Radio brownouts & Reboots - RX, ESC, Servos, or do I just need EBCs?

[gloobnib]

Radio brownouts & Reboots - RX, ESC, Servos, or do I just need BECs?

I need some help figuring out a problem I am having with my ESCs/Servos/Radios. All configurations discussed below are using FrSky S6R RX and 3S Lipos.

Short and sweet version: 1) When should I move to using a dedicated BEC vs the one built into the ESC? 2) Should I be freaked out about flying my HK Walrus 2M motor glider that has a full boat of 6 servos on the HK-supplied ESC? 3) Did I buy a bunch of bad servos, bad receivers, or bad ESCs? 4) How do I go about figuring out the answer to #3 if it isn’t a cut and dried answer?

Longer Version:
I’ve been having mixed success on my journey since re-entering the hobby about 6 months ago. When I got back in, I bought the following as my go-to components when building new planes, which have served me quite well up until about 3 weeks ago:

• 6x Turnigy Plush 30A ESCs
• 8x Turnigy and Hextronic brushless motors
• ~24 HXT900 (9g) and ~10 HXT500 (5g) servos
• 5x Frsky S6R receivers (contains integrated gyro)

About 3 weeks ago I was running low on parts (5 FT planes in the hanger! Yay!) so I went to order a bunch more parts. I was getting sick of the HXT 9g servos stripping (mostly on crashes, but 2x while in flight) so I decided to mix things up a bit. I also had a fairly good sized Amazon gift card, so I wanted to source as many parts as possible there (free 1 day shipping!)…. and this is where the troubles started. I got a new batch of parts:

• 4x additional Frsky S6R receivers (I love these, don’t see any reason to change)
• 4x no-name 1806 2300KV quad motors for Mighty Mini series planes (QUANUM BE1806-2300KV RACE EDITION BRUSHLESS MOTOR 3~4S (CW)
• 4x no-name mini ESCs to go with the 1806 motors (ZTW SPIDER PRO LITE SERIES 20A BEC F390 48MHZ ESC 2~4S BLHELI FIRMWARE)
• Several different NTM Propdrive motors 28mm (PROPDRIVE V2 2836 1800KV BRUSHLESS OUTRUNNER MOTOR) – ALSO THE 2700KV AND 1300KV VERSIONS
• 4x copycat versions of the Turnigy Plush 30A ESC - LHI 30a Brushless Speed Controller Esc Multicopter Kk Quadrotor(pack of 4pcs) Link: http://a.co/0jZZZl1
• 10 Towerpro 9g servos: TowerPro SG90 Micro Servo 2pk - Link: http://a.co/hmmUQ1Q
• 10 TURNIGY™ TSS-10HM DS MICRO SERVO 15T 2.2KG / 0.12SEC / 10G

I immediately started running into issues where my flight controllers would reboot. The worst was in a newly built Bloody Wonder (used an NTM propdrive from previous order, new LHI copycat ESC) which would reboot anytime I moved both servos at once (dual aileron servo + elevator). I had the same problem with the HXT900 servos and the turnigyy TSS servos. It was more intermittent but still occurred with the towerpro servos. I also have a newly built Das Little Stick where I used an existing HXT blue wonder motor, new LHI ESC, new Frsky S6R, and it didn’t matter what servo type I picked.

I was going to throw out the LHI ESCs as junk when I decided to mix things up a bit to troubleshoot further. If I tried the small (undersized IMO) spider 18A ESC in the LittleStick, the problem was much more intermittent (but still kept happening). In desperation I pulled out my 1 and only stand-alone BEC (rated 3A @ 5V) and the problem went away on the LittleStick. I have more BECs coming to see if that becomes a better workaround. But here's the thing: Every single ESC I have says it is rated for 3A at 5V, which is exactly the same rating as the external BEC I used to 'fix' the LittleStick.

I’ve already stated my questions, but the biggest/burning issue for me is “Have I been lucky that I haven’t cratered by Walrus yet?” I ask this because I pile-drove my Bix3 shortly after takeoff on a windy day. I wrote it off as a stall as I entered the downwind leg after takeoff or “pilot error plus flying in too much wind”. But now I am seriously wondering if maybe I had a brownout and my radio rebooted at a critical phase of flight. It had 6 servos (2 flaps, 2 Ailerons, elevator, rudder) and I routinely flew it with the flaps setup as flapperons. At the time of the crash it was in that mode and I was flying fairly slow right after takeoff with about 30 degrees of flaps. It was using a Turnigy 30A Plush ESC, which (until now) I have thought were rock-solid. I am obviously concerned because my Walrus is basically a higher performance version of the Bix3 with the same HXT900 servos / plush esc combination in use.

