Issues Identifying the Root Cause in 8D

[dtann]

Hi Everyone,

My first post here and I would like to seek some help and advice on how to solve a real life issue and where the client has asked us for a 8D report.

D2-A machine we supplied had a hydraulic hose that came loose after 3-4 months of usage.

D3-We replaced all the hydraulic hoses & had them crimped from another reputable company near client's location.

D4-The hose that came loose was supplied by us (bought from a reputable hose company) and crimped by a local supplier near the client's location. Our initial analysis was that the crimping wasn't done according to spec. However, this local supplier had done all the necessary tests and even provided video proof of the tests. The hoses were also used in full production for 3-4 months.

This client is pushing us hard for a 8D but I am kind of stuck.
Give a 8D incriminating us ((the factory manager doesn't like us and prefers other brands, however the global head office likes working with us.) without knowing the exact root cause or don't give a 8D and then the head office gets pissed off.

I can't exactly pinpoint the root cause, although the highest probability is incorrect crimping but I can't really say that given the evidence from supplier.
I also cannot discount mishandling by client since we don't have these sort of issues before.

I am thinking of asking client to wait for a few weeks to see if the D3 works before proceeding further. Even if they agree, I don't think I can be closer to identifying D4.

Thanks in advance.

 

[ChrisM]

What does "came loose" mean? Did you or any of your colleagues see the hose in question whilst still fitted in its "as failed" state?
Possibility of sabotage?

 

[dtann]

What does "came loose" mean? Did you or any of your colleagues see the hose in question whilst still fitted in its "as failed" state?
Possibility of sabotage?

Apparently the hose was connected to the machine and it was dislodged from the fitting.

https://imgur.com/6TRsxZB

So it became something like this.

https://imgur.com/MxTS0wZ

We cannot rule out the possibility of sabotage/mishandling etc which is why we are hesitant to come up with a 8D report that incriminates us. It doesn't help that the plant manager dislikes us and escalates every single fault to global office. Our gut feeling is improper crimping but our supplier claims to have done all the testing at the required pressure and given us videos showing that as well. Having said that, no one was there to witness this test from our side so we can't disprove anything.

 

[OnurSeref]

Hello,
Did you ask for the faulty part for further analyze? I think demanding the piece might earn you week or so to understand and give a statement much clearly.
I suggest that after getting the faulty part, It would be wise to measure the point to match with spec demands, make an report according to that and use a Microscope equipment to see if there is any deformation occurence on the part itself, maybe within that way, You can enhance your hand on the root cause of mishandling/deformation by man-hand.

Also please keep in mind that, There is always a possibility of not finding the correct root cause, As long a the supplier is actful enough to make a progress to help on the subject, usually the OEM's keep it slide if it's not a enormous and future-effective nonconformity.

 

[Randy]

Looks like it was twisted out rather than slipped out because of the ragged edges.

The lines look like they are not stabilized allowing for flexing & movement when pressurized, simply clamping them together near the fittings and a second time a few inches back would add stability.

Verify that you're not having pressure spikes that exceed limits during pressure cycles.

 

[Mike S.]

As Onur suggested, I'd ask for the faulty parts. I'm betting there is someone who can do a failure analysis (the hose mfr perhaps) to determine more clearly what caused the failure.

And as Randy suggested, show the hose mfr or failure analysis house the "big picture" how the hoses are secured/stabilized.

Sometimes the answer is "unable to positively determine root cause" and you can only speculate on possible cause, then determine if you can control those better. For example, if a possible cause is flexed excessively while pressurized, can you limit the flexing?

 

[Golfman25]

Is this a single custom machine? Or do you make many machines? Sometimes stuff happens. Without multiple failures and the ability to recreate them it can be difficult to identify a root cause.

