BEC speed controls

[Ryan]

I am using the Kontronik 3SL 25-6-12 brushless speed control. It has a BEC circuit on it, and I've been using it so far. But just as I was posting my other question, I remembered something. In an article on BEC speed controls on the Ezone Magazine, they said that they didn't recommend using more than two or three mini servos with a BEC circuit. Because the power for the servos has to come through the speed control, it heats it up. If it gets too hot, the speed control will shut down. You lose your engine, but more importantly, you lose power to your radio.

My BEC has worked fine up until now, but I'm wondering if it could be a problem on a warmer day. I really hate to think that I could lose my ECO because of something stupid like this. I think I might put a NiMh battery on board, just in case. Then again, I do love the convenience of never having to worry about or charge reciever batteries!

Has anybody had problems using a BEC with 4 servos on an ECO?

[Tohru Shimizu]

Hi Ryan-san,

> Kontronik 3SL 25-6-12 brushless speed control. It has a BEC circuit on it,

3SL has not BEC.It isn't 3SLB ?

Anyway, rudder servo in EP-heli requires larger e-current, due to the quick gyro control, than airplane and glider. In addition, the reduction from the voltage of NiCd battery to the voltage required for receiver generates large heat loss which may burn the BEC unit.

Hence, in general, if you use BEC in EP heli, most of ESC(electric speed controller) is permitted to be used only in less than 10cells. In fact, two years ago, I had burned Schulze booster40be (sensor brushless controller) using 12cells and BEC.


[This message has been edited by Tohru Shimizu (edited 04-24-2000).]

[Ryan]

Thanks again. You're right, it is the SLB. I took the name from my memory.

I'm using 10 cells right now, so I'm right up against the limits of the BEC. I think I'm going to use a reciever battery, just to be safe.

I'm also going to contact Kontronik and see what they have to say about it, since they know the actual limits of the speed control. I'll repost if I find out anything.

[Ryan]

I contacted Kontronik about the SLB 25-6-12, asking them what the maximum servo/cell count should be.

They responded by saying that for an electric helicopter with 4 servos and a gyro, that you should only use up to 8 cells.

With things like this, they usually play it safe and go conservative on their estimates. That would explain why my speed control has survived up until now. <heh>

I think I'm going to take their suggestion, just to be safe.

[Tohru Shimizu]

Hi Ryan-san,

Without any problem, I fly my ECO8 and EP SR with 3SLB40-6-12, 10cells1250-2000mAh and heading lock gyro which requires large e-current for the strict rudder control. My NiCd are not processed in high voltage, as called as the Zapped NiCd.

Before when the esc's BEC brun due to the BEC's over-load, the esc always becomes very hot. Please touch the running esc, carefully, for your 8cells and 10cells.

P.S. Yours is 3SLB25-6-12 ? The 25A is current draw max limit.It looks too low.

[Ryan]

Yeah, it's the 3SLB25-6-12. Ikarus lists a 3SLB 25-6-18 on their webpages. It's the same one. It's a mistake on either their webpage, or what's printed on the sticker on my speed controller. Either way, I don't think I'm going to use more than 12 cells with it anyay, so it really doesn't matter to me which number is actually correct.

I've flown for a while with 4 servos, 10 cells, and an Ikarus Profi-Gyro. So far I haven't had any problems, but I have noticed that the speed controller is pretty warm. I don't know if it's from the motor current or from the BEC, though.

I can see what you're leading to with the 8 and 10 cell batteries. You mean that I should try the 8 cell and feel how hot it is, and then try the 10 cell and see if it's much hotter, right?

Well, I might save that for some time later. Right now, I can't really afford to lose my speed control (or ECO). While it might be safe to run it with the BEC, and while I might be able to determine that, I guess that it really comes down to me not wanting to take any chances right now.

I'll probably play it safe for now, but I might want to do this later on. What are your suggestions for when I try this later?

As far as the current goes, yeah, it's the 25A version, but I'm not using the standard Ikarus brushless motors. I'm using the Aveox 1409-3Y. It runs at at under 25A. I don't know how it compares to the Ikarus brushless because Ikarus hasn't listed the specs for their motors, but I have a feeling that it is overpowered for the heli (but possibly not, considering the cell count). I say this only because Aveox recommends the 1406 size motor for the ECO 8. I just happened to have this motor. So far, the ECO flys pretty well with this motor, but I really don't have any reference to compare it to.

