# Large formula in align - 2

Good Evening

I'm new to Latex. I am writing the decomposition of an equation about turbulent motion. I have problems with alignment on the left and with numbering of the equations. I do not understand why the alignment direction can not be chosen in the align method. maybe someone can help me to understand what I'm neglecting? For the last two examples I followed the indications of this post

Large formula in align

I have used various conditions, I put the three codes with the results:

\begin{align} % N-S Decomposition 1
\bar{u}_{i,t}+u^{\prime}_{i,t} +{({\bar{u}}_{j}+{u^{\prime}_{j}})}({\bar{u}}_{i}+u^{\prime}_{i})_{,j} =\\
\nonumber
-\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}^{\prime}){g^{\prime}}z_{,i}
-\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}^{\prime}_{,i})
+2\nu({\bar{e}_{ij}+e^{\prime}_{ij})_{,j}}
\end{align}


\stackMath
\def\stackalignment{l}
$$\stackanchor[10pt]{\displaystyle% \bar{u}_{i,t}+u^{\prime}_{i,t} +{({\bar{u}}_{j}+{u^{\prime}_{j}})}({\bar{u}}_{i}+u^{\prime}_{i})_{,j} =\\ }{\quad\quad\quad\quad\quad\displaystyle -\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}^{\prime}){g^{\prime}}z_{,i} -\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}^{\prime}_{,i}) +2\nu({\bar{e}_{ij}+e^{\prime}_{ij})_{,j}} }$$


in this code, everything is marked in red as if it did not recognize the code, but it also creates the pdf

\stackMath
\def\stackalignment{l}
\stackanchor[10pt]{\displaystyle%
\overline{u^{\prime}_{i}u^{\prime}_{i,t}}
+\overline{u^{\prime}_{i}\bar{u}_{j}u^{\prime}_{i,j}}
+\overline{u^{\prime}_{i}u^{\prime}_{j}u^{\prime}_{i}\bar{u}_{i,j}}
+\overline{u^{\prime}_{i}u^{\prime}_{j}u^{\prime}_{i,j}} =
-\frac{1}{\rho_{0}}\overline{\rho^{\prime}u^{\prime}_{i}}{g^{\prime}}z_{,i}
-\frac{1}{\rho_{0}}\overline{p^{\prime}_{,i}u^{\prime}_{i}}
+2\nu\overline{{{e}^{\prime}_{ij,j}}u^{\prime}_{i}}
-\overline{\overline{(u^{\prime}_{i}u^{\prime}_{j})}_{,j} u^{\prime}_{i}}
} \\


for me the red color code is set for: command not recognized

%----------------------------------------------------------

I can not give an answer individually because I find all your solutions very interesting, very eustive. this afternoon I will look at the code more calmly and recompile. if I find any difficulty I will write. I would like to request information from the third code. this belongs to the usepackage {stackengine} library. I would like to know why latex tells me: command not recognized

I solved the problem of numbering. is bound to\numberwithin{equation}{chapter} .I found this command interesting and I will try to use it.

• the red isn't a warning from latex that is just that your editor doesn't know of that package so it isn't a problem. unrelated but you do not need to write u^{\prime} simply u' will make identical output – David Carlisle Feb 22 '18 at 18:10
• it isn't clear what alignment you want but align should only be used if you want to align points (marked with &) on the the different rows. – David Carlisle Feb 22 '18 at 18:11
• @DavidCarlisle I want the formula to be left-aligned. I would like to understand if the second code is correct – Antonio Feb 22 '18 at 18:18
• if you want all equations left aligned just use [fleqn] option to your documentclass (it is better if you always post examples as complete documents that can be tested rather than fragments as above) – David Carlisle Feb 22 '18 at 19:14
• if you want to ask if some code is correct it is best to post a test docuemnt (which for example would show any packages assumed) the markup in both cases is far more complicated than needed, but it isn't really clear what layout you want – David Carlisle Feb 22 '18 at 19:54

Here are three possibilities:

