Are there any Toolmakers in here? Advice Required....

[Sirius2]

This is going to vary between the 'standards' of toolmaking companies out there, and even country to country, but I was wondering if anyone can advise me on the procedures to put in place specifically for toolmaking and fixture work, and general advice on how things should be run, or how you run your operations as Toolmakers.

Im trying to drag the company upto date, and get them out of sloppy attitudes, but I need help and confirmation that its a worthwhile persuit to keep pressing for changes.

We are (allegedly) an ISO9001/9002 accredited company, which means that all our designs for customers tooling and tools leave the factory 'to drawing'........but this never happens.

The situation we have is this.... The manager gets a job request, he discusses it with his family members (family run business) and puts the quote in for the tooling, (usually without consulting the design staff about how they would do the task).

He then passes the work down to the design department (my department) where two of us work. We recieve the CAD data from the customer and any instructions, and design the tooling to make or check the part.

We issue the drawings (Revision A) and thier status is "Working", which means that the job is active and in process. (Whereas when its finished and accepted by the customer as functional, it recieves "controlled" status).

So, we have issued the drawings. The purchasing staff order the materials and any bought out items, and the manager takes a look at the job before issuing it to the shop floor.

The shop floor make the pieces (in our place, people are constantly moved from job to job, so no toolmaker ever seems to be in overall control for the whole project), BUT, they make mistakes sometimes, and sometimes go against the design intent to "get the job out". As there are that many people on and off the job in hand, nobody is ever fully aware of who has done what, what stage things are at, etc etc.

Designs are frequently altered *on the shop floor* to suit the errors of the manufacturing equipment, ie, the EDM machine never wires straight so they add extra dowel holes to details to make sure they can jostle the part on the assembly to 'fit to suit' whats been made. Or they may move holes because of mistakes, or add holes for ease of clamping to machine tables etc etc.

None of this is controlled. Its all word of mouth and 'free range'. We have gone over this many times over the years and things "pull in" for a month or two when somebody kicks a stink, and then it reverts back to chaos again.

Also, for example on a pressed part, we design a tool and for whatever reason the component may not come out properly off the tool at the end of the day, but as nothing is inspected as the pieces are being made, we dont know whether the forming stages are right or wrong to drawing, and then because the parts are wrong and the tool is late, people are given the free range to "hack and slice" the tool about as they see fit, and us responsible for the design dont get to know about it, or have a severe headache catching up with the mayhem.

Also, I must stress, that its not uncommon for the management team to make "inspection fixtures" SUIT whats coming off the tools to fool the customer and deliver on time when these errors occur.
We are then told (which is hit and miss if we are even told or not) to ammend the CAD DATA to relect the "butchering" thats gone on on the shop floor - ie make the CAD look like the WRONG job so that the data sent to the customer 'checks out' on thier inspection.

This leaves me stressed out because things are leaving the factory not to drawing, nobody knows whether pieces of tooling are made right or wrong, and people are compensating for thier lack of accuracy by adding extra features and changing hole sizes of dowels from say 8mm to 10mm as they go along, the part is not conforming, and when things come back to us we really dont have a clue WHAT has happened or what the part should be anymore.

Obviously for the design staff its a nightmare, because the job has become 'organic' and out of control, and we are NOT conforming to the ISO9001 accreditation - which the Management see as a bugbear rather than something to help them.

When it comes to design and manufacture, Im an idealist.

I believe in geometrical tolerances, and proper limits and fits conveyed on the drawing - we currently dont do this, and people arent convinced of it working and are reluctant to change. "Ive been in this job for 20 years, its always been like this and Im not going to change now" etc etc.

The way I see it is that if every piece was made to Drawing, the whole thing should, in theory, fit together perfectly.

The way *they* see it is that things will never go together properly, so why bother putting the effort in to be precise when you can 'make do and mend' to get it to work at the end of the day by any means necesary.

Its like a stack up of errors, say there are 5 pieces that build up a part, and the most crucial part sits on the top....all the parts are dowelled together with various angular faces and features etc etc.......