I have one more question for you. I am using my speed controller in helicopter mode. Because it commands a constant RPM, I hooked it up to my landing gear channel to turn it on and off, and I use the ATV setting to set the max RPM. This way, the left stick only controls the collective pitch, and I don't have to ever worry about setting pitch-throttle curves. How do you have yours set up? I'm just curious.

Thanks for the help!

[Ryan]

Oops. I said that "Ikarus lists the 3SLB 25-6-12..."

I said Ikarus, but I meant Kontronik.

[Tohru Shimizu]

Hi Ryan-san.

Certainly, avx1409/3y may be powerful without overheat problem. But, I think that it may be too heavy for ECO8's active (acro) flights. Rather, avx1406/3y, Kontornik BL50-20 and Hacker B50-xxS look better. In a few weeks, I will examine B50-14,15,17S in my ECO8. It will be reported in my web site.

Comparing with X250-4(H), these are more powerful (like yours), but may easily become hot, depending on your flight style. In this respect, X250-4(H)'s internal cooling fan is very good for overheat problem. Note that if you burn the motor due to overheat, the motor's magnet lose the power, and it is not recovered forever. Your heavy avx1409/3y should be tough for the overheat.

Now, I don't use heli mode (constant rpm mode) in 3SLB40. I am using airplane mode. Because, I dislike abrupt power down of NiCd battery at the end of flight in the use of the heli mode. It is a chracteristic feature of NiCd. In the airplane mode, the power down gradually comes. So, I can always feel the NiCd's condition in the flight, (using rudder mixing by my finger !).

[Mike Sayer]

I am now using an AVEOX 1010/3Y with L160 speed controller and 10 RC2000


The L160 controller BEC is only rated for 3 servos on 10 cells, so I run the gyro and tail servo off a separate regulator which is very simple to build and saves the weight and hassle of an separate RX battery.

The motor runs hot, but the AVEOX motors use cobalt magnets, which should be OK up to 250 degrees C. I read some where that if you spit on it after a flight and it doesn't sizzle it is OK.

I think the magnets would become detached before they lost there magnetism.

[Ryan]

Cool. A new idea! And I was just about to order some NiMH batteries for a reciever! I wouldn't have minded that so much, but it's so nice to not have to worry about charging or checking the reciever battery.

I really like that idea. Now that I think of it, I can remember seeing some high powered surface mount regulators in the Digikey catalog that would probably do the trick. I'll probably get something that's overpowered, just to be safe.

Any recommendations? Also, did you just use a voltage regulator, or did you build something more complex?

[Mike Sayer]

I used a 5 volt fixed regulator in a TO220 package, with a small piecce of aluminium as a heatsink. You need some electrolyticc capacitors on th einput and output. I used LM7805 regulator and 10 and 20 uf caps.

You need to break the 5 volt wire to the servos you are feeding, because you do not want the 2 the regulator outputs in parallel.

I built it on a small piece of Vero board.

I mounted it on the side behind the main shaft. It hardly gets warm so the heatsink was a bit of over kill.

[Ryan]

Sounds good. I've done some research on my own in the mean time. I found the surface mount regulators I mentioned before, but none of them could supply the required current, and adding a heatsink is more complicated. I found a 5V, 2A regulator at Mouser (www.mouser.com), part #511-L78S05CV. It's also a TO220 package.

I also hooked up a servo to my multimeter to determine the typical current use. I found that by moving the servo rapidly, it used around .35A. I then deflected the servo, pushing against it to make it work really hard, and it used around .4A to .5A. So, I figure that a regulator that can supply 4 times the current (provided adequate cooling) will probably be able to handle the job.

That's good to hear that yours didn't get warm at all. That just might have been because of your heatsink. I think I'll put a heatsink on each regulator anyway, just to be safe. One great thing about helicopters is that we don't need a cooling fan! Do you know what the current rating is on yours? I looked it up on the NTE website and their equivalent of that part number is a 1A regulator, but then it could be different depending on the manufacturer. If it is the 1A version, that would be really good since your regulator doesn't heat up.

Since the weight of each one is minimal, I think I might use one for each servo, bypassing the BEC of my speed control all together. I figure that it's getting enough heat from the motor current. The other reason is that I could use all 6V regulators and up the performance of my servos.

