Go Back  R/C Tech Forums > General Forums > Electric On-Road
Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread. >

Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.

Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.

Old 04-09-2009, 12:54 AM
  #1  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.

A couple excerpts from the old thread followed by new text. Note this is

Mattnins dyno that he developed

I helped with the spreadsheet. He is mailing the dyno parts to me to get more testing done out of it. He is out of motors.

richard.bratton

(small edit) It would be useful if you could explain how to use your methods to compare different motors and predict their performance on the track. I don't think it's important to have an accurate power number as we only need power numbers for comparison to other motors. With the introduction of the LRP SPX and SP Gran Turismo 2.0, I'm curious as to how their very radical timing profiles affect performance as well. I think it was John that said that there isn't much difference between the top controllers; does this apply to these later releases?

Additionally, what equipment do I need to replicate your testing? Instead of using a flywheel, would it be possible to just install motors in a buggy with a set gearing and pair of tires and roughly approximate the moment of inertia? Would that be close enough for comparison testing?

You guys are doing some great work here, I'd just like to figure out the application to help me pick the right motor, controller, and gearing for my cars.


mattnin

Richard, the beauty with this custom dyno is that even ESCs can be dynoed if one ESC is compared to another using the same motor. Please refer to my earlier post showing a Novak 13.5 dynoed at 0 degrees boost and then 30 degrees boost using a Tekin RS Pro. At 30 degrees, the motor showed 8000 more RPMs, however there was a torque and power loss.

If you would like to replicate the testing, you must either purchase or build your own flywheel. Often, you will find flywheels for sale here or on ebay, or you can also do like I did and head to your nearest local machine shop and have them make a flywheel for you. The flywheel I made is 3.427" in diameter, and weighs 365g. I came home and balanced it myself using a dubro prop balancer and my drill press. After you have the flywheel made, then you have to make a motor mount. I made a simple box out of pine and affixed a team associated motor mount to it. I then just screwed my test motor into the motor mount using regular motor screws. Add the Novak Sentry and you are in business. Please note that when using the Excel or Open Office spreadsheets, that we are using a mathematical model to calculate torque, power, and efficiency. If you need help using it, I will help you out and I am sure John would help you out as well.

Concerning building a dyno just using a transmission and tires, it can be done, but I feel it would be inconsistent, and consistency is the key with these dyno tests. You will see that even when you have a nice and balanced up flywheel, that even the smallest detail matters, such as timing the runs consistent (i.e. 15 second dyno runs for a 13.5 motor), temperature, and topping off the battery each time. If any of these variables are off, you will see a comparison difference.



John Stranahan

Richard and All

Here are the plans. Matt is going to mail me the Sentry and a steel flywheel or two. I am going to work on procedures to get reproducible results with a LIPO battery. The plan is to make a run only when I get the first green blinking light on my FMA direct Scorpion charger. This will insure that at least the starting voltage of the test will always be the same. This happens at about 90% full. This is a very stable region of the battery discharge curve. I will test only with a cool motor. I going to use a readily available thunderpower gold 5400 mA-h pack. The pack chemistry (another brand) will change the results slightly. These batteries are slow to change internally over months if charge rates are kept to a modest 1C. Thats what I use. At least test results at the beginning and end of the month should be consistent.

I plan to use an LRP competition edition speed control at first. We will likely see an improvement in precision from better more controlled acceleration in the first .4 s. Can you note this on the track. Well yes you can. I'll post some sensorless motor test results from the track in the new thread.

We are going to leave the heavy dyno and motor theory behind in the old thread. Here is a link to the old thread.


as RC guys just need to know the type of question Richard just asked.

I will make a plot of Peak power with motor limited amps vs voltage. This graph may indeed be different than my Competition Electronics Dyno. Well see. The only purpose here is to see if there is an accurate voltage conversion. The Fantom dyno makers suggest that a 5 V test is predictive of all voltages.

One striking thing about boost is the efficiency goes down. Here in Houston we are always up against thermal shutdown issues. Less efficiency means more heat.

When using plastic geared transmissions, you will find enough wear from run to run, and in addition all the vibrations and bearing friction will create havoc with your numbers. It is not like a full size car with all precision roller bearings and balanced tires on a full size chassis dyno. I say go with a flywheel unless your goal is to investigate chassis losses. Matts motor mount sounds like a good solution.

I'll review Richards post when I am up and running here in Houston with the Sentry.

