We need NADCAP certification

A. The easiest way I know of to make SPC meaningless is to start off measuring things because they're "easy". What you want to start with is something MEANINGFUL!

First, take a look at the quality data you (hopefully) have on your chemical processes. Determine which one is in fact causing the most problems. THAT is where you want to start. Now, in that process, what characteristic does the process impart to the product that most determines whether or not it is good. That is what you want to start measuring. The point of this exercise is to determine the variability of the process. In reality, that is all SPC does; it measures the variability of a process' outcome.

Once you've established the variability, you can start to use the data you've collected to reduce the variability through a variety of statistical tools you can get out of the literature.

Software? Why? You don't really need anything more than a good spreadsheet for basic SPC. Software doesn't really become worthwhile until you start using some more advanced things (like Taguchi methods). To start, you basically need to make sure the people involved understand the point of the exercise (and the point should NOT be to make an auditor happy). Making a process perform predictably and consistently good is the goal.

A. If your SPC program is to be effective it must be done manually. The person doing it should be the operator of the process. For your process you'll need individuals and moving range charts.
1) Collect 25 or more data points for each parameter of interest. Determine average. For the data points in in the sequence they were taken determine the difference between adjacent points, ignore signs. Determine average of these differences.
2) With first average add second average times 2.66 to get upper control limit and subtract second average times 2.66 to get lower control limit.
3) Take second average and multiply by 3.65 to get the upper control limit for the range.
4) Draw these lines on a control chart and then follow these simple rules. If a point is outside any of the control limits determine the cause, perform and action the rectify the problem and devise a prevention so it does not reoccur, (CAP). Seven consecutive points above or below the average is also and out of control condition as is seven consecutive points rising or falling.

The mistake most people make is getting the operator to react to out of spec. points. Out of spec. points mean the process is not capable and not fixable by the operator. reacting to these will increase the variability of the process. Leave it to your engineering people to resolve this.

The control charts should be at the operators station posted in such a way that they are easily observable by by others. Peer pressure keeps everyone honest which is why it must be done manually. Out of site out of mind.

If done right this is an extremely effective method of process control.

A. The Appendix A statistical techniques requirements of AC7108 (Nadcap chemical processing) give an implementation period, with a milestone plan done by the initial audit and two years for the first process and additional processes covered at subsequent re-accreditation audits.

Also, it is important (and required) to analyze your nonconformance prior to choosing the first project. You have some freedom here in choosing criteria (part-per-million, scrap or rework costs, customer return data, etc.) The point is not "I think I'll do chem film first", but you do a project that will improve your losses (and therefore of interest to your customers).

Your customers and PRI are not interested if you know how to apply SPC to meet Appendix A, but that you use it as a continuous improvement tool. This is done using the (simple) Pareto analysis and the (perhaps more time consuming) Failure Mode Effect Analysis.

A. As Doug has mentioned, the purpose of Appendix A is to promote continuous improvement. However, in a small shop environment, what it really promotes is companies fabricating data simply to meet the requirement. It's not as simple as finding something that you can improve. It's finding something you can collect #'s on and be able to apply the tools. Eventually one does begin to run out of real problems and just brainstorms to come up with anything they can think of. Complying with this one requirement is the biggest headache with the whole Nadcap process. If it is truly worthy of keeping it should be required by all task groups. This one requirement requires a lot of manpower and time which leads to higher costs. Who does PRI and the Primes think is paying this bill? This requirement puts chem processors at an economic disadvantage over those monitored by different task groups.

A. It's a very big job, Oscar, but start the process either at the NADCAP site at www.pri-network.org/Nadcap,or by retaining a consultant to set you up for it.

A. Hi Joan

The process is not too scary. Ted is right; the place to start is PRI. They don't bite and can be very helpful.

How much work is required depends on how you are currently operating. If you are working exactly to the requirements of your customer specs and recording test results, the answer is very little. Don't forget that the AC7108 requirements are not imposed by PRI. They are collected there by your customers instead of adding them to the specs and have the same authority.

One of the requirements is to carry out your own audit to the 7108 requirements. You will be doing annual audits anyway (I hope) so do this one instead, put right your findings, and you will be in reasonable shape for your first inspection.

Don't be put off by a couple of minor non-conformances.
Major findings usually imply an impact on product quality and if there are any, you should be glad to fix them. Zero findings is an extremely rare event.

You can, of course , employ a consultant but I believe that it is much better to own your own process.

Go for it! Good luck!

