RCCrawler Forums

JohnRobHolmes · 12-22-2010, 02:49 PM

I have been after a good motor dyno for some time now. I have used most of the R/C type dynos, and have been disappointed with their overall design. While the typical inertial dyno that we employ works fine for comparing motors within a small scope, it does not give the full picture of efficiency under various loads and the efficiency of high turn motors is generally way out of whack.

The scope of this project is a full build from scratch. I will build the drive motor cradle, set up the slave motor system, build a digital readout for gathering figures, and as a final system I will have full datalogging on at least a 5hz resolution.

Goals:
One dyno for fractional and sub 2100w motors
One dyno for higher torque and HP motors
universal design with regenerative loads to reduce power supply size
torque, power, and efficiency plotted over the entire RPM range of motor (when datalogging system is complete)
Cost under $1000, including my labor and raw materials.
Easily replicatable

So far I have drafted a few designs and have picked up two load cells (simple Wheatstone bridge devices). The first one is a 5 pound sensor that will handle most hobby style motors. The second load cell is a 22 pound sensor that will take the majority of larger motors. The torque arm can be made longer at a resolution loss for motors that are slower speed/ higher torque like an outrunner.

I have a few data loggers here already, and have been studying AVR programming in case I need to build a data acquisition device myself. I also have some voltage and current monitors here that will stream tab delineated information via serial connections. Add this into some load cell data and we have ourselves a fully functioning dyno!

The next step is to build a stand, calibrate the torque sensor formulas, and get some DROs on it for static point data acquisition.

My most important goal with this build is to show exactly what a dyno is, and hopefully offer enough information for a universal dyno to be reproduced by another if they wish. Whether it be hobby motors or full size vehicles, the principles of a dyno are the same.

JohnRobHolmes · 12-22-2010, 02:49 PM

formula and thought spot

Force = pressure on an object (newtons and pounds)
Work = Force over a distance (newton-meter aka. joule and foot-pound)
Power = Work/ Time (horsepower and watts)
Torque = Rotational work (newton-meter and foot-pound)

Ohm's law- Voltage = Current * Resistance (V=I*R)
Electrical Power = Watts = Voltage * Current (W=V*I)
Instantaneous Power = Force * Velocity

1 hp = 745.699872 watts
1 watt= 1 newton meter per second
1 foot-pounds = 1.3558 Newton-meters
1 inch-pounds = 0.112984829 Newton-meters
0.1 kg-meter = .980665 nm

Horsepower = Torque x rpm / 5252
Torque in ft/lb = (HP/rpm)* 63,025
Torque in N/m = (HP/rpm)* 85,446 --- needs a double check

Test modes:
Steady state RPM
Steady state Torque
Steady state Amp
Dynamic

Brandon · 12-22-2010, 04:41 PM

Subscribed, I love this stuff

wantAsummit · 12-22-2010, 05:08 PM

Sounds like a great project. I can't believe that you are including your labor in the $1000 goal! You must have a very good idea about how this will go together already (and some very clear end goals). Can't wait to see how it comes out.

JohnRobHolmes · 12-22-2010, 05:15 PM

$20 an hour for hiring myself on the cheap, that gives me a full 40 hours of work after materials are considered. It will cut it close for sure, but I see no reason that I can't hit it. This estimation is only for the smaller machine. The slave motor for the larger unit has already blown $1000, but it will take 30hp and is a lifetime investment.

If I don't hit the monetary goal, then I may have to build a few more units to subsidize my work time over a few machines

The graphical interface will be the tough one. I can buy some ready made software for $600 to $1200, or I can just dump the info into an excel spreadsheet.

At any rate, the machine itself is not the biggest hurdle. It is the data gathering and interpretations.

Manning · 12-22-2010, 05:36 PM

Will be watching this thread!

whoodie · 12-22-2010, 05:41 PM

Quote:

Originally Posted by Brandon

Subscribed, I love this stuff

X2!

Harley · 12-22-2010, 06:48 PM

Here! Sorry I'm late for class

Narly1 · 12-22-2010, 08:28 PM

Hey John,

I actually work for a company that sells instrumentation and sensors. Check out www.disensors.com

Data acquisition, torque cells, etc are amoung the products we carry.

Not trying to sell you anything, but you can pick my brain if you like.

The usual place we start on an app is asking questions like accuracy requirements, measurement rate and of course budget.

Seeing as you have already defined the last two parameters, how about the first??

Narly1

Robbob · 12-22-2010, 08:41 PM

Quote:

Originally Posted by Brandon

Subscribed, I love this stuff

Quote:

Originally Posted by whoodie

X2!

x3 and I dont understand half that crap ..........

