General cleanliness call out? What is the best way to do this without going overboard. 2

[dennisbernal91z]

I am product engineer working for a company that makes medical blood analysis equipment. We do not make anything that touches a patient; it is all just lab equipment. My company sells mainly to second and third world countries. The market we cater to is small labs that can't afford big name analysis machines. This is all important because I need my drawings to reflect that in the sense that I can't spec something up because if I do, our price per piece will shoot up and then we need to pass that onto our customers and they can't afford it.

So with all that being said, does anyone have any suggestions about how to call out a clean part? What I am looking for is if a part gets machined, that it gets rinsed off, maybe tossed in an ultrasonic cleaner to get all large particulate off, and then blown dry with filtered dry air. That’s all. Parts don't need to be incredibly clean, just decent. I know this is ambiguous so it makes it impossible to say there is a pass fail, but I need to put something on there so clear things up.

Finally, being a small company, we don't have very much equipment. Just microscopes and magnifying loupes, and those are not even over in out inspection area. We use those things for production, but could bring them over if need be.

Either way, just want to hear some opinions.

Thanks.

 

[IRstuff]

Unfortunately, when one starts talking about vague requirements, one gets vague specifications. Nonetheless, I can refer you to an official documentation of some vague requirements from NASA, so it sounds VERY high tech. Take a gander at AppendixA


TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[LittleInch]

How filtered is the air? Surely that's the first "real" spec you need to add - i.e. filtered to x microns...

Hydraulic oil for offshore umbilicals has some cleanliness specifications which you might find similar - they can be very tight, but come in a range of cleanliness levels. I've never got involved in it, but I'm sure a decent search will provide something or a slightly different post in the controls section asking for details.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[dennisbernal91z]

Thanks for the replies.

I guess I can state to what micron the filter needs to be. That’s fine, but that doesn't really address the issue. I can have the parts air dry, I don't mind. What I really want is parts that don't have finger prints on them, no oil from machining and don't look like they got dragged around outside before being delivered to me.

I need to have a loose spec because the small machine shops I deal just simply can't spend time/money with high level cleanliness specification.

I would like parts to be decently cleaned like I stated earlier. I guess I can simply state that on each of my drawings.

Something like: ALL PARTS TO BE CLEANED WITH ULTRASONIC CLEANER FOR A MINIMUM OF 15 MINUTES WITH ONLY DISTILLED WATER. PARTS TO THEM BE DRIED BY FILTERED COMPRESSED AIR TO A MAXIMUM OF 0.3 MICRONS. NO FOREIGN MATERIAL PRESENT ON ANY SURFACE ALLOWED.

I know this is not a Standard Spec I am shooting for. I don't have the luxury of doing something like that. I need something that is "cheap and works".

I know this is not the world most engineers live in, but it is one I find myself in now and need to try and find a solution to this issue.

Thanks for any insight.

 

[drawoh]

dennisbernal91z,

Are there fabricators out there who specialize in ultra clean stuff?

Since you require all of your stuff to be clean, this might be a task to bring in-house. Then, you know you cleaned it, and you know what you cleaned it with.

--
JHG

 

[Compositepro]

If you are going to low cost suppliers and doing the assembly then you should clean the parts yourself to insure that they are, in fact, clean and to keep to cost down. In small lots your supplier cannot clean at lower cost than you, and there is always the possibility of contamination after cleaning. Go ahead and make it clear to your supplier that cleanliness is important to you, but without a detailed specification, do not count on anything.

 

[MintJulep]

NO FOREIGN MATERIAL PRESENT ON ANY SURFACE ALLOWED.

Well, that's really clean.

You need to specify either the process that you want used, or the result you want. Not both.

A Google search for cleanliness requirements for medical devices turns up a lot of useful looking information.

 

[rb1957]

is it clean you're after, ie de-greased ? simple detergent will do that, no?

or anti-microbial ? exposure to UV ?

i like the suggestion above ... try something, assess it in-house, did it do the job ?

Quando Omni Flunkus Moritati

 

[ornerynorsk]

"NO FOREIGN MATERIAL PRESENT ON ANY SURFACE ALLOWED."

This is a land mine. You're best bet is to create a cleaning/handling/packaging work instruction as part of the drawing.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[dennisbernal91z]

I tried proposing that and my boss said no. He wants to be supplied with clean parts. With that being said, I don't need ultra clean parts and I certainly don't need "medical grade" clean parts. I have Googled that type of thing in the past and that is all overkill. My application is not "medical" in that sense. That word "medical" implies that contact with the subject or very high clean standards, which I don't need. Just "normal clean" in my eyes. I know that is super unclear, but if someone hands you a part in your office, you don't expect it to be coverd in oil, coolant, etc... Just clean. That is all I want. Sort of like how clean a DVD, TV, magazine, shampoo bottle, etc. is when you go buy one, or a rubber spoon for cooking, etc... These are just clean items. Nothing special, not ultra clean, but you could rub them against your white shirt and your shirt will still be white. That is the level of cleanliness I am shooting for. I have made it clear to my suppliers that is what I am looking for. When parts come in dirty, they have no issue taking them back and cleaning them. I just don't like the wasted time and thought maybe there was a clean spec that doesn't dive into the realm of particle size per square inch. I just need some sort of "general clean" spec and was curiouse if anyone here had ever seen one.