John



NEW Thread Starts Here. Homemade Novak Sentry Dyno

Work has begun on the Houston version of the dyno. I will say again that this thread is for experiments with this dyno, How to use the dyno, How to build the dyno, suggestions for experiments (even those to prove a theory). We do not want to be a slave to the dyno, however. Only a few experiments will be performed that we have the most interest in.

We do not want the thread to be theoretical. The procedure used by the spreadsheet is well used and well proven. There is one tiny data smoothing twist I added to improve the accuracy of the calculations. See the other thread for the Theory. I have outlined exactly what we did.

http://www.rctech.net/forum/electric...less-dyno.html

A little bit about myself. I have used the big three of motor dynos (the ones that are real dynos) as well as some motor checkers like the Tekin dyno. These include the Robitronic. I have a novel data analysis subroutine that Robert Schachhuber the designer included on that dyno called partial average power. I have many many hours on the CE dyno which is probably the most precise and reproducible of the three. My brushed motor tune kicked major ass at the touring car regionals. Christian Tabush received it indirectly and drove it. This was the result of a 3 dimensional study using assymetric motor springs. The Fantom Dyno: I have used the Fantom a bit in brushless mode. I have seen plenty of Fantom data files. My experience runs deep here with brushed and less so with brushless although I have a nice section on the Novak 13.5 in my CRC pan car thread. I used a gearing derived directly from this dyno run to not only win the oval series at our track, but I kept from frying the motor by a substatial 40F. Not so for the other guys. We differed by 5 pinion teeth. I think the fallacy here is that a hotter motor is faster, or that you should always gear to maximum heat.

I like to do experiments to debunk current RC practices which are founded on the divining rod principle and to prove those founded in testing. I am very sensitive to inadequate motor theory resulting in outlandish claims by the manufacturer like the early brushless AVEOX. As a retired Analytical Chemist I am fond of measuring things.

I wrote Novak a note and hopefully my Sentry dyno will wear a GTB controller. The entire dyno is going to fit on the antique oval car chassis except for maybe the battery charger.

One weakness that may affect precision on the existing flywheel dynos is that the motors are mounted in stone. Very much unlike our use in RC cars. I decided to use a full suspension to keep the motor bearings from taking such a beating. I will see how that works. When a dyno is emitting 140 dB (decibells, not quite as loud as gunfire) you might expect something is taking a real pounding. I expect the bearings are not suitable for this task with a heavy flywheel. I have about $50 invested in this oval car. (used oval prices went to the basement a couple of months ago).

I have a $25 Astroflight servo tester which is a compact device powered by the speed controls BEC (battery eliminator circuit). This will eliminate the need for a radio and receiver and separate reciever pack to run the sentry. Now once I put all the stuff together it's not going to be cheap. There is no commercial alternative at the time, however. Then you take the motor and speed control and maybe the data logger out and pop it in your car and you have dual use of the major expense. I like to look at cost per use.

The major change I made to the chassis is to reverse the motor plate. I happened to have a cut open motor plate on the left side of the pod. I used it on the right side and got a nice slide into the rear motor change feature. I only have to losen the top plate screws one turn as this motor is slightly larger in diameter than the older brushed. This may work with the standard associated motor plate or maybe not. I have also had to use button head motor screws to insure sufficient flywheel clearance on the inside. I rearanged the center shock a little using existing holes. It came out nice. I will just clamp it to my butcher block cart for tests. My center shock has an internal spring and is actually sprung to center. It has springing in both up and down travel. Any T-plate car will mimic this as the T-plate will provide up and down springing. The speed control goes on that little white pad. the entire right side can hold the sentry and the astroflight servo tester. Thick oil.

pic you can click on the pic a second time to enlarge it.
John
Attached Thumbnails Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-novak-sentry-dyno-chassis-motor-mount-002.jpg  

Last edited by John Stranahan; 04-20-2009 at 10:48 PM.
John Stranahan is offline  
Old 04-09-2009, 05:25 AM
  #2  
Tech Master
iTrader: (12)
 
pejota's Avatar
 
Join Date: Apr 2008
Location: Vero Beach, FL
Posts: 1,022
Trader Rating: 12 (100%+)
Default

Looks good John, can't wait to see some results of these new ESCs with programmable timing. There is a big debate about it on the VTA message board.
pejota is offline  
Old 04-09-2009, 10:47 AM
  #3  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

Pejota-Thanks for the post. Send me one of those puppies and I will give it a test. I possibly have a GTB coming from Novak. The LRP has a first start routine that limits acceleration at the start of a race. There is a work around printed below.

