Science desk
< June 5	<< May \| June \| Jul >>	June 7 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 6[edit]

brushed DC motor control[edit]

if you connect a small DC motor to a speaker and twist the shaft, there's a buzzing sound. could this be used to infer the number of revolutions (i.e., the angular position in multiples of 360° or 120° or 90°) a running motor has made, i.e. like a fork sensor plus disk with holes-type setup, but without it? Asmrulz (talk) 02:15, 6 June 2016 (UTC)[reply]
PS something with comparators perhaps? is this reliable?
PPS is there a single word for the disk with holes that interrupts the light in a fork sensor? Asmrulz (talk) 02:16, 6 June 2016 (UTC)[reply]
PPPS I don't have an oscilloscope (and I'm afraid to use my soundcard because of inductive kickback), but my understanding is that the transient processes inside a motor manifest as a (dirty) sinusoid on top of the DC voltage, i.e., ripple. The amplitude of the ripple gets smaller the higher the mechanical load on the shaft, but the phase and frequency correspond (in one way or antoher) to the motion of the rotor, correct? Asmrulz (talk) 02:41, 6 June 2016 (UTC)[reply]

The classic brushed DC motor works fine in reverse as a generator. A cheap one will generate a pulsating DC signal, but the pulses are far from sharp. You'd probably want to try and clean them up a bit with a Schmitt trigger or maybe even a phase-locked loop before trying to feed them to a digital counter. I wouldn't expect this to be reliable, which is one reason why motors with built-in encoders are commonplace. You might find some reading on the older AC techniques of synchros and resolvers to be of interest.

There's also the problem of speed vs. torque. Brushed motors are low torque devices so are often run at high speed with gearing to their output. The encoder is then placed on the low-speed output shaft. This makes best use of both device's optimum speed: the motor runs fast, the encoder measures position directly. When motors are used directly, they're usually some multi-pole brushless form as they have better torque than a brushed motor.

The disc is often just called an encoder disc, no special name, and may be distinguished as a relative encoder, rather than an absolute encoder. Absolute encoders will tell you position when first turned on (but are more complex). Relative encoders will either tell you only speed, or may infer absolute position if they can pick up some zero point. You will also see quadrature encoders, where there are two sets of holes, phase shifted by 90º (you can find a pair of these in an old mechanical mouse with a ball). This allows direction to be determined, so are more useful for absolute position.

Are you measuring position or speed? The techniques often vary, accordingly. Some encoder types are adequate for speed, but not position.

Try to find an oscilloscope, as you can learn an awful lot from one. Second hand 30 year old scopes are cheap in the West and still very useful - my bench scope is the same 20MHz dual trace delayed sweep design I started with as a telephone apprentice in 1982. It's fine for the sort of hobbyist / Maker stuff I mostly do (If I need better I borrow friend's). You can also buy pocket-sized flat panel 'scopes from China now (try eBay) for £20-30 and these are very handy for field work. Andy Dingley (talk) 10:31, 6 June 2016 (UTC)[reply]

The article Rotary encoder describes devices for encoding the angular position (absolute) or motion (incremental) of a motor shaft. The OP's idea is for a Brushed DC electric motor to serve as its own rotary encoder. If there were no commutator the back EMF from the motor would be a sinusoidal voltage of 1 cycle per whole revolution, identical to a simple AC generator. However the brushes and commutator switch successive windings so that only a part of the cycle is seen. The number of repetitions depends on the number of commutated windings in the motor construction; more windings means a higher ripple frequency (as a multiple of the r.p.m.) but smaller amplitude because a smaller angular range of the mentioned 1 cycle/revolution is sampled. I conclude that the idea requires tracking very small electric load variations that depend on the motor construction, that change greatly with r.p.m. and mechanical load, and will only give incremental position information. To sense absolute position an additional optical or mechanical sensor, such as a limit switch, will be needed. Using a PC with soundcard as a digital oscilloscope has limitations of no low frequency response DC - 50 Hz, no calibration, and the audio inputs should be protected against overload with series resistor 10 kohm, back-to-back diodes 1N4148 and series capacitor 1 µF. AllBestFaith (talk) 10:39, 6 June 2016 (UTC)[reply]

