U5513172@NMSUVM1.BITNET (Thomas L. Baca) (01/23/91)
Hello fellow HP-users, I have what may be a silly question, but I shall ask anyway. WhenI add 25_'C ( read degree Celcius) and 1_'F on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing am I overlooking? My wife tells me that the answer looks like Rankins (sp?). Why is this happening? I heard someone say that they asked HP about it and got a fix. However, I don't really know if this guy knew what he was talking about. I've never seen a reference to this phenomenom, so I'm inclined to believe that the problem is not really a problem. I would appreciate any help on this. Thanks in advance. Thomas L. Baca NMSU
silvert@cs.dal.ca (Bill Silvert) (01/24/91)
In article <2D042619A000047E@gacvx1.gac.edu> U5513172@NMSUVM1.BITNET (Thomas L. Baca) writes: >Hello fellow HP-users, I have what may be a silly question, but I >shall ask anyway. WhenI add 25_'C ( read degree Celcius) and 1_'F >on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing >am I overlooking? My wife tells me that the answer looks like Rankins >(sp?). Why is this happening? I heard someone say that they asked HP >about it and got a fix. However, I don't really know if this guy knew >what he was talking about. I've never seen a reference to this >phenomenom, so I'm inclined to believe that the problem is not really a >problem. What do you expect the sum of two temperatures to mean? The HP48SX doesn't know either. So it converts the temperatures to the absolute scale before adding them. In other words, it converts 25'C to 298.15_K, 1'F to 256_K, adds them, and expresses the result in 'F. But the result doesn't mean anything. The difference of two temperatures is of course meaningful, and you can add two temperatures in order to average them, but here you have to be careful: if you want to average 25'C and 1'F you MUST convert to a common base, add them and divide by two. The average of these two temperatures is 39'F, which you don't get by dividing 538'F by two -- you have to convert to absolute, divide by 2, and convert back to F. This is an inconsistency in the way the HP works. -- William Silvert, Habitat Ecology Division, Bedford Inst. of Oceanography P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2. Tel. (902)426-1577 UUCP=..!{uunet|watmath}!dalcs!biomel!bill BITNET=bill%biomel%dalcs@dalac InterNet=bill%biomel@cs.dal.ca
khbsnsr@nmt.edu (Kenneth Brunell) (01/24/91)
In article <2D042619A000047E@gacvx1.gac.edu> U5513172@NMSUVM1.BITNET (Thomas L. Baca) writes: >Hello fellow HP-users, I have what may be a silly question, but I >shall ask anyway. WhenI add 25_'C ( read degree Celcius) and 1_'F >on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing >am I overlooking? My wife tells me that the answer looks like Rankins >(sp?). Why is this happening? I heard someone say that they asked HP >about it and got a fix. However, I don't really know if this guy knew >what he was talking about. I've never seen a reference to this >phenomenom, so I'm inclined to believe that the problem is not really a >problem. > >I would appreciate any help on this. >Thanks in advance. > >Thomas L. Baca >NMSU Yes, it does return the answer in degrees rankine. I just tried it. If you add degrees celsius to degrees farenheit, it returns the answer in degrees kelvin. In either case, it does not change the units accordingly. It must be a bug. Apparently it neglects to convert back to the units it is supposed to output in. -Ken ------------------------------------------------------------------------------- | _ , __ _ _ | "It's green . . ." | | ' ) / / ) // // | Mr. Scott | | /-< _ ____ /--< __ . . ____ _ // // | | | / ) </_/ / <_ /___/_/ (_(_/_/ / <_</_</_</_ | Loooove that .sig | | <khbsnsr@jupiter.nmt.edu> | | ===============================================================================
rouben@math13.math.umbc.edu (Rouben Rostamian) (01/24/91)