I thought I would copy some information from my CRC thread as it is not so easy to find there. These are some typical uses for the dyno. Is the small or big rotor better for my needs? What is the gear that I should use for 4 cell flat asphalt oval? I had sucssesful track outcomes after doing both tests and convincing myself to trust the dyno and not the experienced oval racers.

John Stranahan

The pic shows the modifications made to the dyno. A receiver from the touring car receives a full throttle signal from my transmitter and finger. The motor is run. I get a good curve and a good number. 58 Watts at 9.1 kRPM. This means the brushless is about 18% more powerful than the stock motor I tested in the previous post. The 13.5 is called a Super Stock. Now the bad news. I had 3 rotors and two cans to test. Each one yielded the identical 58 Watts at 9.1k RPM. Max RPM was the same to 1 RPM. As if a robot was limiting the spin up of the flywheel. And in fact there is a robot running the motor, the sphere speed control. It is possible that a punch setting or just a programming limitation is limiting motor acceleration. The printouts are not identical and include a spurios efficiency reading so I think the Fantom is working OK. The speed control is precisely controlling acceleration no matter what the rotor. Anyway thats my current theory. Time to pull out the LRP manual.

I maxed the punch control out to 6. No improvement. I still get the same numbers.
Attached Thumbnails


John Stranahan

I found a combination that gave me reasonable numbers from the dyno. I had one problem, the 100% throttle was not being reached. I adjusted the tranmitter and made sure I got the blue full throttle light. On the suggested power (12 V battery), I got bigger numbers but still always the same. I suspect that a feature of the LRP that controls punch on first start might be to blame. I hooked the speed control up directly to a charged 4-cell pack gave the motor a few slow warmup gooses of the throttle, Reset the Fantom software to do a full 5 second run. Then I hit motor run and simultaneously pressed full throttle. Since this dyno measures power from instantaneous acceleration of the flywheel any error in hitting the throttle just right on time will not matter to these maximum power numbers. I got variable data now which is more likely to be correct. The 12V to 5 volt part of the Fantom was unused.

Small sintered rotor 66, 69 ,66 Watts for three tests averaging 67 W average 9.6 kRPM

Fat tuner Rotor 64,62,65 Watts for three tests averaging 63.6 W average 8.9 kRPM

So what does that tell me. I have been using the fat tuner rotor in error thinking it was more powerful. They don't tell you jack about it. You can use the RPM average to adjust your gearing to suit if you do a rotor swap.

I have been using 3.2 inch rollout with the fat rotor. I should adjust this with the RPM ratio I measured.

3.2inch/rev x 8.9kRPM/9.6kRPM = 3.00 inch/revolution

The system sits there for a while if you have bright ideas for a test. I won't be running the 3.5 on the machine as the RPM reached are excessive. You will have to take their word that these (and the LRP 3.0) are putting out about 746 W or 1 H.P.

I still want to get a motor curve to find peak amperage and amperage at peak power.
John


John Stranahan

Run Line, Average RPM
One of the benefits of the latest Facts Machine Software is some help with gearing. I ran some numbers through the calculator that were of interest rather than using the software which I do not have.

Our run line is just short of 175 feet by a couple of feet. Average laps are 4.4 s. Average speed is 173ft/4.4s = 39.3 (ft/s) (27 mph) or 471.6 in/s.
If I divide by rollout I will get my average RPM.

(471.6 in/s)/3.2 in/rev) = 147.3 rev/sec

change to RPM

147.3 rev/s x 60 s/m = 88425 RPM or 8.8 kRPM

So where is that in relation to my power peak with the fat rotor (8.9 kRPM).
About as close as possible. I am getting the most out of the motor. If I change to the small rotor I am going to gear lower as calculated in a previous post above. Note I am also coming off the track at 175F but on a 140F track. Motor temperature is somewhat useful for gearing but only if your track temp is the same as that other guy that is giving you temp advice.

John

Gearing an Electric Motor
I thought I would try to explain why it's important to get the gears at the right ratio, on road, or on the oval. First take a look at the left side graph. This is output from the Fantom Dyno, through the Fantom power supply, for a Novak EBX 13.5 brushless motor. The graph plots Amp draw vs RPM. All I want you to notice is the shape. The graph is highest on the left at low RPM. If the motor stays in this range it overheats. At the right the line is low. The motor pulls few amps at high RPM. It therefore runs cool at high RPM. Amp draw is highest with the motor stopped (right at the start of a race).

