[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: level of effort for mass balance?

Under one of the RCRA re-authorizations, Congress required the National 
Academy of Engineering to prepare a report that answers this very 
question.  The results of the report (I cannot remember the year) were very 
interesting.  They recommended that the US EPA should NOT require mas 
balances.  This is where they derived the term "resource accounting."  Mass 
balances require many measurements to be taken so there is an exact balance 
prepared.  Some of the sampling is very expensive and could take a month or 
more to perform and verify at each point in the process.  There was a mass 
balance performed on the Amoco refinery (it is a small refinery) for the 
EPA project and the cost was over $1 million.  They had to design, test, 
calibrate and use a special flux chamber to make measurements on the 
dissolve air filtration system.  I believe the cost for three readings on 
this one unit process was about $250,000.  It was money well spent, because 
they showed the actual benzene emissions represented a 750% reduction from 
what the API estimate was.

A colleague of mine is certified for the use of the UNIDO P2 methodology 
used by the National Cleaner Production Centers.  She tells me that the 
Brazil National Cleaner Production Center is having difficulty finding 
companies to use their services because of the difficulty of doing the 
required mass balance up front.  It cost the company quite a bit of money 
in analytical and personnel costs.  Besides the data needs to be completely 
redone the following year.

I think that if you did a Total Cost Analysis (figuring in company 
personnel time at a fully burdened rate, lost production time, analytical 
costs, engineering costs, and management review) for the complete facility 
that you will find mass balance to be an extremely expensive option - 
orders of magnitude greater in cost than what you are proposing here.

The alternative would be to do what I call, just in time mass balance; 
i.e., only do mass balances around the process map steps that you are 
seeking to improve.  Resource accounting will help determine the flow of 
resources through the operation.  Eco-effizienz in Germany has a great 
method for doing this.  The combination of resource accounting and 
just-in-time mass balance should satisfy even the most devoted chemical, 
industrial or mechanical engineer.  Resource accounting is widely used in 
Europe, Japan and the USA.  I hope they will consider the findings of the 
National Academy of Engineering report.  I am sure it is still available 
through their book store.

Bob Pojasek
Harvard University

>I have been asked by person in the environmental agency here to provide
>estimates about what level of effort is needed to do a mass balance for a
>company.  They are considering making a mass balance analysis a requirement
>for a particular environ regulation.
>Obviously this depends on the number of processes being considered.  I have
>not done process mass balances for several years, not being an engineer, and
>would appreciate any insights.  Here are my basic estimates for a "typical"
>production process.  I am making these pretty generous.  Also I am assuming
>there is some data around already.
>Preparation of process flow diagram:  3 hours
>Collection of input-output data for each step:  1 hour per step in the
>If we estimate that a process has about 5 major steps, this would be 5 hours
>Preparation of the mass balance for key factors:  2 hours
>So roughly I would estimate for a "generic" company, that it would take an
>engineer about a day to prepare a mass balance for one production process.
>Another day to summarize results from all the processes, and maybe another
>day to work out the bugs.
>I realize this is grossly simplified and the world is much more complex.  If
>any of you hotshot engineers can comment on these estimates I will be
>grateful, and also I think the regulators will be interested to know your
>ideas too since mass balance for processes is pretty basic for most P2
>and Happy New Year to all!
>Burt Hamner
>Univ de Pacifico

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
p2tech is hosted by the Great Lakes Information Network:
To unsubscribe from this list: send mail to majordomo@great-lakes.net
with the command 'unsubscribe p2tech' in the body of your message. No
quotes or subject line are required.
About : http://www.great-lakes.net/lists/p2tech/p2tech.info
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *