What's new
What's new

Efficiency suggestions with low volume jobshop work

jash

Plastic
Joined
Jan 9, 2022
New poster here from the UK. Our shop is your typical jobshop, high variety and very low volume work across fabrications, machined billets etc. A lot of made to order 1-10 off, very rare to get anything past 50 off. Size and complexity of work ranging across the whole spectrum as well. The shop has a mix of prototrak knee mills, bed mills and lathes, 3 VMCs, vertical borer, small grinding machines, a jig borer and a wire edm. We have a welding bay for fabrications and an assembly and electrical department which allows us to do more integrated machine or tooling builds/rigs as well as their own standalone work.

For the guys who own or are employed by a shop handling similar work type and quantity, what have been the biggest low hanging fruit type efficiencies you've either seen or implemented in your companies? Can some level of automation apply with this 1-10 off work to leverage the output of each machinist? Just what is the blueprint for an optimised shop catering to this work segment?

Appreciate all input.
 

Job Shopper TN

Cast Iron
Joined
May 17, 2015
Location
Southeast TN
I have not direct experience, but working in a shop that did this kind of work for years, the things I always lamented that shop not taking more advantage of includes:

5 axis and 4 axis milling to reduce the number of ops on a part

Spindle probes and toolsetters on all mills to boost setup times, toolsetters on lathes

More uniform equipment, even in an all Haas shop, when some have 4000 RPM, others 7500, one 12000, one has TSC, one has a probe, one has a 4th axis - that makes it where efficient processes can become a bottleneck.

Standardized tooling is also helpful, if an operator only has to build two tools instead of seven, because five of those seven are kept built and stay in the machine. Some shops break down every tool each job and sometimes you have to, but if it can be avoided it helps a lot.

Organization. Building your workholding and building your toolholding are some of the most laborious parts of setting up. So if all your toolholders are here, but the cutting tools are there, that doesn’t help anything. Same with workholding, if vises are here but stops are here, parallels are there. In my mind, ideally, every machine has its own set of these things. Vises, holders, parallels, basic shop tools.

And ways to quick-change workholding. The next job might be 6 ops in a 3axis mill but held in one hand. But after that could be a plate the size of the table. Being able to switch from one extreme to the other, and in between, is huge because setup time, even if charged for, is still downtime. This comes through smarter workholding, pallets, subplates, fixtures, that kind of thing.

Cycle times only get you so far in the 1-10 piece world, setup times can make a much bigger difference. Not that cycle times don’t matter, and of course, the more parts in a run, the more it matters. 5 minutes each on one part is negligible, 5 minutes each on 10 parts could mean a whole setup.

That’s some highlights that stem from observations spanning 7 years. Some I’ve been able to work on and implement and change, some I haven’t due to budgets and management not buying it (literally and figuratively!)
 

mhajicek

Titanium
Joined
May 11, 2017
Location
Minneapolis, MN, USA
More uniform equipment, even in an all Haas shop, when some have 4000 RPM, others 7500, one 12000, one has TSC, one has a probe, one has a 4th axis - that makes it where efficient processes can become a bottleneck.

This. Bit over a decade ago I was programming for a medium sized job shop. Seems they'd buy whatever used machine they could get a deal on. Over two dozen mills, and each one was different, no two alike. I spent a LOT of time porting repeat jobs over to another machine.

More recently, being a one man prototyping shop, I saved a lot of time using Lang quick change tooling. Put a vise on the trunnion, now move it to a 3 axis station on the table and put a collet holder on the trunnion, etc.

I'd program things to go on the trunnion even if they were only 3 axis, because I wouldn't need to pickup a work offset; COR of the trunnion was always G59 and all the commonly used workholding was in my Mastercam template file. Then of course for the next revision they'd ask for three evenly spaced holes around the outside, and I could just add them, repost, and go.
 

Sol Itaire

Plastic
Joined
Nov 28, 2021
Lean principles

@jash - I am a one-man machine shop inside an engineering company. The company is my customer, and they pay me to figure out machining strategies and produce 1st articles. What they don't want to pay me for (and shouldn't have to) is everything else that is required to prototype a part timely, accurately, and precisely.