Next take a look at the right hand graph which plots power vs RPM for the same dyno run. Notice we have a hump shaped curve. Power peaks at mid RPM. The point of carefully gearing a motor is to try and keep it near its power peak a greater percentage of the time. There are sometimes heat considerations that will foil our attempt but you see my point.

If we gear the oval motor to a higher rollout. It is going to run at fewer RPMs. It is going to draw more current. If we are to the right, of the humps top, this is a good move. If we are too the left of the top you will just heat the motor more and produce less effective power.

I was interested to see these brushless motor curves to see if they were dramatically different than brushed. Maybe the hump was broader or something like that, but no, they are remarkably similar. The proper gearing is still going to give you an edge.

The numbers on these graphs are low for reasons I stated previously having to do with the LRP speed control.

Eventually you just develop a gut feel for what a good ratio is by looking at the car accelerate out of the slow corners of the track and onto the straight. For the beginner there is no substitute for advice from someone experienced with their particular combination as a starting point and then lap times after that. A dyno is not needed just a nice thing to toy with.
Attached Thumbnails

rcbarrmeister

One thing we found using the Facts Oval Gearing facility through the years was that you couldn't use the actual runline to achieve accurate/realistic gearing from the program. The reasoning for this was that your rear tires slip from time to time, so in reality they may have to turn the equivalent of 181 ft, to carry your car around the optimum 175 ft runline.

To compensate for this, we would take a well tuned car with a specific motor and keep on adjusting the runline in the program, until it gave us the proper gearing/rollout we had already predeterrmined through trial and error. Once this was established, the results from the Facts program were much more reliable/accurate.

John Stranahan

Novak 3.5 large vs small rotor
by popular demand. I did some more dyno testing. First order of buisness was the flywheel had about .012 inch side to side wobble from wear. I thought this was too much for a 3.5. Careful application of a prick punch several times near the hole on the worn edges reduced this to about .002 on an dial indicator. This will reduce wobble vibration and stress on the machine. I used the Fantom power supply run by a small 12V gell cell battery. I was getting a good blue full throttle light on the sphere speed control. The power supply by itself will limit the huge amperage that this motor can draw. Fortunately with this motor, efficiency was peaking in the 20,000 RPM range at 5.0 Volts, so I got a good look at peak efficiency. I took 3 runs each on the small and the large rotor. Looking at the raw numbers the small rotor was .186 and the large .183 just averaging the highest readings on the 3 runs. This is inconclusive.

The graphs shows only two of these runs. The first run with the small rotor and the first run with the large rotor.
In the blue are the data points for the small rotor. There are some .18 at the highest point. This is 18% efficiency. Thats why these motors run hot.
The large rotor reached .18 but not quite as long as the small rotor looking at the green data points.
I smoothed the data by adding a polynomial trend line. The green lines peak shows the peak efficiency occuring earlier in the RPM curve but about the same efficiency as the small rotor.

I declare a tie in efficiency, but if you run at lower RPM on a smaller track the big rotor wins.

This motor is really screaming on the dyno. I made a 3/4 inch wood scattershield to place on top in case of a flywheel explosion or motor shaft breakage. The better balance was really appreciated by removing the wobble.

Pic Green large rotor, Blue small rotor both in a Novak 3.5 R, my favorite wide pan motor.

John
Attached Thumbnails

John Stranahan

I thought I would do one more test while this 3.5 was hooked up. This time I used a 4 cell pack. This pack can deliver the huge amps this motor needs to perform at its best. It is not like a 6 cell and at this point I don't recommend running the dyno with a 3.5 and a six cell pack. It is starting to sound like a small jet engine already due to the 48,000 RPM reached. Note that without using the Fantoms power supply you can get no efficiency numbers. The dyno cannot measure input power.

In the red corner the small rotor peaked about 140 W. I got very smooth data an did not need a trendline. In the green corner the large rotor 123 Watts. This large difference did not require replicates on this smooth data. you get more power with the little rotor. This is a repeat of the conclusion with the 13.5.

