Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length.

[Geoff]

Much as the title suggests. When I join several servos or servo sized sail winches together using a "Y" lead they don't seem to operate exactly the same insofar as the movement is not quite identical and/or simultaneous.


In one of my models I use three small servos linked together and when operating they all move but a bit like a daisy chain - there is a visible fraction of a difference in movement and I don't understand why as I thought they should all move together.


I use servo sized sail winches to turn gun turrets by winching them round but if I use two in a "Y" lead they get out of synch. This may be due to small differences in the dead band and to be fair it may be the quality of the winches but the specs show a very small dead band.


Does anyone have a solution to this?


I have a number of ideas:


- the length of the leads is different so there is a slight delay in signal getting through and/or signal strength (max length difference is about 20 inches). I don't see how this can make much difference.
- the dead bands are different causing differential feedback
- One of the winches "grabs" the signal first and then it gets relayed to the other
- more expensive sail winches needed - I used King Max (SW4805-2PA)
- is there really a meaningful difference between digital and analog (do they both use a feedback potentiometer?)


Tha nks


Geoff

[DJW]

Hi Geoff


Assuming the load on each servo is same / very similar, my inclination would be to consider three identical digital servos rather than analog (which can wear / age differently). It just feels to me that as the RX is outputting its pulses, you need each servo to react in the same way and digital is more likely to reliably do that.


If they're older analog servos they may be usable on their own, but no longer matched when put alongside each other, basically what you're seeing.


Best regards
David.

[malcolmfrary]

Manufacturing tolerances, differing wear and tear, after connecting through Y leads, they might not be getting the same voltage.
Signal timing as such won't be affected, the signal travels at the speed of light, and it takes a lot of slowing down to change the pulse length with any significance.  A long led "might introduce some extra capacitance in the signal wire that "might" have the effect of delaying the start and/or finish times of the control pulses by a microsecond or so at the receiving end.
Digital is a marketing word that gets slung around quite indiscriminately by sales and marketing, who generally haven't a clue what it means. 
I have 27MHz radios with "digital" on the label, alongside "AM".  That just means that it contains some digital logic circuits, even if they are not performing any actual digital function.  The PWM signal that is involved somewhere along the line with just about every radio control system is as analog as you get, even if it is switching on and off like something digital.  The pulse length is analog, honest.
Just how digital are digital servos?  I don't know, but they respond to the exact same pulse length signal (length is the analog bit) as everything else that plugs into a radio.
An analog servo converts that pulse length into a value which is compared to that being generated by the position sensor, and seeks to make them equal by driving the DC motor using another bit of PWM.  The nearer equal, the slower the motor, this to prevent overshoot, because motors have inertia.
A digital servo must respond to the same, analog coded, signal.  It "might" convert the pulse length into a number to be compared with a number derived from the position sensor, or do that job just like an analog one.  When driving its motor, it "might" fire the motor at full speed until it gets where it is going, and use a bit of brute force braking to stop it dead.  A full digital one would do both, I suspect that "marketing" digital ones just do the latter.  Advertisers are going to need to find a new word when really digital stuff arrives.

[JimG]

Here's a good simple explanation of the difference between analogue and digital servos.https://www.radiocontrolinfo.com/the-difference-between-analog-and-digital-rc-servos/
Jim

[malcolmfrary]

Here's a good simple explanation of the difference between analogue and digital servos.https://www.radiocontrolinfo.com/the-difference-between-analog-and-digital-rc-servos/
Jim

From the explanation given, all that a digital servo does is use numbers to define position rather than compare a pair of voltages, and as a result of the speed at which digital circuits run, is able to send an output to the motor based on a higher frequency.  It can correct its arm position faster because it looks at shorter intervals.  It whines when doing so because the frequency it runs at is within the band that humans can hear.  The analog ones will hum under the same conditions, but the air masses moved are so small that we don't hear them.  Search for threads about motors whining when driven by almost any modern ESC.  In the loudspeaker world whoofers are bigger than tweeters because at the low frequencies that they operate in they have to move a lot more air.  A digital servo is like a tweeter - it can convert battery power into audible sound.  An analog one does the same, but outside human hearing.
If a digital servo is whining away to itself when unloaded must be trying to correct to a position that it cannot find.  If it had accurately found it, it would be silent.  Analog servos have a response curve which means that the smaller the correction required, the less power if offered to the motor.  It gets near where it is going and slows down to stop at the right place.  A digital one, OTOH, seems to drive to its point at full speed, overshoot, and before the human notices, correct at full speed in reverse.  Probably overshooting that way as well, depending on any gearing backlash.
In practical model boat terms, the boater is unlikely to notice the difference between a response time of 1/50th of a second and 1/500th of a second.  What will be noticed in a situation where operating session time is important, is the reduced sailing time on one battery.


If similarly loaded servos are showing different responses to an in-boat setup, the answer is to invest in a servo tester that allows more than one servo to be plugged in simultaneously.  If the servos behave differently, the problem is to servo or the loading.  If they behave the same, its the wiring.  It becomes possible to test further by using the servo tester and the Y lead or leads.

[JimG]

In practical model boat terms, the boater is unlikely to notice the difference between a response time of 1/50th of a second and 1/500th of a second.  What will be noticed in a situation where operating session time is important, is the reduced sailing time on one battery.


If similarly loaded servos are showing different responses to an in-boat setup, the answer is to invest in a servo tester that allows more than one servo to be plugged in simultaneously.  If the servos behave differently, the problem is to servo or the loading.  If they behave the same, its the wiring.  It becomes possible to test further by using the servo tester and the Y lead or leads.

What the main advantage of a digital servo has over analogue is the holding power, as it uses higher power near the endpoint is is less likely to be moved from the endpoint by a hydrodynamic load. E.G. a large rudder in a tight turn will have a load trying to reduce the deflection, digital servos will resist this load much better. In aerobatic model aircraft you often find that smaller control movements are needed with digital servos due to them holding position better.When using servos in parallel using y leads it is not uncommon to find slight differences between them even the same make and type of servo. Large model aircraft which need 2 or more servos per control surface use some form of balance box which allows for each servo to be adjusted to give equal movements. Many modern digital servos can be individually programmed so that they all give the same movement.
Jim

[roycv]

Thanks very much gents for the explanation, I did wonder what it was all about.

 I often use arm winches in my yachts and try and position the arm at maximum pull in of the sheets to to be almost across the centre of the servo.  I do it because it is less likely to put a strain on the gear train. 

I also use an electronic servo extender to increase the throw of the servo allowing for the stop points on the servo.
Thanks agian
Roy 

[Geoff]

Thank you for an excellent explanation which appreciated. Much to think about.


Cheers


Geoff