Is loading criteria sufficient in quality control of flex shaft core?

[rollingcloud]

Is it a good idea to remove the heat treat process control on the flex shaft core? The new vendor has a heat treat process that is a trade secret and is asking us to remove the standard heat treat process. I proposed specifying a hardness range after removing the heat treat control, but they argued that having hardness control on a flex shaft core (which is made of a bunch of stainless-steel wires) is not an industry standard. Currently, it has a heat treat process control per AMS 2759/3 with a minimum hardness and the following mechanical requirements:
THE TORSIONAL LOAD AND DEFLECTION CRITERIA AT A 7.5 MINIMUM RADIUS ARE AS FOLLOWS:
AVERAGE OPERATING: 11 IN-LBS.
MAXIMUM OPERATING: 25 IN-LBS.
TRANSIENT MAXIMUM: 90 IN-LBS.
TORSIONAL BREAK LOAD: 240 IN-LBS. (IN STRAIGHT CONDITION)

I am thinking maybe it should have a max hardness at very least if we are removing the heat treat process to ensure it's not too hard/brittle? Which would meet the torsional load and deflection requirement but might have premature failure.

 

[SWComposites]

How does the vendor propose to meet the mechanical requirements?
Are they going to test a sufficient number to validate that their process produces parts that meet the requirements?
And what does the vendor propose for quality control check if they don't want to perform hardness tests?

 

[rollingcloud]

They did say that they will be able to meet the torsional load requirement but did not mention any new sampling plan. I know they do check it, just not sure at what AQL.

The raw material AMS specification has two conditions: one is "as received," and the other is a CH condition. They want to remove the CH condition heat treat control. Each condition has a tensile requirement. So maybe I can ask them if they are willing to use a tensile requirement instead of a hardness range? I’m just not sure if the tensile requirement is a range or if it is also a minimum, will have to check tomorrow.

But if they agree to have a maximum hardness value, would this, combined with the torsional break load, be enough to ensure a quality part? Since the torsional break load ensures the material has enough strength and hardness (lower bound), and a maximum hardness value prevents it from being too brittle (upper bound), which affect its fatigue life?

 

[mfgenggear]

Rolling cloud
Run a test sample of 5 parts. Or less, ask for fixed process , and metlab certifications, sign an non disclosure agreement (NDA),
Test the 5 parts for all met lap properties. Inspect and record.
If they don't agree , then run
Must document all procedures.
I don't care if it top secret

 

[SWComposites]

What tests (static and fatigue) were done to qualify the existing flex shaft? Repeat those qual tests for the new vendor.

And find out the sampling plan?

But I am usually very, very leery of a “trade secret process”. Is this vendor reputable? Where are they located? How long in business? Frankly you should insist they set up an NDA with you or a cognizant engineer at your company and then review the process in detail with that person. This is done all the time in aerospace. You can’t take the risk they are hiding some crappy process.

What is your role? QA? M&P? PM?

 

[rollingcloud]

mfgenggear,

Thanks for the suggestion. We do have NDA with them, I misspoke earlier, they are new vendor for this particular part or the next lot, but they are not a new vendor to us. I could ask for a fixed process or a VSR, but what to control in the VSR?

 

[mfgenggear]

Rolling cloud
What is a VSR?
Visual

 

[rollingcloud]

SWComposites,

Thanks for the insights. We do have NDA with them, I misspoke earlier, they are new vendor for this particular part or the next lot, but they are not a new vendor to us. They are good at what they do, I do trust them, but I just don't like leaving things unconstrained and unchecked.

 

[rollingcloud]

vendor substantiation requirement

 

[desertfox]

Hi rollingcloud

I don’t see how the above load criteria affects there heat treat process, let’s face it you don’t care how they achieve the required hardness, flexibility and strength requirements so long as the shaft after manufacture does what it’s supposed to do. Bottom line I wouldn’t remove the load / mechanical criteria information because if you change supplier you will end up putting the same information back on the drawing.

Another good reason for me not to do it -
Many years ago I checked a bracket for lifting some equipment and I quickly realised that the bracket bolts wouldn’t hold if the said bracket was fastened in a certain orientation, which was physically possible for this to happen. So I had notes and a big blow up area on the drawing to show how the bracket had to be positioned and bolted.
The customer didn’t read the drawing and fastened the bracket just as I feared, fortunately the problem was spotted during lifting by an observer who saw the bracket and the item it was lifting separating and the lift was halted. The first thing the customer did was try to lay blame with the company I worked for until of course the studied the drawing. So I could of verbally instructed someone how to fasten the bracket on and left the drawing blank, given that verbal conversations tend to be difficult to prove in court, I preferred to put the criteria in writing and you should do the same. If that shaft of yours fails in service how will you prove that mechanical strength and criteria was specified?






“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[mfgenggear]

Rolling cloud
Stick with a fixed process, and do add as stated above

 

[rollingcloud]

desertfox,
Thanks for the example. While the function tests do not directly control the heat treat process, the tests would indirectly control the hardness of the flex shaft to some degree, the shaft core construction might be a bigger factor though. I don't intend removing the mechanical requirement, just want to make sure the upper bound of the hardness is constrained.

 

[desertfox]

Hi rollingcloud

If you specify a hardness on the shaft made from a bunch of wires how do you verify the hardness?
I don’t know what your heat treatment process was on your original drawing, did it specify a hardness?
If you are going to specify an hardness then what range or tolerance is acceptable? It’s impossible to specify a single hardness and your original post states it had on it an original hardness but of minimum value ? so how come you are now worried about the shaft hardness being to high?

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[mfgenggear]

Test sample of the same material are run with the heat lot. The test sample will provide the proof that the times, controls, and temperature were ran correct. If the wires are thick enough, a micro mount can be made to micro hardness test.
But there woul have to cut up a test part.

 

[MintJulep]

Your mechanical requirements are ambiguous.

THE TORSIONAL LOAD AND DEFLECTION CRITERIA AT A 7.5 MINIMUM RADIUS ARE AS FOLLOWS:

Load and deflection, so I might imagine that one end is held fixed, and the specified load (in this case a torque) is applied at the other, and the deflection is measured. However, you don't give any deflection criteria. So maybe you mean something else, but I can't imagine what.

AVERAGE OPERATING: 11 IN-LBS.

Is this an operating condition as stated, or a test load? Average over what? Over a time, over a number of samples, over the entire population?
Perhaps you mean if one end is free then it can't take any more than 11 in-lbs (average) to turn the flex shaft from the other end?

MAXIMUM OPERATING: 25 IN-LBS.

So, it's ok for some to meet this requirement (whatever it means). How many? If it's ok for some, why isn't ok for all, what's the purpose of the "average operating" requirement?


TRANSIENT MAXIMUM: 90 IN-LBS.

Transient means a range over a time. What's the range? What's the time?


TORSIONAL BREAK LOAD: 240 IN-LBS. (IN STRAIGHT CONDITION)

Does this mean that it must not break at 240 in-lbs, or that it must break at 240 in-lbs?