......To me, each piece has potential to be slightly wrong from 100% perfect limits, so, by the time youve stacked up all the potentials for errors (hole positions, angle face errors etc)there is possibility of the final piece being out of place/wrong. This, I gather is the designers job to apply accuracy to each piece so that, within allowable limits, the thing can assemble to the final item and be correct to drawing.

The way they work is to make all the pieces as they see fit, getting 'somewhere near' on each piece and then for the final part sat on top, make the part NOT conform to drawing so that it functions right after their manufacturing processes and errors have crept in.

This then leaves the CAD DATA wrong for the part on top, should there ever be need for replacement in the future.

This is bad news in my opinion, and making gauges to suit parts rather than fixing the tool to conform to gauges, and management coming down to verbally tell the shop floor not to do this or that on the design, and the shopfloor taking it on themselves to 'modify' the design as they see fit.......well, its just CRAZY in my opinion.

Another thing, is there is no "project manager" here, so - to give one example - its often the case that one person is making the 'baseplate' for a job, somebody is making something else for it, and nobody really studies what the total job is doing, each burrows away at thier own station happily making thier own little piece.
We had a case this week where somebody had spent all day machining chamfers on a plate when they could have been marked off with a scribe and cut on the horizontal miller in about 1 hour.

I feel guilty because I feel the drawing should have stated +/- 5MM on the dimensioning so that it was obvious, but seeing as they dont work like that, and things are given a general tolerance of 0.1, and nobody understood what was and wasnt important, close to a full a day was wasted machining and grinding up angles.

Personally I would like to see full geometric tolerances applied to parts, I would want every piece to conform to drawing from the outset, and any design changes logged and tracked through the appropriate mechanisms and reissued to the shopfloor as a design change mod.

I would like to stop the managemnt instructing verbally to the toolmakers to 'not bother' about this or that which they dont have a real clue about how important 'this or that' is.

I also see no reason why I cant expect parts to be accurate, even over lenths of 1.5 meters, to within 0.025mm on dowel centers or die hole centers, or why a Wire EDM profile cannot be positionally correct to its datums within 0.05mm.

Am I asking for the unreal? Im beginning to doubt myself!.

Any advice would be welcome on procedures and the whole arguments for and against accuracy and geometric tolerances etc in Toolmaking situations.

Thanks for your time

Sirius2

 

[ctopher]

I worked at a company like that once, and privately owned.
Seems to me it's not a design/drawing process problem, but a QA/Eng process problem company wide.
From my experience, private companies will not implement such a process because of cost. They will do what it takes to get the job done.
Sorry, I don't have much suggestion to fix the problems other than leave the company. It is what I did.

Chris
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 10-27-06)

 

[xgrigorix]

The only thing that I can think of is that if a customer ever needs a replacement part, there is no way that your company will be able to provide one that actually works!

Has this ever been pointed out to management?

 

[ajack1]

Wow, that sounds like the company is in a right mess Sirius 2.

I have never seen a toolmaking company that works to fully geometrically toleranced drawings, that would be an extremely expensive way to work and working off the cuff is the norm in most companies.

To use a simple example if you have a flame cut sub plate that is plough ground to do try and make it an exact thickness or is it quicker and cheaper to just adjust a few pocket depths and the height of setting stops to suit?

There is absolutely no reason why wire cutting should not be within 0.005 mm, but do the outside dimensions of a die actually matter, unless of course they butt together or have a cutting function?

The one thing you do need is for each tool to be project managed be this by a toolmaker, a supervisor, a manager or whoever but someone has to do it, that seems to be the biggest problem from what you have said.

Toolmaking will always be done “to suit” as in many cases things like cutting stages are developed using tooling before final cutting is done, the same applies to forming, you will always get spring back or thinning that needs to be sorted as the job progresses. High-end software like Pam stamp can reduce this dramatically but I have never seen it eliminated completely.

So for my views, Toolmaking by its very nature is not suited to hard and fast design, but for this very reason it requires constant and precise project management, this seems to be where the company you work for falls down.

