Hi All,

I finally have some data from my electric experience from last year until now. (I hope that I can get these attachments to work on the forum) I started testing with one of my own old stunters and then went on to design my Electron as a purpose-built e-stunter from scratch. As some of you know, I burnt out a couple of the MVVS 4,6/840 in/out runners plus some Jeti Spin esc's, so I needed to do some more research and fiddling with all of this stuff. I now have an Eagletree data recorder with an additional temperature probe, so I thought that it's time to share my findings.

In case someone needs to know why I chose to start out with MVVS motors and not AXI's, or any other makes, this is quite simply because I'm the South African agent for MVVS, and since they ventured into the electric game, they sent me a few samples to play with. I was most impressed by the quality of workmanship, and they very idea of an out-runner inside a casing seemd to me like a good idea. Just the concept of not worrying about a rotor touching anything in the nose, is a good idea in my book. After burning my motor out plus one of Igor's in the world champs, I was totally puzzled as to the cause. This was a puzzle because it had worked so well for some time before the champs. The only change on the days when the motor's popped, was that it was a few degrees hotter. After each flight I checked the temperature and it was never so hot that you couldn't touch the motor. The MVVS factory said that the temperature must have gone over 130 degrees, but there is no way that it was ever over more than 70 degrees. They did state however that this 4,6/840 is rated as a 4,6 cc (.28 cu.inch) glow motor and that my 1920 gram (68 oz)stunter was too much for it. Maybe they do in fact, know what they are talking about?? Igor kindly lent me his AXI 2826/10 for my final rounds in France, and it certainly felt to me (this is very subjective) that it was not as powerful as the MVVS, but maybe the AXI is also not really made for my overweight model. Just by the way, the target weight was 1700 grams (60 oz.), but I found that making the Electron fully take-apart, this target was difficult to achieve.

I now believe that it was most likely the high current spikes that took out the windings and not that it was too hot. The other problem with this in/out runner, is that I used the Sport version because it has a built-in fan. I thought that this is also a good idea, but the problem is that the fan is in front so the air is sucked through the rear. The Electron's nose is made with cheek cowls to feed air in the area behind the motor, but in hindsight, this does not work well.

I then bought the bigger MVVS 6,5/910 which is their .40 size glow equivalent. It also has an integral fan, but you mount the motor either way around. My first attempt had the motor with the fan in front, because it was easier to mount. It had huge power and felt good, but a little hot to the touch, so I swung it around, and this is when the Eagletree recorder arrived. I put the tiny temperature probe right inside against the windings, and you can see the results in the chart. I do not have any fancy thermal compound, but I don't believe that it will be that much hotter anyway. I found it interesting to see just how much the temperature rises after switch off. I was told by someone that I must always let the motor cool down properly before flying again. This did not seem reasonable to me, so I did two flights with just a short stop to change batteries. The chart shows the 3 flights that I did, and flight#1 used more current and became hotter, due to the battery pack losing voltage prematurely. I run in governor mode, and of course this means that the current will go up to try and keep the rpm constant if the voltage drops. Flight#2 shows that the temperature simply goes back down to normal operating temperature, and starts all over again, so in fact, this is no problem at all. Flight#3 was much later, but for this flight I blocked up 3 small (6mm) holes on top of the nose just behind the spinner. The holes were drilled to see if I could get some more air into the mouth of the motor, and I really thought that were not doing much, so I blocked them. As you can see, the temperature was higher, so like the man says "Heat is your enemy......!!"

After quite a few more flights with the Eagletree, the high temperature after switch-off is around 58 degrees C, and this is with a summer's day of around 30 degrees C. I think that the Eagletree data recorder is a most wonderful toy to have when playing with this electric stuff. I've done so much "surgery" to my poor Electron now that it's a mess, but the testing is certainly worthwhile for me. Once I am reasonably happy with the installation, I intend to re-build the nose section once again.

In the meantime, here are the main details of my set-up with the MVVS 6,5/910:

Line length: 20,750 metres (68 ft) from the centre of  the model to the centre of the handle. The flying line is PAW Staystrate steel line of  about .014” (0.355 mm)
Prop: APC 12 x 6 E set at 9000 rpm with governor mode on Jeti Spin 44 esc.
Lap times: 5.1 to 5.2 secs/lap
Data: Eagletree V3 data recorder with temperature probe inside motor windings
Altitude: 5500 ft. (1676.4 Metres) ASL   Air temperature 30 to 32 degrees C.
Weight 1850 grams (65.26 ounces)

I hope that some of you will find these results interesting.

Keith Renecle

Out of curiosity, did the windings actually fail on your motors?

Hi Keith
THANK YOU for publishing your test and data summaries.  I managed to let the smoke out of a motor last year too - darned hard to get it all back in!  I figured mine was a combination of running the motor at its limits, on a hot day, with inadequate cooling airflow.  I did an almost complete pattern, everything working fine.  DId the first loop of the clover, the motor screeched and shut-off like it was out of fuel...

