Talk:Servomechanism

Power window example

Is it strictly true that a power window control cannot be considered a servo, as some power windows mechanisms automatically cut off when the windows (or in some cases just the drivers side window) reaches the top or bottom of the frame: as in, the button does not have to constantly be held for the window motor to operate. Surely without a system that could be considered a servo, the motor moving the window could simply carry on running endlessly? 86.9.242.214 (talk) 09:24, 15 April 2012 (UTC)

No, power window controls are not a servomechanism. The motors shut off as the result of limit switches, in a fixed position and not in a commanded position. On the other hand, power steering might be a servomechanism, in that there is a mechanism inside the p.s.unit that detects difference between where steering wheel is and the wheel position and opens/closes hydraulic valves accordingly. --71.214.209.23 (talk) 01:38, 16 April 2012 (UTC)

I beg to differ. There are windows (in fact, they may now be standard) that retract if they encounter something between the start position and expected stop position. This was instituted after small children had been asphyxiated when their heads were caught in the windows. (source: http://www.nhtsa.dot.gov/cars/problems/studies/NonTraffic‐NonCrash/Pages/TOC.htm) — Preceding unsigned comment added by 208.40.4.190 (talk) 14:04, 10 April 2013 (UTC)

vandalism

I wasn't login at the time so it appears has my IP. I have reverted both changes by 193.132.24.231 has they were vandalism the page by changing words

Louispq 16:34, 16 April 2007 (UTC)

Inconsistencies in Terminology

In this article, it mentions that "power steering" is not a true servomechanism, but a type of amplifier. However, in the Power Steering wiki page, it mentions prominently that "The power steering system in a vehicle is a type of servomechanism." Is one of these unequivocally right, or is there a blend? 141.212.48.226 (talk) 23:42, 16 April 2011 (UTC)

In my opinion "Servo" denotes a mechanical amplifier, e.g. power-assisted steering or power-assisted brakes, without necessarily including any feedback mechanism. Indeed, power-assisted brakes used to be called servo-assisted brakes. Perhaps there is a difference between "servo" and "servomechanism", with the latter including a feedback system. Biscuittin (talk) 19:29, 29 October 2011 (UTC)

Definitions from Chambers 20th century dictionary 1972: Servo "of a system in which the main mechanism is set in operation by a subsidiary mechanism and is able to develop a force greater than the force communicated to it". Servo-mechanism "a closed-cycle control system in which a small input power controls a larger output power in a strictly proportianate manner". There is no mention of feedback. I would say that a servomechanism may, or may not, include a feedback system. Biscuittin (talk) 19:40, 29 October 2011 (UTC)

I've done some more research and I think "closed-cycle" does imply the use of feedback, see Stepper motor and Open-loop controller. I think this confirms that "Servo" implies an open-loop system (without feedback) and "Servomechanism" implies a closed-loop system (with feedback). Please discuss. Biscuittin (talk) 19:55, 29 October 2011 (UTC)

There is no such distinction. They both use both open loop & closed loop control. If there is any distinction between "servo" and "servomechanism", it's about the system boundary: a "servo" is an actuator that forms part of an overall "servomechanism". Open loop is rare today, as electronics makes closed loop so easy, but open loop certainly was in use for simple force multipliers. You could argue that these go back to the 1860s and Lord Armstrong's early hydraulic engines. Andy Dingley (talk) 20:01, 29 October 2011 (UTC)

So the statement in the article "The term correctly applies only to systems where the feedback or error-correction signals help control mechanical position, speed or other parameters" is incorrect? Biscuittin (talk) 20:12, 29 October 2011 (UTC)

It's hard to give a statement that is both comprehensible to a reader, yet defensible for perfect accuracy against one of those obsessive editors with no real interest in the readership.

A servo or servomechanism is a device for reproducing an action with a different magnitude. It's usually increased, often an increased force, but there are also waldos and micro-manipulators that make smaller movements. To be a servo, there is some "echo" of one movement by another. A car's accelerator pedal isn't a servo - it controls the engine, but the engine doesn't follow the same movement. Probably the earliest real servomechanism I can think of is a ship's steering engine. This reproduces the force applied to the wheel by applying a larger force to the steering chains. This is open loop though: a servo mechanism to reproduce a force on a larger scale, but not a closed loop that tracks the tiller position (rather than force applied). Andy Dingley (talk) 23:41, 29 October 2011 (UTC)

I have added an "accuracy dispute" tag. We need references. Biscuittin (talk) 08:49, 30 October 2011 (UTC)

merging servo and RC servo

yes join them —The preceding unsigned comment was added by 86.142.85.145 (talk) 09:50, 5 May 2007 (UTC).

No don't join them. I'm not an expert, but RC which stands for radio control(led) is a special subset of servos. It seems to me that the author of the RC servo article isn't an expert either and is actually describing wired servos - although I could be mistaken. Ideally, rather than see the articles merged, I'd like to see both articles expanded with a small section on RC servos in the main servo article - linking to the lengthier RC article. I'm not expert enough to make these changes.

Also I think that they should remain separate because wired servos of many different types are used for many different functions in many different industries. The RC aspect of RC servos should be expanded upon in its own article.

