I can imagine getting away with no test on a simple, 2-layer board, but I'd NEVER, EVER do that on a multilayer.
First thing, before stuffing the board, put it under a microscope or strong magnifier, and look for the probe marks on the pads. The flying probe tester leaves a TINY dimple in the center of each pad when it tests the board. If you don't see probe marks, the board was NOT TESTED!As I said above though, you have to pay extra for it (as an option) if you want 100% test. Otherwise they will just do random test of a few panels in the batch.
Check your power delivery.
- Fails on 0.5oz, works on 1.5oz
- Only fails when heavily loaded
Points, at least for me, to insufficiently wide power traces or planes/traces necked down by rows of vias or something. Let me also restate that it points to a marginal design. If tomorrow another supplier has a -10% tolerance on the copper thickness or on the copper width of the power trace or... you may well find yourself in trouble again. You need to understand why it works on a 1.5oz board and not on 0.5oz.
If you don't figure it out, it will remain a project that works by luck rather than by design.
The system can only produce 1 amp of power for the +5v ground plane.
Just ran the calculations for 0.50z internal pour to the +5v plane, and the connections have enough capacity for 1.25 amps with a voltage drop of .019v. Not sure that is the issue there either.
I'm wondering if there was enough copper on some vital via? I am running vias at the minimum for JLC.
"1/2 oz." copper has a range - it's not always 18µm and some fabs do skimp. For an inner layer of 18µm nominal copper thickness, IPC-A-600J Class 1&2 accepts a minimum of 11.4 µm after manufacture, IPC-4562 is 15.4µm min.
What standards are the boards built to, do we know? IPC membership does not mean the standards are followed or independently audited.
I've sent PC boards to Japan to get micro-section analysis and verify a laminate was genuine, and the finished copper thickness, wall plating was in spec. as I was having problems with the controlled impedances and Dk being way off.
in those ways, JLCPCB can make the quality complaint rate to be 0.25%(In the 20,000 orders, there is less than 50 orders with quality complaints).
First thing, before stuffing the board, put it under a microscope or strong magnifier, and look for the probe marks on the pads. The flying probe tester leaves a TINY dimple in the center of each pad when it tests the board. If you don't see probe marks, the board was NOT TESTED!As I said above though, you have to pay extra for it (as an option) if you want 100% test. Otherwise they will just do random test of a few panels in the batch.
Really? I was not aware of that. When ordering from JLCPCB, you can only opt out on testing, but at least for two and four layer boards it doesn't make any difference in price or manufacturing time. Where's the option to have 100% testing?
For the order with a quantity of no more than 50 pieces, we will provide full test for the whole order for free.
For the order with a quantity equal to or bigger than 50 pieces, it will cost you an extra fee for full test, the extra test cost can be calculated on our website.
Check your power delivery.
- Fails on 0.5oz, works on 1.5oz
- Only fails when heavily loaded
Points, at least for me, to insufficiently wide power traces or planes/traces necked down by rows of vias or something. Let me also restate that it points to a marginal design. If tomorrow another supplier has a -10% tolerance on the copper thickness or on the copper width of the power trace or... you may well find yourself in trouble again. You need to understand why it works on a 1.5oz board and not on 0.5oz.
If you don't figure it out, it will remain a project that works by luck rather than by design.
Let's have a look at some of the numbers.The system can only produce 1 amp of power for the +5v ground plane.
Just ran the calculations for 0.50z internal pour to the +5v plane, and the connections have enough capacity for 1.25 amps with a voltage drop of .019v. Not sure that is the issue there either.
I'm wondering if there was enough copper on some vital via? I am running vias at the minimum for JLC.
You have a 25% headroom capacity in your design"1/2 oz." copper has a range - it's not always 18µm and some fabs do skimp. For an inner layer of 18µm nominal copper thickness, IPC-A-600J Class 1&2 accepts a minimum of 11.4 µm after manufacture, IPC-4562 is 15.4µm min.
What standards are the boards built to, do we know? IPC membership does not mean the standards are followed or independently audited.
I've sent PC boards to Japan to get micro-section analysis and verify a laminate was genuine, and the finished copper thickness, wall plating was in spec. as I was having problems with the controlled impedances and Dk being way off.
the '18um' inner layer could be less than 2/3 the thickness and still be within IPC-A-600J Class 1&2. That is more than your headroom allows.
If your design works with companies that have the option of thicker inner layers then it seems very likely that your design is out of spec for 0.5 oz inner layers.
IMO If you beef up your inner layer tracks and maybe didn't use the very smallest vias allowed in the design rules, you would get reliably working boards from JLCPCB.
Personally, I try to never approach the limits of production, I like to have some margin for error in my designs if for no other reason than it gives the product extra reliable lifespan when it starts to show signs of age.
If it is running close to the limits when new then any degradation of function means it will fail early.
Yup, looking like a very good possibility of why JLC's boards sometimes work, sometimes don't with the "allowable" swing of inner layer thickness.
For this design, I will just keep going with another manufacturer with thicker inner layers instead of go over with a board re-design. I generally only use Chinese manufacturers until I'm satisfied with a board design, than switch over to South Korean manufacture for larger lots.
Wow, they still haven't fixed the octagonal pad problem. I changed to round just to use JLCPCB so it doesn't bother me but they really do need to fix that problem.
in those ways, JLCPCB can make the quality complaint rate to be 0.25%(In the 20,000 orders, there is less than 50 orders with quality complaints).That's quite impressive rate
What is actually 'quality complaint'?
Let's say somebody 'played' with stencil apertures or solder masks are way off as requred, so orders back to the house to be redone again - does it qualified for 'quality complaints'?
Maybe it was something in my specs that didn't get along with JLC's tolerances, but I was never given any warnings of the sort when I made the orders.
Maybe it was something in my specs that didn't get along with JLC's tolerances, but I was never given any warnings of the sort when I made the orders.
I think it's a valid point. Each PC board vendor should check the Gerber file parameters against their own allowable tolerances. If it's out of spec, there should be a warning along with an opportunity to abort -- or proceed with the order.
Paul
Maybe it was something in my specs that didn't get along with JLC's tolerances, but I was never given any warnings of the sort when I made the orders.
I think it's a valid point. Each PC board vendor should check the Gerber file parameters against their own allowable tolerances. If it's out of spec, there should be a warning along with an opportunity to abort -- or proceed with the order.
Paul
Maybe it was something in my specs that didn't get along with JLC's tolerances, but I was never given any warnings of the sort when I made the orders.
I think it's a valid point. Each PC board vendor should check the Gerber file parameters against their own allowable tolerances. If it's out of spec, there should be a warning along with an opportunity to abort -- or proceed with the order.
Paul