What to put on a drawing that guarantees that your part is vapor polished? How to inspect?

[dennisbernal91z]

I am a product engineer that makes drawings for parts that need to be vapor polished. In the past I have received parts that were spec’ed to be vapor polished, yet showed up not polished. The parts were buffed slightly so it was a bit hard to tell if they have been polished or not, this was the first time I had ordered the part, so it was hard to tell if it was polished or not. (Very small part, about the diameter of a BIC pen and a thickness of 0.053"). I used the parts thinking they were polished. Later to learn from the vendor they were not. This problem got sorted out, but in the future inspection will not be done by me, it will be done by a non-technical QA employee. How do I make sure they know what a polished part looks like? Give them a polished and a non-polished piece and allow them to do a visual comparison? My company is very small (I am 1 of 2 engineers) so we don't have a lot of cash for inspection equipment.

Thanks for any insight.

 

[dennisbernal91z]

All the clear parts we use are for appearance. If there is a MIL spec, my vendors either won't be able to check it, or we won't be able to afford them. I can't stress that enough. This is simply an appearance thing. The small part I am talking about now, is what I DON'T want to be clear. I had a vendor that made it clear and didn't tell me. In the future I want to make sure that the parts come on NOT polished, yet all other parts polished. How do I get the QA inspection person able to understand the difference appart from training them and leaving an example with them?

 

[dennisbernal91z]

Thats exactly what I thought of yesterday but after thinking about it more thought it would be way to over the top to check how clear something is. I have larger parts as well, around 1" x 3/4" that are also clear. Where do you point the laser? There I want the enire part to be clear, yeet due to the nature of vapor polishing you get some areas more polished than others. We have not had an issue with this yet, nor do I think we will. Just saying that becuase something is a certain clarity in one area doesn't mean it is in another. That is not easy to spec unless to tell someone to check many areas on one part ($$$). When it is cheap part that just needs to be "clear" there is a lot left up to the inspector.

 

[rb1957]

i think QA are the check for how clear your drwg is ! though you can cheat and tell them what you want. but why have a QA department, if you rely on CofCs ? for your own build ?

i think in your case the easiest way is to get a prototype part, a little trial and error, then say "build like that one". maybe let your QA keep the reference sample. if the parts do come polished, you could always scuff them on next stage of ass'y.

Quando Omni Flunkus Moritati

 

[KENAT]

While it's not particularly robust and is somewhat subjective, given you limited requirements and resources having 'coupons' or 'gold standard' for matching may be a reasonable approach.

I suggest having samples sent to the vendor as well as keeping in your own inspection, drawing note something like.

FINISH: PART SHALL BE VAPOR POLISHED ON BOTH SIDES TO MATCH SAMPLE COUPON XXXXX. COMPARISON SHALL BE MADE AT 2X MAGNIFICATION IN A WELL LIT AREA.

As to your in house inspection I'd prefer to leave that off the drawing and record it either in your quality plan or your routing or some such - nail down any other points you can such as sampling rate.

I work at a place that is struggling with inspection and relatively low volumes etc. too and sometimes the practical measures that can easily be implemented are pretty limited. That's potentially OK, just don't kid yourself that they're anything more than getting a warm fuzzy feeling.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[IRstuff]

Getting back to the OP's posted question as written, there is NO way to "guarantee" that ANYTHING ever gets done. Training and repetition can only create some higher probability, but that's not a guarantee. You can at least come close by having a procedure or checklist for Incoming to follow. This will hopefully ensure that the checking is done consistently.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[dennisbernal91z]

Thanks for all the replies everyone.

- After talking over with some other departments and getting other opinions here, someone suggested a photo of the part on the drawing so that QA knows what something looks like. An enginner that works with me brought up metal platting and how QA never knows what "clear zinc chromate platting" looks like over not platted steel. The plating is just a general anticorrosion plating we have applied to out sheet metal parts to keep them from rusting over time. Nothing critical, sort of like plating you would have on your IKEA bed frame, you need something, but don't want to pay tons of money for checking, it just isn't that critical. The photo may be the best "cheap" idea at being able to show QA what a plated part looks like. Only do this for parts that have special not easily checkable features. Apprt from that, there is the "gold standard" sample idea, I would like to stay away from this simply due to the amount of parts we would need to catalouge.