Pic: Note the white scatter shield has a beveled cut so it's supported by the black knob as well as a strip of tape. Aim the exposed part of the flywheel into a safe area. Note that pile of papers; that is two years receipts for RC stuff. Damn. Don't tell the wife.
John
Attached Thumbnails

Last edited by John Stranahan; 04-09-2009 at 11:29 AM.
John Stranahan is offline  
Old 04-09-2009, 01:39 PM
  #4  
Tech Fanatic
iTrader: (3)
 
lutach's Avatar
 
Join Date: Oct 2003
Location: Kearny, NJ
Posts: 823
Trader Rating: 3 (100%+)
Default

Anyway of getting some torque numbers? I would love to see what is the difference in torque from the small diameter rotor to the larger diameter ones. This could be very useful for comparing Inrunner type motors with 2, 4 or 6 pole. Have you done any sensorless set ups on the dyno to see if they actually put out more power?
lutach is offline  
Old 04-09-2009, 02:04 PM
  #5  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

lutach-Thanks for the post. I have done track testing of senorless vs sensored motors. In theory you should obtain the same power, but they are not constructed the same. For the most parts the sensorless are machine wound coils of very fine wire. Three thick coils of this multistrand wire is then pressed and glued into the shape of a motor can. It is quite a messy looking build. There is no iron core. Lots of efficiency and potential power is lost by the fact that the iron core is not there to multiply induction by a factor of 10. A hand wound sensorless motor on an iron core using fat wire would peform well but cost much more.

Here are some questions and comments on torque that might be of help from the CRC thread.

Itchy

Very interesting testing John.

I suspect the different efficiency readings may have something to do with the balance of the rotor. I have made some rude atempts to balance brushless rotors, and i have found that smaller rotors seem more balanced than wider ones.

I still have the impression the larger rotors produce more torque for a given consumption than a smaller one. This can be seen for example by the difference in the brake settings. Tipically i will run a 13.5mm with 25-35% brake. The same motor will need 45-60% for the same braking power. it seems to me that more amps are needed for the same power output.

In any case it seems the most interesting feature in runing smaller rotors is that shift in the powerband you mentioned. For me its way more confortable and the consistency in my laptimes shows that.

Again, thanks for the interesting info

EDIT: just read the last post. I stand corrected. The smaller rotor is more powerfull all the way. I would never guess that by real track feeling though.


John Stranahan

Itchy thanks for the post. You are very welcome.

"I still have the impression the larger rotors produce more torque for a given consumption than a smaller one. This can be seen for example by the difference in the brake settings. Tipically i will run a 13.5mm with 25-35% brake. The same motor will need 45-60% for the same braking power."


I agree with your assesment of braking. Braking has to do with reverse voltage. Under braking the armature acts like a generator; the winds are shorted to produce brakes. The more "reverse" voltage the motor produces the better the brakes. It seems from the testing that the big armature does produce more reverse voltage, this makes it act like .5 more winds on the stator. More winds also produces more reverse voltage. Fewer winds produces the least reverse voltage and let's the motor rev up to these huge RPM's. (Reverse voltage limits motor RPM when it equals input voltage.)

There may be some other tricks played on the newest LRP TC edition speed control that increase brakes to levels that bruhsed motors have. It may be some application of reverse power rather than simply shorting.

I agree that it is a little hard to tell the difference between the two rotors on a wide pan. Yet we still picked the small one. I think this has something to do with insufficient forward traction. On a four wheel drive touring car the difference between winds and rotors is more clear cut.

As far as saying more torque, I would have to make another plot. We often get that phrase completely backwards. Motors with better bottom end punch often have more RPM and less torque. That shifts that motor power peak to the left to lower RPM and more amperage.

Attached Thumbnails
John

New
The pic shows the guts of a Mamba Max left next to the guts of a Novak 3.5 R. There is quite a bit of manufacturing differrence. The novak has more power and is a lesser wind. The mamba has no inductance increasing Iron core. There are quite a few differences on the track. I will discuss those later.
john
Attached Thumbnails Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-mamba-max-7700-kv-guts-cropped.jpg   Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-novac-3.5r-guts-resized.jpg  

Last edited by John Stranahan; 04-09-2009 at 02:15 PM.
John Stranahan is offline  
Old 04-09-2009, 02:38 PM
  #6  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

Sensored vs Sensorless Track Differences

The mamba's sensorless motors are quite a bit cheaper than the Novak sensored hand wound motors.

The sensorless motor obtains rotor direction position and speed from feedback through the power wires. Whenever the rotor spins it acts like a generator and produces back EMF (Electro Magnetic Flux or Volts). This is sensed and a complex software routine then drives the motor. It is difficult to sense the rotor when the motor is stopped or oscillating back and forth (stuttering),

A sensored motor has three tiny sensors visible inside the car that can positively sense rotor position, direction and speed. No back EMF need be produced for it to work.

