Is a BEC really required @ 6V?

[skip asay]

JMW wrote - "However, a

JMW wrote - "However, a servo compares the requested position with it's actual position and moves it's arm accordingly. Now, how long that takes depends I would say. It depends upon the desired and actual position, the servo's max. speed of movement, possibly the load, and whatever else there might be. And that time can definitely be longer than the 20 milliseconds to the next servo-pulse.
So if you now have several servos going towards their new positions they will draw current at the same time, thus loading the BEC."

That sounds good but......
Without a signal, a servo won't go. Period. Unless you take into account the motor's armature continuing to rotate. But that's just kinetic energy and not powered by the battery energy.

Skip Asay

[Guest]

Just be careful if you're

Just be careful if you're using digital servos. They use the 20 msec pulse as a guide only, and will continue to draw power inbetween the pulses, as the microprocessor works with a tighter time frame. This is the main reason digital servos have better holding torque than their analogue counterparts.

Andy

[jmw]

Skip,
good point. Thing is, I

Skip,
good point. Thing is, I do not know what a servo considers to be 'no input signal'.
I would say that the servo pulse of 1-2 msecs arriving every 20 msecs or so is the input signal and as long as that arrives I, as a servo, would keep or goto my position.

If however a servo considers just the 1-2 msec pulse the input then you're right and in between these pulses the servo would be 'inactive'.
In my view as a designer this would produce quite inefficient servos for they would be active only about 10% of the time leaving 90% of the time to drift away as a result of outside forces.

I had a look at some datasheets of servo ICs I have, yet that reveiled nothing.
So, I am gonna have a look. I'll go and try measuring later today at which moments a servo draws current. If that is only duing the pulse or the entire time. I'll keep you posted.

Jan Martin Wagenaar

[jmw]

Skip,
this is good fun!

Skip,
this is good fun! I guess we're both right!
In fact, Andy's remark regarding the digital servos seems correct to me too, thing is I have no experience with them so I can't tell.

The following remarks concerns an analog servo.

A servo apparently waits until the end of the 1-2msecs pulse and then adjusts it output if need be. So in rest there is no current draw to speak of.
When adjusting it's arm, the output current draw starts at the end of the pulse, and returns to zero almost always before the start of it's next pulse. It can however take most of that time!

I did some measurements using a Multiplex Cockpit SX transmitter, RX-7-Synth receiver and a couple of servos I had at hand, of which at any time only one servo was connected. The pictures are made using a Multiplex Mini HD servo.

The pictures are oscilloscope pictures showing on channel 1 (the upper channel) the input pulse to the servo. A 50Hz (20ms) pulse stream as it should be. The bottom channel shows the current in the ground-lead to both receiver and servo. I know, a bit of cheating but this was easiest for me and the receiver current stays constant anyway.
(The current measured is the voltage over a resistor of appr. 2 Ohms)

The first picture shows the servo in idle position; the current stays zero (well, almost zero yet for todays subject let's say zero)
Please excuse the less than DVD quality of the picture, I hope you can still see what's happening:



Now, when I manually try to move the servo arm, the servo gets to work.
A light touch reveals the current to start at the end of it's input pulse.
As I measure the current in the ground lead to the receiver, the input pulse also jumps up during the curent draw, as the ground level is raised.
The last pulse on the picture required no adjustment apparently][/url]

Applying more force on the servo arm shows the current-draw to take longer, about 5msecs or so][/url]

Applying quite a bit of force to the servo-arm reveals that current is drawn almost until the start of the servo's next input pulse][/url]

So, I guess that a servo indeed only draws current when an input pulse is present, yet that the duration of that current draw can take long enough to coincide with other servos also requiring current.
Which still means that the battery/BEC should be able to supply enough current for all servos connected, whether analog or digital.

Cheers,
Jan Martin Wagenaar

[mermaid]

http://www.subcommittee.com/forum/icon_question.gif So far, only servos

So far, only servos have been discussed. There is the ESC and ballast system to consider as well. Is the ESC control circuitry powered from the Rx or does it only require the control pulses with the control power coming directly from the main battery? I plan to control my gas ballast system via an SES-3 e-switch. Does anyone know how much current it draws? Although it's operation would be intermittent, it could come at a time when most or all of the other actuators are active. This could be the straw that broke the camel's back if the BEC is near it's current limit. What about APC and ADC circuits?

[chuck chesney]

Mermaid, this is getting to

Mermaid, this is getting to be one of those situations where folks are laboring mightily to figure out how many angels can dance on the head of a pin. To save yourself lots of needless concern, I suggest that you take Skip's word as gospel. He INVENTED the SES, founded SubTech, and has more real world experience in the model submarine world than just about anybody. There are many people who post on this forum who have only a tiny fraction of Mr. Asay's real world experience, and since your goal is to have a working submarine model, go with what works from the voice of experience.

[himszy]

In fact, Andy's remark regarding

In fact, Andy's remark regarding the digital servos seems correct to me too, thing is I have no experience with them so I can't tell.

I've heard that some 'hardcore' roboteers actually change the 1ms - 2ms pulse train to one of much shorter duration to give improvements in torque.

JMW, were you the one that designed the sub controller for a school/college/uni project?

Michael

[KevinMC]

Beaten to the punch!

JMW- fantastic

Beaten to the punch!

JMW- fantastic job grabbing those scope plots. I brought a couple of servos in with me to work today so I could gather the exact same data...

I've based my measurements on two different servos- a Futaba 3003 (standard analog) servo, and just for kicks-and-giggles I've compared it to a HiTec 5645 (digital, high torque) servo.

