:bigwave: Dear Ladies and Gentlemen,
I have 2 questions about Cmk,

1. One of processes is armature turning, we need to measure 5 independent characteristics:
<1> cutting outer diameter;
<2> cutting length;
<3> commutator roughness;
<4> commutator runout;
<5> shaft runout.

My question is for this one process, WHICH Cmk should I take? Cmk<1>, Cmk<2>, Cmk <3>, Cmk<4> or Cmk<5>? I am puzzled because "each" Cmk is different, can any of it stand for the machine's capability??

2. One of processes is automatically assembling 4 screws, do I need to take the smallest Cmk of tightening torque for the machine? or just take the average value of 4 torques for the machine?


Thanks a lot for your help in advance!!!

Buffalo_Hua

Buffalo_Hua, I am not sure I understand the question but I will offer the following:

By Cmk, I am assuming you are meaning Cpk (the capability index of the various process characteristics being measured). You state you "need" to measure 5 characteristics which indicates to me these all are identified as process control items but not necessarily requiring SPC (just periodiac measurement to be sure nothing has changed). Are you asking for advice on which of the 5 features should you use SPC with to determine machine process capability? If so, I feel it would depend on first - have any of the characteristics been designated as "Key or Significant characteristics" which require SPC, and which do you feel illustrates the machine's capability best? If there are some of the characteristics which are dependant on others (if the first is OK then the other MUST be OK) then you can narrow your choices. Only you know your process and how it would be shown best. Sometimes, you have to track the Cpk on more than one characteristic - again, your process should dictate that.

For your second question, it would depend on how the tightening was done. Are the screws tightened by 4 seperate drivers or does the same driver tighten all 4 screws? How critical is the torque value? In some installations I would want to track the torque value for each screw position and in other applications, that may be over-kill. I may choose to use the lowest torque for a "minimum torque" requirement but I can't think of a situation where I would use the average of the 4 to meet a required value.

Just as an aside, I would avoid any Cpk requirement involving torque like it were the plague. There are too many outside variables beyond your control to reasonably expect Cpk in the range demanded by most automotive customers.

Hope this helps a bit, at least until we can get some more input from our fellow Covers.

Dave

Right on Dave and I'll second your motion on avoiding torque Cpk. It's a no-win process.

Dear Mr. Scott,

Thank you very much for your reply firstly!!

Yes, I agree with your viewpoints concerning the Capability studies.

I stated unclear in my previous post, the purpose of performing machine capability studies is our customer's requirement, as far as know, we must select the machines which there has no (or very little) effect on machine due to the reasons of man, material, method, envirovment. therefore, we select 2 automatic machines, one is automatic cutting machine, the other is automatic screw-assembly machine.

Yes, for question one, <1> cutting outer diameter;<2> cutting length;<3> commutator roughness; <4> commutator runout;and<5> shaft runout all are finished at the same time when cutting the commutator, there does not exist the situation of "there are some of the characteristics which are dependant on others" . In the drawing, items <1>, <2>, <3> all are critical characteristics, we need to perform SPC charts, I really do not know which Cmk I should take...

For my second question, the 4 screws are assembled in two steps at one automatic screw-assembly machine by four screw drivers, at first, the two screw drivers automatically assemble 2 screws which are at the cross position, then, the other two screws automatically assemble the remained 2 screws, there are 4 Cmk values for this process, this is also a critical characteristic stipulated in the dwrawing, we also need to perform 4 SPC charts for No. 1/2/3/4 respectively. I do not clear whether I should take the average Cmk value or what else.

Thanks a lot for your kindly advices.



-----


Dear Sam,

Thank you for your kindly attention.


Merry Christamas to all... ;)



---Buffalo_Hua

Right on Dave and I'll second your motion on avoiding torque Cpk. It's a no-win process.

Merry Xmas for you all...

I fully agree with the above statment. I'ld add that torque
is not suited for SPC studies.
For example, an ISO standard for nuts with prevailing torque
states the same idea.

Celso :agree1:

Buffalo_Hua,
For question #1 - You must study each characteristic to determine the machine capability. And yes they will all be different due to the natural variation of the machine, the cutting tools and the operator.
I would caution against using an average of the capabilities to rate the machine.

For question #2 - As stated earlier, avoid capability study for torque, if possible. If not possible you should confer with your customer as to the best approach to validate torque requirements.

Some customers use the term Cmk. After discussed, I feel it's just similar to Cpk. The formula are the same.

Buffalo, I would like to add some comment on your question #1. A machine(or process) can have several Cmk. If it's the customer requirements to submit the Cpk (or Cmk) data, of course their conern is the product features. They want to know your ability to meet their product requirements. So I think you need to sumbit all the 5 Cmk data, even though it come from the same process.

On the other hand, if this Cmk data is used internally to monitor the machine capability, then the concern should be the machine characteristics. I think in a typical CNC cutting machine, it can be the X, Y, Z motion. The rotating motion can be considered in some case (So the several data about run-out can be grouped together).