Driving the two in a TC. The Mamba Max is only 7700 Kv and the Novak 3.5R is 10,500 Kv. The power edge goes to the novak. On our track in a touring car this gives the Novak about a 60 foot advantage every lap. That is just huge. The Novak makes the TC leap out of the corner. Top speeds on the straight were similar during this test.

Starting the race.
I find the sensorless always confused when starting the car from a dead stop. Half the time or more it starts backwards develops a little back EMF, this is sensed and then it goes forward. An inexperience racer does not even notice this. An experienced racer notices he is 1.5 car lengths behind on the starts that go backwards.

Contact with the boards.
You just have to think Nascar on this. The car gits a tap. No mater how small if you are at the limit of traction you are likely to spin. During a spin the tires will often stop turning forward and stall the motor. A good Nascar driver with deft use of the clutch will restart the motor during the time when the car is rolling forward in the spin, expert use of the steering wheel and throttle can sometimes pull the car out of the spin so it can proceed.

Taps to the boards.
the Mamba is a little like the Nascar but a little worse also. It does not just restart. When its stalled in a spin it is likely to be ossicilating a bit. The car and ROTOR has to come to a complete stop. Then, the motor has to do its little backwards or maybe forwards dance, then you finally get to go. I can pull out of some spins. I can very often proceed from a tap to the inner corner markers, But not with the Mamba. Again an inexperienced driver will not notice the car is doing him wrong. The Novak just keeps on ticking. Absolutely no interruption in power on taps to the board.


Stuttering on difficult traction at light part throttle.
I had the Mamba in my wide pan for quite a while. Sometimes the track is low traction. This is the tiime the pan requires the utmost sensitivity to changes of throttle in the turns. The mamba on occassion will stutter under this condition when you slightly increase throttle. This of course just makes the pan car spin out. Very agravating since you had it under control right at the edge of traction and a motor stutter wipes it all out.

So this is the status of sensorless. They are best on carpet in a touring car where the motor is at high drain all the time. They are poor in a 2 wheel drive pan car on a dusty layout.

Guys that own the Mamba will swear they never experienced any of the above; that is until they read this and start looking for the problems. It only took me two days.

Matt and I are running sensored motors at the moment. The Early Tekin sensorless had teething problems and I think they have seen the light and gone to sensored now.

Last edited by John Stranahan; 04-09-2009 at 03:04 PM.
John Stranahan is offline  
Old 04-09-2009, 06:30 PM
  #7  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

I was changing motors in the TC5 and had a motor unsolder and out of the car. I said, well I might as well stick the electronics on the dyno and give it a spin.

I oiled the motor bearing.

I had a 17.5 motor installed. I just stuck the front of the chassis in the gap between my tackle box drawers. It spun up to full throttle super smoothly. Sound level "and bearing stress" is probably 1/4 what it was before. I'll try some hot stuff later. On the Fantom mount even at 4.5 volts the dyno makes a kind of get the hell out of the room noise.

This mount is so much better having a full suspension. If you need an old pod and oval chassis I have about 3 send me a PM.

Man it's looking really good so far. I plan to mount a magnet driven sensor if possible. This might improve the first few points. It turns out Matt has been using a sensor that plugs into the sensor harness. I did not even know till yesterday his motor and esc was sensored.

Test to come, 17.5, 10.5, LRP X12 3.5, A voltage test with 6 cell 4 cell and 5 cell on the same motor. I need another cell. I have one fourcell pack left of fairly recent vintage. 4600 Energy pack. Anyone have a spare cell to donate give me a private message.

Hopefully we will have a GTB without a soft start routine to use on the beast. Note the balance weight is not needed.

Be sure and click on the pic a second time and get it fullsize. It is much clearer this way. You can see the cut in the motor plate that is reversed. This allows a from the rear motor install with little difficulty. Of course all wires and connections should be top notch just like you use in your 3.5 powered TC.

John
Attached Thumbnails Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-homemade-dyno-motor-electronics-mount.jpg  

Last edited by John Stranahan; 04-19-2009 at 08:25 PM.
John Stranahan is offline  
Old 04-09-2009, 06:42 PM
  #8  
Tech Fanatic
iTrader: (44)
 
Join Date: Jun 2006
Posts: 907
Trader Rating: 44 (100%+)
Default

Got your email John. I will be sending the flywheels, Sentry and sensors out tomorrow, look to be receiving them by middle of next week.

