Talk:Differential (mechanical device)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Technology (Rated B-class)
WikiProject icon This article is within the scope of WikiProject Technology, a collaborative effort to improve the coverage of technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
Checklist icon
 

Loss of traction[edit]

At first I was going to correct the individual errors [...] I would like to rewrite the section. [...] Another Stickler (talk) 04:05, 27 September 2008 (UTC)

[...]
[...]
[...] I still think that less is more so far as this section goes. Here's my initial attempt: "A differential establishes a force balance between the three shafts. This means that if one shaft has no torque on it, in the steady state none of the others do (OK, wrong, the other two could have opposing torques), so all of the available torque is used to accelerate the rest of the mechanism. The practical outcome of this, in a car with an open (or conventional) differential, is that if one tire loses grip, then drive to the other wheel is also lost, and the engine speed climbs rapidly as it accelerates the driveline and one halfshaft and wheel, effectively unrestrained by ground reactions." [...] Greg Locock (talk) 00:35, 29 September 2008 (UTC)
[...]
Here's my draft, taking your "less is more" approach but fattening it just enough to be self-contained:
"An undesirable side effect of using an open differential in a vehicle such as a car is the reduction, even total loss, of propulsion when one of the driven wheels loses friction (by being separated from the ground or slipping). The sharing of force between the open differential's output shafts (when driven by the input shaft) means that if either loses force, the entire propulsion system from motor to tires loses force, just as the entire surface of a balloon loses pressure if any part of it has a hole. (See other sections for descriptions of differential types designed to reduce or eliminate this effect.)"
That should go at the top of the section. I'm hoping it capsulizes the whole issue, making the later description of traction unnecessary and better left to the traction article, where most of it might be moved. I'd keep the part debunking the "one wheel drive" misconception since it applies directly to open differentials. I'm also thinking that the section should be renamed to "Loss of propulsion" since that is the undesirable side effect, not "Loss of traction". -- Another Stickler (talk) 11:36, 22 October 2008 (UTC)

Pinion Gear Rotation[edit]

Neither of the pics makes sense to me. If the pinion gear is indeed interfacing between each side gear, then for one to turn, the other must be turning in the opposite direction. The pic seems to indicate that if the ring gear receives torque in the illustrated direction, the left driveshaft will necessarily spin in that direction. If the pinon gear interfaces with the left side gear, then the pinion gear will necessarily spin counterclockwise (looking from above, as in the pic). If the pinion gear interfaces with the right side gear, this will necessarily drive the right side gear in the opposite direction from the left side gear. What am I missing? JunkCookie 21:58, Aug 20, 2004 (UTC)

  • Yeah, the pics could stand to be clearer. The top image is meant to show that the entire assembly rotates forward; picture it without the ring gear for a moment, and imagine the green pinion gear rotating around the other two gears. The pinion gear is pushing against both side gears, so the pinion gear itself can't rotate unless one of the side gears meets some kind of resistance. When there is resistance on one side (as in the second picture), the pinion gear starts rotating, turning the other side with additional speed. The link to the HowStuffWorks version is much better at explaining this than what we currently have here, unfortunately. -- Wapcaplet 23:51, 20 Aug 2004 (UTC)
I figured out what I was missing. I didn't realize that the cage to which the pinion gear is mounted actually rotates with the ring gear. Somehow I had it in my head that the ring gear was fixed to the left side gear, and that the pinion is used to transfer the rotation to the right side gear. The animated pics at HowStuffWorks gave me the 'Aha!'. JunkCookie 20:59, Aug 23, 2004 (UTC)
And the fact that you needed another site to figure it out underscores that our figure is unclear. The figure does for instance not make clear that the cage is separate from the left side gear. Anybody up for making an animation? - Mglg 23:24, 4 June 2006 (UTC)
The problem with making an animation is that the politically correct thought police will only allow certain file formats to be loaded. I suspect that given the time that it would take to build the model, and then go through all the hoops required to turn it into an acceptable file format that very few people know of (ogg video) will put them off. Leastways it does me.Greg Locock (talk) 22:14, 26 January 2008 (UTC)