In article <1991Jan24.014724.15552@nmt.edu> khbsnsr@nmt.edu (Kenneth Brunell) writes: >In article <2D042619A000047E@gacvx1.gac.edu> U5513172@NMSUVM1.BITNET (Thomas L. Baca) writes: >>Hello fellow HP-users, I have what may be a silly question, but I >>shall ask anyway. WhenI add 25_'C ( read degree Celcius) and 1_'F >>on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing >>am I overlooking? >Yes, it does return the answer in degrees rankine. I just tried it. >If you add degrees celsius to degrees farenheit, it returns the answer >in degrees kelvin. In either case, it does not change the units >accordingly. It must be a bug. Apparently it neglects to convert back >to the units it is supposed to output in. It is not a bug. Just gotta be careful when adding temperatures. Digression: How to boil water cheaply: Two bring a galon of water to boil, heat half of it to 106F, then another half to 106F, mix and stir well. Since 106F + 106F = 212F, then the mixture will be at the boiling temperature! Anoter digression: 32F + 32F = 64F, you say? Convert this equation to centigrades. Recall 32F = 0C, and 64F = 17.8C, therefore: 0C + 0C = 17.8C. Uh... what did you expect 0C + 0C to be? An experiment on the HP48: Try 0C + 0C on the HP48. You will get 273.17C. How does it arrive at this answer? It converts the data to Kelvin first (0C = 273.15K) then: 0C + 0C = 273.15K + 273.15K = 546.3K = 273.15C See page 201 of the owner's manual. The example from the previous posts: Again the data is converted to absolute scale first (Rankine, in this case) then added: 25C + 1F = 536.67R + 469.67R = 997.34R = 537.67F Get it? -- Rouben Rostamian Telephone: (301) 455-2458 Department of Mathematics and Statistics e-mail: University of Maryland Baltimore County bitnet: rostamian@umbc.bitnet Baltimore, MD 21228, U.S.A. internet: rouben@math9.math.umbc.edu
akcs.dnickel@hpcvbbs.UUCP (Derek S. Nickel) (01/25/91)
Thomas,
Temperature addition is done on an abolute scale. Your are adding 25
degrees Celcius to 1 degree Fahrenhiet (not 1 Fahrenhiet degree).
25_'C = 298.15_K
1_'F = 255.927_K
adding these together gives:
554.077_K = 537.67_'F
OK?
Derek S. Nickelfrank@grep.co.uk (Frank Wales) (01/25/91)
In article <2D042619A000047E@gacvx1.gac.edu> U5513172@NMSUVM1.BITNET (Thomas L. Baca) writes: >WhenI add 25_'C ( read degree Celcius) and 1_'F >on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing >am I overlooking? [...] Why is this happening? I've never seen a reference >to this phenomenom, so I'm inclined to believe that the problem is not >really a problem. [This has come up before, so probably deserves a mention in the FAQ posting.] What you're encountering is a consequence of what it means to add temperatures; in short, it means nothing. You're attempting to add a point on an absolute scale (degrees C) to another point on an absolute scale (degrees F). This has no physical meaning. (Note: it would be no more meaningful if both values were on the same scale.) The units system on the HP 48 tries to deliver an answer, and the answer comes about because the 48 reduces both temperatures to the Kelvin scale, adds them, and then converts the result to the units of the argument taken from level one. See 'Adding and Subtracting Temperature Units' on page 201 in the HP 48SX Owner's Manual [edition 4, July 1990], which basically gives the same explanation, but without the statement: "P.S., this is meaningless."(*) The current design of the units system doesn't accomodate the special needs of things like temperature arithmetic; a full handling of this case would need the existence of both degree_X units (where X is the temperature scale of your choice) and X_degree units for representing absolute temperatures and temperature differences respectively, as well as appropriate conversion rules. This may also impact other unit definitions (and could also be used to implement weird things like units for C pointer arithmetic :-)). If you need to perform temperature arithmetic, then the Equation Library card functions TINC and TDELTA represent one way of handling this "problem". - (*) If you don't see why the operation is meaningless, imagine replacing temperatures with dates (another absolute scale); now add January 12th, 1989 to August 28th, 1993. What could the answer be? And what could it mean? -- Frank Wales, Grep Limited, [frank@grep.co.uk<->uunet!grep!frank] Kirkfields Business Centre, Kirk Lane, LEEDS, UK, LS19 7LX. (+44) 532 500303