A further option would be an oscilloscope board which plugs into a PC - like the sound card idea, but with a front end that's better designed for this use. They can be handy for multi-channel inputs where you need to log and compare many signals. Andy Dingley (talk) 11:54, 6 June 2016 (UTC)[reply]

That's exactly right - I want the motor to double as its own encoder. Incremental is ok (I'll have limit switches for self-calibration on power-up.) Thank you for the soundcard tip! Asmrulz (talk) 12:39, 6 June 2016 (UTC)[reply]

I'm measuring (trying to, anyway) position (number of revolutions since the motor was powered.) You are right about the oscilloscope. It's like in Plato's cave without one! Asmrulz (talk) 12:39, 6 June 2016 (UTC)[reply]

Asmrulz, when a brush connects a coil of the rotor, the voltage drops depending on its load and the power supply. When it disconnects, an inductive burst occurs. Operating a hand drill, you see the blue sparks between brush and rotor. On cheap compact cassette players the drive drive motor noise affects the sound output. A small ceramic capacitor located next to the motor blocks much of the noise. A little better design, sometimes without additional componects reduces the noise dramendously. Buck converters rectify this burst for current output, see also current doubler. --Hans Haase (有问题吗) 21:01, 6 June 2016 (UTC)[reply]

thanks! what if there's no space for a capacitor (I understand it must be "electrically close" to the motor)? can I run the leads from the motor through a ferrite bead instead? Asmrulz (talk) 16:38, 7 June 2016 (UTC)[reply]

Asmrulz, see the picture of the tape drive. Some noise is being killed with this capacitor. Else, note how to connect the wires to the supply. Use capacitors with low ESR. Wires are no supra conductors. See each of them as a resistor. Doing so, You kan kill the noise. Old soundcards in the 1990 on the ISA bus, fed the 12 volts to a 9 volts linear voltage regulator to cut the noise from the motors. I guess the load and wiring might be complicated in Your case. On a PC power supply for example, use another arm of cables or if possible choose another rail. A filter behaves like a preresistor to the motor. Too small electrolytic capacitors may get hurt from negative voltage bursts. Smaller capacitors were choosen as ceramic type, resisting a higher voltage and having a low ESR. Ferrite will be overfeed and beginns to be useful in MHz frequency range. --Hans Haase (有问题吗) 19:28, 8 June 2016 (UTC)[reply]

Asmrulz, differing the frequency range of noise and interferences is just a question of the size or way the filters work. Lower frequency range can be filtered with voltage regulators, leveling the noise to the impedance of the supply. This replaces over sitze filters for lower frequency. Small capacitors work a a higher frequency, but no to cover low frequency noise. The next question is the power of the noise, also making the filters mechanical dimensions grow. It can be compared with a windshield wiper to rain or a waterfall, the vehicle is exposed to. When the supply of the motor does not cover the noise, it is getting difficult. An idea is to block next to the H-bridge power input, als have the supply output to a low ESR capacitor as a central supplier of all circuits in the system. The cheap tape drives used the same routed wire on the Printed circuit board for the motor and the amplifier, causing noise from the motor in the speakers. This is caused by using single wire to supply both devices, make the wire become a common preresistor. When the amplifier was able to compensate the noise in the operating voltage, it also affects its reference voltage, modulating the noise audio signal input of the amplifier. Analog oscilloscopes use High speed CMOS logic devices. HC is know for opening both transistors of the output for a short periode of switching. This shortcut was prevented by the CMOS transistors inside, which limits current their own, but an increased power consumption occurs on switching. For this reason, the oscilloscope circuit designers spend the HC-logic device a preresistor of 22 ohms and a blocking capacitor to keep the noise of from the voltage of the power supply rail. HC is known to have low power consuption, but increasing power consumption when operating at higher frequency causing higher number of changing the output signal drivers per time and exceeding the power consumption of TTL or LS (→Low-power Schottky) logic devices at a known frequency.

