Oh, a wise guy, eh? (said like Curly from "The Three Stooges") 
I've had a hidden area at the top of my worksheets for years and years that has my odd units included - hanks, barleycorns, smidgens, firkens, furlongs, fornights, fathoms, hands, cubits, hogsheads and so on. I'll occaionally prank my co-workers with them. I also have useful definitions there, but they've been gradually diminshed with time as Mathcad includes more and more of them.
And your chart brought back wonderful memories of MAD Magazine. I still use "furshluginner" as a descriptive adjective, i.e. "aw, the furshlugginer thing crashed again!"
As to Victor's suggestion below, I don't know how Mathcad can manage to be all things to all people, as so many unit names, abbrevaitions and conventions overlap. Perhaps all temperature units could be two-character, including the degrees symbol (Alt+0176 in English). I don't know how that would wash with the rest of the world.
Best,
Tom
Richard Jackson wrote:
1) As an option, allow us to force Mathcad to track angles and solid angles as real units that can't just disappear from the result.
2) Allow user defined dimensions.
Valery Ochkov wrote:
I must use in my thermal calculations units with same name but it is units of different physical quantilities - kg of steam and kg of air.
Now I do so kgSteam:=kg. kgAir:=cd
It will be good to have more correct tools for this.
See this http://communities.ptc.com/message/52543#52543 (and many, many others 
) for a recap on quantities, dimensions and units.
What I think you mean, Richard, for item 2 on your list is that we should have user-defined base quantities.
Item 1 of your list, might possibly be met by allowing user-defined quantities; it could certainly be met by introducing angle and solid angle as base quantities and then defining a new (to most people) set of relationships that deal with the length/angle, etc that would necessarily require explicit conversion from ratio of lengths to angles.
With varying amounts of work ... sigh ..., either method might solve your problem, Valery, depending upon how you want to use and what equations would have to change to accomodate the distinction (I'm not familiar with the units).
If I seem a bit picky about the terminology (and I confess to having the odd lapse or two myself), it's because there is already a well-documented and established set of metrology terminology (ie, the VIM) and Mathcad should be extended in a way that is consistent with that terminology (see http://communities.ptc.com/message/53940#53940 for a previous comment on this or http://communities.ptc.com/message/53884#53884 in the same thread).
Stuart
What I think you mean, Richard, for item 2 on your list is that we should have user-defined base quantities.
If I seem a bit picky about the terminology (and I confess to having the odd lapse or two myself), it's because there is already a well-documented and established set of metrology terminology (ie, the VIM) and Mathcad should be extended in a way that is consistent with that terminology
You are right. Got to stick to the rules
Item 1 of your list, might possibly be met by allowing user-defined quantities; it could certainly be met by introducing angle and solid angle as base quantities and then defining a new (to most people) set of relationships that deal with the length/angle, etc that would necessarily require explicit conversion from ratio of lengths to angles.
I personally would have no problem having two different types of angles / solid angles, even in one worksheet. If I Mathcad want to track it, I would use one type. If I want an angle to represent a ratio, I would use another type. So if we had user defined quantities I would just define solid angle (the one that really affects me) as a base quantity. Nothing else required.
Philip Oakley wrote:
You may not be surprised, but I'd like:
- Angle dimension (nominally the ratio of Length to Length, as in '3d geometry'), with the option that trig functions recognise them.
- Solid angle as an indicator (in terms of linear powers the dimensionally of solid angle is Angle squared, but see '3d' above!), so that it can't disappear without the user noticing. Trig functions would refuse solid angle. (actually the 'solid angle' here could be either left handed or right handed, I'd use the latter based on current convention)
Unfortunately, that is one of the difficulties with angle in that, by sticking to the standard definition, it is impossible to distinguish between the ratio of 2 co-linear lengths and the ratio of subtended arc-length to radius (angle). IMO, the standard definition completely misses the mark in that it doesn't take account of the fact that angle effectively measures the non-colinearity of lines and is a feature of multi-dimensional spaces (well, OK, in 1-space I suppose it has values of n·π). Angle should be a base quantity with equations modified to take account of its existence rather than making assumptions about it being ratios of lengths. (As I've stated in other messages, I don't see any particular problem with using trig (or any other functions) with any quantities ... IMO, they're just functions of some kind that operate on the numerical values alone, not on the units)
Solid angle has similar problems, as do higher dimensional forms.
User defined Dimensions (at least three), which allow the user to denominate a result as being in a specific reference frame. An example is the distinction between input voltage referred measurements and output voltage referred measurements for an electrical amplifier. (I'm sure the same can be applied to the Money unit with "Now", and "Then" dollars when comparing costs).
"Dimensions"? 
See http://communities.ptc.com/message/185618#185618 ("Dimensions" indeed .. mutter ... grumble)
I´d like to have the possibility to define my own units based on other predefined units like it was possible in mathcad 15:
Could not find another thread on Prime 3.0
Original road map had 3.0 being fully capable up to MathCAD 15 functionality. Is this still valid? Feel free to move to a more appropriate thread if necessary 