We can all count on John to thoroughly dissect this dyno setup and get it to work excellent

Matt
mattnin is offline  
Old 04-09-2009, 11:10 PM
  #9  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

I sanded both graphite plates flush with the motor plate.

I removed the left motor plate and changed it to a simple post. This is also sufficiently strong. It does not have to withstand crashes. This frees up a cheapy used motor plate.

I just stuck the car again into the space between drawers in the tackle box. The car did not move.

On startup there was some pod motion similar to when you take off in a pan car. The rear of the Pod depresses since the front cannot come up. This is well dampened. The startup on part throttle is oh so smooth.

At full throttle with the 3.5, you get all the glorious pitch but none of the excruciating high volume that you do with a standard solid motor mount. I liked it a lot for powerful motor work. By contrast the stock motor did not move the pod at all. You could tell by the sound of the 17.5 that a lot of vibration is dampened and not fed back into the motor bearings.

I did put on safety glasses and got out of the shrapnell path so to speak. Similar to what you do when you cut with a dremmel thin cutting wheel.

Really good so far. I have sentries, sensors, the Astroflight servo tester coming. Then we will do some motor work.

John
Attached Thumbnails Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-homemade-dyno-motor-electronics-mount-003.jpg   Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-motor-mount-modification.jpg  
John Stranahan is offline  
Old 04-09-2009, 11:30 PM
  #10  
Tech Elite
iTrader: (49)
 
andrewdoherty's Avatar
 
Join Date: Jan 2005
Location: ★Wylie, TX★
Posts: 3,815
Trader Rating: 49 (100%+)
Default

Originally Posted by John Stranahan
I sanded both graphite plates flush with the motor plate.

I removed the left motor plate and changed it to a simple post. This is also sufficiently strong. It does not have to withstand crashes. This frees up a cheapy used motor plate.

I just stuck the car again into the space between drawers in the tackle box. The car did not move.

On startup there was some pod motion similar to when you take off in a pan car. The rear of the Pod depresses since the front cannot come up. This is well dampened. The startup on part throttle is oh so smooth.

At full throttle with the 3.5, you get all the glorious pitch but none of the excruciating high volume that you do with a standard solid motor mount. I liked it a lot for powerful motor work. By contrast the stock motor did not move the pod at all. You could tell by the sound of the 17.5 that a lot of vibration is dampened and not fed back into the motor bearings.

I did put on safety glasses and got out of the shrapnell path so to speak. Similar to what you do when you cut with a dremmel thin cutting wheel.

Really good so far. I have sentries, sensors, the Astroflight servo tester coming. Then we will do some motor work.

John
John, have you looked into using the Speedpassion BL analyzer? If so is there a reason you went with the system you have? I don't have quite the technical knowledge it appears you do so forgive my ignorance here. . .

I am able to read RPM, KV, Amps, Voltage, and Watts from the BL analyzer along a variable throttle input. With a load (such as the flywheel you have pictured) attached to a motor, and knowledge of the motors input voltage, current, and the how many watts of power the motor is producing, is there any reason I could not calculate efficiency from that? Thanks for sharing your insights!

Andrew in Wylie, TX
andrewdoherty is offline  
Old 04-10-2009, 12:01 AM
  #11  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

Andrew- If you have access to the instruction manual it would help me answer the questions.

I went here to get some basic info

http://demonpowerproducts.co.uk/prod...oducts_id=3895

This is what I saw

There is no load connected the motor. Once you put a flywheel on the motor then you have to have a strong motor support to hold it. What I have shown so far in this thread is very suitable.

What is the amp range that the system can detect. If it is about 100 amps or so then it will be useful in determining efficiencys.

If the output that is given to the computer includes a Volt, RPM, amp, and time then our spreadsheet would do the calculations. Your output would be torque, power, smoothed RPM, Max Power, Efficiency. The speed passion product would take the place of the Sentry.

There is some dispute on the page as to whether all speed controls will work with it or only just speed passion speed controls.

I would be happy to look at the spreadsheet like computer file it generates from a test or a downloadable instruction manual.