Jake-S@cup.portal.com (Jake G Schwartz) (01/25/91)
The temperature conversion issue in the HP48 was explained rather well at last June's Chicago HP handheld user's conference by Bill Wickes and Eric Vogel. The videotapes of that conference are still available, as usual. We're up to 69 copies sent out at present. Jake Schwartz
akcs.joehorn@hpcvbbs.UUCP (Joseph K. Horn) (01/25/91)
Lines: 13 Seems like the old "temperature bug" is being revisited. For those of you who didn't follow this thread shortly after the 48 came out, just remember this: If you have 75 degrees Fahrenheit, and you want to add 5 more degrees, all you have to press is 75 F, 5 R, SWAP, +. Similarly, if you want to do Celsius (Centigrade) arithmetic, make the second number Kelvin, then swap and do whatever you wanna do. This even makes sense, as per others' observations already posted. -- Joseph K. Horn -- (714) 858-0920 -- Peripheral Vision, Ltd. --
silvert@cs.dal.ca (Bill Silvert) (01/25/91)
In article <38439@cup.portal.com> Jake-S@cup.portal.com (Jake G Schwartz) writes: >The temperature conversion issue in the HP48 was explained rather well at >last June's Chicago HP handheld user's conference by Bill Wickes and Eric >Vogel. The videotapes of that conference are still available, as usual. >We're up to 69 copies sent out at present. Does it explain why 0C+0C=273.15C (this I understand, addition is in absolute scale) but 2*0C=0C (why not convert to absolute first)? The problem with this discrepancy is that it is not possible to average two temperatures to find the midpoint unless you convert to a common scale first. I therefore feel that the inconsistency is a bug. -- William Silvert, Habitat Ecology Division, Bedford Inst. of Oceanography P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2. Tel. (902)426-1577 UUCP=..!{uunet|watmath}!dalcs!biomel!bill BITNET=bill%biomel%dalcs@dalac InterNet=bill%biomel@cs.dal.ca
E7P2BAT%TOE.TOWSON.EDU@CUNYVM.CUNY.EDU (Harry E. Bates) (01/26/91)
Where can the video tapes of the Chicago HP handheld user's conference be obtained? Thanks! Harry E. Bates Towson State University HBATES@TOWSONVX.BITNET
jsims@vuse.vanderbilt.edu (J. Robert Sims) (01/27/91)
Please note that all the temperature conversion behavior is documented in the manual. It clearly states that in cases of multiplication the additive offset is ignored, and in cases of addition it adds the absolute temperatures. The conversions are quite logical, and behave as documented. As someone mentioned earlier, there should be a Celsius degree in addition to degrees Celsius. If you need the Celsius degree in your calculations, define a unit yourself. 1 Celcius degree = 1 Kelvin. Simple as that. 0_^C + 0_^C = 273.15_^C 0_C^ + 0_^C = 0_^C It's probably easier just to use the K directly. My major complaint about the units management is its lack of support for fractional exponents and _its lack of documentation in the manual_. I did find brief mention of it in the EQ Lib card, however. I spent hours trying to find the problem with one of my programs, including a couple of calls to HP. It was a design decision to have this limitation, but it should have been in the manual. Rob jsims@vuse.vanderbilt.edu
egraeler@uceng.UC.EDU (Eric S Graeler) (01/27/91)
In article <2D042619A000047E@gacvx1.gac.edu>, U5513172@NMSUVM1.BITNET (Thomas L. Baca) writes: > Hello fellow HP-users, I have what may be a silly question, but I > shall ask anyway. WhenI add 25_'C ( read degree Celcius) and 1_'F > on my HP 48SX, I get an answer of 537.67_'F. What terribly simple thing > am I overlooking? My wife tells me that the answer looks like Rankins There is nothing wrong with this. This is perfectly correct. What you are over looking is absolute 0. Temperature starts at 0_K which is equall to -459.67_F. Therefore 25_C is 298.15_K and 1_F is 255.93_K for a total of 554.08_K which is 537.67_F. You can do these conversions easily by have the number you want to convert on the stack and then press the left shift (orange) and then the units you want. Eric Graeler egraeler@uceng.uc.edu