Anyways, I think I have a fairly good idea as to what others may be doing and what my options are. Also learned that trying to describe a part and the way my company works is quite hard to describe over the internet. haha. I will make sure to start off all my future posts with "OK so can't afford just about anyting..." and go from there.

Thanks for the help eveyone.

 

[dgallup]

I do not think putting a photo on a drawing is a good idea at all. There are so many variables involved in taking and reproducing the photo that it would be worthless. Are they going to look at the drawing on a monitor or hard copy? Photo quality paper or typical low grade laser printer paper? Will the monitor be color calibrated? I have two identical monitors on my desk side by side and the difference between their color rendition is pretty large.

I think KENAT's suggestion of keeping masters on hand is good, provided these are properly stored and handled. Ive seen 10 or 20 year old "masters" on somebodies window sill that have yellowed and warped.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[KENAT]

Agree with dgallup that the picture probably isnt' a good option for the reasons he lists.

Visual comparison to a known good part is already pretty marginal in its usefulness for quality control, comparison to a photo is worse.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[dennisbernal91z]

Photos don't wear out or get dusty and warped. Color calibration is a non issue at all. I am not having pantones cross referenced. I am simply showing what a part should or shouldn't look like. Like I said in my example before, if you were working at IKEA incomming inspection and had no technical background at all and say that a bracket had zinc plating and you were in charge of checking this (with no testing apparatus) you would just want to look at a photo to see that it looks like zinc plating, and not raw steel. That is the level of inspection I am going for. Saying that the screen needs to be color corrected is simply a level of precision that I do not need. Our drawings that show anything in color are looked at on a monitor, not printed.

With that said, I agree that "masters" are a good idea. We have thousands of parts that we machine, blow mold, injection mold, rubber transfer mold, sheet metal stamp, cicuit boards, and buy as stock items. Saying that we will have a place that can house one "master" of all these doesn't sound like a completely reasonable solution.

 

[KENAT]

I'm not sure anyone proposed that you have masters for every part, just in cases where it makes sense.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[ewh]

One master can cover many parts, as you are only interested in the finish. A square coupon with the appropriate finish will suffice for a multitude of parts that are to have the same finish.

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV

 

[3DDave]

Several things that can happen to comparison pieces -

One, the QC item is altered and unusable parts become accepted
Two, production is short one item, but they know where a good one is...
Three, an 'extra' item is found in production, and discarded, because it wasn't needed
Four, the inspection item is moved to production and an incoming piece is substituted.

A photo is OK subject to a bunch of assumptions. If it was that easy everyone would do it.

Perhaps a drawing note refering to a controlled QC video that is used to teach inspectors is possible. The tough part is determining what the limits are at which you want to reject parts.

***

For appearance control the usual methods are to either spend a lot of money on inspection equipment or to qualify the process, and sometimes both. For example a weld callout goes on the drawing. The welder is typically required to create a weld of the same type, which is tested, for approval before creating deliverable welds. The inspector gives a visual on the welds to see that they appear to be the same as the qualified welds. Even then, welds may require dye-penetrant or x-ray inspection. Paint is another one - samples are done for gloss and adhesion and the process adhered to.

Here's a strong article on why depending on visual inspection controls can be costly and why understanding process is critical to success:

http://edn.com/design/analog/4427151/The-last-half-century--Wirewound-resistors-Part-one

 

[MikeHalloran]

A lot size of a couple hundred gets into the range where low volume injection molders can be more than competitive with machining and bonding.

So take a look at molding in one piece what is now an assembly.



Mike Halloran
Pembroke Pines, FL, USA

 

[Ripz]

1. Specify a Roughness Average Value.

2. Inspect using one of these. Link