\documentclass{article}
\usepackage{mathtools}

\begin{document}

\begin{align} % N-S Decomposition 1
& \bar{u}_{i,t}+u^{\prime}_{i,t} +{({\bar{u}}_{j}+{u^{\prime}_{j}})}({\bar{u}}_{i}+u^{\prime}_{i})_{,j} =\\
\nonumber
& {-}\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}^{\prime}){g^{\prime}}z_{,i}
-\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}^{\prime}_{,i})
+2\nu({\bar{e}_{ij}+e^{\prime}_{ij})_{,j}}
\end{align}
\bigskip

\begin{aligned} % N-S Decomposition 1 & \bar{u}_{i,t}+u^{\prime}_{i,t} +{({\bar{u}}_{j}+{u^{\prime}_{j}})}({\bar{u}}_{i}+u^{\prime}_{i})_{,j} =\\ & {-}\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}^{\prime}){g^{\prime}}z_{,i} -\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}^{\prime}_{,i}) +2\nu({\bar{e}_{ij}+e^{\prime}_{ij})_{,j}} \end{aligned}
\bigskip

$$\begin{multlined} % N-S Decomposition 1 \bar{u}_{i,t}+u^{\prime}_{i,t} +{({\bar{u}}_{j}+{u^{\prime}_{j}})}({\bar{u}}_{i}+u^{\prime}_{i})_{,j} =\\ {-}\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}^{\prime}){g^{\prime}}z_{,i} -\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}^{\prime}_{,i}) +2\nu({\bar{e}_{ij}+e^{\prime}_{ij})_{,j}} \end{multlined}$$

\end{document}


only first equation:

with using split environment:

\documentclass{article}
\usepackage{amsmath}

\begin{document}
$$% N-S Decomposition 1 \begin{split} & \bar{u}_{i,t}+u'_{i,t} +{({\bar{u}}_{j}+{u'_{j}})}({\bar{u}}_{i}+u'_{i})_{,j} = \\ & {-}\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}'){g'}z_{,i} -\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}'_{,i}) +2\nu({\bar{e}_{ij}+e'_{ij})_{,j}} \end{split}$$
\end{document}

• @Mico, thank you . corrected. difference is evident :-) – Zarko Feb 23 '18 at 11:49

The first two-line formula below reproduces your first align example, while (a) adding alignment points (&) at the start of each line and (b) getting rid of quite a few redundant pairs of curly braces and replacing all instances of ^{\prime} with '. Note that I encased the first - (minus) symbol of the second line in curly braces, so that LaTeX treats it as a unary rather than as a binary operator.

The second formula applies three minor tweaks: the = symbol is moved to the start of the second line, the second line is indented relative to the first (via a \quad directive), and the vertical separation between the lines is increased slightly. Observe that because the second line now starts with a = symbol, it's no longer necessary to do anything special with the subsequent - symbol in order for it to be treated as a unary operator.

\documentclass{article}
\usepackage{amsmath} % for 'align' environment
\begin{document}

\begin{align}
&\bar{u}_{i,t}+u'_{i,t} +(\bar{u}_{j}+u'_{j})(\bar{u}_{i}+u'_{i})_{,j} =\\
&{-}\frac{1}{\rho_{0}}(\bar{\tilde{\rho}}+\tilde{\rho}')g'z_{,i}
-\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}'_{,i})
+2\nu(\bar{e}_{ij}+e'_{ij})_{,j}  \nonumber
\end{align}

\begin{align}
&\bar{u}_{i,t}+u'_{i,t} +(\bar{u}_{j}+u'_{j})(\bar{u}_{i}+u'_{i})_{,j} \\[1ex]
-\frac{1}{\rho_{0}}(\bar{\tilde{p}}_{,i}+\tilde{p}'_{,i})
+2\nu(\bar{e}_{ij}+e'_{ij})_{,j}  \nonumber
\end{align}

\end{document}

• \multline could also be an option. – user94293 Feb 22 '18 at 18:37
• @user94293 - multline might be an option for other two-line equations. However, for the present one, using a multline environment would produce a fairly substandard result. (Give it a try if you're skeptical about this claim.) – Mico Feb 22 '18 at 19:30
• @Mico I had read that for a better compilation it was more appropriate to use {prime}. objectively I find ugly _ {\ tilde} _ {prime}, the first graphically should be smaller and more detached from the symbol. if the file is printed in small font, the job does not come out good. – Antonio Feb 23 '18 at 8:02
• @Antonio - I don't know where you read that "for a better compilation it was more appropriate to use ^{]prime}". It's simply wrong. There's a lot of TeX-related disinformation (as well as all kinds of other disinformation, of course!) out there on the Internet. – Mico Feb 23 '18 at 8:14
• @Mico - you are right. much misinformation. for this reason the best solution is to talk with you all. Thank you – Antonio Feb 23 '18 at 8:17

I can not give an answer individually because I find all your solutions very interesting, very exhaustive. This afternoon I will look at the code more calmly and recompile. Iif I find any difficulty I will write. I would like to request information from the third code. This belongs to the \usepackage{stackengine} library. I would like to know why latex tells me: command not recognized.