If we are talking about the H-bridge ciruit of Your last question, make sure the H-bridge is protected against shoutcuts by turning on all transistors same time or still opens the complementary channel while the frist has not shutdown completely. Power transistors have higher capacitive ballast and operate slower than logic devices like microcontroller or similar. Shortcutting the H-bridge also occurs when connected without and protection circuit to a microcontroller. An failure in software, operating voltage, clock generator, electromangtic interference or electrostatic discharge is able to cause garbage to be written to the H-bridge input, causing the shortcut of the H-bridge and torque it or to the power supply. Note the load and shortcut behavior of the controlled motor. A stepper motor driver device should cover such requirements and support the hold voltage level, giving the motor a brake power at the position where it became stopped. This hold voltage is lower that the operating voltage due the magnetic field on the motor coil has already beeing establistablished, reducing the impedance of the motor coil to the electrical conductivity of the cooper wire of the coil in this motor. The modified H-bridge in a stepper motor device might be designed by a timer or monoflop which is being reset by change of the inputs. When its time exceeds, the H-bridge input voltage is turned off from the operating voltage and feed by the hold voltage, least to protect the motor coil. --Hans Haase (有问题吗) 13:00, 9 June 2016 (UTC)[reply]

I know that about the H bridge, the ATTiny (which I'll be using) even has a programmable Dead Time generator that introduces a short delay (phase shift) between the complementary PWM outputs exactly for that purpose Asmrulz (talk) 00:30, 10 June 2016 (UTC)[reply]

Falcon Rocket Guidance for landing[edit]

I'm wondering what the Falcon rocket uses for guidance to land so accurately on "Of course I still Love You". Isn't it too high to use GPS? If it is homing onto the drone ship, does it's own engine firing distort the homing signal? Thanks 49.197.26.168 (talk) 07:56, 6 June 2016 (UTC)[reply]

I think this is the article you want: attitude control system. Maybe inertial navigation system too. As for GPS, the GPS satellites orbit way up there. See picture. "ISS" is the space station. Why do you think engine firing would affect the signal? It's a chemical rocket. The engines don't put out any significant electromagnetic interference, I think. --71.110.8.102 (talk) 08:36, 6 June 2016 (UTC)[reply]

Interesting question. GPS has capabilities (high speed, high altitude) that are only available to the Western military, to avoid North Korea etc. using them to control missiles. It would be interesting to know if Falcon is permitted to use these. Andy Dingley (talk) 10:33, 6 June 2016 (UTC)[reply]

I assume this is about Falcon (rocket family) ? One problem is that they probably can't use most off-the-shelf GPS systems, which would assume that they were on, or near, the surface of the Earth, and may well give an error if it calculates a position in low-Earth orbit. So, they would need to use, or create, a GPS system which allows for that. StuRat (talk) 13:15, 6 June 2016 (UTC)[reply]

How about some references? First, as our article (and many others around the internet) notes, GPS is involved in the guidance process. The US government maintains a website full of info about GPS; here's their page about space-based applications, which specifically notes usage on things like ISS return vehicles. StuRat is correct that a space-qualified system would need to be used, but that's probably more for safety certification than it is for capability set (though perhaps that, too). Remember also that the highest precision for landing a Falcon is needed at the end of the landing, when speeds are lowest -- while there's probably an additional error component at high speeds, the corresponding high distance from target will also raise the tolerance for error. Further, we have an article on the various types of GPS signals. While there is a military component to the GPS system, it's used for anti-jamming and anti-spoofing purposes; it does not materially impact accuracy. Selective Availability, the capacity for the US government to degrade the accuracy of civilian GPS receivers, has not been used since the Clinton administration and post-2007 satellites do not even have the capability, so tracking accuracy is not a concern for the North Korean missile program -- the far harder engineering matter is making use of that tracking data. Finally, I would expect that there's some sort of radar-type backup or parallel guidance system for the final approach. As above, at orbital speeds, positioning is largely a matter of the pre-planned orbital mechanics (and radar can work out to that range, too) and so centimeter-level accuracy isn't needed at that phase of the mission. — Lomn 14:37, 6 June 2016 (UTC)[reply]

