Science desk
< March 26	<< Feb \| March \| Apr >>	March 28 >

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

March 27

Maglev train?

If you were to guess, does this description below describe that of a Maglev train, like perhaps that seen at Disneyland?

... a miniature railway an elliptic curve of about twelve feet long and six wide around which traveled a miniature magnetic engine drawing at high speed a miniature car...

--Doug Coldwell (talk) 20:21, 27 March 2022 (UTC)[reply]

Well you've just primed us to think of a maglev, so of course it does now. I can't picture what "six wide around" means, it's not geometrically possible to have an elliptic path that's half as wide as its circumference. (Oh I get it now, it's just saying the major axis is twice the minor axis. Probably the writer should not have used the word "around", it might confuse a stupid person.) The crucial thing here is "magnetic engine". That could be a maglev, sure, or it could be an "engine" (or just a block of metal) on wheels, which is dragged around by electromagnets embedded in the track - so, a mag-non-lev. It could also just be an electric motor described in, I don't know, 1840, like how push-buttons used to be called "springs". Card Zero (talk) 21:06, 27 March 2022 (UTC)[reply]

(edit conflict) It would not be my first thought; the text does not suggest levitation is involved, which I think is something one would note, well before going into the shape and dimensions of the track. We can only guess at the meaning of the vague phrase "magnetic engine"; whoever wrote that was probably not an engineer. --Lambiam 21:09, 27 March 2022 (UTC)[reply]

I got it from here. It's an 1869 Journal about a person that was an expert in electricity experimentation and knew all about electromagnetism. It got me thinking that perhaps he hit upon the idea of a Maglev train.--Doug Coldwell (talk) 21:33, 27 March 2022 (UTC)[reply]

The article seems to refer to around 1836, it is a retrospective of Page's achievements. On the previous page the writer refers to Page "taking up voltaism and constructing batteries" and eventually "producing motion by magnetic power". I'd guess therefore that the "magnetic engine" was a simple DC motor. Any sort of induction motor or maglev would require AC sources and I'd think this was a bit early for that. Martin of Sheffield (talk) 22:24, 27 March 2022 (UTC)[reply]

Probably not this (World's Simplest Electric Train - YouTube), then. It can also be implemented as an actual railway, with a coil placed between the rails, and a bar magnet on a flat wagon, powered either by an on-board battery or by pick-ups from powered rails. --Verbarson ^talk_edits 08:44, 28 March 2022 (UTC)[reply]

O.K. I seem to get the idea now that it was a DC motor as a train engine pulling another car - AND they were all on wheels on a track. In the article Charles_Grafton_Page#Scientific_accomplishments it is show such an electric locomotive. Apparently this link above I pointed out is describing a DC motor locomotive. Everything being wheels on a track = no levitation. --Doug Coldwell (talk) 11:04, 28 March 2022 (UTC)[reply]

The description in the article comes from the mouth of "Gen. H. K. Oliver", indeed, as I surmised, not an engineer, but having enjoyed an illustrious career. From being a (popular) instructor at the Salem schools,^[1] he went on to become the Salem mayor.^[2]

How about Thomas Davenport (inventor)? Built an battery-powered model car running on track in 1834. (Wikipedia:WHAAOE)--Verbarson ^talk_edits 16:18, 28 March 2022 (UTC)[reply]

Maglev doesn't use a locomotive pulling a wagon. The propulsion always acts on every wagon of the train. I assume the miniature magnetic engine is what we would nowadays call an electric model locomotive (although it may not have been a model if it wasn't based on a full-size original).

For actual maglev trains, you need strong electromagnets and a fast feedback system for stabilisation (requiring at least triodes, invented in 1906), or spin-stabilized magnetic levitation (invented in the 1970s) and strong permanent magnets, or superconductors (discovered as a complete surprise in 1911) and cryogenic tech, all combined with 3-phase variable-frequency AC power (invented in the 1880s) for propulsion. That doesn't appear feasible in the 19th century. Spin-stabilized magnetic levitation can be explained with classical physics, so a smart person could have invented it earlier, but stability of this method is marginal. There are simpler methods to magnetically levitate trains, but those only work at speed. PiusImpavidus (talk) 09:14, 28 March 2022 (UTC)[reply]