Narly1 · 12-22-2010, 09:07 PM

John,

Re-reading your original post, look at digital read outs (DRO's) that have the capability to provide excitation voltage for the load or torque cells that also include a reasonably fast analog output (to send to a logger).

An instrument like I've described + Winwedge software and you'd be off to the races.

Narly1

JohnRobHolmes · 12-22-2010, 09:41 PM

Sounds like a DRO that can take the 0-10v load cell output and send it to the logger in a 0-5v signal would be the tits! I will certainly pick your brain on the matter, because that solves a few issues for me.

Harley · 12-22-2010, 09:43 PM

I hear you guys talking but what I hear is...

JohnRobHolmes · 12-22-2010, 09:48 PM

Me stickum motor in machine. Me getum numbers from machine that say good.

rcjunky3 · 12-22-2010, 10:18 PM

thanks for simplifying that for us, i mean oh ok soundum good

Narly1 · 12-22-2010, 11:04 PM

Quote:

Originally Posted by JohnRobHolmes

Sounds like a DRO that can take the 0-10v load cell output and send it to the logger in a 0-5v signal would be the tits! I will certainly pick your brain on the matter, because that solves a few issues for me.

Maybe send me the datasheets on the load cells you bought when you have a chance. Somehow I thought they were mV/V output. But if they already have an amplified output you would be better off connecting them straight into a PC-based analog card.

Narly1

JohnRobHolmes · 12-23-2010, 06:39 AM

I may be wrong about the output. I need to test them first.

Hulksta · 12-23-2010, 08:43 AM

Subscribed!

TEDROCKZ · 12-23-2010, 04:07 PM

Hell! I'd buy one. A grand would not be a bad price...ETA?

JohnRobHolmes · 12-23-2010, 05:56 PM

I'm not pushing too hard on it. Once I get the DAQ figured out it will be a few days work to actually build it. Either I have to purchase a signal conditioner, or I will need to make a chip to take the torque sensor output and convert it to 5v analogue.

12-22-2010, 02:49 PM	#1
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	Universal Dyno project with streaming data I have been after a good motor dyno for some time now. I have used most of the R/C type dynos, and have been disappointed with their overall design. While the typical inertial dyno that we employ works fine for comparing motors within a small scope, it does not give the full picture of efficiency under various loads and the efficiency of high turn motors is generally way out of whack. The scope of this project is a full build from scratch. I will build the drive motor cradle, set up the slave motor system, build a digital readout for gathering figures, and as a final system I will have full datalogging on at least a 5hz resolution. Goals: One dyno for fractional and sub 2100w motors One dyno for higher torque and HP motors universal design with regenerative loads to reduce power supply size torque, power, and efficiency plotted over the entire RPM range of motor (when datalogging system is complete) Cost under $1000, including my labor and raw materials. Easily replicatable So far I have drafted a few designs and have picked up two load cells (simple Wheatstone bridge devices). The first one is a 5 pound sensor that will handle most hobby style motors. The second load cell is a 22 pound sensor that will take the majority of larger motors. The torque arm can be made longer at a resolution loss for motors that are slower speed/ higher torque like an outrunner. I have a few data loggers here already, and have been studying AVR programming in case I need to build a data acquisition device myself. I also have some voltage and current monitors here that will stream tab delineated information via serial connections. Add this into some load cell data and we have ourselves a fully functioning dyno! The next step is to build a stand, calibrate the torque sensor formulas, and get some DROs on it for static point data acquisition. My most important goal with this build is to show exactly what a dyno is, and hopefully offer enough information for a universal dyno to be reproduced by another if they wish. Whether it be hobby motors or full size vehicles, the principles of a dyno are the same. Last edited by JohnRobHolmes; 05-15-2012 at 11:23 AM.

12-22-2010, 02:49 PM	#2
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	formula and thought spot Force = pressure on an object (newtons and pounds) Work = Force over a distance (newton-meter aka. joule and foot-pound) Power = Work/ Time (horsepower and watts) Torque = Rotational work (newton-meter and foot-pound) Ohm's law- Voltage = Current * Resistance (V=IR) Electrical Power = Watts = Voltage Current (W=VI) Instantaneous Power = Force Velocity 1 hp = 745.699872 watts 1 watt= 1 newton meter per second 1 foot-pounds = 1.3558 Newton-meters 1 inch-pounds = 0.112984829 Newton-meters 0.1 kg-meter = .980665 nm Horsepower = Torque x rpm / 5252 Torque in ft/lb = (HP/rpm)* 63,025 Torque in N/m = (HP/rpm)* 85,446 --- needs a double check Test modes: Steady state RPM Steady state Torque Steady state Amp Dynamic Last edited by JohnRobHolmes; 09-02-2013 at 08:04 PM.