As for the packaging of the regulators, I figure I might just solder the components directly together without a circuit board, and just have plugs for the servo to plug into. These plugs would be connected to the ground (common to everything), voltage out from the regulator, and the signal from the reciever. And yes, I'll be keeping the voltage output separate from the reciever power pin.

I had another idea. (Been doing a LOT of thinking) I am planning on moving my reciever and gyro to the back end of my ECO to reduce the chance of radio interference. That leaves only the speed control in the front. I was also going to chop off the "shelfs" on the front of the fuselage, leaving the front section level with the motor mount.

I was considering getting an aluminum motor mount made for me. One where it would extend out to the front of the fuselage, with no hole for the aileron servo. I would then mount the speed control to the top of this large area using zip ties and heat transfer compound. So the whole motor mount would act as a heat sink. Grooves could be machined on the bottom side to lighten it up a little more and also add more surface area.

So, getting to the connection between this and the voltage regulators... I could mount the voltage regulators directly to the motor mount wherever there's room, and forget about the individual heat sinks.

I probably won't go so far as making an aluminum motor mount, but even if I don't, the individual heat sinks should be fine.

As I was writing this, I was thinking that I might have gone too far in my planning on this one, as far as the custom aluminum mount is concerned. (It's easy to think about things, but harder to actually do them) I think it's the nature of RC helicopters that we're always looking for some way to modify them. I've never thought this much about modifying a fixed wing aircraft! I guess it's because heli's are so mechanical, and there's so many little parts that you could add if you wanted to.


[This message has been edited by Ryan (edited 05-09-2000).]

[Mike Sayer]

I don't think the maximum current rating of the regulator will be the limiting factor. Using 10 cells and generating a 5 volt rail means over 5-7 watts dissipated within the regulator. At 2 amps this is 10-14 amps.

A typical small heatsink for a TO220 has a thermal resistance of 20 degrees C per watt.

So 5 watts would give a temperature rise above ambient of 100 degrees. At 30 degree ambient this means the cchip temperature will be about 130 , and this is about as hot as you want to go. Airflow will give you better cooling than this, but I don't think you want the average current any more than 1 amp. When I measured the current on my ECO while stirring the sticks as much as possible the total current was about 0.75 amps. I think 2 regulators would be safe. But my speed controller/BEC does not get very hot, so I see no reason not to use it.

By the way my regulator is out in the draught not under the canopy.

cheers

Mike

[Ryan]

Actually, I didn't mean to imply that I was going to run 4 servos on it to run it at the maximum allowable current of 2A. I was planning on running one servo per 2A regulator. I figured that if the regulators cost and weigh the same, then I might as well go with the one that can handle more current, to give me a better margin of safety.

But after reading your message, I realized that my thinking was off track. I was thinking electrically and not thermally, and as you pointed out, it's a thermal problem. I had assumed that the 2A version would be able to dissapate heat at twice the rate of the 1A version, thus allowing it to output more current. But the two regulators have the same physical shape and size, so there's no reason for this to be true. I even looked up the thermal resistances of two regulators with different current limits, and they were about the same. So, a 2A regulator would do no better than a 1A, as far as dissapating heat. The only difference is that the 2A can supply more current, but you'd still have to provide the means to dissapate twice as much heat. Right?

You can put these in parallel with each other, right? (Up until now, I was thinking of one for each servo)

If that's the case, then I'll probably use 6V, 1A regulators with heatsinks, and connect all the output voltages so that they are common. Then it would be a matter of determining the total current through the servos, and then the total power put into the voltage regulators. From that value, and the thermal resistances of the regulators/heatsinks, I could determine the number of regulators to use.

This is a lot more comforting, knowing that if one happened to shut down because it overheated for some reason, the others would take up its slack, providing there's enough of a margin of safety. I was a little worried with my original idea of one per servo, since the results could be catastrophic if one failed or shut down.

Also, after looking into these thermal issues, I've decided that the aluminum motor mount idea is probably more trouble than it's worth. Partly because it would take some effort to build it right, and partly because it's nice to be able to look up the specs of a regulator and heatsink and know that it won't fail, rather than finding this out after hours of work building the aluminum mount.



[This message has been edited by Ryan (edited 05-12-2000).]