 

[Sirius2]

Thanks for your quick replies!

Ctopher, Im not sure where youre based, maybe the United States or Canada?....but to be honest, over here in England, Engineering is "on its arse" generally, and finding another job as good as I have got (built myself up), is going to be difficult!.

I know what you mean though, as you can only beng your head against a wall so many times before you simply stop as it hurts.

xgrigorix, Youre right, theres a snowballs chance in hell that anything we get back to modify in the future, or need replacement parts to be made for, wil fit the job sraight off as it should. Or its a case of "oops, we cant do that alteration we've just designed beacuse actually theres now a hole that runs right across where we need to change"...

It has been pointed out to the management, but it gets political because they kind of rub thier hands together becuase they can charge the company say 20 hours to sort it out rather than the say 5 hours it should take.
Me, as a designer, want to get things right.....sometimes, companies want to make money by relying on things being wrong.....do you know what I mean?. Its sometimes a conflict of interest between a designers will to do the job right, and what, in various ther ways, will make the company money, and in some examples, the fact we are inneficient makes them money.
Its crazy really, when you think about it! But, if they didnt do this, the company could probably fold and Id be out of a job!....hmmn. But then people wonder why jobs are going to India and Poland. China etc.


Ajack1, I know fully geometric parts for all pieces of a stamping die would be unrealistic really because yes, the cost would be too high.
What I envisage though is applying things sensibly to datums (ie point out that dowel holes need to be within true position of 0.025mm), or that forms need to be within 0.1mm of true shape, etc etc. The rest would be left to general tolerance figures and some common sense.

We generally apply a decimal procedure, where 3 places equals 0.01, 2 places equals 0.03, 1 place equals 0.1mm and zero places equals 1mm.

As we dont apply proper limits to every plate etc, we generally, for speed, dimension most things to 1 decimal place, and give 'tied up' features 2 places (0.03)....and thus rely on the toolmaker to 'know' that such and such an edge is in "fresh air" and an "open limit".

I always ask myself, 'how do you know its right?', when theres no criteria for it to match?.

If I design a plate with two tennon slots in it, at right angles to each other, the company works on the principle that the toolmaker will know that those tennons have to be perpendicular within such and such an accuracy, whereas, in reality, to the drawing, the could be allover the place as nothing stipulates perpendicularity, and as such I dont see how people can say whether its right or wrong.

For one example, say its a long progression tool for a pressed part, maybe 2 meters long. They tell me that they cannot rely on the accuracy to achieve correct postional tolerances for punches and dies, die punch profiles, dowel holes etc etc, pitched in the plates. I dont quite believe this, or, if they cant do it because the machines we have are too old, we need to sort things out so we can do it.

Say we have a die plate with a fancy shaped profile in it, which is held to the dieset with 6 screws and two dowel holes. They ream the holes in the toolsteel, send it for hardening then wire the profile inhouse.

Its claimed (Im not saying theyre wrong), that the tool steel distorts that much in the hardening process that they cant guarantee those centers, and the wiring program is "picked up" from one of those said holes as a datum. This means that the profile and hole relationships could be out of sync to the mating parts.

They should (in my opinion), wire the fixing holes at the same time as the profile, to ensure proper geometric relationship as decribed on the drawing (ie to 0.03mm). I dont see this being too much to expect, but Im told that we cant wire pitches of holes over a distance of a meter accurate enough to match the existing tooling. I find that bizzare, and something amiss.

However, Im out of touch with whats expectable on machine shops. I dont know, for example, how accuarate I can ecpect a conventional univeral milling machine to pitch at. The CNC should be within 0.01 though I would have thought. As for parts fitting together, I dont fully understand why relevant grinding allowance isnt left on, and the part ground to achieve the 'ideal' drawing specification, etc etc.

I dont know if other toolmakers wire thier holes to tie it to the die shape aperture, etc etc.

Ive seen a few companies that do toolmaking use Geometric Tolerances, but it all depends on how 'anal' the company standards are, and whether they are gritty toolmakers or prestine toolamkers....if you know what I mean.

