Copper Alloy Machining

[Mr.Bronze]

Hi,

Long time reader, first time poster here. PM has always been a great resource for me in the past, for both machining issues and shop management thoughts.

Here is a little history of what my shop does:
I run a foundry/machine shop that produces bronze and brass bushings for heavy industrial applications. Manganese bronzes, aluminum bronzes, leaded and tin bronzes, etc. through green sand casting. It is a job shop, running from 1 to 100 parts, up to 600lb finished parts, depending on the alloy. (C863 C865, C903, C905, C907, C911, C932, C937, C954, C955 are the commmon alloys we work with, some are free machining, some are just HATEFUL to work with)

Here is my dilemma:
I feel that the time it takes to machine the parts - on the mills- is 2-5 times long than would be quoted by my competitors, I have even seen some empirical evidence of this on some work orders for castings that I sold to be machined elsewhere.
Our lathe times are respectable, obviously can be improved, but the mill work just is excruciatingly slow.

I talk to my guys, ask them how long they can do a part in, and say for a part 10" long x 5" wide x 2" high, maybe with some grease grooves and some chamfered/Cbored holes, with 0.125" of stock on each surface. A good ball park estimate for their quote is it would take them 8 hours to do one, 4 hours to do many. This is just for a simple block, the times just get outrageous (like 24 hrs for one, 16 hr each for many) when there is any kind of features or undercuts or contouring.

The mills that are being used are a Fadal 15xt, Fadal 5020, and Haas VF-6. we also have some 4th axis rotary heads and right angle/adjustable angle heads.
Also have mastercam X7 mill3 with solids module

I have them walk me though their process, and it just seems like they are too conservative in their approach.
While I totally agree that letting mastercam dictate how to make chips is a great way to ruin parts, I don't think my guys are using it to its full potential as far as achieving a higher spindle up time - removing more metal in one clamping of the workpiece, using more effective roughing/ finishing mechanics, better utilization of cutting edges.
Beyond their hesitance to use CAM software a little more effectively, which I don't think will make up the difference in machining time for my parts, I don't know how to address this issue.

Any thoughts?

 

[wheelieking71]

Send me a solid-model of the part, and an .stl of the stock, and I will "sport-quote" one for you.

I can maybe see 8-hrs for one if it is a part# that has never been done before, and they are squaring up an as-cast casting.
MAYBE!

But, if there is enough meat to grab the casting, and hit 4 sides, and the top, in one operation, that is a 3-4hr job tops. Including set-up, and programming.

24hrs? Hell-no

 

[Limy Sami]

''Any thoughts?''

Plenty, your post screams to me, ''the tail's wagging the dog '' IMO they're taking the piss, so it's time to start being the boss.

If you don't know the ins and outs of sorting out the mess (which must be costing you a fortune) hire in a guy or gal who does, eg someone who's been in the job shop side of the game and knows what they're doing.

From your brief description, (any more information would be pretty superfluous as we really have to be their right on the ground to fully understand the situation.) you are in a heap of trouble, of the kind that's exceedingly difficult to deal with.

IME in these situations just ''re educating'' by an outsider is usually a waste of time, the only viable solution is to take an outsider on full time, give him the authority to kick ass, and keep kicking it until said ass toes the line or walks.

 

[Madis Reivik]

Have You thought about "fear of broken tool" ? If the workers will be punished for broken tools OR they feel they might be punished or they have crashed the spindle - then its quite common to reduce feeds and speeds and be "conservative".
Trochoidal toolpaths + HSS tools (if carbide breaks) and perhaps the mills need some software upgrade to better handle "continous curved" toolpaths.
Just my 2 cents
Also, perhaps the workers are not aware that "lowest tool cost" is not the "lowest final cost". Very often people tend to underestimate the value of their time and overestimate the price of purchased tooling.

 

[wheelieking71]

From his description, These sound like pretty basic parts. No fancy-pants tool-paths.
If they are having tooling issues, and that is the reason for 3hr jobs taking 8hrs.
Then somebody needs to pipe-up and show the boss a better way.

Limy pretty much nailed it.

 

[wheelieking71]

Wheelie,

Here is a perfect example of the simple things that end up taking forever here. View attachment 143122

We recently made both parts, 8 pcs of the the one on the left, 4 for the one on the right. They took 3.5 hrs each and 10.5 hrs each, respectively (although there were 6 castings started, 2 scrapped for the latter).


thats a horrible picture of the print.. let me fix that

You need new guys. Limy Sami is right on the money!

