BATT_MAN

YOU ARE 100% RIGHT...

People can get carried away with their dynos and stats, but if they do not do what you said, the motor will not make a difference...

Good stuff EddiO

Hyper1

Isaac,I need to understand this better! as I need a better method of grading my motors,especially at the track.

Chris Reilly

This is good stuff

BATT_MAN

hyper1,
I will be glad to explain this to you.
I will be sending you in a few days a package. In it, I will include some of the papers that I have written over the years on motorr design, testing, and parametrizing. Take a look at this stuff, after you read it give me a call and I will go into further detai to explain this to you.

The entire principle is to find the REAL operating torque that a car operates at a specific track. You will need to get one of our Axxis data loggers to install on your car to do real track testing. This only used when you practice, not when you race because of the added weight...

I know that many tuners are using an average torque that they are assuming based on operating current levels, but that is just an average. You need to know the PEAK torque needed for your car to accelerate at the rates you want. Averages just don't do it. I broke a track record at a track in Mexico a few months ago by 2 laps using a hand out motor. At the end they opened my motor because they thought I changed the armature or timing. All I did was tune my motor as per track/car requirements. I found the required torque and got the motor to operate at its best possible efficiency at that power bandwith. By choosing the right brush/spring and current draw I was able to do that....

You will need to do lots of testing to find the Kv curves of all your motors...
Some persons claim that this does not work, but this is the same procedure we use to design motor applications for machines, servos, and believe it or not, for REAL CARS. I used to own a GTP car team in the 80s and I was the chief engineer and aerodynamicist, I used the same principles to get the best transmission /motor for a track.

Our r/c transmissions have fixed gear ratios so the only thing left to figure out is the right pinion/spur ratio (final rollout). There are ways to fine tune this by reading back emf numbers from your motor. I will elaborate on that when we talk.

Later,

Isaac

burgboyz

iTrader: (70)

bat man, I checked out your site...looks promising.
please keep us informed when the catalog is completely loaded.
sounds like you have an enormous amount of knowledge and the experience
to back it up.... I look forward to learning more!!! thanks alot, jon

Yankey

batt_man
I am interested in finding more information on motor theory as well. Please either post it, email, or point me in the right direction. Also, is the data logger available? What does it record and how much is it?

To be honest, I have forgotten the setups on those two motors. They were examples of what I was seeing with various motors. A Motor 1 setup always looked better to me using amp steps, but Motor 2 always seemed to be faster when geared right. That's what has me confused.

Looking at the graphs using torque steps, Motor 2 is shown to be the better motor. I agree that torque steps seem to be a more apples-to-apples comparison, and it seems to show what is happening on the track. Torque is the force required to move the car and that seems like it will be the same no matter which motor is used, so it makes sense to me to use that for the steps.

Thanks for all the input!

Yankey

BATT_MAN

I have close to 2000 pages on motor design, analysis, and testing.

All this info is related as close as possible to r/c fixed magnet motors.
I will be posting on an ongoingbasis thisstuff on my website. However, I will be glad to copy most of the material I have and I can get copies to you.

What I need to know is what is the highest level of algebra/calculus you know. I ask this because a lot of what I have is in diff equations and if I have to I can simplify them for all of you. Although it is much easier to understand what is going on if you understand calculus.

Everything in a motor is related to something else, so differentiating shows the relationships...but I can get it down to plain algebra.

Knowing how a motor operates and understanding how the motor constants relate to each other will open a new world of motor tunning to all of you.

For example, we are very limited to what we can do to a motor because the actual geometry of the motor plays a huge factor and that cannot be changed. The geopmetry is the way the armature is built, its moment of inertia, the way it is wound, and the can... The air gap also plays a HUGE role in the performance. The few things we have left that we as users can manipulate is current draw. THTS ABOU IT.

Now to get that to change we have lots of ways to influence that. It starts with the brushes, the way they are seated, aligned, and compounds. The best way to analyze brushesis by knowing their density and coefficient of friction. No manufacturer gives us that information but there are ways to figure this out. Density is easy to measure...it is how many amps a brush can safely deliver per square inch of surface area... The coefficient of friction is more difficukt to obtain, but you can set a reference guide line by which you test your brushes and use index numbers to control that. Like I said, current is a major thing that we need to comtrol in a motor, but friction is a very important because it waists power. There are different types of friction that affect a motor. For example, there is brush friction, torque friction, bushing friction, and other mechanical issues thataffect friction. By reducing friction you can improve motor performance by as much as 10%...

