dual switching BECs into Power Expander?
 

If I need to reduce the voltage that I'm running servos at, one of my options might be to put two BECs between the two 2S lipo batteries and my Power Expander 14MZ. Some switching BEC manufacturers (notably Castle Creations on the CC BEC Pro) caution people not to run their BECs in parallel. Do you know if there is likely to be a problem when switching BECs are used with the power expander in this manner (will the circuitry in the power expander isolate the two supplies such that there won't be a problem)?

RE: dual switching BECs into Power Expander?
 

Hi,
I am not sure why they say not to run them in parallel but the circuits in the MZ will not help the situation. Hooking them both into the MZ is the same as running them in parallel until one fails catestrophically. It would isolate a catastrophic failure but running them in parallel on the MZ may cause a failure if this is result of running them in parallel.

RE: dual switching BECs into Power Expander?
 

Thanks. Looks like I should avoid trying that in that case. :)

The reason I ask is that I'm having problems with some Hitec 7940 servos glitching when run through the power expander and the problem seems to go away as rx pack voltage drops from its 8.4V peak.

Another option that may work might be to power the rx with the full voltage - will I do any harm if I power the RX using the full voltage off the servo rails instead of using the regulated 5V lead? The idea here is that I want higher signal voltage. The aforementioned glitching also disappears if I put an Equalizer 3D inline between the extension and the servo I was hypothesizing that the equalizer 3d when run on 8V is going to send out 8V pulses rather than the +5V pulses that would be coming from the regulated receiver. A bit of a long shot but I'm kind of desperate to make this work and won't maiden the plane until it is rock solid on the ground. :-(

RE: dual switching BECs into Power Expander?
 

Hi,
Running the receiver on anything but the 5V we supply will damage the servo signal buffers in the unit. Running the Equalizer 3D on 8V will not do what you want, the control signal to the servo out of the Equalizer 3D will always be 3.3V. If you were going to run some BECs why not just run a TurboReg and set it at 6.5V?

RE: dual switching BECs into Power Expander?
 

Since the problem seems to go away with the Equalizer 3D I am thinking that Hitec did not terminate the signal line correctly (Futaba made the same mistake with the 9156 and 9157 which it seems they corrected as the newer ones shipped don't exhibit the bad behavor the originals did). You are probably seeing "ringing" on the servo control line. The Equalizer 3D has less drive than the MZ so you would get a slower rise and fall time on the signal edges and not as much ringing. Not much you can do about this except to put a 220 - 330 ohm resistor in the signal wire near the MZ end of the extension on all the servos.

RE: dual switching BECs into Power Expander?
 

I will definitely try the resistors! If it works that will be a super-cheap solution.

The problem with using a normal regulator is merely one of availability - I could have a CC BEC really quickly because they are stocked everywhere.

RE: dual switching BECs into Power Expander?
 

Unfortunately the resistor did not work, I tried 200 Ohms and 300 Ohms and neither had any discernable effect. I also tried running the extension direct to the rx (knowing that I never draw more than 0.4A per servo during the test because I'd measured it) and the glitching disappeared. I'm thinking that running the rx at higher voltage wouldn't work anyway because the servo buffers would drop the voltage down anyway?

What if I were to 'Y' all the servo wires with ground and power going to the power expander servo connectors, and signal going straight to the rx, with the rx powered also from one of the unused power expander channels? (i.e. each servo connection to the power expander has only red and black connected, each servo connection to the rx has only signal connected, and the batt port on my rx is connected red and black to an unused servo connector e.g. ch 14 on the power expander)?

I have a 'scope so it may be time for me to pull that out and have a look at the signal. I'm done for tonight though.

Thanks for your support, it's really great to have you here on the forum answering questions!

RE: dual switching BECs into Power Expander?
 

Well, this is a little weird but it looks like the frame rate is lower with the equalizer 3d in place, and the signal voltage is lower not higher. Does this make any sense to you at all? Also the signal is clean with or without when the servo is not connected. With the sero connected I get noise when I move the controls, i.e. when the servo is actively moving, which seems to cause glitching.

RE: dual switching BECs into Power Expander?
 

