0.1 micron finish on 7075 aluminum

Good morning All:
I got an RFQ recently for an aluminum turned shaft about 7" long and about 0.1" diameter at the skinniest point (right in the middle of the length).
There are some conical features on the shaft and the engineer is specifying a 0.1 micron RA finish on those surfaces in order to bounce a laser off them onto a target.
I assume but do not know whether the shaft rotates during function.

The finish callout strikes me as extraordinarily ambitious given the material, the likelihood that the shaft will move all over the place from stress release, and the inability of the engineer to specify just how they intend to determine that I've actually achieved the finish they've called out.

When I remarked that this finish means you don't get to wipe off the dirty fingerprints with toilet paper without fucking it up, I got a dead silence at the other end of the phone.
When I pointed out that if there are geometric inconsistencies in the features, the laser will bounce to an unintended location and may not hit the detector, I got another dead silence.

BTW, there is no concentricity callout, and the overall dimensional tolerances quoted in the title bar are very generous (+/- 0.1 mm, and +/- 0.5 degree).

I smell a clusterfuck of monumental proportions coming my way if I accept the job and the part doesn't work.
They've come to me because others have wisely refused to bid on the part.
It sounds like I have carte blanche from the purchaser on this project if I agree to do it.
They sound desperate.

So what would you do?
I don't need the hassle of a failed part.
The customer has been good to me over the years.
The engineer is a young guy just out of university...bright kid but rather inexperienced.

My personal temptation is to let this one go and lose the customer, but I do like a challenge.
So I ask myself, (and you guys): is this a legitimate challenge or a fool's errand?

I eagerly await all opinions.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

"Arguing with an engineer is like wrestling a pig in the mud.
After awhile, you realize the pig enjoys it."

Apologies to the engineers here, they have proven they are more "Down to earth".

Have you thought about offering them Design For Manufacturing services that will help them reduce the cost of their part or at the very least get it made? Just based on reading here you are probably the right guy for the job, but they just need some direction back to reality. Plus if they are already an established customer paying a couple hundred bucks to drop the price of some fancy one off piece shouldn't be that hard to swallow.

Isn't that just a 4 micro finish? Easily doable with a burnishing tool.

Maybe they sent the young engineer to deal with you because they know you will help him come up with a workable solution?

J Gilles said:

Have you thought about offering them Design For Manufacturing services that will help them reduce the cost of their part or at the very least get it made? Just based on reading here you are probably the right guy for the job, but they just need some direction back to reality. Plus if they are already an established customer paying a couple hundred bucks to drop the price of some fancy one off piece shouldn't be that hard to swallow.

Click to expand...

In some companies, it's known as the "Producibility" office.

Hi again All:
I appreciate the replies so far; and I'll speak to mtndew's post #4 specifically.
He wrote "Isn't that just a 4 micro finish? Easily doable with a burnishing tool"

Achieving the finish isn't the problem; I know of a few ways to do that and roller burnishing is certainly a candidate process...rather the fundamental problem is that I smell an engineering catastrophe coming
The mismatch between what's on the drawing and what the stated function is has me suspicious.

I've been embroiled in hassles over bad outcomes even when the part meets the specs before...it's never pleasant, and everyone always leaves the table pissed and aggrieved regardless who's legally "right".
I love to try to help out; it's part of my nature, but I hate confrontation and conflict; it's part of my nature too.

Keep the opinions coming please...they're valued and cherished, even the ones that are more amusing than practical.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

If you have a solid relationship with the customer i would highly recommend asking them to reevaluate their tolerancing. If you will lose their work entirely based on if you refuse to do this part, then it cant hurt to ask for more info. If they hold tight to what they want, run away.

Booze Daily said:

Maybe they sent the young engineer to deal with you because they know you will help him come up with a workable solution?

Click to expand...

Additionally, if you have him pegged as a young and bright (yet inexperienced) engineer, in my opinion the silence you are receiving when pointing out design concerns is a good sign. It means he is processing the information you are giving him instead of getting defensive. It's not my hair that's going to be torn out, but I say go for it

few concerns come to mind:
1) corrosion resistance of clean 7075 surfaces is low, which means they will tarnish and lose reflectivity,
2) very low thickness anodizing might fix it, then again, it will offer quite low corrosion resistance, but it may be enough for the application, some reflectivity will be lost due to the way how 7xxx is anodizing in general,
3) the way 7075 usually machines is that it has some "streaking" on very smooth surfaces (which may show up after anodizing), hence 7075 wouldn't be my material of choice, 3xxx would be better fit, but it is gummy, but easier to achieve high polish (for reflectivity) and retain it after anodizing (afair some automotive headlights used or still use 3xxx series for this exact reason),
4) perhaps stainless tube section pressed on, turned, sanded and polished for the section where the reflectivity is necessary might be a solution? then again stainless on 7075 is asking for trouble depending on the atmosphere it will be sitting in,

3xxx may be substituted with 6061 or 6060, the amount of silicone in the particular batch of material will have an effect on how it will polish and if it will lose some of the luster after anodizing

Metal optics work of this sort is done by diamond turning. For some applications it may require machining the support structure, plating it, and then finish diamond turning.