Well, I am just talking and talking... or writing..

In my experience the only way to test a motor is by torque and not by current. Although the two ar dependent of each other and they are linear as long as you stay within the motors operating torque range everything will add up. No two motors have the same Kt (torque constant) and the best way to find the torque constant of a motor is by getting the torque at no load and then you go all the way to stall torque. by having RPMs on the Y axis and torque on the Y axis you van plot this two points and get the Kt. Kt is basicaly the slope of the line. Once you know the Kt, you van figure operating torque at ANY amp draw....

Regarding the data logger. I am using a data logger that I made and its the size of a AA battery. I install a current shunt from the ESC to the motor and I measure the viltage across there. That will give me the amp draw. I am sapling at 512 times/second in order to get accurate readings. Hopefuly I wil be able to produce and sell the data logger with the shunt for around $110. You only need to run two lapos with this hooked up to the motoir to get accurate readings. Then if you know the Kt of that motor you can figure real torque and under load RPMs.. That way you can find the right motor for your application. The info that dynos give is only usefull for comparison purposes unless a person understands what the numbers mean and how they relate to each other...

I am sure that once I figure out a good way to teach this to all of you, your understanding of motors and your motor tuning abilities will expoand exponentialy. YOu will start to realize that most of what you read on the forums is just plain WRONG!!! and I am sorry to say that.

The one thing that everybody needs to understand is that we need to operate our motors within the right parameters. If you exceed this parameters the motor performance will go out the door, yet the CE dyno will still give you readings that shoe that the motor is OK...
I have seen it, tested it, and proven it may times.

I hope I can help all of you. I am willing to share and pass all my knowledge and expertise.

Hopefully we can get together with Eric and EddieO and they can help us with their hands on experience. They have a vast amount of knowledge and experince (that I do not have) to share regarding the assembly and tunning issues... Together we can figure out better ways to tune and design motors. These two gentlemen know how to wind and tune motors better than anyone I have ever seen...I have tested Erics motors and they are AWESOME...out of the can...

I have a motor dedsign already...I am just waiting to see how the industry is shifting and once I make my decision I will go to Japan to have this new platform built. I have the capital and know how to do it, I just need to get Axxis Racingbig enough so that I can introduce the motor and compete against the established motors...

We will see what happens with time... I am already getting abused and bad mouthed all over the place for not agreeing with the way batteries are matched and sold by voltages...but then again...I have to look at where the rumors and bad mouthing is coming from...

One thing is for sure, once you will understand how motors work and operate you will see what I mean about the voltages... THERE IS NO DIFFERENCE IN PERFORMANCE WHEN WE ARE TALKING MILLIVOLTS...and GOOD BATTERIES
The difference between a 1.17 and a 1.20 battery on a stock motor is less than 110 RPMs/cell 650 no load RPMS for a 6 cell pack... 26,640 RPMS v 26,000 RPMS that is it... no one can feel that difference and this is NO LOAD RPMS. CURRENT is what makes our motors work... as long as a battery has a GOOD voltage, the battery that can deliver current the fastest possible way will be the best battery... and the motor will perform better... As long as a motor can develop the torque needed to operate and a oerson knows how to gear a motor, you will have a WINNING COMBINATION...

Isaac

Yankey

Hi BATT_MAN,
I should be able to figure out the math. I have an MSME and have taken lots calculus, ODEs and some PDEs. Thanks for the information!

Yankey

outlandr91

opps

outlandr91

Lookin' forward to reading more of your posts, BATT_ MAN

BATT_MAN

Yankey, outlandr91;

I will do my best to get as much info to you ASAP. Just give me a few days as I am busy...and I mean busy.

Additionaly, I am working on our motor program with our winders and suppliers. If all goes as planned, we should have our motors out by January.

I would like to get some of you to work with us in our development program. I am looking for experienced racers with enough motor and engineering know how that can give me accurate information. Our goal is to provide some of the best motors out there and the only way we will be able to achieve that is by listening to what racers have to say and report to us.

