Talk:Process window index/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

I found this article interesting, but I do have one serious concern that I'd like to see addressed before reviewing it in detail. I'm worried about what appear to be chunks of text copied almost verbatim from third-party sources. A couple of examples that jumped out at me:


 * The first part of the lead is a copy of what is found here.


 * The example above may plausibly be because that site copied the wikipedia entry, without attribution (?), but most of the middle paragraph of PWI in electronics manufacturing appears to be a more or less word-for-word copy of what's found here.

A couple of other initial points:


 * It would be clearer to use the term "mean" or "arithmetic mean" instead of average in the sentence "The control limits are set at three standard deviations on either side of the process average."


 * It needs to be mentioned somewhere that the raw data used to make up the PWI is normalised to a range of 0–100, to make sense of the bullseye graphic description.

I'm placing this article on hold before completing the review, so that I can be reassured about the possible copyright problem.

--Malleus Fatuorum 15:48, 30 December 2008 (UTC)


 * Thanks Malleus! I've just shot an email using the boilerplate templates to KIC Thermal asking them if they could release that work under the GFDL license. . In the meantime I'll try copyediting the paragraph to make it less of a copyvio. I've also tried to make those textual changes you suggested, but I will wait for further response. Kind regards, Zithan (talk) 18:36, 2 January 2009 (UTC)


 * OK, that's great. There's no rush. --Malleus Fatuorum 21:17, 2 January 2009 (UTC)

Received permission for the kicthermal.com content at 2325182. Stifle (talk) 21:23, 5 January 2009 (UTC)


 * I don't have an OTRS account, but obviously I'm happy to take your word for that. I'll try to finish off this review tomorrow, with luck. --Malleus Fatuorum 23:10, 5 January 2009 (UTC)
 * Yes, I managed to successfully secure their work under the GFDL 1.3 license. I think it should be good enough to resolve all copyright issues. Good luck with the review. Regards, Zithan (talk) 13:53, 6 January 2009 (UTC)

Continuing with the review ...

 * "Thermal process", which is redlinked in the lead, needs to be explained. Could just be as simple as saying that it's a manufacturing process which involves heating and cooling, but something needs to be said. Nuclear fusion is a thermal process, for instance. As is snowballs melting in Hell.


 * In the chart in Statistical process control, LSL and USL aren't labelled. Is that deliberate for some reason?


 * "... a thermal profile is a complex set of time-temperature values for a variety of processes such as slope, thermal soak, reflow, and peak." What's complex about a set of time-temperature readings? I think these processes need just enough explanation so that readers don't have to follow the link to understand what's being said.


 * The bullseye graphic is clipped on the righthand side. Also, it displays peak, soak, slope and TAL. As opposed to the "slope, thermal soak, reflow, and peak" mentioned above. What's TAL? What are TC #1, TC #2, and TC #3 showing?


 * Where else is PWI used, apart from electronics manufacturing? Is it only used for processes involving soldering? The statement "The electronics manufacturing industry has developed a specialized specification limit ..." would seem to indicate that PWI's are only used in electronics manufacturing. Is that right?


 * "The Process Window Index for a complete set of thermal profile statistics ...". What's the meaning of "complete"? Are there PWIs, for incomplete sets as well? Would "thermal profile data" not be better? What's the statistic?


 * "The Process Window Index for a complete set of thermal profile statistics is calculated as the worst case (ie highest number) in the set of statistics." The raw data is an array of time-temperature measurements. What does "highest number" in such an array mean?


 * "The formula for PWI is calculated as follows ..." It isn't the formula that's calculated, it's the PWI.


 * "Each thermal profile is ranked on how it 'fits' in a process window ...". Why is "fits" in quotes? The concept of a "process window" has not been explained.


 * The lead needs to make the subject more accessible to the non-specialist reader.


 * "... provides a quantifiable and reproducible measure of how well a thermal profile performs relative to critical process limits...". Thermal profile has not yet been defined, but when it is in the following section it is described as "a complex set of time-temperature values", not as a process that can "perform".


 * "To avoid the possibility of thermal shock affecting the output, the steepest slope in the thermal profile must be determined and leveled." The word "output" here is being used in a rather unclear way. Is this the output from the manufacturing process? Because of the previous sentence it looks like it might be the output of the computation.

I think there's still significant work to be done on this article, so I'm leaving it on hold.

--Malleus Fatuorum 20:38, 7 January 2009 (UTC)

Thanks for the review Malleus. It's a good review, and some of the questions raised are not very easy to answer.
 * 1) I'm not sure why USL and LSL are not marked. The image has been created by someone else. Let me know if it's necessary to label it; I can then make the suitable modifications. (It's PD)
 * 2) To answer your next question on the electronics industry: Well theoretically, it can be modified for any thermal process. However, as far as I know, PWI is currently used in the exclusively in electronics manufacturing industry. Any electronics manufacture requires soldering work to be done&mdash;solderless electronics are not available&mdash;therefore its use in soldering. PWI.svg
 * 3) Soldering is not a single stage process. There are several zones (preheat, soak, reflow, cooling) that control the rate of heating. For example, too much heat at one stage could lead to oxidation of flux, and you don't want that to happen! Take a look at the graph alongside. The graph alongside is the time-temperature values that you need to follow when you "bake" the soldering job in an oven. There are several variables that come into play: these include the thermal mass of the PCB, thermal mass (or loading) of the oven (ie the more items you put in an oven the more mass &mdash; no mass gives you a hotter oven; conversely, greater mass, a cooler oven.) Also, heat transfer inside the oven (how heat is moved around the zones), the stability & tolerance of the oven (older, cheaper, and broken ovens may function outside of tolerance values) play a major role. To keep all these variables in check, a indexed "process window" was invented that is used to check the slope of the graph at all stages.
 * 4) I hope this is not too heavy, I can go on if required. In the meantime, I will work on modifying the changes as suggested. Zithan (talk) 19:26, 12 January 2009 (UTC)
 * It's not too heavy at all. I think that I understand the process, but I'd like our description of it to be more accessible to the general reader. Don't be concerned about having to get things done quickly; I'm quite happy to leave the review open for as long as it takes while you're still working on the article. --Malleus Fatuorum 19:58, 12 January 2009 (UTC)
 * Hi Malleus, could you please review it again? I tried to fix most of the review comments. Thanks Zithan (talk) 11:36, 22 January 2009 (UTC)
 * OK, I'll take a final look through later, and then hopefully we'll be able to wrap this review up. --Malleus Fatuorum 13:21, 22 January 2009 (UTC)


 * I'm afraid that I'm going to have to close this review as "not listed". From the lead alone, the statement that "In the electronics industry, these specification limits are known as the process window, and values that a plotted inside or outside this window are known as the process window index" is inconsistent with the later description of how to calculate the PWI. It's the worst case, so there can't be any value outside the window. The prose still needs attention too, as in "PWI values are used to calibrate the heating and cooling of soldering jobs (known as a thermal profile) while baked in a reflow oven." Why are we baking PWI values in a reflow oven? While I congratulate you on your efforts to date, the article still needs some significant work to reach the good article criteria, in my opinion. If you don't agree with my decision then you can appeal at WP:GAR. Thanks for the work you've done so far. --Malleus Fatuorum 21:33, 23 January 2009 (UTC)