You are asking the right questions. The name of the game is reducing waste in your processes and continuously improving the way you do things (which is the core message of Lean manufacturing). Waste is basically any activity that doesn't contribute to achieving the final product. Some examples:
- movingstock out of the way
- looking for tools
- waiting for a delivery
- making the same toolpath with the same tool in the same material again and again
- installing and setting up a rotary in a VMC for the millionth time
- waiting for an operator to arrive at a machine after a cycle completes

Look up a chart of the 7 deadly wastes. These compartmentalize sources of waste so you can address them in your shop. Solutions which start to address the above examples:
- build a stock rack, or at least put the stock on wheels
- stock common tools at each machine
- swap out your slow but cheap supplier for a slightly more costly one that gets you material next day
- carefully specify tool defaults, standardizing stick out, tool holders, tool #, step over, feeds and speeds, etc. in a blank CAD model or whatever your CAD/CAM system works on. Have all programmers contribute to this over time and pull toolpaths from the same "library"
- install quick-change studs on the rotary and the corresponding baseplate in the VMC, tramming and locating the baseplate so the rotary pops in and out quickly. Get an engine hoist or gantry to move a big rotary in and out quickly
- switch a VMC out for a 4- or 5-axis mill, and start cutting 3 or 5 sides of a part in one setup. The time the operator would have spent doing mindless part changes can be spent learning to program or addressing other wastes in the shop

I would not recommend automating any process that's full of wasted time, material, or personnel bandwidth. Be open to your personnel recommending and carrying out improvements, as they have the best eyes on the day-to-day wastes.

Take it one week at a time and dedicate a few hours to addressing one of these wastes. Encourage your personnel to do the same, at the short-term expense of slower production time. This is not a "on your own time" thing, or an "OT opportunity". This directly impacts your future earnings and the ease with which you all do your work.

Good luck.
Sol
 

BluishInventor

Aluminum
Joined
Jul 7, 2020
New poster here from the UK. Our shop is your typical jobshop, high variety and very low volume work across fabrications, machined billets etc. A lot of made to order 1-10 off, very rare to get anything past 50 off. Size and complexity of work ranging across the whole spectrum as well. The shop has a mix of prototrak knee mills, bed mills and lathes, 3 VMCs, vertical borer, small grinding machines, a jig borer and a wire edm. We have a welding bay for fabrications and an assembly and electrical department which allows us to do more integrated machine or tooling builds/rigs as well as their own standalone work.

For the guys who own or are employed by a shop handling similar work type and quantity, what have been the biggest low hanging fruit type efficiencies you've either seen or implemented in your companies? Can some level of automation apply with this 1-10 off work to leverage the output of each machinist? Just what is the blueprint for an optimised shop catering to this work segment?

Appreciate all input.

We have a very similar shop and workload. Add a couple lathes, a waterjet, a few high end 3d printers, and a CNC pressbrake and you got our shop.

When I first got to this shop 4+ years ago, organization was a complete mess. And my background being purely aerospace, I was used to having my most used hand tools at each work stations. So, I had to change this as not a single drawer was organized. I can still hear the opening and closing of drawers as folks searched for simple things that they used every day.

I standardized every machine with the tools required to operate, setup, and deburr at that workstation. Used kaizen foam and organized the drawers to have very similar layouts between machines types and identical layouts if the machines were identical. Not having to search for 5-10mins for a simple vise wrench, stone, or set of parallels is SO nice. And the fact that the tool has a specific home in a specific labeled drawer helps guide people to put things away or helps them find what they need. Organization of standard workstation tools in any shop is a must.

Tooling is another thing to organize. Vises, cutters, hardware for toe clamps. This for us was all over the shop. I spent about $5k on new drawers for an old lista wall system and consolidated all this stuff into a single area and then organized it. Finding the right cutter or fixture is as easy as it could be. Big time saver there.

When discussing cutting tools in a prototyping shop, there is generally like 1 of every conceivable type of cutter. You get such a diverse variety of jobs and you just collect odd stuff over time. But, that doesn't mean you shouldn't have a Standard tool inventory. Since we work on all materials from metals to plastics to composites, it's hard to really build something that is all encompassing. So, the best we can do is at least have this:

- Drills up to .500" in both machine and jobber length. (maybe 12 or 14mm for you)
- End mills in typical sizes in short LOC and Long LOC. For us thats 1/16, 1/8, 3/16, 1/4, 3/8, 1/2, 3/4. We carry this in both 3fl ZrN for aluminum and 4fl for Steels and such. If you do more plastics I would recommend 2fl or even 1fl for slotting sheet type work.
- Our face mills are the same product line across the board. So, inserts are the same size. We keep uncoated bright inserts for plastic and aluminum, and coated for steels or hard metals.
- Small variety of 2fl and 4fl ball mills in common sizes. We have maybe 20 total on hand. Not super common, but just enough that we know we have something in the drawer if we need a general ball mill.
-Spots, c'sinks(82,90,100)
-Taps in Coarse and Fine, both plug and spiral bott/form. Form for small, spiral for larger than 1/4".
-And standard turning tools following a similar pattern. Sharp for plastic/aluminum, coated for steels and hard metals. A few tool nose radii that make sense to us. A couple thread inserts, profiling inserts, grooving, and part off.