12-22-2010, 09:07 PM	#11
Narly1 Quarry Creeper Join Date: Jul 2008 Location: Erin, Ontario, CANADA Posts: 471	John, Re-reading your original post, look at digital read outs (DRO's) that have the capability to provide excitation voltage for the load or torque cells that also include a reasonably fast analog output (to send to a logger). An instrument like I've described + Winwedge software and you'd be off to the races. Narly1 Last edited by Narly1; 12-22-2010 at 09:11 PM.

12-22-2010, 04:41 PM	#3
Brandon Im Brandon Join Date: Feb 2005 Location: Colfax, CA Posts: 8,156	Subscribed, I love this stuff

12-22-2010, 05:08 PM	#4
wantAsummit Quarry Creeper Join Date: Jan 2010 Location: USA Posts: 402	Sounds like a great project. I can't believe that you are including your labor in the $1000 goal! You must have a very good idea about how this will go together already (and some very clear end goals). Can't wait to see how it comes out.

12-22-2010, 05:15 PM	#5
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	$20 an hour for hiring myself on the cheap, that gives me a full 40 hours of work after materials are considered. It will cut it close for sure, but I see no reason that I can't hit it. This estimation is only for the smaller machine. The slave motor for the larger unit has already blown $1000, but it will take 30hp and is a lifetime investment. If I don't hit the monetary goal, then I may have to build a few more units to subsidize my work time over a few machines The graphical interface will be the tough one. I can buy some ready made software for $600 to $1200, or I can just dump the info into an excel spreadsheet. At any rate, the machine itself is not the biggest hurdle. It is the data gathering and interpretations.

12-22-2010, 05:36 PM	#6
Manning RCC Addict Join Date: Nov 2006 Location: Lowell, Arkansas Posts: 1,307	Will be watching this thread!

12-22-2010, 06:48 PM	#8
Harley 20K Club Join Date: Jul 2004 Location: Sending illegals home one Hayabusa at a time. Posts: 22,981	Here! Sorry I'm late for class

12-22-2010, 08:28 PM	#9
Narly1 Quarry Creeper Join Date: Jul 2008 Location: Erin, Ontario, CANADA Posts: 471	Hey John, I actually work for a company that sells instrumentation and sensors. Check out www.disensors.com Data acquisition, torque cells, etc are amoung the products we carry. Not trying to sell you anything, but you can pick my brain if you like. The usual place we start on an app is asking questions like accuracy requirements, measurement rate and of course budget. Seeing as you have already defined the last two parameters, how about the first?? Narly1

12-22-2010, 09:41 PM	#12
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	Sounds like a DRO that can take the 0-10v load cell output and send it to the logger in a 0-5v signal would be the tits! I will certainly pick your brain on the matter, because that solves a few issues for me.

12-22-2010, 09:43 PM	#13
Harley 20K Club Join Date: Jul 2004 Location: Sending illegals home one Hayabusa at a time. Posts: 22,981	I hear you guys talking but what I hear is...

12-22-2010, 09:48 PM	#14
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	Me stickum motor in machine. Me getum numbers from machine that say good.

12-22-2010, 10:18 PM	#15
rcjunky3 RCC Addict Join Date: Jul 2006 Location: barstow Posts: 1,303	thanks for simplifying that for us, i mean oh ok soundum good

Sponsored Links

12-23-2010, 08:43 AM	#18
Hulksta RCC Addict Join Date: Dec 2009 Location: Newcastle, Australia Posts: 1,784	Subscribed!

12-23-2010, 06:39 AM	#17
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	I may be wrong about the output. I need to test them first.

12-23-2010, 04:07 PM	#19
TEDROCKZ I wanna be Dave Join Date: Jan 2008 Location: Arat Alabama Posts: 2,678	Hell! I'd buy one. A grand would not be a bad price...ETA?

12-23-2010, 05:56 PM	#20
JohnRobHolmes owner, Holmes Hobbies LLC Join Date: Nov 2004 Location: Volt up! Gear down! Posts: 20,290	I'm not pushing too hard on it. Once I get the DAQ figured out it will be a few days work to actually build it. Either I have to purchase a signal conditioner, or I will need to make a chip to take the torque sensor output and convert it to 5v analogue.

RCCrawler Forums

Thread: Universal Dyno project with streaming data