Hi everybody,

I am new with SPC. I am working with a company that assembles spring brakes.
In SPC we control the following items:
• clamp angles (45º),
• port angles (90º), and
• clamp torque.
The tolerances are the following:
• angle is +/- 10º and
• clamp torque is Min = 180 in/lbs, target= 300 in/lbs, and not max.

We use quickie98 software for SPC, and almost every time we log the information in the software it shows that we are out of control for different reasons.
• Average OOC above
• Average OOC below
• Range OOC

I don't know what to do in this case to have the process in control.

The way we check the angles is by using a protractor so it is not easy to get an accurate angle. I made a comment in the file I attached where we put the protractor.

My question is what do I need to do? I am asking because we are within tolerance. I don't know what I need to say to the people who assembe the units. I don't know if we chose the wrong item to control. This company use SPC in those items before I started. I don’t know what process I need to use SPC.

The same happens with the clamp torque. The range is big and for the product it is ok, but for the process it is not good for SPC. Do you have a recommendation for me?

I don't know if I was clear enough.

Thank you for your time

Rosana

It appears to be a rather small amount of data, but the more recent points appear to have changed. It might help if you could supply the raw data also.

But in these cases, if this has truly gone "out of control", then start asking "what changed"? Since all the early values appear to be exactly the same value, that may be actually more the place to look. Are all the data from the same inspector? Why were the early points the same? Rounding off?

I think you have sort of answered the question yourself when you mention that "it is not easy to get an accurate angle".
IMO, the out of control instances may be due to the measurement system rather than your manufactruing process.
I guess that the protractor process can just about measure to the requirements of the tolerance, but I think the mfg process is much better than that, and the protractor can't realistically detect process changes.

Might be completely wrong! Why not try doing a Gauge R&R study to confirm?

Dave

thank you both

I will prepare R&R study this week

Rosana

thank you both I will prepare R&R study this week
Rosana

Rosana,
If you are interested, provide details with
Better picture (enlarged portion of the feature being measured) and current measurement method (with measuring equipment details), I can provide my views/assessment on measurement method.
You can approach me through PM.
Regards,
Govind.

Hi,

What is PM? :o

I will attach some of my Charts

Port Angle: Line 1, Line 2, Line 3 & File 4 - File #1
Clamp Angle: Line 1, Line 2, Line 3 & Line 4 - File #2
Clamp Torque: Line 1, Line 2, Line 3 & Line 4 - File #3
Push Rod Length: Tolerance 7.8+/-.2

I will try to explain later how we measure all I have for SPC

thank you,

Rosana

Hi again

File #1 Port Angle

Rosana

File # 2 Clamp Angles
Line 1
Line 2
Line 3
Line 4

Rosana

File # 3

Clamp Torque
Line 1
Line 2
Line 3
Line 4

Rosana

File #4
Push Rod Length

Rosana

File #4
Push Rod Length

Rosana
Do you have the raw data that goes with these files?

Hi again
I am sorry I didn't reply sooner. I had problems with computers
I attached the information requested.
Thank you for your help

Rosana

Out of interest, why are you doing SPC on this feature ?

One thing you might want to consider is, rather than the traditional x&R charts, use pre-control charts. There are several articles available on how to use precontrol charts, or if you can't locate any PM (private message) me, and I will be happy to share my knowledge (admittedly limited) with you.

One thing that is critical to ANY process monitoring, however, is knowledge of the accuracy and repeatability of your equipment. If you can't repeatably measure the angle (for example) then all the data collection in the world won't help you control your process.

Hope this was of some help.

One thing you might want to consider is, rather than the traditional x&R charts, use pre-control charts. There are several articles available on how to use precontrol charts, or if you can't locate any PM (private message) me, and I will be happy to share my knowledge (admittedly limited) with you.

One thing that is critical to ANY process monitoring, however, is knowledge of the accuracy and repeatability of your equipment. If you can't repeatably measure the angle (for example) then all the data collection in the world won't help you control your process.

Hope this was of some help.

Just a dissenting opinion - I am very wary of the so called pre-control charts as they (at least what I have seen) are not statistically based, but are based upon the specifications, thus confusing the "voice of the process" with the "voice of the customer" (credit to Dr. Wheeler on those phrases).

Also, if you review your data and are doing good statistical analysis, and are willing to take action, the data would send you to repeatability problems with the measurement equipment. I am afraid the logic would turn into "we aren't sure of our equipment, so we won't do any analysis" and then nothing would happen. Even bad data is of value, because you can learn why it is bad and that should provide value as you can then correct the "badness".