Hi,
You could try running the servo control of the receiver with the power from the servo outputs on the expander.

The Equalizer 3D will output 3.3V on the signal wire always. This is because the microprocessor runs at 3.3V. Also, the 3D frame rate is fixed at about 2.5ms per servo so you are getting 15ms frame rate always. This will not change with a different frame rate from the receiver.

I am not quite sure what you are saying about the noise and "with or without"? The basic circuit inside the MZ is a 74HCT541 driving a 51 ohm resistor that then goes through a ferrite bead and out to the connector. Sounds like some of the noise is coming off the servo?

RE: dual switching BECs into Power Expander?
 

OK, wrt the 3D signal @3.3Vthat explains why the voltage was actually lower on the signal line and also the reduced frame rate I was seeing on the scope (the Airtronics is very fast).

By 'with or without' I was referring to whether the equalizer was in place or not between the extension and the servo. :)
It does look to me like the noise is coming off the servo. Ah well time to experiment to see if there's something I can do to smooth that out (maybe add a cap). Maybe Hitec will have some suggestions.

Thanks for the incredible level of support!

Eric

RE: dual switching BECs into Power Expander?
 

Just a quick update for the benefit of anyone else that has similar problems and happens upon this thread after a web search, it looks like the problem has nothing to do with any interaction with the Smart Fly power expander because the same thing happens if I take the power expander out of the circuit and power the rx and servo directly from a 2S lipo.

So far the only remedies that have worked for the HS-7940TH servo glitching are

+ add a big elecrolytic cap between power and ground, at the servo end
+ put in an equalizer 3D at the servo end
+ lower the voltage (e.g. use nimh rx pack or a regulator)
+ helps a lot but not completely eliminates the problem - direct connect servo to smart fly unit with no extension (obviously this can't work in practice because the servo lives in the tail)

Things that did not make a difference:

- directly connect servo to RX running on a 2S lipo with no Smart-Fly Power Expander device in the circuit
- add a small 500pf cap between ground and signal lines at the servo side
- swapping servo extension for different servo extensions
- adding a 2nd 14GA heavy ground line from the smart fly unit to ground on the servo lead
- add a big electolytic between power and ground at the rx end
- add a 200 or 300 Ohm resistor to the signal line at the servo end (to help with possible ringing)
- add a ferrite ring at the servo end

RE: dual switching BECs into Power Expander?
 

Hi Eric,
Thanks for the post. Sorry you are having such a time with these. Have you tried to contact Hitec yet?

RE: dual switching BECs into Power Expander?
 

Also, you would probably be best with a very low ESR capacitor like a tantalum or MLC. I don't know if you want to go down this route but these would probably give you the best results.

RE: dual switching BECs into Power Expander?
 

Thanks for your help and advice. Yes I'm in contact with Hitec and had done the test direct to the rx at their request but beyond that they haven't had any suggestions to offer me. :) A smaller cap (the 500pf/0.5uF) across the power leads does nothing at all, I don't have any larger capacity low-ESR types (like a 220uf 10V tantalum), I have lots of electrolytics (200-400uF/50-60V) because I'm an electric flyer and we use these to resolve back EMF issues on long leads between the battery and ESC (which is what gave me the idea to try it in the first place).

RE: dual switching BECs into Power Expander?
 

Let me know if you get any suitable resolution to this. I try to keep up on these things so I have an idea of what to ask when trying to solve a problem. Thanks,

RE: dual switching BECs into Power Expander?
 

I just went out and bought some 22uf tantalum caps and these _also_ work. Again, the position for these is between the +ive and -ive power lines (_not_ negative and signal, this does not work) on the servo side of the extension. There is a thread on FG about problems with the HS-7950TH on 2S lipos with extensions and I may post some of this info there, see what people think.

RE: dual switching BECs into Power Expander?
 

Hi,
My presumption is that they did not decouple the power supply well enough and they are getting big enough power glitches to cause the microprocessor inside the servo to get weirded out and reset. There should also be a significant cap across the motor and maybe they miscalculated this. Seems all the radio people have hired new kids and these are making errors in designing new products that should not be made if they had some experienced guidance. I am seeing this happen on all the brands and it really disturbs me.