Current state of R&D machining

I'm curious about what everyone's perspective is to the current state of R&D machining. I went to trade school in the early 90's and worked as an R&D Machinist for the next 15 years. Quite a bit has changed since then. I remember in 2002 or so an Engineer came down with a prototype made from an SLA type process of a new product he designed. He was pretty blown away by the technology (3D printing). This was a part I would have normally made. That's when I knew it was the beginning of the end. I worked soon there after at a reputable local CNC shop for a few years. The owner didn't want to have anything to do with small quantities of anything cause there was no money in it (which makes sense). I since then moved into a new line of work. Fast forward to today and it seems like virtually all prototypes are 3D printed and when it needs to be machined, it goes to ProtoLabs (which is an amazing company - not sure how anyone can compete with those guys for R&D type quantities). The company I use to work for that I did a ton of R&D prototyping for has recently eliminated the machine shops from their locations (this is a multi billion dollar company).

So my question is, is this type of work a thing of the past? If you work at a company that designs/sells their own products, how has R&D machining changed? If you're a shop owner, is it accurate to say that most places pass on low quantity machining (less that QTY 5 for instance)? Do you quote that type of work or has that fallen off as well (due to lack of demand for it)?


Not complaining - things change. Just curious how things are these days.

i have work in many factories where maintenance machinist every day are asked by engineers for big multi story machine to run different. design, make install test and debug something
.
rarely does maintenance machinist have new equipment to make parts often its 30 to 60 years old and long ago written off with zero asset value. new tooling, boss would complain about spending $100., metal go look in scrap lugger for some metal.
.
since maintenance guys time is already paid for parts were made for free or a few dollars out of materials available. used to have a pile of drawings or parts to be made in my spare time about a foot high. why ? cause regular machine shop would charge $10,000 for a small tray of parts where i would make parts for about $100.
.
we ever had anything 3D printed ? no. usually cause actual final costs were way to high. laser engrave parts ? they wanted $1000 to write a program. i could walk to crude cnc program engrave mill a part clean up walk away in 30 minutes with costs basically about $1. in tooling......... just saying many places dont spend a lot of money. not like everybody being the government and can spend $1000. for what can be done for a $1.
.
i see just as many jobs going the other way. looking for people who can research and design and make and install and test parts as cheap as possible. boss use to say he was not looking to hire a "expert" he wanted a jack of all trades that could do most anything. maybe not the best or the fastest but get the job done.
.
used to make parts made with simple 2D drawings done by a technician. why ? cause "real" engineers with "real" cad was so complicated they would charge over $1000. to make any drawing. technician could make a drawing on a napkin if he had too. a lost art. guy can design and make simple drawings with simple software and make a drawing in 5 minutes. in many ways i had a lot of respect for the technician. many of the "real" engineers wouldnt make any drawings at all for years. beware $10,000 or $100,000 cad software so complicated a factory with 10,000 people and only one or 2 can actually figure out how to use the cad software.

guy2600 said:

I'm curious about what everyone's perspective is to the current state of R&D machining...... <SNIP>

So my question is, is this type of work a thing of the past? If you work at a company that designs/sells their own products, how has R&D machining changed? If you're a shop owner, is it accurate to say that most places pass on low quantity machining (less that QTY 5 for instance)? Do you quote that type of work or has that fallen off as well (due to lack of demand for it)?

Click to expand...

I work for a company that designs and builds prototype laser systems. While we will occasionally use Protolabs, Protocase, Go Proto, and one other I can't recall right now, for simple quick turnaround parts. But they can't do it all. And have constraints that too often we can't live with. Protolabs won't even install heli-coils for example, they tap the STI sized hole, but we have to install the inserts. And they won't do chem conversion on alum parts. We have to do it in-house. So we end up buying 98% of our machined parts from the domestic based shops we have been using for years.

We are beginning to dabble in additive metal parts. We have our own SLA (plastic) printer that we use for doing fixture-type parts, and proof runs. And we have a pretty talented manual machinist on staff for creating "emergency parts", fixtures, and modifications. When he is out, I get called to fill in for the "EZ" fix-it type jobs. I have but a fraction of his skill & knowledge (I'm a mech engr).

So I would say prototype machining is still going on. But not like it used to. Everything has gone CNC mostly.

Where's Johhny LaWhoo ?

Probably a special case here (watch business, tiny parts) but I frequently do 1-5 parts for customers. When the watch biz tanked with the economy a decade back it was hard for a while as everyone tried to either skip the prototype phase or bring it in house, it's better now. I'm really in a niche inside a niche as many of these products might see from a dozen to low hundreds total production. I guess like anything it's good to find your niche. Probably countless shops ready to do aluminium widgets that weigh a pound or ten, fewer doing really big/small/ complex stuff.

I know a guy in my niche who'll do anything, but only if you've been turned down several places due to complexity. Not that he digs the knife in on price, just keeps the work interesting for him.

