User talk:Bobblewik/units of power

units of power

Hi Bobblewik. Your edit of Rotary Engine leaves us with "Originally a 5-cylinder 50 hp (37 kW) engine, the production versions were scaled up to a 50 hp (40 kW) 7-cylinder design". What should the second hp figure be? cheers. Moriori 22:42, Apr 6, 2004 (UTC)

Horsepower

Horsepower is a complicated issue. Part of it is to do with there being two definitions of horsepower: Imperial horsepower (1 hp = 33,000 ft·lbf·min^-1) and German pferdstarke (nowadays defined in terms of SI units: 1 PS = 75 kp·m/s = 735.49875 W). The two thus differ slightly; by my calculations, 1 hp = 1.014 PS.

Fairly often, in non US markets, a power rating is given in 'hp' in English-language publication, but the values are actually 'PS' unconverted. This is even sometimes true for British or American products sold on the world market.

In the US market for automobiles, true hp is used.

The second problem is how the power is measured. In the modern day, power measurements are in SAE net horsepower in the US, and DIN net (kW|horsepower) or JIS net (kW|horsepower) in most other markets. The testing method for both is the same; horsepower is measured at the engine flywheel with all engine accessories attached, alternator fully loaded, emissions-control equipment installed, and the production exhaust system attached. Thus, the difference between SAE and DIN/JIS ratings is largely explainable by the hp/PS difference above.

Before 1972, US manufacturers used SAE gross horsepower. This tested the engine's power at the flywheel without most accessories attached, no pollution control equipment, and an open, unrestricted exhaust system. This gave power figures approx 20% higher than SAE net, although there is much variation. Thus power figures for 1972 and later US cars cannot be directly compared with those quoted for earlier vehicles. A failure to understand this change exacerbated the impression left on the US consumer that emissions and fuel-economy laws in the 1970s ruined US cars. There was, indeed, much of a reduction in power, but the change in measurement method exaggerated it.

Added to all this, of course, is the fact that horsepower figures in the press are advertising numbers rather than engineering numbers, and thus are often rounded roughly, exaggerated or even minimised. It has in fact been quite common for mass-market manufacturers to understate the power of their high-performance vehicles, perhaps to avoid bad publicity or high insurance charges for the new owners. This affects figures in both kW and hp.

I would suggest that when we quote horsepower figures in car articles, we should state which hp is being talked about. For US-market automobiles in the modern day, we should state hp (SAE net) or SAE net horsepower, I think. It is safe to perform conversions to kW using the standard ratio, since SAE hp is measured in the same manner as DIN kW. Older US-market automobiles should be quoted as hp (SAE gross). We should also flag kW conversions for those vehicles as (gross) as a warning that they are not directly comparable to DIN kW figures.

In general, we should obtain the original kW figures for vehicles sold outside the US, since in most modern cases this will be the original measurement and the hp a conversion. We should also flag hp conversions as to whether they are DIN or JIS net hp (which are generally actually PS) or SAE net.

I am not sure which figures are generally used in the UK market; the figures are normally quoted in 'brake horsepower' (bhp), but what this exactly means, I'm unsure. I suspect it is probably quite equivalent to SAE net, but more research is probably needed here.

I hope this complicated explanation helps more than it hinders! —Morven 13:34, May 26, 2005 (UTC)

Gosh! I hope all of this is recorded in horsepower or Pferdstarke (or is it Pferdstaerke ?) ... -- ALoan (Talk) 15:57, 26 May 2005 (UTC)[reply]

Would have to be Pferdestärke (redirects, as it should), but it belongs into horsepower (and quite a bit already is). Rl 16:36, 26 May 2005 (UTC)[reply]

Power plant GWh does not equal GW

Hey Bobblewik: I think you've missed the boat here. It's usual practice to give,say, a hydroelectric plant rating as both the peak power (megawatts) and the energy produced in a year (say in GWh/year). These are almost never related by the factor of 8760 hours per year because a power plant almost never operates at peak power for a whole year. Especially for hydro plants, the annual energy production in GWh is an intersting statistic as it tells you how much water the plant has to work with, if it can be used for base load or peaking only, etc. I've reverted one edit to Kariba Dam and as time permits I'll look at others. A base-loaded CANDU plant might have a 100% load factor for a year, a hydro plant can range from a few per cent to perhaps 80% for a year, and the best wind turbine plants run around 35% load factor - so a 100 MW wind plant will make about 100 *8760*.35 = 306.6 GWh/year, say. --Wtshymanski 16:19, 19 September 2005 (UTC)[reply]

Hmm. The article originally said: The Kariba supplies 1320MW of electricity to parts of both Zambia (the Copperbelt) and Zimbabwe (6400GWh per annum).

