also provide gearing guidance too. Even still, these are just starting points. They let you know that some small adjustments might be necessary.

I love the way a car drives with dynamic timing. It's incredible how much you can increase the performance of the motor, while wasting less energy as heat. I get the appeal of the simplicity of a blinky setup. Dynamic timing is something that was just too difficult for most people to get right. Set it up wrong and you could burn up your stuff very easily.

Because I use an aluminum gearbox the heat from my motor dissipates into the chassis. After a run my chassis is warm to the touch in the back.

There are possible outliers that will impact the perfect gearing and timing. For example, the largest track where I frequent usually has a flowing layout with a few slow corners. If they did a track layout with one long straight and a bunch of tight corners and short straights, the taller gearing would be slower on track. Similarly, this track had a main water pipe burst in the winter, which meant they couldn't water the track. This made the dirt dry out, which made the grip decrease significantly. The track surface got really rough because of the cracks in the clay. These outliers shift the optimal gearing and timing, and it is something that would be hard to quantify and predict the new optimums. Sometimes there are jumps that have really short run ups, and making that jump consistently is one of the most important aspects of a good lap time. In that case the timing and gearing might not be optimal for any other section of the track, but the small trade off everywhere else is worth being able to make the jump throughout the race.

​​​​​Based on the weight limit you mentioned earlier I would say you run touring cars. Touring cars would probably be the best platform to extrapolate data. The grip level is much more consistent than off-road, which makes calculating your target load much easier. Until you mentioned that weight I had no idea what type of car you are running. I think you could use the telemetry data to make your adjustments more effectively.

I track my mAh used very carefully. One time I switched from the really soft stock A-arms for hard arms. I could drive the car so much harder that I used 100mAh more than usual in a 5 minute run. That is probably enough of a load increase to warrant a timing or gearing change, just from different parts that moved the physical limit of the car.
waitwhat is online now  
Reply With Quote

Thread Tools
Search this Thread

Contact Us Archive Advertising Cookie Policy Privacy Statement Terms of Service