Servos Chattering.

[David Davis]

I am entered in a light-hearted competition called "La Coupe Des Barons" on 12th June, Covid regulations permitting. The event is for a three channel French trainer called a Baron and rather than describe the event itself in detail, I've incorporated a video of the 2017 event here: https://www.youtube.com/watch?v=IwWg0Ad5cKs Suffice it only to say that two parts of the competition require competitors to fly accurately at low level, which is why a high percentage of the models are wrecked on the day. All good fun! In order to have a spare I have built two Barons, one in Russian colours which I call Boris, the other in RAF markings, is called Bertie! Pictures below.

 

I have been practising with Bertie until recently but one day when I switched on the servos started to chatter. I.e they were oscillating without any signal from the transmitter. Bertie is equipped with two Corona CS 939MG servos actuating the rudder and elevator and originally, a nondescript analogue servo taken from some long-deceased foamie ARTF for the throttle. I thought that this servo might have been worn out so I replaced it with an old Futaba S148 servo. No improvement. Since then I've changed the receiver, changed the model memory on the transmitter, a Spektrum DX9 and by-passed the switch by plugging the battery directly into the receiver. I'm still left with the same problem. The battery, a 2000 mAh NiMh, is showing 6.1 volts on my NiCad checker.

 

Before anyone says that the problem is caused by having both digital and analogue servos in the same model, Boris, the Russian Baron, has two Hitec HS5085 servos actuating the rudder and elevator and an unlabelled, presumably analogue mini servo operating the throttle but I have no problem with oscillating servos on this model.

 

Could it be interference from the neon light in my workshop? If not, I'm stumped!

 

The advice of the cognoscenti is urgently sought.

 

 

[Cuban8]

When you say "no signal from Tx", I presume the Tx is on but just no stick inputs?

If the Tx is indeed on, does the chattering stop when you give all the Tx sticks a brief 'flick'?

Does the problem occur with just one servo plugged in?

Had a very similar problem to what you're describing a while back.

[Dickw]

Have you checked the Frame rate? The Corona are analogue servos so should be operated at 22ms not 11ms.

The only other suggestion I have is to try disconnecting all the servos and then plugging them back in one at a time to see if that shows up any problem with a particular servo.

 

Dick

[ken anderson.]

could it be something as simple as the TX too close to the RX...are you using 2,4 or 35 mhz gear?

 

ke anderson...ne..1...simple dept.

[Nigel R]

ken

 

3 hours ago, David Davis said:

changed the model memory on the transmitter, a Spektrum DX9

 

[Nigel R]

Cynical hat on.

 

Those are £5, sub micro - 12g - servos. 

 

Used in an IC model - subjected to some vibration, for some time.

 

Personally, I would not waste much time digging.

 

Perhaps replace them with something a tickle larger and from a more reliable brand.

 

Incidentally, HK says they are analogue.

[Ron Gray]

CS - analogue, DS - digital.

[Denis Watkins]

A Voltage reading is never the full story, but a partial test of power

Double, double-check the nimh by substitution

Chattering is also an indication of a low power supply.

[John Stainforth]

Have you disconnected and reconnected all the electrical connections, including extension leads, to eliminate the possibility that this is caused by a poor connection somewhere? 

[David Davis]

Thank you gentlemen.

 

I disconnected everything then put the battery on charge. I am not clever enough to understand the graphs and figures which my charger displays at the end of a charge but when I put everything back together everything functioned normally. So it was either a poor connection or the rx battery. I may well treat the model to a new battery for the competition.

[Steve Hargreaves - Moderator]

I think I'd side with Nigel on this one....& change the servos...

 

I had a similar issue with some HS81s on the ailerons of my lovely Pitts Special....they started chattering & after a bit of messing about with connections etc they stopped even though I never actually found "the problem". Sadly it came back whilst the model was in flight & I lost the model....still annoys me to this day...

