Error Proofing / Machine Setup Responsibility - Where to draw the line

ncsteinb · Sep 2, 2021

Hello, I have a quandary regarding error proofing our machines. We recently had an issue where a lot of parts were drilled too deep. What controls the drill depth is the drill length (which may vary, as they are long and get reground often) and a prox switch to indicate the drill depth (which is adjustable). One idea to error proof this issue is to make the prox switch location harder to adjust by using a special bolt (like a tamper-'proof' bolt, or whatever). Another idea is to ensure that the operators understand how to adjust it properly (this is usually only adjusted on machine setup or drill changes) and ensure that the setup parts are inspected to catch this condition.

My concern here, is that our QM wants to error proof and lockout as much as possible, whereas I think that locking out machine adjustments will just make setup harder and not really solve the issue of incorrect drill depth. I would rather address the issue by educating the setup machinist and operators of this adjustment, and inspect parts for this issue, and let them make the adjustment as necessary.

I know there is a fine line here... But I just don't know where to draw it. Any ideas??

(This isn't specific to one machine or process. I'd rather address this as where does the responsibility lie, on the setup machinist, or just error-proof into oblivion?)

boslab · Sep 2, 2021

Don’t you have first off part inspection, machine gets set, run, first part 100% inspection before running then inspect as per your QA plan, 10, 50th etc
If your just setting a machine and letting her rip your kind of begging for a complete clusterfuck to occur imho
Who sets it checks it signs for it and walks if it’s wrong,
Mark

Lewie · Sep 2, 2021

There are LOTS of things you cant "Legislate" (like error-proofing) :-) So Educating is probably the best that can be done in most cases.
...lewie...

FredC · Sep 2, 2021

I took it the drill adjustment was changed after the first article was approved. Drills do not last forever, they get dull and push back or break, they will need readjustment and making it harder or impossible for operators will be counter productive. You make an adjustment you have to wait till you get a finished part and check it. On multiple spindle automatics that might mean making 5 or 7 parts till you see the results. I remember the days when I drilled through 1200 feet of material with a single drill, that has not happened again for 30 years now. Seems to me you will be more likely to make scrap with a machine that is difficult to adjust.

ncsteinb · Sep 2, 2021

boslab said:
Don’t you have first off part inspection, machine gets set, run, first part 100% inspection before running then inspect as per your QA plan, 10, 50th etc
If your just setting a machine and letting her rip your kind of begging for a complete clusterfuck to occur imho
Who sets it checks it signs for it and walks if it’s wrong,
Mark

Well, I'm told we are doing this... However, we still managed to make about 300 parts wrong... So, QM is asking engineering to error-proof the machine to prevent the setpoint from changing. But my argument is, "Well why wasn't it caught and corrected?". I see this as any other tooling/work offset. Why lock this one out, and not the other drills/inserts/taps/etc.?

Joe Gwinn · Sep 2, 2021

Is there a way to "touch off" the drill point, eliminating the effect of varying drill lengths?

ncsteinb · Sep 2, 2021

No, this is a gun drilling operation, with 6 heads. The adjustment between drills are done with an adjustment screw at the base of every gun drill.

Aaron B · Sep 2, 2021

I too agree this is an operator education and discipline issue. It's not going to be corrected by welding the prox switch in place.

Being in maintenance this is a common thorn in the side I deal with a lot. Production and quality don't want to deal with problems at their root, so they task maintenance and engineering with trying to idiot-proof the problem. In 20+ years I've never seen it succeed.

Sent from my iPhone using Tapatalk

mhajicek · Sep 2, 2021

Do a root cause analysis. Ask "Why?" five times. Why were the parts bad? The drill was set to the wrong length. Why was it set to the wrong length? (Maybe it wore too short and someone adjusted to compensate but did it wrong?) Then why did that happen? Keep going until asking "Why?" makes no sense and you have all the causes. Then the solution will be obvious.

CAMasochism · Sep 2, 2021

mhajicek said:
Do a root cause analysis. Ask "Why?" five times. Why were the parts bad? The drill was set to the wrong length. Why was it set to the wrong length? (Maybe it wore too short and someone adjusted to compensate but did it wrong?) Then why did that happen? Keep going until asking "Why?" makes no sense and you have all the causes. Then the solution will be obvious.

Management doesn't like it when your root "why" is "Because you hired a bed wetting idiot who still relies on mom to tie his shoes."

DDoug · Sep 2, 2021

ncsteinb said:
Hello, I have a quandary regarding error proofing our machines. We recently had an issue where a lot of parts were drilled too deep. What controls the drill depth is the drill length (which may vary, as they are long and get reground often) and a prox switch to indicate the drill depth (which is adjustable). One idea to error proof this issue is to make the prox switch location harder to adjust by using a special bolt (like a tamper-'proof' bolt, or whatever). Another idea is to ensure that the operators understand how to adjust it properly (this is usually only adjusted on machine setup or drill changes) and ensure that the setup parts are inspected to catch this condition.