Light weight stress free support structures for the reflecting film may be made by electroplating on a temporary mandrel.

There are many internet references and a number of companies that specialize in this work.

Diamond Turned Optics - NiPro Optics

Low Cost Very Large Diamond Turned Metal Mirror | SBIR.gov

Hi jz79:
You make some excellent practical points; I am very sure the engineer responsible for this part of the project hasn't even considered any of them; and this is yet another point in support of my suspicion it will all end in tears.

From a purely practical perspective, your comment about pressing on some stainless sleeves and polishing those is excellent, but if that were a viable option, maybe just turning the shaft from 420 M stainless, hardening it, grinding it and selectively polishing it would work just fine too...maybe even better.

Problem is we cannot tell, and unless the customer lets us into the engineering mysteries surrounding their product we can NEVER tell, regardless how smart we are as designers.

I am not a qualified engineer...not by a long shot, so it's a domain in which I am willing to handhold a bit for a young guy just getting in, but I'm NOT willing to accept responsibility for the fitness of my suggestions...that's the engineer's job.
Additionally, if I spend a hundred un-billable handholding hours to do a 2 hour job, I just gave myself a good screwing for nothing.

So yeah, your commentary is very valuable and I learned useful things from it...thank you for that.
But the question remains...do I jump in or do I stay out?

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Hi Robert R:
That's it...right there...you just saved my bacon!!!

What you just gave me is a graceful way out...I can present this as a viable path forward if the young engineer decides his specs are absolutely necessary to his design.
I don't have to tear my hair out tackling a project better suited to someone with different gear.
All are happy, I save face, and I can move on with my life.
You sir, are a genius and I owe you one!

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Robert R said:

Metal optics work of this sort is done by diamond turning. For some applications it may require machining the support structure, plating it, and then finish diamond turning.

Light weight stress free support structures for the reflecting film may be made by electroplating on a temporary mandrel.

There are many internet references and a number of companies that specialize in this work.

Diamond Turned Optics - NiPro Optics

Low Cost Very Large Diamond Turned Metal Mirror | SBIR.gov

Click to expand...

I just tuned into this thread and "Ditto".

Diamond turning with a near zero - runout air spindle.

There could be many reasons also the form tolerances don't match a 0.1 micron Ra - lasers can be used in many different ways without needing a geometrically coherent surface.

Issues of tarnish/ oxidation versus plating are very valid.

my first thought is high surface finish then plating and polishing after would have worked best.

An update for all:
I just had a pleasant conversation with the engineer again; he estimates he needs the surfaces to be within a micron for overall geometry, so his title block tolerances are the bullshit I expected they would be.

Per Robert's recommendation I forwarded the link to Nipro and I breathe a sigh of relief at not having to immerse myself in yet another of these "glory projects".

All is good, he thinks I'm a hero, the company he works for thinks I'm a hero, no feelings were bruised, nobody needed to be called an idiot, and we're all happy happy happy.
A good outcome I'd say!

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Just wanna say, threads like these are why I keep reading PM. Totally outside my expertise and regular activities, but I learned something from listening to knowledgeable people working at a high level. Please, more threads like this and less OT flamebait and newbie pile-ons!

so what kind of plating will be done? there is electro and chemical polishing too.

that is the thing - 7075 is not suited for highly reflective surfaces, and reflectivity is important, that is what I gathered from the surface finish requirement, and burnishing will most certainly not be the solution especially for one off, it will be sanding/lapping/polishing sort of finish work

and when they said they need single digit micron precision on a 180mm long part with a 2.5mm section in the middle and that is an aluminum part for some odd reason... that is just a recipe for a relationship disaster between the engineer and the one who will try and make the actual part

I've seen these fantastic tolerances here also and had to let a customer go (this wasn't the only reason) who complained that I'm not anodizing to the required thickness since when they check their part after anodizing they are undersize, and the argument is about 1-2 missing microns in thickness, I then check next batch of parts that I receive, and the critical dimension on those parts (around 20 of them) was in the range of -0.02mm (correct allowance for anodizing growth) to -0.10mm! yes, comma is in the right place and the number of zeroes is also correct, and in the reality all those parts would work perfectly well in that assembly, there wouldn't be any issues at all, but they bought a CMM and they have a young guy running it and red looks very disturbing on the test reports they get when they measure those parts

sometimes, when these things are explained, the customer/designer/engineer will adjust, but sometimes they are just too arrogant to understand that learning didn't stop with the university diploma, best avoid those types unless you enjoy "wrestling in the mud"

Reflectivity at what wavelength? Ask. My go-to on this would be a PCD insert.