If any of you are interested to work with us please let me know. You will get first look at the motors, get to test them, and your feedback will help us get the best possible results.

Yankey, you knowledge should be WAY up there since you are an ME. I am looking forward to working with you. I hope I can provide you with enough information to help you master your motors... BTW, are you a practicing engineer and if so in what area do you work. I am interested to know more about you.

This should be very easy for you to understand...

I look forward to helping both of you.

Thanks for your confidence in me. I will do my best not to let you down

Isaac

John Stranahan

iTrader: (27)

Batt-Mann-You might describe what you put on the x axis here"by having RPMs on the Y axis and torque on the Y axis you van plot this two points and get the Kt. Kt is basicaly the slope of the line. Once you know the Kt, you van figure operating torque at ANY amp draw"


Yankee-I did some more tinkering with your stock motor data. I took the torques and multiplied by the car overall gear ratio that you might use for these two motors.

I made the wheel torque at the average amp draw equal by adjusting the ratio of motor two to make it so. Now I always adjust the final gear ratio with lap times, but the plot that resulted was instructive. I plotted Power, which is not changed by the cars gear ratio, vs rear wheel torque, which we can change at will by adjusting the gear ratio slightly.

Hmm- Starting to look like a power vs amp draw graph. Motor two looks like it would do better on a road course in a heavy car where high torques (high amp draws as well) are required. But now maybe we can also tell that Motor I might be better on an oval or in a light car where amp draws are less and there are low rear wheel torques, and low amp draw.

If you would like to tinker with this Microsoft Ecell spreadsheet. You can request a copy by e-mail to [email protected]

Now whether you use amp steps are torque steps you can plot this data anyway you want. You can run amp steps and plot torque on the X axis. What you cannot get without another type of run is Kv or the RPM / Volt. I have used this latter approach with manufacturers published data on an Aveox brushless motor to plot a power curve. I got rather ridiculous power numbers which the company actually used to promote the product. The motor was rather ordianry on the track at the time. I think it's a better approach to actually dyno a motor and measure power. Then We don't have a dyno for these brushless motors available at present as far as I know. An acceleratin run measured by radar would probably show the mod brush type motors still have a slight edge in raw power.

John Stranahan

iTrader: (27)

BTW- All the spreadsheet, I described above does is lets you input the cars final drive ratio and calculates rear wheel torque from your dyno data. I also calculate a suggested final drive ratio automatically, but you have to know a good ratio for the first motor. You enter your two dyno sheet worth of data by replacing Yankees data. This is easy. All the graphs types that I have displayed in this thread and the slope of Torque vs Amp Draw (Kt) then self generate. You do have to have Excel.

More early Aveox brushless DATA
The other approach I referred to above calculates power using Kv (RPM/Volt), Rm (motor resistance, Ohm) Kt (Torque/amp), Current (Amp), and Io (no load amperage)

You find the RPM/Volt by spinning the motor at a known RPM and finding the back EMF just by measuring the voltage it produces.

You find the slope of the Torque vs Current graph from the run where you use amp steps.

You measure the resistance of the motor Rm. This might not be as easy as it sounds. There is a description of how to do this at the link below.

And then you measure Io (No Load Current, Amps)

For this early Aveox these values are
Io, Rm, Kv, Voltage,
4, 0.018, 4150, 7,

Power is then = (Voltage -Current *motor resistance) (Current - No Load Current)

Power at 45 amps and 7 volts for this early brushless, which is where I used to test my mod motors, calculates out to a phenominal 253 Watts. The motor was just a weak mod. If you carry the calculations even further the power kept rising to over 700 Watts. No way. So the theory must break down at amp draws that we normally experience in RC.

The bottom of this tutorials first page shows how to measure some of these constants with a drill press, power supply (for example a turbo thirty discharing a battery, this will get you a reasonable voltage) and voltmeter. Sounds interesting. A drill press dyno.

Motor Tutorial

Here are the values for a 15 turn brushed motor
15 turn Brushed Motor
Io, Rm, Kv, Voltage
2.55, 0.074, 4182, 7
Power at 45 amps and 7 volts was 156 Watts. I believe this one.

and as usuall you can just skip this stuff, if you are not interested.

BATT_MAN

Hi John,

The reason I plot torque on the X axis and all the other data (RPM, Eff, Power, current) on the Y axxis is simply because you can see all therelationships as they relate totorque and it is very easy to visualize what is happening with the motor.