I did not reallize the MVVS was a Hybrid motor, also known as an "outrunner killer"  Mike Palko over here has been using a Steve Neu "ORK" motor in his excellent Mustang.

Anyhow, since my burn-up last summer, I have been very concerned about cooling path design.  I have been working on arrangements that try to force air through the motor - but I will need some in-flight data like what you have recorded to see if I am on track.

I am disturbed by your observation that the heat is retained flight to flight - I have been counting on flying back to back flights but if the cooling isn't adequate then this will not work.

I wonder if we should be using a cooling fan on the ground betwen flights?

I've sold a few of the 4,6/840 motors to the glider folks and you are dead right, they climb up at full power, and then shut off, so cooling is really no big deal. I did in fact, fly the model one day with no spinner but this was before I had the Eagletree, but I will do the test again, and see what happens. I am really considering a radial cowl like the Yak and Sukoi for my next model.

Dennis, thanks for your comments as well. I don't believe that it's a problem to fly back-to-back flights at all, the motor should be able to handle running at its operating temperature all day.......if it is within its own limits. This is the data that we seem to have to find out.....the hard way, but basically I believe that the principle is not to operate close to the limits. It's like the glow stuff, the OS.46 on a pipe is enough power......but the PA.61 just cruises. On hot days, the motor that is on the edge will be pushed over its limit. These electric motors are absolutely amazing because even the smaller motors will swing a 12 x 6 prop at the same rpm and use the same average current, but then they cannot do this for very long. This makes it easy for us to believe that we can get away with a smaller, lighter motor.

One of our local hobby importers brings in Scorpion here and they are selling like hotcakes, so I intend to buy a couple of them to check out. They look really good, and the spec on the wire and magnets is also impressive. Right now though, I am fiddling with controlling the motor more from the timer side. I used Igor's timers, but I popped the one that I got from him (I'm pretty good at blowing this stuff). Igor does not mind sending me stuff at all, but I just wish that he would move a little closer to me.......the post is just too slow.....and the winters are just too cold for me to move to Slovakia! I have been an electronics technician and hobbyist for most of my working life, but I never got involved with the programmable stuff like the PIC chips. I bought myself the PicKit 2 USB programmer and managed to make myself a few of Igor's timers by just downloading his software. Soon I needed to fiddle with my own idea's....as usual, so I tried to learn some assembler language. It was just too much like hard work and I could not get my head around it quickly enough. I did some BASIC programming on the old Apple 2.......in the steam-driven days, so I did a search and found a neat program called GCBasic that is designed to do robotics with these PIC chips. Within a couple of hours I had my first timer working, and although this BASIC compiler is a bit limited, it can certainly allow me to do this basic job quite well. I'm now playing with my own rpm checking timer that does not need any fancy ESC with it's own governor. The FAI rules really need you to stop the prop before landing, so we need the brake function. Flying without the governor mode, is like flying with a hot 2-stroke R/C glow engine that changes speed all the time through the maneuvers. Most of the available ESC's for heli's have governor mode, but without the brake function because they do not want the heli guys to stop the rotors by mistake. The ESC's that do allow this, cost a lot more than the basic ones, so this is why I'm having a go at this. The real reason is of course, that I like to experiment...........just like I did with the diesel stuff. My wife just prefers me to do this with electric...for obvious reasons!

I'll try the heat thing without the spinner and then post the results soon as I get a chance. It seems like I don't actually need to fit the sensor inside the motor for the flight because it keeps the max temperature for a few minutes anyway. Maybe I can get some of that thermal compound to try. We shall see!

Thanks Alan, Good comments and I trust your experienced judgement as well. Igor says very much the same in fact, in that you need more metal to dissapate the heat. The more that I listen to your comments, the more I see the importance of getting through the motor and not just over the casing. How about a liquid cooled stator?? mmmmmmmm....... P-51 with a real radiator??

We have some lousy weather right now, but I will do the test without the spinner soon and then post the results. Thanks so much for the feedback.

Here's another issue in cooling.

As the windings get so hot that the insulation starts to melt, the permanent Neodymium-xxx magnets--which block the heat getting out of the case) actually start to lose their strength. When this happens, the torque constant "k_t" (~10/k_v) drops (torque=k_t*current). Well of course your governor tries to keep the rpm constant so it ups the throttle--making more current flow, which of course heats the windings even more, which......., in other words you get a thermal runaway  , stopped only by your ESC perhaps cutting back on over-current, or the winding burning up.

According to the Scorpion motor "propaganda", their wire insulation is good up to 180C and the magnets are good up to 200C. They claim this is higher than "other" motors, but for the sake of argument, it gives you  an idea of upper limit of the acceptable temperature range.

I don't know if this happened to you when your motors burnt out or not. Since the magnetization loss is permanent, the motor basically is toast, even if the windings survive.

The thermal runaway that Alan describes does not even need the governor: operation at a constant duty cycle or throttle setting will produce ever increasing currents as the magnets lose field strength and the Kv rises, and that produces higher temps and eventually you start blowing ESCs while flight performance hardly suffers. Don't ask how I know!
Dean P.