Schiit Happened: The Story of the World's Most Improbable Start-Up
Dec 15, 2016 at 1:55 PM Post #15,091 of 151,275
  WOW, all this talk about tubes. Perhaps someone should start a separate thread about tube rolling in Schiit equipment........................................wait a minute.
mad.gif

 
Is there a Freya Speculative Tube Rolling thread?  I'm holding out for glowy 6SN7 lava lamp tubes.
 
Dec 15, 2016 at 2:17 PM Post #15,093 of 151,275
Dec 15, 2016 at 3:54 PM Post #15,098 of 151,275
 
Is this fact or speculation? A "loose" part is easily found in visual inspection.

 
As someone who has spent my career in quality engineering in manufacturing environments, I can state that it is generally accepted that inspection is rarely more than 85% effective at catching flaws.
 
This is why we have the phrase "you can't inspect quality in" and you need processes and procedures to design it in, both in product design and in the manufacturing process, to use statistics to validate your key process variables in your initial runs and then monitor your subsequent runs using statistical process control to make sure they don't drift away from their nominals.
 
I'm not a pick and place or wave solder expert, so I don't know what type of process validation is typical for these types of operations, but yes, it is very common for vendor process faults to ship out their door and arrive at the customers door (the final assembly location) with issues.   You try to inspect for these things as they come in the door, but remember, inspection is rarely more than 85% effective at catching issues, so usually the faults are only first discovered when you go to assemble and test the damned things.
 
Sometimes - again - since inspection is only 85% effective, you don't catch the issue on the first several of a new lot of products.  Many may slip through and wind up in active inventory, ready to go out the door until you discover the issue on a product in production and then realize "holy ****, this impacts everything we made from the last lot" and you have to suddenly and quickly stop product that was about to go out the door, and sometimes even call stuff back that already shipped,
 
If you have an existing product and do your planning right, so you are building your next run well before you actually drop dead need it, you have the time to stop the line, and work with the vendor to fix the problem without running out of inventory and going into back order.
 
Unfortunately when something like this happens near a product launch, you don't have any existing inventory to run down, so you are immediately going to start impacting shipping product, especially since it's a new budget product, so right off the bat it's going to have higher demand than previous products because it is cheaper, and also happening as it is, right in the holiday rush.
 
Chances are keeping up with demand on a new product like this during the holiday rush would have been difficult to begin with, but a vendor production issue makes that even more challenging.
 
Being a small company more than likely makes this more difficult than if they were a huge corporation, but even so, huge corporations have delays and vendor issues like this all the time.  It's really just a matter of fact of manufacturing, unfortunately.
 
People who don't understand manufacturing love to complain when issues happen, but the truth is, manufacturing is very complex and fraught with complications and errors, and even in the best large world-class corporations with armies of optimizing quality engineers doing statistical validations on processes, things often will and do go wrong.
 
Its one of the big problems with working in manufacturing.  You don't get noticed and encouraged when everything is going right, regardless of how unlikely it is for everything to go right for long stretches of time.   You only get noticed when something goes wrong, and it's only a matter of time until something goes wrong, even at the best manufacturers.
 
All this being said, I trust the Schiit guys to do the right thing,and fix it.  It may result in a bit of a delay, but they will make it right.  They have been very good to deal with every time I have done so (even when the issue I was dealing with them about was entirely my fault, I've actually never had an issue with them that was their fault)
 
Dec 15, 2016 at 4:27 PM Post #15,099 of 151,275
   
... Being a small company more than likely makes this more difficult than if they were a huge corporation, but even so, huge corporations have delays and vendor issues like this all the time.  It's really just a matter of fact of manufacturing, unfortunately.
...
 

 
 Samsung 7 battery anyone?
 
I've got 4 pieces of Schiit and all have been flawless so far (as I type this on a wooden desk). I usually don't buy new releases out of the starting gate. I prefer to let the pre-order end users stress test new products and, hopefully, uncover that remaining 15%.
 
But subsequent production runs could always experience a bug in a part as well, it's all a part of the process.
 