We have a biomedical research and clinical support machine shop but I'm also around a lot of other projects and spend time mentoring young engineers on design projects. My design consultancy friends use proto labs quite a bit with rapid turnaound at amazingly high prices. They do that for jobs with huge companies and it's fine, but the engineers I know there really want to have their own machine shop to modify parts or do the simple stuff. That would be for the engineers/designers to use though, they wouldn't hire a machinist unless the person was also a designer. Our shop just got a new CNC (subject of a new machine day slide show when I get around to it) and it is busy all the time with projects where material properties are key. But we also have a waterjet and a large fraction of projects go through the waterjet, with either some sheet metal bending after or other simple secondary ops on a mill. We also have a good 3D printer that is used regularly but mostly for exotic human anatomy related things you would never try to machine. CNC machining, waterjet and 3D printing are actually all quite complementary as there is almost always an obvious choice of what to do, especially related to the size of the part. You try to minimize the CNC jobs as they take the longest but even so one of our guys is full time doing CAM and CNC while lots of other stuff flows through the waterjet pipeline.

Overall this means that our R&D shop staff are more productive than they used to be but they also need to be more capable of operating different machines and doing a certain amount of design work themselves. They also need to go back and forth with designers to identify issues with models and in some cases iterate on the fly. What we don't need so much are traditional machininsts who work from carefully prepared drawings.

rcoope said:

You try to minimize the CNC jobs as they take the longest ...

Click to expand...

Were you including 3D printing in that comparison? Because if you were, I'd really like to hear a bit more about the complexity of your CNC jobs!

I mean with CNC the CAM takes a long time. Sometimes the machining take a while too; currently we are doing the first of a set of human vertebrae but made out of 316. They are being surfaced pretty fine but it's not like when you 3D print a human skull full sized and it takes an hour from CT DICOMS to starting the printer and then 60 hours to print!

Edit: The larger point is our CNC lead times are a few weeks because we're doing difficult job after difficult job, so if your little Instron test bracket or cyclotron maintenance thingy can be all waterjet, bent welded and powder coated, then we can divert to a different queue in the shop schedule and do it this week.

We are very nearly 100% R&D - all local companies. Still pretty steady.

Was trained 40years ago on manual and still use them today as a R&D gunsmith for the US Army. I do a lot of one off prototype parts. No CNC

I do design and 5 axis milling as part of a small medical device company. For the vast majority of our parts Protolabs can't handle the complexity and tolerances we need. Metal printing can't hold the tolerances either. The nice thing about running the prototypes and the production on the same machine is you can just keep refining the same program through revisions, and when it comes time for production you're ready to go with the process already developed.

guy2600 said:

...it seems like virtually all prototypes are 3D printed and when it needs to be machined, it goes to ProtoLabs ...

Click to expand...

This is a grossly inaccurate statement. I have worked all my life - and still do - in R&D in medical, optical, nuclear, mechanical, electro-mechanical etc. fields, designing, making and bringing to production hundreds of prototypes of every kind and size. In the last decade in my shop only three parts for one medical device (out of maybe 60-80 in the system) were done by 3D printing in plastic and 2 parts out of a large and complicated nuclear medicine system were ordered to be made by metal printing. All other components are made by conventional machining, TIG welding, laser welding, EB welding, silver soldering, grinding, wire EDM, wtaerjet, and so on; both on manual and CNC machines. If anything the demand for a knowledgeable manual-machines machinist or toolmaker is so great that since the time my best toolmaker retired a few years ago I cannot not find one to replace him.

R&D machining will always be needed, IMO. I made quite a bit that way. Many shops really HATE small quantities, so it works for your benefit.

billzweig said:

This is a grossly inaccurate statement. I have worked all my life - and still do - in R&D in medical, optical, nuclear, mechanical, electro-mechanical etc. fields, designing, making and bringing to production hundreds of prototypes of every kind and size. In the last decade in my shop only three parts for one medical device (out of maybe 60-80 in the system) were done by 3D printing in plastic and 2 parts out of a large and complicated nuclear medicine system were ordered to be made by metal printing. All other components are made by conventional machining, TIG welding, laser welding, EB welding, silver soldering, grinding, wire EDM, wtaerjet, and so on; both on manual and CNC machines. If anything the demand for a knowledgeable manual-machines machinist or toolmaker is so great that since the time my best toolmaker retired a few years ago I cannot not find one to replace him.

Click to expand...

I think this tends to be somewhat company/product specific. Perhaps in the industries and companies you work with locally this holds true. Consider yourself fortunate. Not the case out here. I know people in a number of industries locally and this is the clear trend. Are there exceptions? Sure but they're exceptions. Last week I spoke with someone who does R&D exclusively for med device. He said now most everything he gets is post machining ProtoLabs parts. ProtoLabs is soon offering their own 2nd op service for tight tolerances. That will eat some of the business for him.