Torque[edit]

When the ring gear causes the planet gears(green) to push the sun gears(red & yellow) is there a ratio involved?Does the torque on the axle as a whole equal to the torque on the ring gear(purple) multiplied by the sun to planet ratio?(only on a straight road)

I don't know anything about torque, but the ratio of ring gear to axle as a whole is 1:1. Ring gear turns once, both sides also turn once; when both sides are going at the same speed, the effect is the same as having one long, solid axle, since none of the gears are moving with respect to one another. -- Wapcaplet 16:41, 22 Aug 2004 (UTC)
There is a ratio not shown in the diagram, the ratio of the drive pinion to the ring gear, but the drive pinion is missing from the diagram (and the teeth on the ring gear are on the wrong surface). But as said above, the ratio of ring gear to axle is 1:1. (The ratio of drive pinion to ring would typically be in the range 3.5:1 - 4.5:1, and yes it multiplies torque by reducing speed.) RB30DE 21:37, 20 June 2006 (UTC)

"Gear" vs. "Device"[edit]

I don't see any reason that a differential has to be composed of gears. Saying that "...a differential is a gear..." kind of implies that it's one gear, rather than several; while I am not well-versed enough in mechanics to know for sure, I can imagine that a differential could just as easily be implemented using a set of disks that interact through friction, through some kind of fluid coupling, or through a magnetic field. The term "device" seems much more general, and thus a better choice. I have changed the article to read "...a device, usually consisting of gears..." -- Wapcaplet 22:54, 28 Aug 2004 (UTC)

I believe that in fluid coupling limited slip differentials, the gears are eliminated, and exactly what you describe occurs. Thanks for the edit! Scott Paeth 07:09, 8 August 2007 (UTC)

What is this moon thing?[edit]

Some one put something about the phases of the moon in text. That has to be fixed.

Why? It seems like a perfectly acceptable addition to the article to me, if you actually read it in context. Graham 22:52, 8 August 2005 (UTC)
I agree. Please read the article Antikythera mechanism to see why it makes perfect sense.

Torsen[edit]

How about mentionning and linking to the Torsen --Olivier Debre 12:08, 30 August 2005 (UTC)

The article in question should be limited slip differential - looks like Torsen should be merged in there as there is already some duplicaiton. Graham 23:24, 30 August 2005 (UTC)

Rename "Differential (mechanical)"?[edit]

I propose this page be moved to "Differential (mechanical)". The current name implies it's about "Differential" in the field of mechanics. (I can only guess that would be a differential element of volume.) In reality, this article is about the mechanical device known as a differential. Comments? —BenFrantzDale 23:37, 13 January 2006 (UTC)

I agree with this idea. Any objections? Geometry guy 13:47, 21 February 2007 (UTC)

