HP Forums
Too hot for Prime? - Printable Version

+- HP Forums (https://archived.hpcalc.org/museumforum)
+-- Forum: HP Museum Forums (https://archived.hpcalc.org/museumforum/forum-1.html)
+--- Forum: Old HP Forum Archives (https://archived.hpcalc.org/museumforum/forum-2.html)
+--- Thread: Too hot for Prime? (/thread-252391.html)



Too hot for Prime? - calculatethis - 10-07-2013

Hello,

Am I doing something wrong here, but in either RPN or textbook modes I can't add 20 degrees celcius to 30 degree farenheight.

Typing the following

20 [SHIFT] [UNITS] [Select Temperature] [Select degrees C]
ENTER (put number on stack)

30 [SHIFT] [UNITS] [Select Temperature] [Select degrees F]
ENTER (put number on stack)

Then try to add [+]

Calculator returns

Error in RPN mode

or "Error: Invalid input" in texbook mode.

Perhaps I'm not doing something correct or need something reset somewhere?

Many thanks in advance,


Re: Too hot for Prime? - Paul Dale - 10-07-2013

Not having a prime, I'm probably speaking untruths and errants here.

How do you add two temperatures?

1oC + 1oC = 2oC ?
Sound easy? A temperature plus a change in temperature.

Should we convert to the base unit and add and convert back? 1oC + 1oC = 274.15K + 274.15K = 548.30K = 275.15oC. At least both are temperatures here, rather than a temperature and a change in temperature.

Which is correct? Are they both?

Now add oF to the mess. Should we convert to oC, K or the even Rankine scale?


The 48 series device had difficulties with addition of temperatures and it seems the prime is just banning them.


- Pauli


Re: Too hot for Prime? - Michael de Estrada - 10-07-2013

My HP 50g gives the message "+ Error: Inconsistent Units"

However, if I convert the entries to consistent units, it does work.

For example: 20 degC 30 degF > -1.11 degC + > 18.89 degC

You can, however, multiply different units to obtain compound units:

20 degC 30 degF x > 600 (degC x degF)

Which is nonsense in this case, but can be useful in cases such as:

Torque = distance x force

20 ft 30 lbf x > 600 (ft x lbf)

Edited: 7 Oct 2013, 9:50 a.m.


Re: Too hot for Prime? - calculatethis - 10-07-2013

Hello,

Many thanks.

The issue is that in the Prime you can't really (that I can see) do conversion between units in the same (old familiar) method as in the 50g.

So to convert between units you basically add two units together and the result is in the format of the first one.

An example:

Convert 15 gallons to litres

Add 0 litres to 15 gallons and the Prime responds with 68.88 litres. UK Gallons BTW.

This works (to me it's a hack, really and far from intuative) with volumes but not at all for temperature

Unless I've missed something obvious here...

Many thanks,




Re: Too hot for Prime? - Tim Wessman - 10-07-2013

Quote:
Not having a prime, I'm probably speaking untruths and errants here.

No, you are speaking complete truth. As you've so concisely stated, it seems a trivial thing but actually can introduce very serious and wrong calculation errors. This is a very difficult problem that has no real solution and was "solved" by just not allowing it. The nspire series does exactly the same thing.

I never really understood back in chemistry why the professors we always so insistent on "you must convert all temperatures first before starting". When I finally understood the underlying problem with C and F, it all finally made sense.

TW


Re: Too hot for Prime? - Tim Wessman - 10-07-2013

You can use either the convert command, or the sto operator to directly convert.

15_galUK STO _l, or else CONVERT(unit,unit)

Yes, I agree units is not as nice for simple conversions as the 48 series (yet), nor has it surpassed it for more complex conversions and operations (yet).

TW


Re: Too hot for Prime? - calculatethis - 10-07-2013

Quote:
15_galUK STO _l, or else CONVERT(unit,unit)

TW


Thanks Tim. "15_galUK STO _l" this works for me in textbook but is this possible in RPN? No STO softkey is showing in RPN mode.

Many thanks,




Re: Too hot for Prime? - Manolo Sobrino - 10-07-2013

As for the Nspire, temperature conversions are done with the same commands of the TI 89. tmpCnv() for temperature values, and ΔtmpCnv() for temperature intervals. There's the conundrum, there are two possibilities for a temperature number depending on your intention, so the ordinary procedure (_unit1 ► _unit2) would be ambiguous, that's why there is an "Inconsistent units" error if you try. As they won't let you use the relational operators on numbers with units, that's the end of the problem.



For the sake of comparison, the 50g has the cryptic TINC and TDELTA commands. For the relational operators, numbers are assumed to be temperature values if they just have simple temperature units, otherwise they are treated as temperature intervals. Arithmetic operators assume them to be temperature intervals... Yeah, it is tricky.


Re: Too hot for Prime? - Tim Wessman - 10-07-2013

Well dang... :-(

TW


Re: Too hot for Prime? - cyrille de Brébisson - 10-08-2013

Hello,

As Pauli said, °C and °F are NOT units...
Units, by definition start at 0 and "define" the 1.
°C and °F do NOT start at 0, which is what is making them "not units".

However, since everyone is so used to think that they are units... confusion occurs...

Prime does handle °C and °K like units in some ways, it does have some special cases to allow you to convert to and from them, but not every unit functionality works with them for this exat reason.

use °K or °R :-)

Cyrille


Re: Too hot for Prime? - Marcus von Cube, Germany - 10-08-2013

The Prime should be able to add K to °C but refuses to do so (at least from the CAS screen which I chose because it does not use RPN which is my default elsewhere).

I tried in home, too. The message isn't "incompatible units" as in CAS but "illegal input". Alas, the result is the same. :-(

Edited: 8 Oct 2013, 5:25 a.m.


Re: Too hot for Prime? - Walter B - 10-08-2013

Quote:
use °K or °R

Shouldn't it read: use K ?

Since the interval 1 K equals 1 °C, a tolerant Prime could support adding x °C to y K. And it must be tolerant since it supports °F and similar Imperial stuff two centuries after SI was invented, doesn't it?

d;-)

Edited: 8 Oct 2013, 3:47 p.m.