It is true that Selective Availability is no longer used, but there is still ... some technical capability that allows planned, scheduled activity that may make GPS unreliable or unavailable. For example, tomorrow, June 7, 2016, GPS will be unreliable for most of the Western United States (NTTR GPS 16-04). This activity will recur for several days over the next month. It is not particularly uncommon for GPS "outages" to take place within about a 400 mile radius from "that area."

Nimur (talk) 21:16, 6 June 2016 (UTC)[reply]

@Nimur: Are you sure it's Groom Lake (or even something to the northeast of it?) rather than the Nevada Test Site? You know, if your county had been hit by that many nuclear weapons from the air, they'd be messing with GPS too! Wnt (talk) 23:05, 6 June 2016 (UTC)[reply]

The specific notice I linked earlier - like many similar GPS outage notices - was published by FAA. It references the Nevada Test and Training Range (NTTR); its GPS coordinates, or rather, its latitude and longitude, are explicitly specified in the published NOTAM. You can also refer to !GPS 06/016, KZOA A0292/16, using your favorite database tool, such as FAA's NOTAM search engine. As they say in all the textbooks, always review all available resources prior to using GPS. There have been plenty of other center-points for GPS outages - in places like China Lake, Beatty, and elsewhere; it just so happens that the one in effect tomorrow is centered here.

Groom Lake is an unofficial official name for the dry lake bed that shows up on charts of that region; you may know the area by many other names, depending on your attention to detail.

Nimur (talk) 00:58, 7 June 2016 (UTC)[reply]

All the photographs of the barge shows two domes. I can't see a 'differential GPS base station' (which gives cm accuracy) but they are very small toroidal antenna. So most likely the barge had both GPS and radar. Think the thing to do is email Elon Musk directly. He is the sort of guy that would even remember off- the-top-off-his-head which walmart store he bought the rocket's paint from and would be happy to reply to such a simple question.--Aspro (talk) 00:49, 7 June 2016 (UTC)[reply]

Those domes are the satlinks for live video. This is why the video usually drops out then the stage is nearby, vibrations throw off the satlinks and connection is lost. This has been discussed to death, there is no feedback from the OCISLY, it is inertial+GPS from the stage to predetermined coordinates. In flight 15, when the sea conditions were too severe for landing and there was a thruster problem with the barge, it was moved out of the way, and the stage made a water landing, exactly the same trajectory, there just wasn't a barge at the predetermined coordinates. Fgf10 (talk) 07:06, 7 June 2016 (UTC)[reply]

Thanks for all your comments and refs, particularly Fgf10. Question answered! Original Poster49.197.104.167 (talk) 07:15, 7 June 2016 (UTC)[reply]

an organism feeding by rotten organic matter which their name suffix with "phagus"?[edit]

What is the name of the organisms feeding by rotten organic matter? It should be something with suffix of "phagus". I have tried to search this question on Google, but nothing I found. 93.126.88.30 (talk) 22:22, 6 June 2016 (UTC)[reply]

See http://www.onelook.com/thesaurus/?s=*phage.—Wavelength (talk) 22:42, 6 June 2016 (UTC)[reply]

Saprophagous seems to match what you're asking for. Other terms (without the -phagous suffix) that may be relevant are scavenger and detritivore. Deor (talk) 22:45, 6 June 2016 (UTC)[reply]

Thank you! 93.126.88.30 (talk) 23:20, 6 June 2016 (UTC)[reply]