Time to merge! I think we can have a section in servomechanisms about the RC portion. --72.241.176.124 23:47, 28 May 2007 (UTC)

Agree...

also merge Linear actuator, Servomotor, Servo drive, and any others... let's have one decent article rather than several shite ones.

[User: Macdaddy] 87.114.239.214 (talk) 16:57, 3 March 2013 (UTC)

• Oppose Rather than working to expand articles on distinct topics, why do you think that simply copy-merging them would improve anything? Andy Dingley (talk) 19:31, 3 March 2013 (UTC)

Proposed change to intro

This intro is too narrow. Servos are more broadly used. I propose something like the following:

A servomechanism, usually shortened to servo, is a device used to provide control of a desired operation through the use of feedback. Feedback is the term used to describe the use of a sample of the current output from the system which is "fed back" to an earlier part of the system (earlier in the forward signal flow) to bring the system's output closer to the desired result. An example of a servomechanism is a home furnace controlled by a thermostat.

INFO: The R/C servo is indeed, only one type of (position control) servo. Others are Furnace (temperature servo) Automotive power steering (force servo), Phase-Lock Loop (phase servo), AM radio AGC ( amplitude servo), FM radio AFC (frequency servo).

INTERESTING: The term servomechanism was originated in H.L. Hazen's 1934 paper Theory of Servomechanisms, Journal of the Frankiln Institute, from the words servant (or slave) and mechanism. This from Feedback Control System Analysis and Synthesis", D'Azzo & Houpis, 1966 McGraw-Hill.

I'll make these citations legal if they are used.

-- Steve -- 13:04, 20 May 2007 (UTC)

The section on control inputs for an RC servo, while interesting, seem to be describing a certain model, which isn't identified. If it's really a general standard, someone should say so, but otherwise, the info should be removed or, if it's a very common model, qualified. 207.179.248.39 23:43, 12 August 2007 (UTC)

That PWM format (1 to 2 ms pulse width, about 60Hz rep rate) is a de-facto standard in hobbyist R/C servos. It's not single-vendor. But it's not formally standardized, as far as I know. --John Nagle (talk) 04:31, 23 December 2007 (UTC)

I'll back up the above user, having tested a range of servo-transmitter pairs, I found 1.5 mS, 55Hz to be the norm across the range Jdedmond (talk) 15:17, 23 March 2008 (UTC)

I'd certainly agree on the change to the intro. This is important if the intent of the article is to eventually build a relationship with the article on control theory (leading to root locus, Bode plots, and all the rest). One minor complaint might be that the example of a furnace is really that of a regulator, not a servo. The difference being that a regulator has a generally fixed command point while a servo has a dynamic command point. Never the less, some example that shows the scope of servo theory goes far beyond position or velocity is important to put all of this in context. RC servos are a very narrow subset and their dominance in the world of servos is minor. --LifelongEngineer (talk) 02:48, 23 September 2009 (UTC)

Proposed merging Servo drive into here.

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

The result was do not merge. Wizard191 (talk) 14:57, 1 October 2009 (UTC)

---

I've proposed merging Servo drive into there. That article isn't very good, duplicates content here, and may just be a placeholder for an advertising link. --John Nagle (talk) 04:33, 23 December 2007 (UTC)

Good idea

Against merger: A servo drive is the part/amplifier that regulates power supplied to the servo motor. It is therefore not a servomechanism per se. Of course, for small RC servos the amplifier and the motor are boxed together. However, for large industrial motors the servo amplifier is a separate component because much heat must be dissipated (e.g. when breaking the motor). I apologize for not signing this comment; I work for a manufacturer of servo drives.

Against merger: The anonymous post above explains why, for clarity sometimes a certain amount of duplication is required. However, it is good to see that there are people monitoring - the proposed merger was suggested nearly two years ago! Bloodholds (talk) 02:25, 15 June 2009 (UTC)

Against merger: Against merging it into this article, anyway. If it gets merged with anything, electronic amplifier would probably be a better choice, though it is fairly bulky as it is. Steelerdon (talk) 14:00, 1 October 2009 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

PWM Operation

Would other users find it helpful to have a (oscilloscope-like) trace of the PWM'd waveform used to control the servo? The control of servos via PWM (pulse-width-modulation) is mentioned, and discussed briefly, but no diagrams are present. I certainly feel that it would help to improve the readability and ease of understanding. Jdedmond (talk) 15:05, 23 March 2008 (UTC)

PWM is a simple technique for digital to analog conversion and has no special relationship to the general theory or practice of servomechanisms. This is not an article to explain the operating electronics of RC toys. Servos in general can operate on PWM voltage command, pneumatic pressure, hydraulic pressure, analog voltage, analog current, or fiberoptic communications. Detailed techniques of implimeneting theory are not relevant to the development or understanding of the theory. If you want to get into details, lets explain PID control and state space control before PWM. --LifelongEngineer (talk) 02:56, 23 September 2009 (UTC)

PPM. — Preceding unsigned comment added by 212.95.244.2 (talk) 17:28, 21 July 2011 (UTC)

I agree with LifelongEngineer that PID control and state space control are important, so I added a paragraph mentioning them and linking to the PID control and the Control theory#Modern control theory articles that go into more detail.