Disambiguation terms are generally plural nouns, not adjectives. "mechanics" is used elsewhere for this purpose, e.g. Transmission (mechanics). The term differential element is clearly part of mathematics, even though it may be used in mechanics. I don't see a great benefit in making this change.--agr 13:57, 21 February 2007 (UTC)
How about Differential (mechanisms)? This is one of the categories. Geometry guy 14:10, 21 February 2007 (UTC)
Are disambiguation terms really plural nouns? I can't think of one... Some disambiguators are descriptive nouns—words such as "band" and "artist"; others are field names such as "physics" and "mathematics"; none of these are plural nouns. —Ben FrantzDale 23:58, 21 February 2007 (UTC)
Further to this, I checked out WP:disambiguation: a disambiguation term should usually be a class, subject or context (hence a noun), but they rarely seem to be plural; when they are, it is usually because they are referring to a subject area like mechanics. I guess you are using the term "mechanics" to mean "The body of theoretical and practical knowledge concerned with the invention and construction of machines, the explanation of their operation, and the calculation of their efficiency; mechanical engineering." rather than "The branch of applied mathematics [and physics] that deals with the motion and equilibrium of bodies and the action of forces, and includes kinematics, dynamics, and statics." (Both from OED.) However, the mechanics article concerns the second definition and differentials are used all of the time in this subject. The disambiguation term (mechanics) is also used in a number of classical mechanics articles. So I think it is unfortunate that a number of articles about mechanical devices refer to "mechanics" or even link to the mechanics article.
This suggests "Differential (mechanism)" would be better, but the term mechanism is still ambiguous, so perhaps the clearest disambiguation would be "Differential (mechanical device)". Geometry guy 15:21, 21 February 2007 (UTC)
I just noticed that "mechanical" is used for Bearing (mechanical); how about we just go with that? I can't see that being confusing. —Ben FrantzDale 23:58, 21 February 2007 (UTC)
I found eight "(mechanics)" articles and have left messages on some of the talk pages. I think we should use a noun (mechanical device is still my favourite choice) if at all possible. On the other hand, I also found eight "(mechanical)" articles. Both lists are in my user space User:Geometry guy/Mechanics. I'll wait for other responses before moving. Geometry guy 09:56, 22 February 2007 (UTC)

Active Differentials[edit]

This section is unclear, and should among other things explain what "DOF" means. Now DOF links only to a disambiguation page that does not list any relevant page. - Mglg 23:24, 4 June 2006 (UTC)

Based on context, I'm going to change it to Degrees of freedom (engineering), as that seems to be the correct interpretation. Still, it's a poor paragraph, butr at least I've resolved some of the confusion. --Bdoserror 07:48, 4 October 2006 (UTC)

Three wheels?[edit]

I know that saying differentials are only found on four-wheel vehicles is incorrect (cf three wheeled car) but I'm not sure how to reword the summary without destroying what it's getting at. Axled vehicles? Do motorcycles have axles? It's not something I specialize in. Scott Paeth 07:17, 8 August 2007 (UTC)

Changed video link from YouTube to Internet Archive[edit]

Changed the link to "Around the Corner" (1937 Chevrolet film explaining how a differential works) from YouTube to the Internet Archive. The film is rather overproduced for the topic, but anyone who watches it will understand how a differential works. --John Nagle (talk) 02:15, 16 April 2008 (UTC)

Recent edits; wrong illustration[edit]

The article was fascinating to find; I've been a mechanical technician, and very interested in mechanisms all my life. Indeed, a differential (especially as a "torque splitter", as used in cars) is a tricky thing to describe. While searching for an illustration of a spur-gear differential, I came across several descriptions.

As I found the article, it implicitly assumed automotive emphasis, which bothered me somewhat. Differentials are found in many places, most of them not well-known, however, but, their "torque splitter" use in cars is atypical, if one were to make a comprehensive list of their applications. I tried to rewrite the introduction while also trying not to eliminate the original wording any more than seemed necessary; that's an editing courtesy to the author. As it is, I wish it were simple to keep more of the original at that point.

The planetary (epicyclic) differential (power split device) in the Prius drive train interconnects three major entities: the engine and one motor-generator, another motor-generator, and the differential for the front wheels. It's not quite simple. A really first-rate description is by Graham Davies, at <http://www.ecrostech.com/prius/original/Understanding/Contents.htm> He assumes some technical and scientific education, which permits him to offer a really-good explanation. Perhaps this link could be included in the Differential article, but it seems mostly off-topic, although the planetary differential is quite essential.

Perhaps, not incidentally, the Prius does not contain a CVT; that's technically incorrect. A CVT is a very different, variable-ratio mechanism. However, the Prius apparently feels as though it has a CVT.

As to a Google image search for a good, clear illustration of a spur-gear differential, I looked at maybe 25, perhaps more pages of images, and they became increasingly irrelevant. There were some images, but they were engineering drawings that were really not suited for the article.