That part on the left is maybe 4hrs to get first part through inspection, and 30-40minutes for each one after, TOPS.

They could be tooling limited. If so, they should tell you there are more efficient ways.
If they don't know, you have the wrong guys.
Or like Sami suggested, you need one good guy that knows "whats up".

 

[Mr.Bronze]

Wheelie,

Here is a perfect example of the simple things that end up taking forever here.


We recently made both parts, 8 pcs of the the one on the left, 4 for the one on the right. They took 3.5 hrs each and 10.5 hrs each, respectively (although there were 6 castings started, 2 scrapped for the latter).

 

[Mr.Bronze]

I think there is a bit of everything that has been brought up here at work towards keeping these machines and employees from being as productive as they can be.

And above all, yeah I am in a heap of trouble. Fortunately, this business has been successful up to this point (since 1895) even with all this going on. I am in the process of taking over the business from my father and uncle, and I have been working hard trying to pick off all the low hanging fruit towards improving productivity.

As far as kicking out these bums... that would be my last resort, considering they have great attitudes, show up every day, and take pride in their work. I think the biggest thing is, concerning personnel is their lack of experience in seeing things go quickly and smoothly. They definitely err on the side of conservative machining, not because they are afraid of punishment for breaking tools, more of scrapping parts, and not being as confident as they should/could be in the capabilities of their machines.

I hate the idea of trying to find someone experienced with this stuff (the alloys/castings) because it is so rare to find someone that is familiar with this stuff.

My guys definitely are held back by their unfamiliarity with modern tool-path generation/ high speed machining technique; I think that is probably the biggest hurdle, and least difficult path for making progress with this. I am an engineer by training, with lots of experience in solving problems with no clear cut solution, but that doesn't mean anything when I start trying to twist a machinists arm over how I think thins should be done.

I can't count now many times a tool rep has come into the shop grand standing about how this new tool will revolutionize how we do stuff, and within 20 minutes of demoing the tool on our alloys has broken their tool, because they do not have experience with the metals we work with. It's hard to make a theoretical argument for them to do it some way, when my guys can just rattle off the last several times they have embarrassed a tool rep.

 

[Mr.Bronze]

On the bright side, I did just hire a new guy who raised his eyebrows when he saw some of the methods we used in machining, so hopefully he brings some fresh ideas to the table and from a better position of leveraging them.

Madis- because of the abrasiveness of castings and some alloys themselves, HSS is not generally used for what we do here, uncoated carbide is currently the clear winner in tooling life and repeat-ability.
From what I have gathered a diamond variant PCD/CVD/PVD tool has a huge lead on carbide as far as tooling life, but is more well suited for high run parts in more consistent material due to substantially higher tooling costs. There are inclusions and other defects that sometimes occur, and it would take some number crunching on that before I could justify buying $1500 end mills only to have them break 3 parts in.

 

[Mr.Bronze]

Here is a more complicated one,
Rectangular block with ruled radius surface on top, some grease grooves and holes, approx 4" thick x 10.5" wide x 14.5" long



8 pcs, 20 hrs each

 

[wheelieking71]

Don't buy $1500 end-mills. Buy $300 end-mills for finishing operations. All your roughing should be getting done with inserts.
Yes, it is nasty material. But, your not the first shop to work with this stuff.

I'm just stabbing in the dark here, because I have no clue how your guys attack this stuff now.
But, I will say this, that left part on that print you posted, is two operations. Top & Bottom. That's it.
If there isn't enough stock to grab in order to profile the sides full depth? Add it.
The stock is waaaay cheaper than the spindle/set-up/man hours it takes to complete that part in more than two operations.

IF! They are attacking it in two OPs, and it still takes 8hrs for the first piece?
You have work-holding/tooling issues. Inserts are cheap when shop rate is $75hr.

 

[litlerob1]

Mr, Bronze I think you have viable concerns, there are any number of reasons your guys are running slow; Inappropriate tooling, comfort, straight up lazy, tradition, conventionalism, etc, etc. I won't go into the repair of those things, you seem smart enough. I will add though that bull shit about "MasterCam7 Mill3" not being 100%, is just that BS. Lathe, I would agree but not Mill, it's 100%, sorry but maybe retraining is in order.

Unfortunately showing up on time doesn't cut it in the world we live in, they have to compete or go home. Is it worth losing business because you like the guys? I know that seems harsh but, it is what it is.