Anyhow, it's good to see some companies are doing well.

mhajicek said:

I do design and 5 axis milling as part of a small medical device company. For the vast majority of our parts Protolabs can't handle the complexity and tolerances we need. Metal printing can't hold the tolerances either. The nice thing about running the prototypes and the production on the same machine is you can just keep refining the same program through revisions, and when it comes time for production you're ready to go with the process already developed.

Click to expand...

That's awesome! I never had the chance to learn 5 axis. That seems to be the sweet spot. Working for a small company that manufactures their own products. Minnesota is one of the hotbeds for med device. Good place to be (except those winters - brutal over there).

Screwmachine said:

Probably a special case here (watch business, tiny parts) but I frequently do 1-5 parts for customers. When the watch biz tanked with the economy a decade back it was hard for a while as everyone tried to either skip the prototype phase or bring it in house, it's better now. I'm really in a niche inside a niche as many of these products might see from a dozen to low hundreds total production. I guess like anything it's good to find your niche. Probably countless shops ready to do aluminium widgets that weigh a pound or ten, fewer doing really big/small/ complex stuff.

I know a guy in my niche who'll do anything, but only if you've been turned down several places due to complexity. Not that he digs the knife in on price, just keeps the work interesting for him.

Click to expand...

That's really interesting. I know nothing about the watch business and how that stuff is made. Are watch parts machined in production? Honest question, know nothing about it. Those must be interesting parts to machine. How has the trend towards iwatch type watches effected you guys? I'm guessing they still need some machined parts (cases and the such).

guy2600 said:

That's really interesting. I know nothing about the watch business and how that stuff is made. Are watch parts machined in production? Honest question, know nothing about it. Those must be interesting parts to machine. How has the trend towards iwatch type watches effected you guys? I'm guessing they still need some machined parts (cases and the such).

Click to expand...

The finest of mechanical watches will always have a good deal of craftsmanship in them because that craft is essentially ALL the customer is paying for. Even the cheapest of cell phones keeps better time off the back of extremely costly external resources, of course, but there you have it. They are "there" most places it matters to anyone.

Maintenance and repair survive off the back of "need it YESTERDAY" as much as "no budget just FIX it".

"R&D in general" has been making progressively less sense to do manually when doing the prototype with CNC can also vet the code as well as the part itself, thereby streamlining a later shift into pilot, then volume production.

IOW it is worth some extra hassle because there is a payoff in real money and in shortened time to volume delivery. There HAS to be. Products of all types have shorter profitable life-cycles than was once the case before they are knocked-off cheaply.. of just left for dead as some newer product takes their place.

No technology ever TOTALLY vanishes, as if it were a dodo bird. It simply gets marginalized. Leather work still goes on. More Bronze is used now than in the "Bronze Age". Stone spear points and bone needles are still made. Dai Katana are still forged by hand, and intentionally with the oldest methods they can still remember how to apply. It is a major part of their value, much as with fine mechanical watches.

These thing just don't get a lot of shelf space at Walmart, the automobile dealership, under your ass on a flight to another continent, or in the nuisance at the end of your arm with the annoying ring-tones.

Heinlein nailed it one time as a side note. I can't quote it accurately, but the gist of it was that the single absolute BEST person in even the most arcane of trades will always find work. The question then is whether there is support for multiple thousands more to also thrive.

Or not even ONE more.

"Manual" R&D does seem to be narrowing down, at least as a percentage of all work in a more populous world. After all, "teach in" assisted CNC hybrid options have been growing for a long time now, too.

The top people will continue eating well for longer than actually matters to their personal mortal span.

Perhaps BOTH of them? The people. Not the lifespans.
Or even 2 thousand. Or Twenty-two thousand.

But the OTHER seven-plus billion?

Lets hope it doesn't degenerate back to all-manual labour..without any machines at all.

thermite said:

"Manual" R&D does seem to be narrowing down, at least as a percentage of all work in a more populous world.

Click to expand...

Part of this is companies are doing less and less research over the years. There just aren't that many Bell Labs
out there anymore. Yes there are lots of semiconductor fabs being built. But that research was done a long
time ago.

jim rozen said:

Part of this is companies are doing less and less research over the years. There just aren't that many Bell Labs
out there anymore. Yes there are lots of semiconductor fabs being built. But that research was done a long
time ago.

Click to expand...

Yah go a point there, Jim.

Bells labs. IBM Research. Key Universities. Several others.. were "national treasure" class entities.

The names on research papers now? Chinese. Indian. Not as often publishing from a US seat as was once the case, either.

You don't need to be RMS Titanic to guess that where the ten-percent tip of an iceberg is visible, a far greater mass of "goings on" - and a risky one to be unaware of - remains out of sight.

MilGunsmith said:

Was trained 40years ago on manual and still use them today as a R&D gunsmith for the US Army. I do a lot of one off prototype parts. No CNC

Click to expand...

Similar situation for me except different products.
We can often have a single part already made, while the CNC operator is still pushing buttons. Not long ago, those skills did not have the value that they do right now. Interesting isn't it?