Custom Unit Stream Normalization Error

[jasper]

This looks like it might be a symptom of the same problem. Can I see the fsd?

[jasper]

I - I believe you have no fsd available that does not contain sensitive information.

My comments merely suggested that I could not reproduce it. As I have no clues to the nature of the problem at this point, I am not sure what to suggest next, but in general I need to be able to reproduce a problem to be able to fix it (unless for obvious coding errors).

[nrgeng]

I could email the shareable fsd file for the overall flow sheet that solved. It is structured identically as the other sensitive flow sheet. You might use it to verify the stream array element structure taking note that the last two compositions did not transfer correctly giving the "non-normalized" error. Would this fsd file be of benefit to you?

[jasper]

Does in that case the composition not transfer properly at any point? Does it allow me to reproduce the (underlying) problem?

[nrgeng]

The shareable fsd file solves so the compositions must transfer properly. The error is not reproduced in the shareable fsd file. My point was that the sub-flow sheet structure is the same. So I will accept your answer that the shareable fsd file would not be useful to you. Thought that I would give it a try since the Custom Units function in this instance.

I unraveled the sub-flow sheets in stages in what follows.

Test 1
moved sub 1 into sub 2 flow sheet
now new sub 2 flow sheet
connected streams
result was same error message as when
subs were two levels deep
see below

error: calculate failed for unit OEA Module System:
Failed to flash internal material: CalcEquilibrium failed:
Failed to wrap v1.0 CalcEquilibrium of type "TP" to v.1.1:
CalcEquilibrium failed: CalcEquilibrium failed: TP flash failed:
composition not set for one or more compounds

Test 2
moved new sub 2 into overall flow sheet
sub 1 & sub 2 are both at top level now
(no Custom Units are employed)
connected streams
No error messages
the flow sheet solved without any assistance

During this effort, I eliminated the Custom Units and successfully solved the flow sheet. I hope someone can help determine the problem(s) with the Custom Unit because I would like to use it but cannot for this flow sheet project. I will try to use the Custom Unit again in the future. Please inform me of any developments. Thanks for the help.

[nrgeng]

I may have found the problem. Two sub-flow sheets operate in series. It seems that the first sub-flow sheet produces an output material stream whose component mole fractions do not sum to 1.0. Could this be true? Is this the solution to the problem? See original error message repeated below.

solving Sub-Flow Sheet 2
error: calculate failed for unit Sub-Flow Sheet 2:
compositions for stream 15x are not normalized
incomplete specification for input stream 15x

Stream 15x is the feed stream to the second sub-flow sheet for which the error occurs. If this is true, what can be done to correct it?

Upon further investigation, the output of the flow sheet is a material stream whose component mole fractions do not sum to 1.0. This clears the Custom Unit because the problem is with the Units of the flow sheet wrapped within the Custom Unit.

Output Dew Point (C): 61
Summation of Mole Fractions: 0.9999999980
Status: Solved

Output Dew Point (C): 62
Summation of Mole Fractions: 0.0.9999999959
Status: Error

How many decimal places are being checked by each Unit's mass balance? What can be done to correct this problem so that I can finish the project?

[jasper]

Does the unit that produces the non-normalized composition (in the embedded embedded flowsheet) not give an error on calculation?

What kind of a unit is it? Why does it not produce normal compositions?

[nrgeng]

Does the unit that produces the non-normalized composition (in the embedded embedded flowsheet) not give an error on calculation?

The Unit does not generate an error message on calculation.

What kind of a unit is it? Why does it not produce normal compositions?

I am wondering why the Unit produces abnormal compositions too. The Unit's mass balance should check that.

I have spent time in creating a shareable fsd file and an accompanying spreadsheet which calculates the sum of the component mole fractions for the Stream in question for two Conditions. Condition 1, a PDP of 61 C, solves, but Condition 2, a PDP of 62 C, fails to solve giving the noted error message. The Stream information matches that given in Posting 9. I have emailed the fsd file to you. Hope that it is useful in determining the problem.

solving Sub-Flow Sheet 2
error: calculate failed for unit Sub-Flow Sheet 2:
compositions for stream 15x are not normalized
incomplete specification for input stream 15x

I hope that this information leads to the discovery of the problem. Of course, the problem my be due to a configration error made by me. Whatever it is its discovery is hopefully imminent. Thanks for your help.

[jasper]

I do not see that; both the "old" and "new" COFE seem to solve this file fine, both case 1 and case 2. But I am a bit confused, as both cases are already solved when I open the file, and it indicates in the log (when opening the file) that there was no problem solving. How do I reproduce the problem?

Also the composition for stream 15(1) is normalized.

[jasper]

I may have found the problem. Two sub-flow sheets operate in series. It seems that the first sub-flow sheet produces an output material stream whose component mole fractions do not sum to 1.0. Could this be true? Is this the solution to the problem? See original error message repeated below.

solving Sub-Flow Sheet 2
error: calculate failed for unit Sub-Flow Sheet 2:
compositions for stream 15x are not normalized
incomplete specification for input stream 15x

Stream 15x is the feed stream to the second sub-flow sheet for which the error occurs. If this is true, what can be done to correct it?

Upon further investigation, the output of the flow sheet is a material stream whose component mole fractions do not sum to 1.0. This clears the Custom Unit because the problem is with the Units of the flow sheet wrapped within the Custom Unit.

Output Dew Point (C): 61
Summation of Mole Fractions: 0.9999999980
Status: Solved

Output Dew Point (C): 62
Summation of Mole Fractions: 0.0.9999999959
Status: Error

How many decimal places are being checked by each Unit's mass balance? What can be done to correct this problem so that I can finish the project?

The "compositions are not normalized" error originates from validation, and the tolerance is the "Mass balance relative tolerance" in the flowsheet preferences. Perhaps you have set this tolerance really low on one of your documents?