 

[ChrisM]

If you've not already done so, you need to involve the company that performed the crimp for you; they will probably have more expertise in determining likely cause. If practical and cost-effective, try to have one of your relevant engineers visit during their investigation, or at least to deliver the part in person to them to explain to their relevant staff the background to the failure and the need to establish a root cause if at all possible, to explain to your customer why there has been a failure

 

[dtann]

Hello,
Did you ask for the faulty part for further analyze? I think demanding the piece might earn you week or so to understand and give a statement much clearly.
I suggest that after getting the faulty part, It would be wise to measure the point to match with spec demands, make an report according to that and use a Microscope equipment to see if there is any deformation occurence on the part itself, maybe within that way, You can enhance your hand on the root cause of mishandling/deformation by man-hand.

Also please keep in mind that, There is always a possibility of not finding the correct root cause, As long a the supplier is actful enough to make a progress to help on the subject, usually the OEM's keep it slide if it's not a enormous and future-effective nonconformity.

Hi yes I have just asked for the status of the faulty part and what the team did with it. Will update once I find out more.

Looks like it was twisted out rather than slipped out because of the ragged edges.

The lines look like they are not stabilized allowing for flexing & movement when pressurized, simply clamping them together near the fittings and a second time a few inches back would add stability.

Verify that you're not having pressure spikes that exceed limits during pressure cycles.

Thanks, from what I understand the machine operation was normal and within acceptable range. Having said that, this is an area which is worth a further look.

As Onur suggested, I'd ask for the faulty parts. I'm betting there is someone who can do a failure analysis (the hose mfr perhaps) to determine more clearly what caused the failure.

And as Randy suggested, show the hose mfr or failure analysis house the "big picture" how the hoses are secured/stabilized.

Sometimes the answer is "unable to positively determine root cause" and you can only speculate on possible cause, then determine if you can control those better. For example, if a possible cause is flexed excessively while pressurized, can you limit the flexing?

Yeah I will try and get back the part into my hands and then get the hose manufacturer to provide their analysis. I used to think that a root cause must be identified in an 8D(new to all these) but a speculated one might be feasible provided our client accepts it. Will look into the flexing part as well.

Is this a single custom machine? Or do you make many machines? Sometimes stuff happens. Without multiple failures and the ability to recreate them it can be difficult to identify a root cause.

We have made many machines and I was informed that It's not a very regular occurrence. When it happens, usually the easiest and laziest way is to blame the crimping company. However we have made it a point to improve the situation by using a good brand of hose and requesting for proper testing and video proof of the test at a given pressure for a period of time. The thing is, this happened after 3-4 months of production so it is tough to blame the crimping contractor.

If you've not already done so, you need to involve the company that performed the crimp for you; they will probably have more expertise in determining likely cause. If practical and cost-effective, try to have one of your relevant engineers visit during their investigation, or at least to deliver the part in person to them to explain to their relevant staff the background to the failure and the need to establish a root cause if at all possible, to explain to your customer why there has been a failure

Thanks I will involve the company that performed the crimp again. In fact, I might go one better and involve the other company that we bought the new replacement hoses from and get a second opinion.
The biggest issue we face is that we cannot get a pre-crimped hose as site conditions always change and the length of hoses has to be cut at site. Then we will have to look for one of the crimping company available locally and ask for their service. In this situation, the quality of the companies vary depending on location. I wish to change this practice by getting pre-crimped hoses but everyone in the company tells me it will lead to a lot of wastage.

I really appreciate all your inputs and advice and I have gained a lot of knowledge and clarity on this. I will update more once I finish going through the suggestions all of you have put forth. Hopefully we can get a proper root cause for the 8D report.

Thank you very much.

 

[greatwhitebuffalo]

I built hydraulic hoses in a past life. It's not rocket science. Select proper die, set crimp pressure, slide fitting into hose (this may lead to premature failure if hose is not pushed far enough into fitting), press button, measure collar diameter to ensure proper crush. Might be worth trying to determine if the hose was not pushed far enough into the fitting, that picture makes it look like it worked its way out, cutting the collar off of the fitting might tell you something.