You see, Im sat there, quite happy in my job, I cant really grumble, but all this stuff keeps going on and I see no reason why we shouldnt "tighten up" ship and implement better practices, better machinery if necessary, and better documented designs, as Im sure it could atleast pay for itself - it may not make profit, but we could surely be more efficient and perhaps spend less time 'chasing our tails' un necessarily.

The tail is ever wagging the dog. It used to be worse, and we had lazy design staff too, but I like to think Ive improved things in some ways over the years, but whether thats the case I dont know. Things are better than they were, but I think things need to go even futher......

......but the hard part is knowing what is and isnt realistic, and how far to take documentation, and how much 'logging' of "free range" activities we can cover without it becoming a rod for my own back, and a costly paper trail because the underlaying problems are never sorted out.

You see, I may be well off the mark and should stop trying to aim for 'normal' engineering practices in the Toolmaking arena, and Im willing to accept Im wrong on this, as I see the complications from both sides. Its the intermediate I have to find and its just knowing where that boundary lays between proper 'first class' methodology, and 'get the job done as cheap as posible by any means necessary'.

We do need some sort of a project engineer really, but because people are chopped and changed so much arounf jobs and tasks, you could spend an hour devising a plan of action and going through the design with a toolmaker, and he spends two days getting into it, then he is taken off to do somthing else 'more priority' and then somebody else is given the job to do in the meantime, etc etc........

But thats why I see a proper structure and proper drawings as crucial because as far as Im concerned I should be able to send 5 mating parts to five different companies and be able to assemble them together - ie, it shouldnt matter who's making them, things should be plain and obvious enough and to proper limits to allow it to happen.

Anyway, its good to hear your thoughts. Its no biggie, I was just curious of how "right or wrong" some of my thinking is in the wider world.

Thanks

Sirius2

 

[ctopher]

Sirius2,
I'm in the USA.
I suggest having meetings with management and dept heads and discuss your intentions. I did it, it took a couple years and a lot of grey hairs, but it eventually worked. You have to be persistant and have documentation to back you up.
I think your company needs someone to manage Configuration Management and have some sort of control over processes.
Good luck.

Chris
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 10-27-06)

 

[MikeHalloran]

I just got ejected from a similar company. They don't make tools. But everything they make is late and/or wrong, and they are constantly in chaos.

The source of the chaos is the owner, as in your case.

You can drive yourself crazy trying, but

YOU CAN'T CHANGE A DAMN THING.

Live with it, or leave.

;--

When I learned GD&T in 1967, my teacher said that it came out of toolmaking, so it shouldn't have to be reintroduced. I'd like to think that your outfit is an exception. I'd like to think that about my last outfit, too. ;-)



Mike Halloran
Pembroke Pines, FL, USA

 

[kbro151]

I hear you, Sirius2. My company makes tools for use in the UK and we have been extremely disappointed in the quality, lead time, and costs of the suppliers we try to use.

 

[Sirius2]

What would people recommend as a general operating procedure, starting from when the job is given us to quote?

If thats too labourious, is there anywhere I can look up general 'chains' of command in such companies, or what recommended practices should ideally be?

Im thinking perhaps a flowchart to represent smooth running and complaince to good practices....

Cheers,

Sirius2

 

[Sirius2]

Yeah Mike, Im quite placid, but Im also a stickler for the rules .

I dont let it agitate me to the extent of leaving, but it just frustrates me how people arent striving to improve anything. Things take AGES to move along. If customers hadnt have stipulated we CNC parts back in the early 90's, Im sure we would still be on drawing boards because {to quote} "we have always done it like that".

Kbro151.....Im sorry that youre dissapointed with UK manufacturing. Not everywhere is chaotic, we have some first class tooling shops, and although I gripe at times, we do turn out some good stuff.
We recieve tools from allover the world, and we find shoddy work from China and Japan, and recently a local company which 'emigrated' to eastern Europe ended up sending work back to the UK because they couldnt get the quality over there.

