Originally posted by landru
mmormoto,

What motor efficiency do you reckon you're getting with the 7035 prop producing 165g static thrust on a power input (to motor) of 20 W?

Andrew

I am not sure, I didn't measured torque. Probably near 50%, as it was a motor with stock grey magnet.

Probably near 50%...

Wow! Only 50%? I was thinking it would be quite alot more efficient than that.

I've read a number of posts claiming 70% or higher efficiency for converted cd-rom motors.

An input of 20 W producing 165g thrust gives 8.25g/W. If the motor were 70% efficient, we would expect to get 11.25g/W, or 231g thrust for the same 20 W input to the motor.

For a point of comparison, the GWS EDP-400 is listed on the GWS website as producing a paltry 4.18g/W at 7.2 v with the same 7035 prop. Wouldn't you expect the direct drive EDP-400 motor be working in the same 50% efficiency range?

So if the s400 motor is working around 50%, the cd-rom motor must be quite a bit more efficient. Either that, or the s400 is working at a really low efficiency with that 7035 prop.

Am I missing something here?

Originally posted by landru
I've read a number of posts claiming 70% or higher efficiency for converted cd-rom motors.

An input of 20 W producing 165g thrust gives 8.25g/W. If the motor were 70% efficient, we would expect to get 11.25g/W, or 231g thrust for the same 20 W input to the motor.

My prop calculator (I found it's quite good) gives me the following result:

Prop = 7x3.5
Rpm = 7150
static thrust = 5.9 oz (165 g)
mechanical power = 17.7W

So if electrical power is 20W the efficiency is 88%. Either data are wrong (electrical power or static thrust) or you just have a perfect motor.
I'm trying to get the same thrust from mine, by I aim around 4A at 7V (at efficiency around 65%)

RysiuM

I am pretty sure this GWS EP-7035 needs much less power then 17,7W to produce 165 g static thrust, I estimate about 10W.

My motor's efficiency must be below 70% at the measured current. The DC resistance of the winding is high, the copper loss alone is more then 20%.

I measured some GWS propellers, the other's parameters are more or less similar to the calculator results, but EP-7035 is an exception. The shape of this propeller is very different too.

I am just making efforts with a very simple torque measuring device.
The motor is fixed to a salvaged hard drive drum, the shafts are horizontal and parallel. A thin rope is fixed and turned around the drum, the other end of the rope pulled the digital scale (loaded with something heavy). The torque of the motor turns the drum, the drum pulls the rope, the scale displays the force. The torque is the product of the force and the radius of the drum (plus the radius of the rope, to be accurate...). With my small CDRom motor the force is in the range of 0...100g, my scale is very accurate in that range. The drum has very good ball bearings, so I hope the measurement is accurate.

Here is a picture of the GWS EP-7060 (upper) and the EP-7035. The shape is different. The calculators are more accurate in case of the 7060 then the 7035.

Originally posted by landru
Wow! Only 50%? I was thinking it would be quite alot more efficient than that.

Am I missing something here?

This motor is just a stock CDRom motor, original grey magnet ring, original winding. The resistance of the winding is high, that is the main part of the loss. The motor is much more efficient, if:
- the battery voltage is higher
- the rpm is very high
- the load (and so the current) is not so high

With this motor constants Motocalc estimates a peak efficiency of about 82%, but not with this "big" propeller. This 82% semms nice, but it comes at rather low power and extremely high rpm - practical use of this motor needs different settings...

I hope (with good reason) that replacing the stock ring to NdFeB magnets improves the situation. Stronger magnetic field lower Kv, I need less turns, it is possible to use thicker wire etc.
I need quantitative results, so torque measurement is a must.
I am interested in the following:
- same stator
- parameters with stock magnet
- parameters with iron fluxring and 1,5*5*5 neo
- parameters with smaller gap
- as above, but aluminium drum

The last question is very interesting. We all know that iron is better, but it is heavy as well. Exactly how much is it better? Worth the extra weight (and more complicated machining)?

who have contributed to this thread.

last night i was able to get my first cdrom prototype working. it took several attmepts to get the winding correct w/o shorts. i think this was because when i took the stator off the plastic covering on the bottom plates came off. without the covering the wire kept shorting. thanks to this thread i was aware of the need to check for shorts so this likely saved me a cc phx 10!!!!

the cdrom is just a generic type very similar but not exact to mmormota's. i wound it with 20T of radio shack red wire (30 gauge) and connected it using the star approach. without a prop i measured about .9 amps at full throttle on a kokam 1020 2s1p pack.

i then put a gws 3x2 prop on the shaft (directly with ca). this gave me about 1.9 amps at full throttle. i don't have the motor mounted so i couldn't get a tach reading. it was certainly in the 1,000's of rpm, but i doubt much past 9,000 rpm.

it had a bit of trouble starting, i think due to the fact that the stator and bell got out of whack when i took it apart. i found the cd over a year ago before i read mmormota's guide and did more damage to the mechanical parts than i probably needed to.

i have 5 motors coming from nippon dave so i'll continue onward. my 2 goals are to: 1) wind a cdrom motor that can fly a gws moth direct drive with a 6 or 7 inch prop and kokam 340 3s1p and 2) wind a cdrom for a hobbico ff ww2 fighter w/ kokam 340 2s1p.

thanks again guys,

jeff

About insulating the stator, it's about a lrk motor (different winding diagram!) but the method is the same:
http://translate.google.com/translat...known&oe=ASCII

Originally posted by burgan
...i was aware of the need to check for shorts so this likely saved me a cc phx 10!!!!...