[I just realized that I have a computing differential, could take a picture of it, and add it to the article. No promises, though! It doesn't look dramatically different (it's much like the one at the reference I gave). The spider is very easy to see.]


[Wrong illustration; sorry!]

As to the illustration "Reconstructed 19th century carding mill differential", it looked pretty odd. When I clicked on the little icon to learn more, it turns out that this is really, truly unlikely to be a differential! Its title is "Ox powered treadwheel". It's a rugged, strong turntable, and you can just about see a set of "bevel" gears near the ground. The gear on the shaft has individual pins that point down and mesh with a smaller gear on a shaft that looks like a log. That shaft goes down to the right, to power the mill.

The cage-like structure at upper left is probably where the ox was tethered; as he walked on the turntable, he made it rotate -- one ox-power.

I didn't want to act like a bull (or ox?) in a china shop and delete the image, but I do hope that someone who's been more connected with this article will do so!

Regards, Nikevich (talk) 08:53, 9 May 2008 (UTC)

I removed the ox treadmill photo for you since you were not bold enough. Try to donate your computing differential picture and add it to the "Non-automotive applications" section please. -- Another Stickler (talk) 07:46, 13 October 2008 (UTC)

Removed material on railroad wheels[edit]

I backed out some material on railroad wheels by an anon. It's not that relevant. Also, wheels on most freight car trucks aren't keyed. For historical reasons, most railroad wheels are a simple press fit. The standard for railroad wheels predates the milling machine. --John Nagle (talk) 20:47, 20 June 2008 (UTC)

Regarding http://en.wikipedia.org/w/index.php?title=Differential_(mechanical_device)&diff=next&oldid=220584803 I think the main point by anonymous is that using railway wheels as an example of slippage is not valuable if they don't slip. To me, that's a good argument to remove all reference to railway wheels, since they are only brought in as a demonstration of slipping, which is probably what anonymous should have done (with an explanation in the edit summary rather than in the article). Your undo was not helpful in that it neither removed all references to railway wheels, nor let stand the well thought out and linked correction by anonymous of the bad assertion (that they must slip) in the version you reverted to. -- Another Stickler (talk) 21:13, 6 October 2008 (UTC)
Never mind. The most recent version has morphed to a compromise version anyway. It's hard reading through the history and doing a manual mental compare. I lost track. I think we still might want to remove references to railway wheels if we don't need them to explain slip. -- Another Stickler (talk) 21:25, 6 October 2008 (UTC)

EV direct drive[edit]

in case of some EVs with direct drive, the differential is eliminated by using 2 motors at the back (preferably dc motors as speed versus current is linear) and sending different amounts of current based on the turning radius to each wheel —Preceding unsigned comment added by 192.18.192.76 (talk) 06:56, 17 December 2008 (UTC)

Can we have a clearer image?[edit]

The images included in this article, Differential_free.png[1] and Differential_locked.png[2], are well done and make understanding this mechanical device easier for visual learners. However, there is one part of Differential_locked.png[3] that I think could be improved. If the red left side gear encounters a lot of resistance and stops moving, then the right side yellow gear will turn twice for every one turn that the blue ring gear turns. This can be an important fact in visualizing this mechanical device. To better illustrate that point, I think it would be better to have the right-most red arrow that is placed on the yellow right gear rotate around that gear about 1.5 times, instead of the 0.5 times that it is now. That would show that the right yellow gear is rotating faster than the blue ring gear when the left red gear stops rotating.

I will see if I can modify the image myself while maintaining its quality, and post the edit and see if it sticks. Pgn674 (talk) 18:28, 24 February 2009 (UTC)

Re: Traction-adding devices[edit]

So, here's what we have in article text:

Traction-adding devices

There are various devices for getting more usable traction from vehicles with differentials.


  • A simple solution for a standard "open" differential is to partially apply the vehicle's handbrake when one wheel is spinning, as this often provides sufficient resistance to increase the overall torque and allow the other driven wheel to move the vehicle. This only works where the handbrake acts on the driven wheels, as in the traditional rear-drive layout. Naturally, the handbrake should be released as soon as the vehicle is moving again.