Many companies here are feeling the pinch though, and without investment being made in manfacturing, it does get to the stage where some companies are ever finding 'other ways' to produce on budget.....

But mike, I would be interesting to see the comparison in real terms between an outfit thats 98% "all singing and dancing", and a similar one thats "rough around the edges" wouldnt it, lol.

Ah well. I will put it all out my mind for now.

 

[MintJulep]

Wow, what a mess.

There is an old saying: "In chaos there is opportunity." If you have the balls, quit and set up your own shop in competition.

Back when I worked designing machines, the toolmakers I worked with took great pride in making each part accurately. I had excellent relations with the shop floor, and the toolmakers had no problems coming to me with suggestions for improvement, or questions like "did you really need the default tolerance for this, as it looks like a non-functioning edge, is it alright if I just whack it off on the band saw?" Most of the time they were right. Sometimes I was was right, and after a confirmation or explanation they gladly went back to the shop and made the part to the print. So it is not the same everywhere, not that that helps you.

The owners were crooks, but that's another story.

ISO certs need reaffirmation periodically, and sometimes the auditor likes to talk to engineering. Of course telling the auditor the truth could backfire.

 

[ctopher]

We have procedures for this kind of stuff, but unfortunately I can't pass them on.
Do a Google search or maybe someone else can send you some.

Chris
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 10-27-06)

 

[MikeHalloran]

My experience with toolmakers parallels Mint's, sort of. I like to put huge tolerances on stuff when I can. Guys who are new to my prints will come back incredulous, and I'll tell them that when I put a loose tolerance on a part, I really mean it, and conversely. They always make the parts way better than the loose tolerances, and when I use a tight tolerance, I get it, every time.

I.e., in my experience it's hard to get a toolmaker to make a part that's not perfect, or damn near. There have been odd occasions when I wanted imperfect parts, and I have had to specify exactly how imperfect and in what direction. No kidding.

I'd be disinclined to use extensive GD&T on a drawing to be executed by a toolmaker, not because it adds expense, but because he won't use any bonus tolerance, or any regular tolerance either, unless he's having a really bad day.

Either your boss is hiring machine operators and calling them toolmakers, or he has conditioned actual toolmakers to be sloppy (unlikely), or ... he is driving them crazy too.



Mike Halloran
Pembroke Pines, FL, USA

 

[ajack1]

Sirius 2 I to am in the UK and share your concerns about engineering over here. I was laid off a while ago when the (smallish) company I worked for sold their land for property development, the boss walked away with £10 million and we walked away with minimum redundancy, life’s a bitch.

I then started up a contract design house that specialises in tooling and machining consultancy and programming, our last few projects have gone to Germany, USA, Poland, Turkey and the USA again. So whilst times are tough there are still some good tool rooms over here.

The guys on the shop floor are correct things will distort in hardening, the “normal” way to overcome this is to put small dowels in prior to hardening along with screws etc and wire all apertures and dowels to size after wiring. I would also suggest hard machining any complex shapes after hardening. We design transfer tooling up to 7 metres long and this works fine.

Some of the “smarter” companies we work for now work solely to 3D models but this does require heavy dependency on CNC machining, but the companies that do this seem to get the cream of the work for the world wide market.

As I said originally most tooling has “live” drawings due to changing developments and the like but it does require strong project management, this seems to be where your company falls down. It sounds like you require a major change in attitude from someone up the ladder, good luck.

 

[thundair]

Sirius 2

Can you document a cost savings in building to print?

I would think that with production parts it is more important to control the end item.
On the other hand tooling needs to produce the acceptable part and that is the bottom line.
I use to be where you are, with a QA background and years as a checker, I was frustrated that I have all this good information and no one was paying any attention to me.

Well take a deep breath and get over it. Your job is to do the design and take all the money you can get from that company.. Their job is to make money ...

As far as drawings go (let me get back on my soap box) is to switch to Model Base Definition and with FT&A you can call out the critical requirements and let all the rest fall into +/- .1

Cheers

I don't know anything but the people that do.