Only other thing I can correlate this to is how I get parts machined out of metal and tell the machinist to break all edges or smooth them. They don't ask how much material to remove when breaking an edge, they just know to take off a minimal amount so that edges that are 90° (or whatever angle) are not going to hurt you if you grab the part.

If there is no spec like this, please let me know. As for the breaking edge bit, what do you guys have on your drawings for something like that?

 

[MintJulep]

"No visible staining or discoloration when wiped with white filter paper."

http://www.carolina.com/science-lab-filtration/filter-paper-qualitative-15-cm/712804.pr

 

[rb1957]

does yor boss have pointy hair (a al Dilbert) ?

Quando Omni Flunkus Moritati

 

[LittleInch]

Edges are normally something like, "All edges to be machined or ground smooth with a 5mm radius."

If you can't get into precise numbers, I don't think there's too much wrong with words like clean, smooth, dry, free from grease, metallic particles or any other contaminent. They have common meanings which are well understood and would hold up in any dispute.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[ivymike]

We have a series of technical specifications that we will reference on drawings as appropriate. As you might expect, there are a series of specs around the cleanliness of components (some parts are more sensitive than others). Suppliers need to have a copy of the relevant specs, and they can look up which kind of cleanliness they need on a particular part based on the code on the print.

The main things we control are
- Largest allowed particle
- Maximum no. particles per size range of particle
- Maximum mass allowed in total
- Particles/materials of particular kinds (allowed/not allowed)

The different levels of cleanliness of course have different bogies for each. There are also some special requirements for certain component types which are listed separately, and some tips for how to correctly measure the above parameters.

In your case you would probably need to start the development of such a spec with a benchmarking excercise to determine what is/isn't acceptable in your case(s). No point in trying to control something if you don't know where to set the knob.

 

[IRstuff]

Wow! It's so refreshing that so many people are into quantifiable and verifiable specifications!

However, quantifiable and verifiable entails cost, plain and simple. If you don't want the cost, then you have to have a dodgy spec. That's just the way it is. If you impose a numerical spec, the inspection costs will increase 10-fold, unless your supplier already does that level of testing as a matter of course.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[IRstuff]

Some of you probably didn't bother looking at the NASA spec, so here's a sample:

GC (Generally Clean) (1) Freedom from manufacturing residue, dirt, oil, grease, processing debris or other extraneous contamination. This level can be achieved by washing, wiping, blowing, vacuuming, brushing, or rinsing. The GC level shall not be designated for hardware that is sensitive to contamination.

And this is from the Johnson Space Center 1998 for spacecraft.

Not a single quantifiable thing, anywhere in that paragraph. And in case you think that was an anomaly, here's the next level:

VC (Visibly Clean) (2) The absence of all particulate and nonparticulate matter visible to the normal unaided (except corrected vision) eye. Particu- late is identified as matter of miniature size with observable length, width, and thickness. Nonparticulate is film matter without definite dimension. This level, with the exception of the Orbiter payload (cargo) bay, payload canister and payloads, requires precision cleaning methods, but no particle count.


Only in the NEXT level are there quantifiable things:

VC + UV (Visible Clean Plus Ultraviolet) (2) Visibly clean (as defined above) and inspected with the aid of an ultraviolet light (black light) of 3200 to 3800 Angstroms wavelength (3.2 x 10-7 TO 3.8 x 10-7 meters).

And that only specifies the test method, not the results. Only on the next page do they start to get into "precision particulate levels."

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[MintJulep]

However, quantifiable and verifiable entails cost, plain and simple

True. But vague and unquantifiable creates risk, which also entails cost.

 

[LittleInch]

If you develop your own "spec", why not do a picture spec showing for each of your definitions acceptable and non acceptable. I read the NASA one which is one reason I said that words in common parlance can be acceptable if you have no specif measurable requirements.

If it (pictures) is good enough for the Swedish standard for how to rate blast clean standards for pipe it's good enough for your purposes.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[IRstuff]

Future cost is a risk, not a cost, and that's only a probable/possible thing. As with the thread posted by an SE about inadequate design on a house, the low, immediate cost appeals to everyone all around. So long as everything runs smoothly, the additional cost/risk is an intangible.

Lots of people want quantifiable requirements, but are unwilling to pay for the implementation costs, so they cheat elsewhere, by either not testing, or testing inadequately, thereby not only incurring the risk anyway, but also creating an illusion of rightness.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[IRstuff]

and don't forget, that's from NASA, an organization willing and able to spend millions of dollars. But, even they know when to just let things be.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529