[T. Shimizu]

Hi Mike-san and Ryan-san,

Your mails are too lo--ng ! I don't read the whole, now. But, your discussion is very interesting for me. Now, I will try 11-12cells in my EP-heli but I don't know the upper limit of my ESC's BEC regurator (Kontoro 3SLB, Schulze booster40be and so on) in heli. Do you know the 'real (without margine)' upper limit of the Wattage permitted to be dissipated in those BEC.

P.S.

What is your ESC ? It is L160 ?
In my L160C's manual, at 10.5W heatloss, auto-cut function works. For your saying 20degC per watt, how long time the temp. rearch the 20degC ? I think that duration time is important in the comment.

For Cobalt, Nedium ,,,,, certainly, the Curie? critical limit temp for the overheat may be your saying temp. But, note that when magnetic force induced, the critical temp may be decreased somewhat. At this point, even if it is less than 50degC (which is usual temp), when your induced magnetic power is too large, the magnet loses the power, forever.

shimizu@cs.ehime-u.ac.jp
http://www.dokidoki.ne.jp/home2/junkos/



[This message has been edited by T. Shimizu (edited 05-12-2000).]

[Ryan]

Sorry for being so long. I tend to be confusing when I try to keep things short. But I can also see how writing long messages can make things difficult if English isn't your primary language.

In one of my earliest messages on this topic, I mentioned that I asked Kontronik about the limits of the 3SLB. I asked specifically what the actual limits were, but they didn't say anything more than 8 cells, 4 servos. I think I got that simple answer because the average modeler wouldn't know how to use the actual specifications properly. (I didn't at the time) I think I'm getting a feel for it now. As far as finding out that information, I wonder if it's a matter of talking to the right person, or if they would give out that information at all?

About the 20 degC/W... That message is talking about using a voltage regulator, separate from the BEC of the speed controller. Here's an example of what we're talking about:
http://www.nteinc.com/specs/900to999/NTE962.html

The 20degC/W is a steady state value. That example value was for a voltage regulator with a heat sink. In my experiences with voltage regulators, they heat up to that steady temperature pretty quick. 10-20 seconds.

[This message has been edited by Ryan (edited 05-13-2000).]

[Tohru Shimizu]

Thanks for your reply.

I also think that production companies of speed controller must announce the real upper limit wattage of the BEC dissipation. Their saying 8cells limit, which is basically un-clear announcement, looks much lower than my flight impression, in which it seems that 10cells is OK in all seasons (0-30degC, about 5min flight).

I may also try regurator. I will research about it.

By the way, eventually, I began to examine hacker B50-xxS, in my site (klick above buttun for my profil). In my preliminary impression, this motor was actually good. If you use it in your 10cells ECO8, B50-17S or more than the coil turn number will be available. My the examination is still continued.

[Mike Sayer]

Hi,

This is a rather late follow up, but you must not connect the o/p of the regulators in parallel. If you do you will probably find that one supplies all or most of the current .

The input 5 volts and grounds are common, then each regulator must have a separate o/p, only connect to the servos it drives and not to the Rx.

This is the information I have found on various types of magnets.

Material Temp.Curie Temp.working max.

Ferrite 450 300
Cobalt 750-825 250-300
Neodym 310 150

I am not sure how the performance of the magnet changes as we approach the maximum working temperature. I have sent an email to Aveox to find out their thoughts.

[Tohru Shimizu]

Mike-san,

The data is interesting. But,
the Cobalt's data....How can I read that ?

P.S.

Hacker brushless motor B50-xxS's test was mostly finished.The data is partially reported in my web site. It is very good for 8cells ECO8's acro flight. By 10cells, acro is .......????

[Mike Sayer]

Hi ,


Hope this explains the info on the Cobalt magnets.

Cobalt
Temp.Curie 750 to 825
Temp.working max 250 to 300

Cheers

Mike

[Mike Sayer]

I have just had a reply from Aveox about the working temperatures of the 1010/3y. They say 80 C max. I assume this means that it uses Neodym magnets.

I have fitted a 14g computer fan (brushless!!) and that seems to keep the temperature down. Aveox suggested an 11 tooth pinion, I guess that will operate with higher voltage and less current for the same head speed.

It looks like I might have to try another solution. This is getting expensive !!

cheers

Mike