[Jon H]

Any servo that behaves suspiciously gets removed and a loose knot tied in its lead. I then sling it in the box for faffing with at a later date. If i forget its misbehaving, the knot in the lead tells me something is wrong with it and i need to investigate before using it again. If it then donates its innards to another servo it gets another knot to tell me it has two problems and not one! 

[Andy Stephenson]

1 hour ago, Steve Hargreaves - Moderator said:

I think I'd side with Nigel on this one....& change the servos...

 

I had a similar issue with some HS81s on the ailerons of my lovely Pitts Special....they started chattering & after a bit of messing about with connections etc they stopped even though I never actually found "the problem". Sadly it came back whilst the model was in flight & I lost the model....still annoys me to this day...

Steve, it seems like your problem was with the connector not the servo itself. If the connector had some oxidation or dirt creating a high resistance in the power circuit it could quite easily seem like it was not getting enough voltage to operate properly. When the motor started to turn it would pull more current causing the voltage to drop and the servo to "turn off" at which point it would stop pulling current allowing the voltage drop at the connector to disappear and turn back on. You can see from this it is a feedback loop causing an oscillation which causes the servo to chatter. 

Edited May 13, 2021 by Andy Stephenson

[Nigel R]

6 hours ago, David Davis said:

I disconnected everything then put the battery on charge. I am not clever enough to understand the graphs and figures which my charger displays at the end of a charge but when I put everything back together everything functioned normally. So it was either a poor connection or the rx battery. I may well treat the model to a new battery for the competition

 

Right now you only have a hypothesis that it was a poor connection or rx battery.

 

Where could the poor connection be? You've swapped RX and removed the switch, problem remained, you're left with the battery or servos being root cause; the common thread across all things being battery, and with it arriving on both servos simultaneously. Which is highly suggestive of battery but not conclusive.

 

If you're attributing fault to state of charge, then prove it, discharge the battery, show the fault again, charge the battery, show the fault goes. Otherwise it could easily be within the battery wiring itself; that isn't fixed by charging.

 

For my money, with no further information and the absence of the above test showing the cause - new battery and two new servos, put the old kit aside marked "to be debugged later" as Jon suggests.

 

In the end we either track down these problem and conclusively fix them, or live with the fact they are liable to return and gift you a bin bag of former model.

 

As a PS, would suggest learning what your charger is telling you.

[Piers Bowlan]

I know you said that you swapped out the 'nondescript analog throttle servo' with a Futaba S148 with no improvement but did you do so with the Corona CS939MG servos too? May not have proved anything but it might have been worth a try. A worn feedback pot or even a missing gear tooth can cause a servo to chatter but when all three are doing so something else is going on. Pure speculation on my part but if one servo was drawing heavy current for some reason (partial short?) then the receiver battery voltage would crash which conceivably might cause the symptoms you discribe. So even if the battery is new and shows fully charged off the charger it may still be a servo or a wiring problem - or the battery! So I am with Nigel, buy a new battery and some new servos so you can relax and enjoy the flying. (After you have a replacement battery i'd strip the shrink wrap off the old one and check you don't have any of the dreaded black wire corrosion)

 

By the way, nice pictures of La Coupe Des Barons, David.?

[David Davis]

Thank you gentlemen. I fully intend to fit a new battery to the model before flying it again.

 

When I started flying r/c models in the late 1980s we nearly all used 500 mAh NiCads as receiver batteries. A year or two later and 600 mAh NiCads were all the rage. Strange how we all seem to need at least 1600 mAh NiMhs today.

[Don Fry]

Or as the instructions in a Flair quarter scale Fokker DR 1 says,( 120 size aircraft), fit a 1000 mah battery for safety. 

[Martin McIntosh]

I concur with DD above as a 500 Deac was all that was available at the time and could be stretched to five flights with four servos. I do not see a battery make stated but if it is of the super high capacity type like 2600 then it is only suitable for a torch or maybe a Tx. Having eliminated a poor connection or switch then it really must be the battery. They just cannot supply enough current for even a few analogue servos. If you still use NiMh`s then Eneloops are by far the best, and always use a 6V pack, not 4.8 providing that the servos can handle it. I speak from bad experience.