By definition average is less than peak. However, neither '6400 GWh/year' nor '730 MW' refer to peak. Both statements clearly indicate average.

As you suggest, I do not quite get it. Furthermore, I interpret The Kariba supplies 1320MW similarly as an average, not a peak. I do not mind you reverting but can you go into more detail and also explain why 1320 MW is correct and 730 MW is incorrect? Many thanks. Bobblewik 17:11, 19 September 2005 (UTC)[reply]

Certainly. For example, suppose I add up all the nameplate generator ratings at the new wind farm just south of Winnipeg. There will be 60 units, each rated 1650 kW, so the total peak rating of the wind farm is 99 megawatts. Now, 99 MW * 24 h/day *365 d/y is 867.24 GWh/year. But the wind doesn't blow all the time - I dont' know the exact capacity factor but it will be around 33%. The annual energy production is likely on the order of 286 GWh/year. The peak power is proportional to *cost* of the installation - even if you only make a megawatt for a minute, you must supply all the wiring and hardware required to transport that megwatt. The annual energy production is proportional to the *value* of the installation - how much it will return for the investment. All else being equal, you'd be better off spending $1 on a plant with a 0.6 capacity factor than on one with a 0.2 annual capacity factor. To express the annual energy output as an average MW is misleading since it would give a lower value for an estimate of the capital cost and physical size of the plant. Where we have both a power rating and an annual energy production value, we should preserve both ratings. --Wtshymanski 19:28, 20 September 2005 (UTC)[reply]

You must think me really stupid but I still don't get it. Let me try talking it through. You said that you have a peak output of 99 MW but it is only used at 33%. Multiplying 99 MW by 33% produces 33 MW.

I understand that to get that 33 MW, we paid for a capacity of 99 MW. It seems straightforward to me that peak output and average output are different. All we need to do is specify which is peak and which is average. Perhaps it would help if you clarified the expression: Kariba supplies 1320 MW. Is the 1320 megawatt figure a peak value or an average? Forgive me for being so slow on this. Bobblewik 00:37, 21 September 2005 (UTC)[reply]

The standard way of expressing power plant output is to give both the peak rating (sum of nameplates) and the annual energy production. It is misleading to express the annual energy production as an "average" power output. I think where we have both figures, we should leave them both in and not do the unecessary conversion of annual energy production to some fictitious average value - this is not the way plant outputs are usually expressed in the industry. I expect the 1320 MW given for Kariba is the peak (nameplate) power capacity, and the GWH per annum figure is consistent with the capacity factor I'd expect for a hydroelectric plant. Can you trust me on this, or do you want me to point at a bunch of references?--Wtshymanski 12:44, 21 September 2005 (UTC)[reply]

I trust you on it. I just misunderstood what you were saying earlier. If I understand it correctly:

Instead of saying 'xx MW (peak)', people write 'xx MW'. They do not use the term 'peak' but assume that the reader will know that they mean peak.
Instead of saying 'xx MW (annual average)', people write 'xx GWh/year'.

It seems a bit odd to me but I can live with it. I wanted to go into detail with you because I was on a completely different train of thought. I thought you were somehow suggesting that power.time/time does not equal power. Now that we have cleared that up (I hope), I am ok with however you think the page should look. Bobblewik 15:00, 21 September 2005 (UTC)[reply]

I think you are finally figuring it out, Bobblewik--then you back off from your sensible suggestion. What Wtshymanski needs to realize is that whatever is traditionally done in the power industry isn't all that important; this is a general encyclopedia, not an article written for use by power industry engineers who know what is going on. A simple "peak" or "maximum" output figure could clarify things a lot. I think that as far as the average goes, the time period of a year is sufficient to make that clear. Not reducing that to a simple watts figure by dividing out the two time units is a clue that the year is the time period over which it is averaged. I disagree with Wtshymanski in his claim that there is anything misleading about calling that an average, but there is no crying need to point out that that is what it is. But there is nothing making clear that the power given simply in watts (with an appropriate prefix) is not the average power, but the sum of the maximum rated power of the generators. That should be clarified; it doesn't need to be any elaborate explanation; even "rated power output" or some term along those lines would probably be sufficient. Gene Nygaard 04:40, 22 September 2005 (UTC)[reply]