John
John Stranahan is offline  
Old 04-10-2009, 12:57 AM
  #12  
Tech Elite
iTrader: (49)
 
andrewdoherty's Avatar
 
Join Date: Jan 2005
Location: ★Wylie, TX★
Posts: 3,815
Trader Rating: 49 (100%+)
Default

John,

Here is a link to SP's website concerning the analyzer. I wasn't able to find an electronic version of the manual like the one that came with the product.


http://www.speedpassion.net/product.php?lang=&c=46


I assumed I would need to fabricate a suitable motor mount for use with the flywheel. I am confident that the SP BL Analyzer will function properly with any sensorless or sensored motor and any sensored ESC.

I only have a hard copy of the instructions and no way of scanning the information. However the manual says the analyzer will read from 0-100 amps in .01 amp increments and 0-27.5 Volts in .01 increments, and from 200-200,000 RPM on a 2 pole motor with .6% accuracy. At first glance I can not tell if the analyzer will output anything other than max, in, and avg data. hopefully the data is sampled and recorded in small enough increments to be useful. Once I have some motors to test and have run some data through it I will see about sending you some useful information. Thanks!

AWD

Last edited by andrewdoherty; 04-10-2009 at 01:12 AM.
andrewdoherty is offline  
Old 04-11-2009, 02:32 PM
  #13  
Tech Adept
 
Join Date: Mar 2009
Location: Metro New York
Posts: 139
Default

John,

I'd like to create a track diagram of your 17.5 Oval Gearing problem using Google Sketchup 6.

I usually use the Arc tool to reproduce a "racing line" with constant radius r:

http://mathforum.org/library/drmath/view/55037.html

so if you provide the width of the turn (2d) and height (h) I'll estimate the turn radius r using the formula, and draw the two turns in the oval with the Arc tool.

Also, what is the length of the straights? I'll post my map as a reference on this thread, or start another, if you think that's useful.
SystemTheory is offline  
Old 04-11-2009, 03:09 PM
  #14  
Tech Elite
Thread Starter
iTrader: (27)
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,780
Trader Rating: 27 (100%+)
Default

Here are the details on our oval. Maybe Joe can post his on the Velodrome oval over in California.

Track space 75 ft long by 36 ft wide. Apex to outer barrier 11 ft. Turn radius maximum is 18 ft. Run Line is 173 feet. Inner corner marker is a semicircle with 12 ft diameter. The corner markers on our track are extended with flexible matirial to a drop shape and attach to a single center board. This gives a lot of runoff room for cars in trouble to straighten up. The actual path for best lap times seems to be a little diamond shaped. We use slightly less radius in the turns than is available. If you go to the Snowbirds, Thunder Ranch website and look at one of the Carpet nationals videos you can see what I am talking about as far as shape of the runline. Here is a video of a mod A-main. At the start you can see the cars almost exactly on the fast runline (inner lane) for most cars and classes. The runline is also a darkened part of the carpet. We use a very similar shape on the oval but the radius is just a little bigger than the start but very similar to the first few minutes of the race. We are near the side boards on the straight for only 3-5 feet.

Now the routine I used is not complex at all. We calculate average speed based on a few of the hot laps. Then I geared the motor to produce maximum power at that speed. The RPM at maximum power was determined from a dyno run of a 13.5 motor on a 4 cell pack. You will note in Barrys post above and also in the Fantom software there is a fudge factor thrown in that is supposed to allow for tire slip. I did not need nor use the fudge factor. You do try your best to control tire slip by reaching full throttle only at mid straight. You roll on throttle. Especially in a mod class.

Last edited by John Stranahan; 04-11-2009 at 09:11 PM.
John Stranahan is offline  
Old 04-12-2009, 08:08 PM
  #15  
Tech Adept
 
Join Date: Mar 2009
Location: Metro New York
Posts: 139
Default

John,

I didn't initially understand the term run line. My racing line in Sketchup is used to get an image.

Each 18 foot radius turn has a displacement of 38 feet. Each 164 foot radius "straight" has a displacement of 47.5 feet. That's 2 feet short of your 173 foot run line. I suspect this sketch is off a bit from the shape you would draw.

Where is your throttle position at the center of the sharp turn, when you first begin to roll up throttle?

I figure it would take 2.72 g's of average side bite to hold 27{mph} on the 18 foot radius, and about 1{s} in each corner, but you're obviously going a bit slower there than after accelerating hard in the long sections.
Attached Thumbnails Dyno, Homemade, Using a Novak Sentry Data Logger, Continued, The Experimental Thread.-oval_diag_1.png  

Last edited by SystemTheory; 04-12-2009 at 08:10 PM. Reason: clarity
SystemTheory is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service -

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.