Sorry for the longwinded post, but I wanted to get as much info out there as I could up front, in the hopes I could get some answers in time for next weekend’s flying sessions!

TIA for advice, suggestions!

 

[quorneng]

You seem to have ordered 'quad' ESCs. A quad does not normally use servos as in plane so its BEC only has to supply 5V for the radio system.
I suspect this is your problem.
Even 'no name' plane ESCs are likely to have 2 or 3 A BECs which should be able to handle 4 of your 9g analogue servos provided the linkage geometry do not cause them to stall their limit of travel.
A 6 servo set up really needs an ESC with a more powerful BEC or disconnect the ESC red wire and use a stand alone 10A UBEC.
Then of course if you have really free running linkages and control surface hinges you could use smaller servos that are less current hungry.
You can stall 6 x 3.7g servos all at once and it does not take more than 3A.

 

[gloobnib]

You seem to have ordered 'quad' ESCs. A quad does not normally use servos as in plane so its BEC only has to supply 5V for the radio system.
I suspect this is your problem.
Even 'no name' plane ESCs are likely to have 2 or 3 A BECs which should be able to handle 4 of your 9g analogue servos provided the linkage geometry do not cause them to stall their limit of travel.
A 6 servo set up really needs an ESC with a more powerful BEC or disconnect the ESC red wire and use a stand alone 10A UBEC.
Then of course if you have really free running linkages and control surface hinges you could use smaller servos that are less current hungry.
You can stall 6 x 3.7g servos all at once and it does not take more than 3A.

Thanks for the quick supply. I guess I was under the impression that "amps is amps" so the "quad" designation didn't scare me off. The bigger LHI servos LOOK exactly the same as the Turnigy Plush ESC and are even programmable using the Plush programming card. I take your point that using them in a quad shouldn't be an issue. So maybe they are using a bit of poetic license when they say the BEC is rated for 3A and they tested it quads with no issue so figured "passed QA"?

I will definitely look into External stand alone BECs for my bigger planes. Also sounds like I need to reevaluate "beefier is better" engineering that has served me well in my non-airborn engineering past. It is probably overkill to be using 9g servos in a LittleStick with its relatively tiny control surfaces. I guess I over corrected from my problems with gear stripping on my FT22.

Is there a quick and dirty way to quantifiably measure that these no-name ESCs are being advertised as having bigger BECs than is reality?

 

[TEAJR66]

You did not mention prop sizes for the differnt motors. Or weather your problem occurs on the bench or in flight. If it is while you are flying, i would suspect you are overpropped and browning out the esc for either amp draw or thermal protection. If you over amp the esc, bec size to the rx does not matter. the esc is greedy and wants its amps first. It only gives up what it has left over. All this can be checked with a watt meter. Too many amps drwan heats up the esc and some have a thermal cutoff that prevents buring them out.

I had a couple planes that were overpropped and every minute or so, i would lose control for a few seconds. Once the ESC cooled down and the battery sag recovered, the plane would spring back to life.

I'm not saying this "IS" your issue. Just one more route to explore in your troubleshooting.

 

[gloobnib]

You did not mention prop sizes for the differnt motors. Or weather your problem occurs on the bench or in flight. If it is while you are flying, i would suspect you are overpropped and browning out the esc for either amp draw or thermal protection. If you over amp the esc, bec size to the rx does not matter. the esc is greedy and wants its amps first. It only gives up what it has left over. All this can be checked with a watt meter. Too many amps drwan heats up the esc and some have a thermal cutoff that prevents buring them out.

I had a couple planes that were overpropped and every minute or so, i would lose control for a few seconds. Once the ESC cooled down and the battery sag recovered, the plane would spring back to life.

I'm not saying this "IS" your issue. Just one more route to explore in your troubleshooting.

Thanks, I should have specified. The scenarios I mentioned (with exception of the Bix3) were all on the bench with throttle at zero/minimum. Some of them were even without any loading on the servos at all (with the servos just sitting loose on the bench). Move the right stick from center to any of the corners (2 to 3 servos moving at once depending on the plane) and then BAM the RX reboots.