You get the gist.

Just find the most common tools used in your shop and keep them stocked. No point in needing something that may make it to you next day. Anything outside this kit that are only for a specifc job, just order it as you need it. Keep leftovers as anyone would. Or if you're really stingy, you might be able to return unused products back to your Tool Supplier which makes sense on expensive cutters that they may stock and you can't afford to.

Having Trak mills(which we do as well) you're programming conversationally alot. But we also have a CAM system that we can program with and use with the Traks, but it's more often used for our 5x mills, 3-4x haas, and mill turn. Either way, if you have a CAM system to compliment your conversational, its worth setting up your tool library with all the standard tools that you stock. Most CAM packages these days allow for assigning feeds, speeds, and strategies for the tools in the database. If you take the time to build this database up, you can significantly improve programming times. The less data you have to input again and again, the better.

Others have already mentioned hot swap vise type tooling. This makes a big difference, but sometimes we need a 5C collet on the trak mill or a manual indexer. We keep these extra types of fixtures and all our vises in our lista wall i mentioned earlier which is right next to the machines. You don't have to walk more than 20 steps from any machine to grab a tool. It wasn't always like this, but we had an opportunity to do some big shelve and machine re-organization to help accommodate a new machine. If your shop layout isn't the most functional, at least have a draft of what a better layout would be so that when the time comes you can just implement it or a part of it.

Continuous improvement can go overboard too. So, don't be afraid to roll back changes that make you less effective. At the most organized shop I worked, really 5S/kaizen heavy, adding a new tool to the CAM library meant I had to add it to an auto tool crib and a tool pre-setter too. It took WAY too long to do this(like 2hrs) and my programming times suffered significantly when getting a new tool for 1 job. It wasn't a good situation.

One last thing. Sometimes folks that are in an environment for long enough will think things are just fine. Then you bring someone new in they are like wtf is this? That's when it pays to listen. Then you can decide if the ROI on the feedback is worth it. Will this change only give me 1hr just this once? or will it give me 5 minutes 300 times a year?

Our track mill drawers 20180216_103938.jpg20180215_075051.jpg20180119_052515.jpg
 

jash

Plastic
Joined
Jan 9, 2022
Thanks for the comprehensive and detailed responses guys, keep them coming.

Currently mulling over all of this and how we could implement in our shop. I'm not the owner but I believe the shop is profitable so currently things kind of 'work' but we are sorta old school (only started using cadcam this past year, most stuff is prototrak conversational at the machine) and there are constant aggravations and issues that keep coming up (productivity, deadlines and deliveries, quality, scrap and rework etc oh and of course wages.. the usual stuff). Some can be attributed to things outside the company but it's clear we are running nowhere near 'optimal'. Standardising and documenting all spindle tooling and workholding is something I was keen on doing anyway since we are a shop built on decades of tribal knowledge and make-do. A chunk of that will be walking out the door in the next few years as the old guard retires. New inexperienced guys will need to know what we have, what it does and where to find it to make it much easier to plan out and complete their work.

One more question though. What are your approaches to high complexity and/or high value work in the 1-10 off range? A lot of this ends up being put on the shoulders of the more experienced guys on prototrak 2 and 3 axis mills and lathes with all the traditional manual equipment. Had some real nightmare setups and processes for some stuff as you can imagine. Some guys are used to it but I don't think anyone likes it for the obvious reason that it's high risk and low reward for both the company and the machinist (no bonus on these jobs if they overrun which they tend to do, and if it's a scrapper most likely from human error everyone feels it). What processes, procedures, software, machinery, equipment etc do people use to make these jobs worth doing? I'm guessing improvement here will come from all the previous suggestions being implemented but at the end of the day a lot ends up on the shoulders of the machinist here.
 