Always test your motors from a current limited power supply or put a powerresistor in series with your battery.

mmormota,

I agree with you that accurate quantative results for the power output of our cd-rom motrs would be wonderful to have. Your torque meter sounds very interesting.

Because I don't have access to quanatative mesaurement of power out, I've attempted to calculate it by assuming that the power out is equal to power in minus copper losses (I^2 * Rm) and minus iron losses (Io * U). I have measured/calculated the constants for my motor (19mm stator, Neo magnets):

Kv = 1727
Io = 0.42
Rm = 0.49 (difficult to measure -- this is a Motocalc value)

Using the above figures and equation, I calculate ~64% efficiency with motor drawing 4.00 A at 7.62 V.

That's still a long ways from 70% unfortunately, but much better than 50%. But perhaps my assumptions about power and copper and iron losses are wrong?

P.S. Your point about considering power levels when talking about efficiency is good. After all, we do want to use the motors we make to fly planes. Well, most of us do. Right now I've gotten more interested in making and analyzing the motors.

Originally posted by Cue
I have a problem with my motor, its a 9 pole and I rewound it a while back so I don't remember how many turns, its the green wire from radio shack, 26 gauge. I think it was around 14 turns. It is a star winding.

Well with the stock magnets a while back I remember running it an it would stop after full throttle or so and jitter, probably due to the weak magnets. Well I just put in 12 5x5x1mm magnets, polarities are correct, and I tried to run it with no load and it just would stutter like it was trying to start but wouldn't. I changed the timing on the phoenix 25 to the high setting and it still stuttered but if i spun it with my hand i could get it to run. It got pretty warm though and isn't right.
Any ideas on why it wont start properly? I think Im going to rewind it, whats a good gauge and turns to start with for say a gunther prop or a 5x5?

Nobody has any thoughts???


Well I rewound the motor with 20 turns of 26 gauge and it runs great now with no load on it, has full throttle range. If I put a prop on it it will reach full throttle at about 3/4 stick travel, anything after that it will slow down. With a 6.6x4 prop it pulls around 3.5 amps on a 3s etec pack.
Anyone know why it wont run right with a prop on it? I have also put a wheel collar behind the shaft so I know the magnets are staying in the proper place, and I have tried the standard and advanced timing on the phoenix 25 esc.

does the controller have the latest software uodate?

@Cue

If I put a prop on it it will reach full throttle at about 3/4 stick travel, anything after that it will slow down...

Did you try programming your ESC to low timing advance?

Also, how old is your Phoenix 25? I recently tried an older Phoenix 10 with a cd-rom conversion. With the older esc (purchased about 9 months ago), I found the same problem you describe above. Also, the same motor runs noticeably hotter and less powerfully than with my new Phoenix 10. I think CC has done some re-engineering to improve the performance of the Phoenix 10 with this kind of motor.

I tried the low setting and it is worse. I have 2 25's and they are only both about 2 months old. I tried a hacker 40 amp esc and it runs perfectly, so its something with the CC ESC's for sure. I hate having to send CC controllers in every other month for software updates.

I got 6 amps on the 6.5x5 prop and a 3s pack of lipos, anyone know how much this motor can handle roughly amp wise with 20 turns of 26g star wind?

Originally posted by Cue
I got 6 amps on the 6.5x5 prop and a 3s pack of lipos, anyone know how much this motor can handle roughly amp wise with 20 turns of 26g star wind?

Hi Cue,

6 amp on 6x5.5 prop with 3s (60W?) is too much to me. I had tried CDROM motor with 0.4 mm (~26g) 20T~23T. I found it is a bit too hot for 3S and 6x5 prop. 0.35 mm 29T was better, or smaller prop like 5x5 or 6x3 is better depends on airplane design.

I think about 40~50W input is the upper limit of my motor and efficiency will decrease drastically above. I noticed hotter wind will give more heat than rpm increase. (With 0.4 mm GWS 6x5 ET1200 3S, the motor will melt the soldering of mountin 30 seconds! )

I started to double the stator and the motor accepts more power with less heat. I assume you have similar sized stator as I have. By the way, I still use CC Phoenix-10 with pre August software.

Satoru