The Bix3 was definitely at heavy amp draw at WOT when it went in. I'm not 100% convinced the Bix3 was a brownout; it definitely could have been an abrupt stall induced by a change in wind; it winged over into the turn and dove straight down. However, if that is the case, it would be the first time I've ever experienced ANYTHING close to that behavior on that plane with maybe 50-60 flights logged. It normally stalls nice and gentle and pretty much self-recovers after losing maybe 10-15 feet of altitude. This went straight in from about 30 feet up. I was flying a clockwise box pattern, and the crash occured halfway into the turn from the crosswind to downwind leg and rolled right (into the turn). According to Windy.com, the wind was at about 11 knots with gusts up to 17 knots. I've flown in those conditions quite a bit and (prior to this) never had any issues at all maintaining control.

But to get back to the original question/point, if the ESC says its BEC is rated at 3A/5V (15 watts), I should be able to test them by putting some test loads (resistors?) in line and seeing at what point the RX reboots, correct? I should also be able to repeat the tests on some of the older ESCs and see if I get the same/similar results. If that is the case, I am "fortunate" to have both one of my original ESCs and one of my original RXs free now that the BIX is in a large plastic bag awaiting a Steve Austin type rebuilding.

Thoughts?

 

[Hai-Lee]

Just as a bit of general info the common 9 gram servos can draw 0.37 Amps at stall and so always go for enough current from the BEC to give yourself some headroom.

Additionally most BECs have overcurrent and over temperature protection. Some ESCs if they get too hot from the motor drive circuitry can actually shut down the BEC due to over temperature.

If possible try to avoid SBECs because their 5V regulators get hotter in operation than the UBEC regulators for the same current draw.

Finally ESCs that get or are run hot often seem to become more temperature sensitive and give BEC issues, so for large servo numbers a separate BEC is recommended and mount it away from the ESC which drives the motor.

Sadly the above is the resultant information from investigating many lawn darts, (Far too many:black_eyed.

Have fun!

 

[gloobnib]

Just as a bit of general info the common 9 gram servos can draw 0.37 Amps at stall and so always go for enough current from the BEC to give yourself some headroom.

Additionally most BECs have overcurrent and over temperature protection. Some ESCs if they get too hot from the motor drive circuitry can actually shut down the BEC due to over temperature.

If possible try to avoid SBECs because their 5V regulators get hotter in operation than the UBEC regulators for the same current draw.

Finally ESCs that get or are run hot often seem to become more temperature sensitive and give BEC issues, so for large servo numbers a separate BEC is recommended and mount it away from the ESC which drives the motor.

Sadly the above is the resultant information from investigating many lawn darts, (Far too many:black_eyed.

Have fun!

So after a little more research, it turns out that both the Turnigy Plush ESC (the type I've been fine with) and the LHI knock-off (the one that started this whole mess) are using LINEAR BECs, not a switching or Ubec. If I have correctly processed the reading I've done, the linear step-down process basically accomplishes the step-down by turning the excess voltage into heat. Further, the more the supply voltage is above 5v, the less efficient it will be AND the less less amps it can supply at 5v. So it may be able to supply 3a@5v when running on a 7.2v 2s lipo (30%ish step down), it will provide considerably less amps when run at 10.8v (53%ish step down). And it will generate more heat.

Still gonna test the plush and lhi side by side to see I'd the knockoff is considerably worse performing than the plush, but I also now know I need to put some external BECs in my bigger planes. I was planning on upgrading the RX in the walrus this weekend anyway so I could get independent channels for each flap and ailerons (hello crow!), so now is as good a time as any to upgrade the BEC.

 

[gloobnib]

Still gonna test the plush and lhi side by side to see I'd the knockoff is considerably worse performing than the plush, but I also now know I need to put some external BECs in my bigger planes. I was planning on upgrading the RX in the walrus this weekend anyway so I could get independent channels for each flap and ailerons (hello crow!), so now is as good a time as any to upgrade the BEC.

Adding more detail/resolution to the post in case anyone else is reading this thread after a search and finds themselves in the same boat I was in previously:

Basically the LHI (cheap knockoffs of Turnigy Plush) ESCs were total garbage. Over the following winter, I had three of them catch fire will on the bench and connected to a 2700 KV motor. The first time was while measuring AMP draw on the bench with a new plane (severely damaged the plane). 2nd time was just the motor/prop mounted to a board trying to figure out what happened. Third time it was a different make/model 2700KV motor with no prop. Basically these ESCs would self-destruct if you got them up to north of 20K RPMs (give or take). It doesn't surprise me that they were also probably badly spec'd in the BEC department too.

Lesson learned - don't skimp on electronics trying to save a buck.