Job Shopper TN

Cast Iron
Joined
May 17, 2015
Location
Southeast TN
One more question though. What are your approaches to high complexity and/or high value work in the 1-10 off range? A lot of this ends up being put on the shoulders of the more experienced guys on prototrak 2 and 3 axis mills and lathes with all the traditional manual equipment. Had some real nightmare setups and processes for some stuff as you can imagine. Some guys are used to it but I don't think anyone likes it for the obvious reason that it's high risk and low reward for both the company and the machinist (no bonus on these jobs if they overrun which they tend to do, and if it's a scrapper most likely from human error everyone feels it). What processes, procedures, software, machinery, equipment etc do people use to make these jobs worth doing? I'm guessing improvement here will come from all the previous suggestions being implemented but at the end of the day a lot ends up on the shoulders of the machinist here.

Take a long hard look at if that work makes you any money. If it doesn’t, and you have no other reason tying you down to it, don’t take on that work. If you do have to do it (part of a package or something), jack your rate up to compensate. If $70/hr is good for basic jobs, like simple milling, loose tolerances, drilling and tapping laser burnouts etc., then it should be $120/hr for parts that take your highest skilled staff just to struggle through.

After years of taking on crap jobs like that, that you can’t possibly put the right time on, we took a hard look at it and got out. It was either that, or ante up bigly on equipment. Given the skill pool in this area, I think we made the right choice. Now if an RFQ comes through that makes us shiver, we let the next poor sap take it. No-quoting is seen as a sin in some places and I never understood that.

As an aside though, 5-axis can really open you up to do tougher jobs better. But it goes with better holders, a more accurate machine, so on.

I’m in the same boat as you man, I don’t own the place, never will, but I try to act and work like I do. People will get in your way, I’m dealing with 5 layers of budgetary constraints when I can see us bleeding off money through some of these many inefficiencies. But, it is what it is. If it gets frustrating enough, there’s always another shop to ply your juicy brain. Not settling for good enough and wanting to do better is getting rarer and rarer, fight the good fight.
 

barbter

Banned
Joined
Oct 27, 2007
Location
UK
This. Bit over a decade ago I was programming for a medium sized job shop. Seems they'd buy whatever used machine they could get a deal on. Over two dozen mills, and each one was different, no two alike. I spent a LOT of time porting repeat jobs over to another machine.

More recently, being a one man prototyping shop, I saved a lot of time using Lang quick change tooling. Put a vise on the trunnion, now move it to a 3 axis station on the table and put a collet holder on the trunnion, etc.

I'd program things to go on the trunnion even if they were only 3 axis, because I wouldn't need to pickup a work offset; COR of the trunnion was always G59 and all the commonly used workholding was in my Mastercam template file. Then of course for the next revision they'd ask for three evenly spaced holes around the outside, and I could just add them, repost, and go.

This.
IMHO, Individual machines with different controls, pullstuds, tool holders etc is a nightmare.
You at least need cells of at least 2x machines, which are identical in all ways. Makes planning of work so much easier.
Different machines (travel etc) isn't a bad thing, but again standardise.
My shop ended with 4x Chevalier #40 machines - all 10k spindle, all 24tool, all the same spindle, all the same grid plate on the bed, all with the same XY centre location in the table and Z spindle to bed identical.
Easy peasy to run the same job on 2x machines, or if a spindle seized, swap the setup to the adjacent machine.
Also had 3x #30 machines, all the same with 15k spindles and 30 tools etc all set the same way as the #40's.
So 2x cells - the #40 larger machines that can cut anything, and the smaller #30 machines that were primarily for non steel.
Lastly 2x siemens lathes - identical make/model so spindle liners, jaws, tool holders, programs could swap etc etc etc.
Standardisation is a wonderful thing, if only people would do it....
 

ttrager

Aluminum
Joined
Jul 23, 2015
There is advise in the thread representing experience way beyond mine. Some good stuff.

As a miscellaneous call out:

Look for opportunities to eliminate "walk aways" necessary to complete an operation.

One of the things we've improved on is making sure all positions have all necessary inspection fixtures so we reduce/eliminate the need for people to walk away from the position to complete machining or inspection of features. Where ever it is possible and makes sense.

Examples:

Between Center Checks. Instead of having multiple positions walking away to check runout "over there", we've deployed between center beds at each relevant position. No more walk-away to get that snappy-quick check done.

Concentricity fixtures. Historically we used v-blocks and a dial indicator, but several years ago we invested in St. Mary's Spinrolls. We have a bunch of these distributed through the shop so that people aren't walking away to borrow one.

Exception Example - Shared: Optical Comparators. Obvious is obvious: These are expensive and fairly large, so not everyone gets their own. But we do have three throughout the shop in strategic positions.

Bottom Line: Look over the dynamics at your positions and see if there are ways to eliminate "walk aways" to complete operations. (within reason & budget).
 








 
Top