.... (skipped a few items) ....

  • In the absence of any of these devices, it is sometimes possible to slow the spinning wheel by partial application of the parking brake in a rear wheel drive vehicle.[1]

My question is: aren't those item both telling the same thing?

--K001 (talk) 20:14, 12 March 2009 (UTC)

Yes, but neither is correct for an open differential because the gain at one wheel position in the brake also appears at the other brake and not at the wheel. It only works for limited slip differentials. That has since been changed in the article.04:38, 20 October 2009 (UTC) —Preceding unsigned comment added by 121.218.77.253 (talk) Sorry, I forgot to sign.121.218.77.253 (talk) 01:30, 22 October 2009 (UTC)

Differential without large carrier wheel[edit]

I noticed that most differentials use a large carrier wheel (blue) in http://en.wikipedia.org/wiki/File:Differential_free.png However, in case the space is limited near the axle, isn't it possible to include another type of differential (without a large carrier wheel). Do differentials like this exist; ie without such a large carrier wheel. Ideally, the carrier wheel would be best only slightly bigger than the other wheels. 91.182.219.189 (talk) 08:04, 22 July 2010 (UTC)

First off, do you mean "differential" (mechanism) or "differential" (car transmission component)? The transmission component is both the final-drive reduction gear (hence the big crown wheel) and also the differential. The crown wheel itself isn't part of the differential mechanism as such. The reason it's so big is to provide the gear reduction ratio - if you want that reduction, it has to be provided by a big gear somewhere and this is just convenient. If you look at the Stanley steam car it had quite a neat arrangement that was large diameter, but short, as the differential bevels were mounted through holes in the crownwheel. if you look at differentials for all-wheel drive cars, especially the centre diff, these often don't have attached crownwheels and so the "differential" is quite small and compact.
To see a range of differential images, take a look at Wikimedia Commons, Automobile differentials and Differentials. Andy Dingley (talk) 09:35, 22 July 2010 (UTC)

Antikythera mechanism[edit]

Wright, M T. (2005). "The Antikythera Mechanism and the early history of the Moon Phase Display". Antiquarian Horology 29 (3 (March 2006)): 319–329.

This reference is used to support the statement, "the assumption of the existence of a differential gearing arrangement was incorrect". I scanned the referenced article [4] and it does not mention differential gearing. Please improve the reference to point to the exact phrase which supports this statement. --beefyt (talk) 21:46, 10 December 2010 (UTC)

There are plenty of refs out there that categiorize the reconstructed Antikythera as a differential. I think the old Sci Am article (a significantly early mention) did this. The claim that modern analyses shows it to not be a differential certainly needs better sourcing. Andy Dingley (talk) 22:25, 10 December 2010 (UTC)
This is weird and almost certainly wrong. There is little doubt that the mechanism does contain differential gearing, and there is a diagram here. Someone in the past edit history of the article seems to have a bee in the bonnet and overstated or misrepresented the sourcing on this issue.--♦IanMacM♦ (talk to me) 21:09, 25 December 2013 (UTC)
I found this article on the history of the Antikythera mechanism to be convincing PHASES IN THE UNRAVELING OF THE SECRETS OF THE GEAR SYSTEM OF THE ANTIKYTHERA MECHANISM. Prof McCarthy (talk) 22:28, 25 December 2013 (UTC)
This would be best presented by giving both sides of the argument, and pointing out that the conclusions of Derek J. de Solla Price have been questioned by more modern research. Previous edits to the article have either been uncited, poorly cited, or stated that the modern research was correct or better, without explaining why.--♦IanMacM♦ (talk to me) 22:52, 25 December 2013 (UTC)

Year 1926 missing in history[edit]

There is missing the year 1923 in the history, when was in Czechoslovakia was invented concept of jointless gearbox see... http://www.youtube.com/watch?v=CRAqzv-bHS0&feature=related —Preceding unsigned comment added by 147.228.209.164 (talk) 22:26, 21 February 2011 (UTC)

needs non-automotive information[edit]

We open with this marvellous teaser in the introduction:

in one way, it receives one input and provides two outputs—this is found in most automobiles—and in the other way, it combines two inputs to create an output that is the sum, difference, or average, of the inputs.