My static tests with the 3003 produced the same results as Jan's, but I also took a grab of what happens when the servo's command position is changed. (What happens when you move the stick from one position to another.)



I've stepped down the frame rate of the scope to capture the complete waveform. The command input is represented by the upper trace (in yellow) and the current draw by the servo is represented by the lower trace (in blue). The initial peak in current just hits hits 500mA, and drops to less than 300mA within a couple of frames. For a small sub running two or three servos, I agree with Skip that a 1A BEC will suffice.

To answer the question of digital servos, these guys are completely different animals. The HiTec I benchmarked contains a micro which remembers the command input and will hold to that command at a rate of roughly 600 times per second. It's this higher control rate that affords a digital servo its speed, tremendous holding power, and funny whine when in motion. Even under a light load, their current draw is essentially constant across the entire frame. Here's what a minimal static load looks like][/url]

And the same servo with a very heavy load.



I'd definitely want more than a 1A BEC with a rack full of these guys. But then again, I can't make a reasonable argument for why I'd need to put a digital servo in any of my subs, so I guess it's a moo point... (You know, a cow's opinion... )

[jmw]

Michael,

no, school was a long

Michael,

no, school was a long time ago for me....
The basic idea will be similar though.

As my old profession of electronics is still one of my hobbies, I decided to design my own controller with pressure sensor, pitch and roll sensor, outputs for lights and stuff and (eeehhmm, being a technician I find it hard to stop when it is actually quite enough) IR-detectors for additional remote-control adjustments, memory-card to log several parameters, and an internal communications bus for future modules.

It'll probably be something which already exists with the added bonus of sub-veteran-experience, I just like doing this.

Jan Martin Wagenaar

[Guest]

But then again, I can't

But then again, I can't make a reasonable argument for why I'd need to put a digital servo in any of my subs.

Some people recommend them for pitch control, as they exhibit less deadband. Would be especially useful in a very fast submarine.

Andy

[KevinMC]

Careful there Andy, don't perpetuate

Careful there Andy, don't perpetuate the "because it's digital it's better" mentality...

Some people recommend them for pitch control, as they exhibit less deadband. Would be especially useful in a very fast submarine.

I've not seen many subs with linkages that were slop-free enough that the difference in servo deadband would be noticed. Further, for tight pitch control you will benefit far more from the fast acceleration afforded by a good quality coreless analog servo than what you'll get from a "comparably priced" cored digital servo.

I agree that if making a sub for speed the additional performance of a digital servo may be noticed, but given that I was talking about the value of digital servos for my subs (and my OSCAR just doesn't travel that fast) I stand by my earlier position of "no justification"...

[Guest]

The thing that annoys me

The thing that annoys me is the premium manufacturers place on digital servos. They should be less expensive, as they surely have a lower component count than the older servos that used the dedicated servo chips.

Coreless servos are best for torque and response, agreed. But don't most of the digital servos feature coreless motors?

Here's a good site for some really fancy servos-

http://www.openservo.com/

Andy

[KevinMC]

Agreed- I've long since grown

Agreed- I've long since grown tired of paying for hype. Yesteryear it was premium for coreless and micro. Now it's premium for digital.

The only digital servos I have any first-hand experience with are HiTec. Their digital servos are available in both cored (56XX) and coreless (59XX) models. I've found them to be great servos, but they have their place.

Fantastic link on the DIY-digitals!

[himszy]

This came up on USENET

This came up on USENET and thought it may be of some relevence:

As a follow up on a previous post (http://groups.google.com/group/
comp.robotics.misc/browse_thread/thread/d22215745d7a0e80/
ab5b25aaa4ce18cc) I wanted to mention that today I implemented a
feature where I can adjust PWM pulse frequencies on the fly. Whereas
before I had found that my servos (Hitec HS-81MGs) would not be able
to stabilize on a position at frequencies above about 80Hz, today I've
found something slightly different. That limit only holds true when
the servo's output shaft is a long ways away from the pwm signal's
pulsewidth's matching angle. Thus when you first turn the servo on (as
I was doing in my first experiment) - at too high of a PWM frequency
they will indeed oscillate pretty terribly and sometimes never
stabilize (well, at least not in a minute or so - I lost patience
after that). But if you first let it stabilize at an angle (by using a
lower PWM frequency), and then increase the PWM frequency, it will
hold the angle beautifully. Once it has reached a steady state for a
constant PWM pulsewidth at a high frequency, if you gradually adjust
the PWM pulsewidth it will track the signal beautifully. Thus I'm
going to be changing my code so that on power up the PWM signal is at
50Hz, but as soon as the servos have stabilized I'll increase the
frequency.

I've found that reference tracking is much, much better at higher
frequencies, and disturbance rejection is also much, much better at
higher frequencies. Whereas before at 50Hz when I would torque an
output shaft I could feel it vibrating, now the vibration is pretty
much gone. The shaft is just solid. Similarly, when I'd move the foot
of a leg on my robot in a straight line I could see it jerk around a
little bit, but now, it moves almost perfectly straight. Power
consumption is definitely higher, but oh well. Glad I'm using li-
polys I'm going to try to get some quantitative results with
regards to power consumption tomorrow. I wish I knew more about how
this affects the life of the servo though - as that is my only worry.
It'd be a tad spendy to replace all 18 of my servos...

I went up to 100Hz and everything just kept getting better and better.
The way my PWM pulse generation is set up I can't go much higher than
that without making some fairly major changes to my code, as I
originally designed it to only drive the servos at 50Hz.

Hope somebody finds this information useful.

-Mike

Michael