Dec 15, 2016 at 5:03 PM Post #15,103 of 151,275
   
As someone who has spent my career in quality engineering in manufacturing environments, I can state that it is generally accepted that inspection is rarely more than 85% effective at catching flaws.
 
This is why we have the phrase "you can't inspect quality in" and you need processes and procedures to design it in, both in product design and in the manufacturing process, to use statistics to validate your key process variables in your initial runs and then monitor your subsequent runs using statistical process control to make sure they don't drift away from their nominals.
 
I'm not a pick and place or wave solder expert, so I don't know what type of process validation is typical for these types of operations, but yes, it is very common for vendor process faults to ship out their door and arrive at the customers door (the final assembly location) with issues.   You try to inspect for these things as they come in the door, but remember, inspection is rarely more than 85% effective at catching issues, so usually the faults are only first discovered when you go to assemble and test the damned things.
 
Sometimes - again - since inspection is only 85% effective, you don't catch the issue on the first several of a new lot of products.  Many may slip through and wind up in active inventory, ready to go out the door until you discover the issue on a product in production and then realize "holy ****, this impacts everything we made from the last lot" and you have to suddenly and quickly stop product that was about to go out the door, and sometimes even call stuff back that already shipped,
 
If you have an existing product and do your planning right, so you are building your next run well before you actually drop dead need it, you have the time to stop the line, and work with the vendor to fix the problem without running out of inventory and going into back order.
 
Unfortunately when something like this happens near a product launch, you don't have any existing inventory to run down, so you are immediately going to start impacting shipping product, especially since it's a new budget product, so right off the bat it's going to have higher demand than previous products because it is cheaper, and also happening as it is, right in the holiday rush.
 
Chances are keeping up with demand on a new product like this during the holiday rush would have been difficult to begin with, but a vendor production issue makes that even more challenging.
 
Being a small company more than likely makes this more difficult than if they were a huge corporation, but even so, huge corporations have delays and vendor issues like this all the time.  It's really just a matter of fact of manufacturing, unfortunately.
 
People who don't understand manufacturing love to complain when issues happen, but the truth is, manufacturing is very complex and fraught with complications and errors, and even in the best large world-class corporations with armies of optimizing quality engineers doing statistical validations on processes, things often will and do go wrong.
 
Its one of the big problems with working in manufacturing.  You don't get noticed and encouraged when everything is going right, regardless of how unlikely it is for everything to go right for long stretches of time.   You only get noticed when something goes wrong, and it's only a matter of time until something goes wrong, even at the best manufacturers.
 
All this being said, I trust the Schiit guys to do the right thing,and fix it.  It may result in a bit of a delay, but they will make it right.  They have been very good to deal with every time I have done so (even when the issue I was dealing with them about was entirely my fault, I've actually never had an issue with them that was their fault)


I agree completely. My initial post was directed at claims of "parts loose or not present" and whether this is factual or not. Product development and transition to production is a non-trivial activity best not experienced by the faint of heart. As a retired aerospace program manager, I am conversant in MIL-Q-9858 and related quality and manufacturing standards, and have managed these transitions from development to production. My post was in response to some reactions that appeared to be unsupported by fact. I have no reservations that Schiit will apply the required corrective action to assure that the problem, whether  design, manufacture, or test related will not recur. My complaint is toward broad pronouncements that are not supported by fact. I rest my case.
 
Dec 15, 2016 at 5:08 PM Post #15,105 of 151,275
 
I agree completely. My initial post was directed at claims of "parts loose or not present" and whether this is factual or not. Product development and transition to production is a non-trivial activity best not experienced by the faint of heart. As a retired aerospace program manager, I am conversant in MIL-Q-9858 and related quality and manufacturing standards, and have managed these transitions from development to production. My post was in response to some reactions that appeared to be unsupported by fact. I have no reservations that Schiit will apply the required corrective action to assure that the problem, whether  design, manufacture, or test related will not recur. My complaint is toward broad pronouncements that are not supported by fact. I rest my case.

 
Well, Jason has been lurking in this thread today, so maybe he can put your mind at ease.
 

Users who are viewing this thread

Back
Top