Maglev § Technology sketches a rather variegated field, with for example a choice between electromagnetic and electrodynamic suspension, and between propulsion powered by an onboard linear motor versus the track serving as a linear motor. --Lambiam 12:19, 28 March 2022 (UTC)[reply]

The Disney monorails are not maglev. Hayttom (talk) 05:43, 29 March 2022 (UTC)[reply]

Estimating thermal resistance for some cheap heatsinks

I bought some cheap heatsinks without any kind of rating and decided to try to figure out what their "thermal resistance" is. I used some cheap thermal grease and mounted a heatsink to a 10 ohm TO-220 resistor using a little machine screw and nut. I connected my bench power supply to the resistor and allowed the resistor time to settle at different voltages before measuring the temperature using a thermocouple pressed against the copper mounting tab of the resistor. I realise that the temperature of the actual resistor will be a bit higher but I can't get at it without drilling or something.

I notice that the value I calculated for thermal resistance changes as the dissipation increases, maybe as radiation becomes a bigger factor at higher temperatures. It just made me wonder: when I see heatsinks which do have thermal resistance ratings, should I assume that it's at the highest temperature, e.g. 175 °C so the manufacturer can give the best-looking rating? What would you say is the thermal resistance for my heatsinks? 46 or 38 °C/W?

Ambient temperature (°C)	20.5	20.5	20.5	20.5
Voltage (V)	2.4	3.4	4.3	5.4
Current (A)	0.23	0.33	0.425	0.53
Power (W)	0.55	1.12	1.83	2.86
Resistor temperature (°C)	46	68	93	129
Temperature difference (°C)	25.5	47.5	72.5	108.5
Thermal resistance (°C/W)	46	42	40	38

78.150.168.49 (talk) 21:05, 27 March 2022 (UTC)[reply]

Heat sinks I looked at typically have a falling charcateristic like you have shown. eg https://docs.rs-online.com/b558/0900766b81581ebb.pdf i don't know what spot they pick on that curve, perhaps 60 deg C. Greglocock (talk) 00:55, 28 March 2022 (UTC)[reply]

I will start with a pedantic remark, but it matters: what you are really measuring is the thermal resistance between wherever the thermocouple sensing element is (assuming the heat source is not too far from there) and the ambient air temperature. Possible reasons that this resistance can vary with temperature are:

The thermal conductivity of the heat sink increases with temperature (see for instance the entries for aluminium in List_of_thermal_conductivities up to 400K)
Radiative cooling effects (as mentioned in the OP)
Better thermal contact (lower contact resistance) between the thermocouple and the copper, the copper and the resistor, the resistor and the heat sink (thermal dilation usually benefits thermal contact)
An increase in the natural convection heat coefficient as temperature increases (= decrease in thermal resistance between the heat sink and faraway ambient air). Frustratingly, I could not find a graph plotting heat coefficient vs. temperature for a horizontal surface in air, when I have definitely seen such a graph on the internet before, and I doubt anyone has better keywords than me. I refer you to the adimensional correlations you can find in Natural convection#Behavior or in a Google Scholar search for "nusselt rayleigh correlation natural convection" - what correlation you can use in what geometry is a mess and still a very deep subject of research, but all of them say that Nu (proportional to h, the convection coefficient) is an increasing function of Ra (proportional to T, at least for ideal gases).

My educated guess about the relative importance of those effects is that (1) and (3) are the most important, (4) matters but not all that much, and (2) is negligible.

I would advise running the test again with and without a fan blowing air above your test device to control for the effect of air convection. I guess^{[citation needed]} the air-blowing condition is what manufacturers do anyway - it gives the best number, and it’s true since you are trying to measure the thermal resistance of the heat sink, not that of the heat sink plus air. Improving the solid contacts (resistor/thermocouple/sink) would also make the whole thing more accurate, but it’s harder to do.

As to the question of what is the "true" rating, I would say the thermal resistance of the sink is the lowest number you manage to achieve at power dissipation rates similar to your application (otherwise you’re cheating on the thermal conductivity aspect) but while optimizing other parameters (such as air flow). Tigraan^{Click here for my talk page ("private" contact)} 12:52, 30 March 2022 (UTC)[reply]