... and then there's no further discussion about using a differential to produce different types of output from multiple inputs, and most of the article just talks at length about specific automotive uses. Great if you care about cars, but useless to anyone else.

I'd expect a _(mechanical_device) article to focus on the fundamental concepts of the mechanical device. If anything, all the automotive-specific detail seems as though it'd fit better in a Differential_(automotive) article.

59.167.122.237 (talk) 08:26, 23 February 2012 (UTC)

Interesting, I seem to remember that an earlier version of the article had a section dealing with the South-pointing chariot... —Elipongo (Talk contribs) 08:38, 23 February 2012 (UTC)
There's still a paragraph about the south pointing chariot. It's the second paragraph in "non-automotive applications". DOwenWilliams (talk) 15:56, 23 February 2012 (UTC)

There's a good reason for the emphasis on automotive applications. It's the only purpose for which differentials are used nowadays. Long ago, they were used to perform analog addition and subtraction, and were used in equation clocks, analog computers, possibly south-pointing chariots, and so on. But those uses have now become obsolete. Digital technology has taken over. So all that is left is the use in automobiles. DOwenWilliams (talk) 21:33, 23 February 2012 (UTC)

Are you proposing to delete escapement on the same grounds that all clocks are now electronic? How about Detroit? - that's pretty obsolete too. Andy Dingley (talk) 21:42, 23 February 2012 (UTC)
I don't think he's saying anything should be deleted or not be added, he's just explaining why there's the preponderance of writing about that one application. I should think expansion of the historical uses would be welcome if properly sourced. —Elipongo (Talk contribs) 08:41, 24 February 2012 (UTC)

Right. I didn't mean to imply that the article should focus exclusively on automotive uses of differentials. I just meant that, since most people nowadays, including Wikipedia authors, encounter differentials only in automotive contexts, it is understandable that most of the article emphasizes automotive uses. I am sure that, if anyone feels like adding some paragraphs about other uses, they would be well received.

I have made some edits to the introductory paragraphs, in which, among other purposes, I have tried to give a balanced account of the different uses of differentials. I hope these edits help.

DOwenWilliams (talk) 22:04, 24 February 2012 (UTC)

Revised Overview[edit]

The first paragraph of the overview described properties of a planetary gear train which are more general than the specific features of a differential, and therefore seemed confusing to the reader. I commented out the paragraph. Prof McCarthy (talk) 06:24, 27 December 2013 (UTC)

rear wheel drive[edit]

Sprocket35b.jpg

The description relates to a rear-wheel drive differential, and so does the photo of an actual example of one, but none of the diagrams do. One of the key things about the differential of a real wheel drive car, is that it transfers the direction of the axis of rotation of the driveshaft/propellor shaft by ninety degrees angle, to turn the shafts connected to the wheels. How does this work ?Tallewang (talk) 20:38, 19 January 2014 (UTC)

That's actually the function of the final drive, rather than the differential. They're often confused as both of them are inside the rear axle casing. The final drive is there to reduce the engine speed to match the wheel speed (in gears below top gear, the gearbox or transmission also help with this). The propeller shaft drive used for most rear-wheel drive cars uses a hypoid bevel gear to do this. Bevel gears are well known, but the hypoid form is used so that the drive shaft pinion can be mounted below the centreline of the crownwheel, allowing a lower floorpan. This requires curved teeth on the bevel gear. Within the crown wheel of the bevel gear the differential is mounted.Andy Dingley (talk) 23:49, 19 January 2014 (UTC)