Yes. Kt is the slope of the line and it does not matter how you plot as Kt=T/I or simplysaid, Current through the motor=torque produced/torque constant.

John as you know all the motor constants are essentially the same number, just in different units of measure.
For example, Kt=Ke=1/Kv as long as you work in the SI system.
Kt is in N-m/A and Ke is in V-sec/rad

You mentioned the Kv constant and how to obtain Kv by plotting....Try this:
Being that all the constants are basically the same number but in different units of measure, in the Imperial system we can calculate Kv as follows.
As long as you use the following units of measure:
Kt in oz-in/A
Ke in mv/RPM (Back EMF)
Kv in RPM/volt (Revolutions/volt)

Then:
Kt=1352.4 / Kv thus Kv=1352.4/Kt
Ke=Kt/1.3524
Ke=1000/Kv
Kv=1000/Ke

So what good is this...if you want to figure out in a quickly way what your motors constants are do the following:

If you have a source of rotational speed (like a Dremel or a drill press) you can compute Ke for ANY motor. Just attach the motor shaft to the known rotational source and measure the open circuit voltage across the motor terminals. This will give you the Ke constant (BACK EMF) For exapmle, conect a constant source of speed at 1000 RPMs and measure the voltage across the motor terminals, this will give you Ke as 1000 RPM/mv... then just do the math using the formulas I gave you.

SO, since you have Ke you can now figure Kt (the theoretical torque at any given current), and Kv (you can now compute the motors maximum no load speed at any voltage). If you can measure the stall torque, you can the calculate the motors efficiency (measured torque expressed as a % of the theoretical torque)

This is an easy way to start understanding a motors relationships...

Now, your spreadsheet takes all this into another step, which will tell you how each motor will perform under load.

I hope this short explanation wil simplify the understanding of how the motor constants work and how you can start using them to do your calculations.

By knowing these constants you can figure out anything you want to know about a motor as long as it operates within the motor limits.

Try using theabove mentioned values to compute the other constants.

And BTW to calculate the STALL TORQUE with the given data just use this equation: YOu can also use this formulas to draw an actual speed-torque-current curve... as long as you know the no load torque

Stall torque= [(Kt * VOLTS/R) - No load torque] = [(Kt * Ia) - No load torque]
=[Kt * Volts/Rm] - [Io * Kt]
Ia = armature current @ load
Io = no load current (from Johns data)

No load current Io = No load torque/Kt

No load RPM = [volts - (Io * R)] / Ke (sorry, I fixed it, I had made a mistake)

Stall current = volts/Rm

Also, since John is giving the armature Resistance Rm you can figure out the Slope of the Speed-Torque Curve in KRPM/oz-in
STC= Rm / 1.35Ke^2 .......Ke(squared)

You can calculate RPMs at any given load, efficiencies, power, etc. I will elaborate more on this later...

BTW, I have tuned motors just by working with the Ke...BACK EMF... YOu can use a DVM or a scope... By analyzing the EMF you canoptimize the performance of a motor (up to certain limits) but it is A GREAT and QUICK starting point. Just experiment and you will see the results...
I hope this will help.

Isaac

BATT_MAN

John, it makes sense because you are analyzing POWER not RPM or efficiency...and you are not taking Kt into consideration on your analysis. However, try plotting the POWER v TORQUE and you will see the results...Basically the same, just different units of measure. After all the current is proportional to the torque, but I think that if you use the actual torque it is easier to do the analysis.

After all, the mechanical POWER of a motor (WATTS) is the product of speed and the torque load...so even if you are using current (I) as a differetial, torque plays a role in the equation.

If you multiply torque * RPM * (adjustment factor) you get POWER IN WATTS.
For torque in oz-in we use 0.00074 as the adjustment factor...
i.e WATTS = torque(oz-in) * RPM * 0.00074

As you see torque always plays a direct role in motor calculations, no matter how we come out with an answer.

So whenever you are looking at POWER in a motor torque plays a direct role...

It does not matter how you plot the data as long as you know how to interpret it, and you know how to do that. Thus, for you it will be the same because you know exactlly what you are looking for. However, torque is always a factor when analyzing motors...

Just my opinion!

Isaac