My concern here, is that our QM wants to error proof and lockout as much as possible, whereas I think that locking out machine adjustments will just make setup harder and not really solve the issue of incorrect drill depth. I would rather address the issue by educating the setup machinist and operators of this adjustment, and inspect parts for this issue, and let them make the adjustment as necessary.

I know there is a fine line here... But I just don't know where to draw it. Any ideas??

(This isn't specific to one machine or process. I'd rather address this as where does the responsibility lie, on the setup machinist, or just error-proof into oblivion?)

"If you machine operators don't start making good parts, we will automate you out of a job"
These people aren't held accountable ?

TeachMePlease · Sep 2, 2021

Make a jig to pre-set the tool lengths sticking out of the drilling head, so that prox switch can stay stationary for the entire run. More points of adjustment = more points of failure.

Dan from Oakland · Sep 2, 2021

Why are you not checking parts at the machine? Make it easy for the operator to check the hole depths.
Make some simple go/no go gauges for the operator to use to check your hole depths. Implement a quality check at the machine where you check hole depth after X number of parts and enter the go/no status on an inspection sheet at the machine- pretty simple to do.

Address the root cause- you can't blame the machine for making bad parts.

ncsteinb · Sep 2, 2021

Everyone, thanks for your input.

Ultimately, it sounds that the right approach is to make sure that our operators are trained and have the tools to inspect/adjust for these conditions. I wanted a good "gut check" on some ideas that I had, and it seems that I'm on the right track.

I designed some Go-NoGo gauges to check for this condition, but ultimately, the operators need to check for this, and also ensure that the parts are being fully inspected after a machine setup/change. As a manufacturing engineer, I would prefer to not 'idiot proof' the world, as this get complicated rather quickly. I would hope that we have operators and machinists capable of checking their parts, and making changes if necessary, as do ask them to do with other offsets on other machines. I also trust that our operators and machinists aren't hap-hazardly changing machine settings without checking their parts afterwards.

Thanks for your input!

Doozer · Sep 2, 2021

Also....What happened to first article inspection signoff ? ? ?

--Doozer

Rick Finsta · Sep 2, 2021

I'm with TeachMePlease. I've set up jobs that had issues with drills breaking periodically (stamped parts getting secondary ops). I'd set up a jig to make sure the new drill was set to the same length as the original. Even with a toolsetter in a CNC, on the particular parts if the drill was set too short the toolholder would crash into the fixture.

I color coded the drills to the setting jigs.

CarbideBob · Sep 2, 2021

ncsteinb said:
.... As a manufacturing engineer, I would prefer to not 'idiot proof' the world, as this get complicated rather quickly.....

As a manufacturing engineer this is your job in your small piece of the world and why you get paid more than others.
What I see here is "educate" or more training which is just another way of blaming the employees and saying "Not my job".
Yes one has to think hard as employees can be far above average idiots at times.
Sorry to be brutal.
Bob

chuckg7442 · Sep 3, 2021

SOP should be to check the dimensions, especially after a tool is sharpened or replaced. When the drill gets replaced or sharpened, the operator needs to verify that the offsets are correct. That is basic machining 101. One cannot idiot proof complacency. If your shop doesn't have this in the SOP, it needs to be added ASAP.

boslab · Sep 3, 2021

With a six bar auto once the cams are set, the machine stays where you put it, tool wear formulas tell you when to change out tools, so far so good, without a QA plan and at least a rudimentary understanding of statistical quality control aka SPC your going to run into trouble, every time someone has a tinker prepare for 100% rejection, sometimes your lucky and rework is possible but that’s rare
You have what’s called an out of control production process, you need to rectify it or the alternative is grim
Mark

Mtndew · Sep 3, 2021

ncsteinb said:
Everyone, thanks for your input.

Ultimately, it sounds that the right approach is to make sure that our operators are trained and have the tools to inspect/adjust for these conditions.

It sounds like they hire people off the street to just load parts and push a button?

I've seen deep hole drill houses and plating places that have this kind of hiring philosophy.
It usually ends up biting them in the ass.

Can you fab up some kind of fixture that checks the switch location before drilling? Some kind of no-brainer depth gage that is light and easy to use?

Error Proofing / Machine Setup Responsibility - Where to draw the line

Plastic

Titanium

Cast Iron

Diamond

Plastic

Stainless

Plastic

Aluminum

Diamond

Stainless

Diamond

Diamond

Titanium

Plastic

Titanium

Stainless

Diamond

Aluminum

Titanium

Diamond

Similar threads