A. Hi Oscar,

You must start with downloading the AC7108 checklist (and applicable slash sheets). You can find them on www.eauditnet.com.
Then you must make sure that you are compliant with ALL criteria. The only answers allowed are YES and N/A.

even if you are fully compliant to customer specifications/requirement, then still it is a lot of work. Do not underestimate this, you probably cannot do this on your own. Run through the checklist and find out how compliant you are. Depending on the scope of the audit, I would suggest preparing a year before submitting for the first accreditation. Costs are several thousands of dollars, also depending on scope. The audit will take 3-5 days and is very intensive/detailed.
Then ask for a Nadcap audit. I would advise you to do one of PRI's courses on Nadcap audit preparation. See PRI website for more information.

Good luck.

Best regards
Jeroen

A. I have been through the NADCAP Chemical Audit twice. Contact PRI. Print off the audit, take your procedures and go line by line through each question of the audit and find where in your procedure the question is answered. Then note next to the question the procedure number and the section of the procedure where the answer can be found.
Make sure that you have objective evidence of all parts of the operation, calibration records, solution analysis, records of tank additions, etc. Make sure your operators are following your procedure(s) exactly. The preliminary audit will probably be 5 days the first re-accreditation audit will probably be 3 days. Audit all your operations and operators and make sure what you say is what you do.

Remember any NO answer to a question is a finding.

A. Got to love Nadcap and all that it makes us go through like reading specs cover to cover and actually complying with all the requirements rather than choosing which ones we mostly meet! All kidding aside, we've chosen to SKIP altogether becoming accredited for gold - too many combinations and permutations. My take on it, and we've been audited to the MIL-G-45204 spec by many, many primes, is that your approach using coupons is acceptable. The one thing you may run into is that if the lab you're sending out purity and hardness isn't Nadcap accredited, they won't accept the results. We were sending our gold and silver coupons back to the supplier's lab and it didn't fly with Nadcap. Good luck; your first audit is always torture only to be outdone by the torture you'll experience when 1/2 of your initial responses get bounced back.

Thank you for your words of encouragement :)

Seriously though, thank you for your input. The lab we use is Nadcap accredited so we should be good there.

To make this more challenging, we do not supply any primes directly, so our company is not incredibly familiar with their specific requirements. I would think if a similar practice to ours is accepted by the primes than no doubt Nadcap will accept it without hesitation :sarcasm:.

Thanks again for the input.

A. Hi Richard,

You need to get yourself registered with Nadcap so that you can access their documentation.

Once registered you will have access to the AC checklists (for plating/anodising/conversion coating it will be AC7108, for paint AC7108/1) you will need to carry out a gap analysis (basically audit yourself to the AC checklist), then keep auditing yourself until you are happy all the gaps are closed.

Personally I would hire a Nadcap consultant, they can take a lot of pain and time out of the preparation for a Nadcap audit. A lot of these guys are ex-Nadcap auditors so have a good idea about what the Nadcap auditors will be looking for.

A. Hi Lizzy
Ask PRI.
You will find them very helpful.

A. Hi Lizzy,
The cost depends on many factors. For example, you may need to write a quality manual and document all processes and implement a QMS first and possibly add a staff member to help handle it. Most firms bring in a consultant and if they are traveling and staying in hotels you will pay for it. You have to train all the workers and managers and you may have to buy an ERP software to the tune of 25-100k dollars. I would budget a few hundred thousand or half a million if you can including all the extra payroll for training for a smallish company. It will probably take 2+ years.

A. Hi Lizzy
Blake's response is unrealistically pessimistic and the old threads are years out of date.
If you are complying with the contract you have with your customer to process aerospace components to an agreed specification and recording the process on a simple route card, the requirement is minimal.
Yes, you do need an approved overall quality system approval but if you are in the professional plating industry you need one anyway. It is very common for companies to question the need for a quality system and be surprised by the savings made from reduced rejects and an increased customer base.

A. Hi Lizzy,

As Geoff said Blake's response is a little on the negative side. I worked with an Indian company about 5 or 6 years ago to get them up to Nadcap standard and in one year they went form having no quality system in place to having both AS9100 quality system and Nadcap, it also didn't break the bank.

I won't lie, if you are not already working for Aerospace customers you may have quite a bit of work to do to get to where you need to be, but it is not beyond anyone to do.

Talk to PRI, they are really helpful (I say this from an auditee position, not an auditor), employ a consultant if you really feel out of your depth, there are plenty out there with experience in getting companies to Nadcap standard and if you are sensible it shouldn't cost you a fortune.