Robert, my ±2

 

[Limy Sami]

Here is a more complicated one,
Rectangular block with ruled radius surface on top, some grease grooves and holes, approx 4" thick x 10.5" wide x 14.5" long

View attachment 143131

8 pcs, 20 hrs each

A thought, is that the actual time taken or the quoted time?

If actual, (and in a former life 40 + years back I machined a lot of bronze castings,) I'd be going ape shit ;- How f'n long?

Like 8 of those is 4 weeks (@40hrs / week) work? like f'k it is okay might be a bit of a bitch to set up and prove,- so we'll be generous and say 3 of the 8 at the most to get the job sorted, then you start motoring.

I want to say forget the retraining and ass kicking, just reach for the captive bolt gun and put them out of their misery.

Very old Brit saying ;- Are me making money, or are we making work?

 

[Mr.Bronze]

So, I just brought up all these comments up with my guys, and it did not go well. They basically looked at me like I was calling them liars. I tried to stress the fact that I am not accusing anyone of anything, but I am also not looking for excuses, I am looking for solutions; I want to know what they need to perform the job, not why they can't.
I think this discussion is distilling the situation down to the root causes- tooling and training, both with the thought process on setting a job up, and with fully utilizing mastercam.

Also, I am glad to hear someone speak highly of mastercam, as that is one of the biggest reasons for resistance I get with my guys. They like to think that mastercam is only for operators, not for machinists, and that they will have to pour over every one of the lines of code to ensure it won't crash. They also like to talk down on mastercam because "it makes 100k lines of code, and they could do it in 50 lines of code". That last part bugs the hell out of me, cause the number of lines of code that is required to write a program has negligible impact on the cost of the parts being produced.
The only part that hits the bottom line is how many parts are made per day; spindle up time, fewer times chucking the part up, faster metal removal rates.

I also would like to re-state the fact that getting rid of these guys is the last resort. I really don't think it is their intention to be slow, I think they are ignorant of the potential of these machines and the programs, and they are conservative because they have never seen it done correctly.

 

[wheelieking71]

There are several ways to attack that second print you posted.
Depends on how often you plan to make them.
But even the long way is less than 4hrs machine time per part.
20hrs each is ludicrous.

Your guys are stuck in 1980.

(here comes the CAM vs. finger-CAM argument!)

 

[david n]

How many years do some of these guys have with the company? Your guys are comfortable doing things the same as they always have been done. Bad habit to get into.

The times given for those parts is unacceptable.

 

[wheelieking71]

Mr Bronze, here is an example of a low quantity job I did a while back, just to give you an idea of time-frame:



Now, this isn't exactly apples to apples. Aluminum vs. Bronze castings, and so forth. But, its just to give you some perspective.
That stock was 14" x 14" x 3". The centers were slugged out on a lathe Not sure how long it took, maybe 3hrs? I farmed it out.
But, the rest of the work took me 24hrs total, from raw stock, and a print, to finished parts.
That includes programming in CAM (far more complicated part than your example).
And, all set-up/machine time. There were some tight tolerances there as well. That helps to even out the easy to work with material.
There was a flatness of .001 on the bottom of the part. I had to flip it a couple times to maintain that.
The large pocket on the bottom was +.002 -.000 on the dia. and concentricity between it, the through hole, and the outside of that wall was .005

What I am saying is, there was a whole lot more work there than your 8pcs that took 28hrs.

 

[SND]

If one of them ends up 5.9983 will it be rejected? got any extra material to make up for any scrap pieces of they actually try go to fast and something goes to shit?

Anyhow, if they do them often enough and with over a 100yrs in business I'd think there would be a process and knowledge pretty well nailed down by now.

I'm assuming most employees have only ever had that 1 job eh?

 

[wheelieking71]

comfortable doing things the same as they always have been done. Bad habit to get into.

Boy isn't that the truth!
I find myself looking at every single job, new jobs, and repeat jobs, differently, trying to advance my efficiency!
I learn "things" from every single one too.

 

[WILLEO6709]

There is a lot involved in quoting, and part of that is your equipment.... everyone quotes wrong, but hopefully not wrong enough to lose too much on the job. Generally if a job takes longer in a machine there is not a machininst sitting there watching, they are running 4-6 machines in long cycles and constantly have something to do for at least one of them. What a brand new $400k vmc can do may not be what a 15 year old machine can... and different alloy can play a part too, some of those copper alloys are not easy to maintain good finish.