As a side note to this, many models will have been laid up over the past year or so and if the packs have been left connected (like mine!) there could well be black wire corrosion. I always cycle packs that have not been used for a while.

[FlyinBrian]

On 13/05/2021 at 10:55, Jon - Laser Engines said:

Any servo that behaves suspiciously gets removed and a loose knot tied in its lead. I then sling it in the box for faffing with at a later date. If i forget its misbehaving, the knot in the lead tells me something is wrong with it and i need to investigate before using it again. If it then donates its innards to another servo it gets another knot to tell me it has two problems and not one! 

In similar fashion I scratch a big X on the side of any suspect servo

[Don Fry]

I cut the plug off, used for fitting trials in new airframes.

[RottenRow]

On 12/05/2021 at 11:25, David Davis said:

The battery, a 2000 mAh NiMh, is showing 6.1 volts on my NiCad checker.

 

It has been mentioned before but is worth repeating. If you use one of the small black 'Capacity Controller' type battery checkers to test a NiCd or NiMH it will nearly always tell you that it is well charged as it doesn't apply any load when measuring. Something like the HiTec PowerMate II, which applies a load (500mA or more) is much better, and can be switched between 4.8V and 6V for use with different packs. Alternatively, apply your own load using a suitable resistor. The PowerMate II has a 7.5 ohm 7 watt resistor inside it.

 

Actually some of those small black checkers are pretty useless for measuring LiPos as well. I have compared several within our club and found some of the unbranded ones to be rather optimistic in their readings. I think there are a lot of poor copies around, which isn't good if you are relying on one to check that you are not over-discharging your batteries.

 

As regards chargers, the modern multi-purpose types can switch off the charge early due to them misinterpreting the peak in the battery voltage that indicates that the battery is fully charged. For NiMH batteries I still use a normal constant current charger, set to 1/10 of the capacity of the battery, for 14 hours. Even if the battery is only partially discharged, this will not over charge it.

 

Brian.

[Martin McIntosh]

If you must use a capacity controller then you must scroll through it to get an accurate reading of the open circuit voltage. The first reading means nothing. I only use one to get an indication of the state of charge of lipos.

 

Anything further on the o/p`s type of pack?

[David Davis]

Thank you for all of the advice gentlemen. The electrical side of the hobby is not my major interest but I've only had one crash in thirty-three years of flying r/c models which was caused by the rx battery running out of electricity. That was when I was very much a novice in the days of 500mAh receiver NiCads.

 

So the battery is a 5 cell 2000 mAh Sanyo Eneloop and the charger is a Hobby King X120. The tiny handbook only tells me how to charge different types of battery and is otherwise not very helpful. It refers to the IMOS system on the first page without defining what IMOS is. However, I have long since stopped expecting to get much help from the literature supplied with electronic components. Nobody seems to be able to write in clear, concise, jargon-free English these days. It seems that many people get their information from You Tube.

 

However, I have taken Martin Mackintosh's advice and cycled the battery. The usual coloured lines appeared on the charger's screen. On the right hand side of the screen are four coloured squares: red, green,orange and grey. The red, green and orange squares correspond with the colours of the lines on the screen. Having set a charge rate of 2.1 amps and a discharge rate of 1 amp, at the end of the cycle the following information appeared in the coloured squares.

 

Red 8.8 V

Green 1.69 A

Orange 1934 mAh

Grey 26C

 

I suspect that the 26C refers to the temperature of the battery but I really don't know what the other figures refer to. I have somehow acquired three different little battery checkers over the years. They all show that the battery is charged to 6.9 volts.

 

I will still fit a new battery for the competition.

 

[Nigel R]

On 13/05/2021 at 12:01, Nigel R said:

discharge the battery, show the fault again, charge the battery, show the fault goes.

Have you done this yet DD?

[David Davis]

9 minutes ago, Nigel R said:

Have you done this yet DD?

Isn't that what cycling does?

 

I think that I had simply let the battery voltage fall too low without realising it.