I think there's room in this encyclopedia for describing every standard servo control interface protocol -- but perhaps in some other article that this article could link to. I agree with Jdedmond that a picture of an o'scope trace (or some other form of digital timing diagram) would be helpful to explain the "PPM/PWM" control interface used in hobby RC servos -- but maybe it would be better to explain such details at radio-controlled model. --DavidCary (talk) 04:24, 1 November 2011 (UTC)

Off topic/incorrect section

The following is off topic for an introduction:

On the other hand, the line between servo control and negative feedback is in general blurred in the context of cybernetics, where it is thought that a human being can actually serve as a part of a servo loop, and that quite a number of human behaviors can be explained as being instances of such control within the context of the organic world.

I don't think there is any line between servo control and negative feedback. Negative feedback is used to make the servo control mechanism "stable." Further, the argument about the human role in a feedback system isn't informative here: is a car an open loop or closed loop mechanical system? Does the presence of a human make it a closed loop system? The answers aren't directly related to what a servomechanism is.

I recommend deleting or moving the paragraph.

Serrano24 (talk) 21:49, 23 June 2008 (UTC)

Yes, merge them. Clarify and point to RC servo and PWM. takeitupalevel March 2009 —Preceding undated comment added 20:07, 5 March 2009 (UTC).

Redundant Images

Do we really need two pictures of exploded RC servos? They are both good pictures, but I suggest one should be removed. I vote for removing the first one and leaving the second one (leave the one that is numbered and labeled.) (Steelerdon (talk) 21:48, 3 February 2010 (UTC))

Agreed. I've removed the non-numbered image. Wizard191 (talk) 13:51, 4 February 2010 (UTC)

Needs a section in more detail and more advanced concepts

Ouch, an enormous subject with an enormous wedge of some very complex and subtle mathematics, - trouble is I don't see much of it in this article. What is here looks pretty good, it does describe the basic concept quite well and there is a little history and so on - but the article needs to be at least twice as long.
On history, robotics like many subjects has a somewhat complex disputed history, emphasizing one set of authors and not mentioning another might be a little close to POV. Desperately needs to mention Norbert Wiener and could do with a link to Cybernetics which is pretty much a different name for the same subject (actually servomechanisms are a large sub-area within cybernetics). Then there is Marvin Minsky et al and AI and Perceptrons (neural networks) again sometimes pretty close relatives to servomechanisms. I'm sure there are a few others but this isn't quite in my main areas of expertise, someone with a background in electronic engineering or robotics or best of all cybernetics might be a good start. Hope that helps. Lucien86 (talk) 11:14, 12 July 2010 (UTC)

Bad example

For example, an automotive power window control is not a servomechanism, as there is no automatic feedback that controls position

This is not true as the as the window control has feedback that controls position albeit only at both ends of the range. BerlinSight (talk) 22:14, 8 December 2011 (UTC)

This Article Needs Major Revision

Hello editors, The article title refers to a broad area in science and engineering. It as a long history going back to ancient Greece and China. It is both funny and interesting that modern "hobby servos" appear fairly prominently in the article. To an ancient Greek engineer,the hobby servo would be magical indeed. And they are, in many ways, the apotheosis of modern technology.

Here are some early references to help us find out when and where the term servomechanism originated. Merriam-Webster says 1926 but I think a bit earlier.

http://books.google.com/books?id=VhtQAAAAYAAJ&pg=PA813&dq=servomechanism&hl=en&sa=X&ei=7QVUU5KIC4ahyASfp4DwBg&ved=0CC4Q6AEwAA#v=onepage&q=servomechanism&f=false

http://www.merriam-webster.com/dictionary/servomechanism

Codwiki (talk) 17:51, 20 April 2014 (UTC)

Webster says 1889 http://www.merriam-webster.com/dictionary/servomotor

Codwiki (talk) 18:09, 20 April 2014 (UTC)

It predates that. Look at sources for the history of engineering in the French navy. The mechanism is earlier than that (and English, AFAIK), but the word as we know it today seems to be French. Andy Dingley (talk) 21:15, 20 April 2014 (UTC)

Assessment comment

The comment(s) below were originally left at Talk:Servomechanism/Comments, and are posted here for posterity. Following several discussions in past years, these subpages are now deprecated. The comments may be irrelevant or outdated; if so, please feel free to remove this section.

I think Servos, have amplification of the input from the controller in the forward path.

Last edited at 17:05, 16 June 2008 (UTC). Substituted at 05:49, 30 April 2016 (UTC)

Damping

Servo mechanisms need some form of damping to avoid over-reacting and oscillation. This is not mentioned in the article, but ti seems like it should be. — Preceding unsigned comment added by 194.100.8.226 (talk) 19:56, 22 August 2017 (UTC)

Epic

This is so epic oh my god I’m in heaven reading this all the time. ItalicConfucius senses are tingling 😂😂😂 2600:8806:501:FC00:354C:DE66:EC74:6CF7 (talk) 05:03, 8 May 2023 (UTC)