I think it would be a good idea to have the option to “Add the Drill Map” and “Add Drill Chart” in the “Place” menu in the same way that there is the option of “Add Board Characteristics” and “Add Board Stackup Table”.
The drill map and drill table is usually information included in the fabrication drawing of the bare PCB.
At the minute the only workaround is to export the Drill map to a DXF or SVG file and then import back into the pcb file.
I just wanted to post this here to see if other users think this will be useful before adding the feature request to gitlab.
Thanks,
Jesus
As you call your picture ‘fabrication drawing’ than I assume it is for PCB manufacturer.
We always send to PCB manufacturer each data only once. That is if we send gerber we don’t send a pdf with picture how this layer should look like. If we send them a drill file we don’t send them a picture how drilled PCB should look like. PCB manufacturer never asked us to show them how PCB will look like in our opinion.
The data that are not contained in files of course have to be delivered to them separately but repeating information from files probably makes no sense.
The drill files on their own are usually not sufficient to capture all the requirements. They only specify nominal hole size and position. That’s why it’s important to add a manufacturing drawing to capture tolerances of the hole sizes and position and the datum.
Didn’t know that. I was sure that PCB manufacturer specifies minimum annular ring (something like 5 mils) and sum of hole position tolerances and radius tolerances are always smaller than this. I didn’t sow specifying annular ring as depending on hole size so I assumed tolerance is the same for all holes. I don’t see the idea in specifying higher tolerances than they offer ‘by definition’.
So the only sense would be in specifying more accurate manufacturing, but are they able to do that?
I don’t know what hole information you mean with datum.
Every feature in a PCB requires a nominal value and a tolerance to account for the fabrication allowances.
The wear of drill bits, the inconsistentency of the plating of plated holes across the whole fabrication panel or the misregistration of the drill procress. All this causes that the holes in the final unit are not “perfect”, meaning they are not in their true position and their size is not nominal.
Tolerances determine if a unit is fit for purpose or not.
Of course, many don’t define these tolerances and trust that the manufacturer will produce a “good unit”. But that’s not always the best engineering approach. It depends on the type of product and industry.
I simply think in opposite way. Manufacturer specifies tolerances of his process and if they are not good for me I have to search another manufacturer then send to this one my tolerances I want from him.
I think the key is in my first sentence:
You didn’t questioned it so I assumed it is true.
But if it is not true and what you are thinking about are the specifications for your own staff to be able to select right PCB manufacturer then we speak about two different subjects.
This does not happen. To prevent issues with broken or worn out drills, they are replaced after a certain amount of drilled holes. Modern CNC drilling machines have magazines with several hundred drills and the machines replace them fully automatically.
The biggest cause for deviations in drill sizes is probably drill substitution. PCB manufacturers apparently substitute drills for others when their diameters are “close enough”. Metric sized drills may be substituted for inches / banana units, or vise versa. For soldering it usually does not matter much, but for press fit connectors, hole diameter tolerance may be tighter. As far as I know KiCad does not have specific tolerances for holes, but just generic diameters. The only solution I know is to use a “special” diameter for such holes, and communicate with the PCB manufacturer when some holes have a tighter tolerance.
That I believe.
In the page you showed, it is about drill breakage, deflection and misregistration. Surface finish of the hole will also be worse when drills wear out. But I do not see any mention of the hole diameter changing because of this. No proper PCB manufacturer will ever let it go that far.
The video below shows how PCB drilling is done these days. 3 or more parallel heads, and each head drills 2 to 4 holes each second. There simply is no tolerance for worn out drills. You can also see the drill magazines for each of the drilling heads. Each head has 400 drills in their magazine.
Also note: The PCB’s themselves are below an aluminum sheet. This is to cool the drills, and to reduce burrs on the copper. This is explained in other video’s.
Well, the thing is you can specify tolerances and give detailed engineering drawings to fab houses. But this will generally make your process more expensive. Because now the house need to bid and do quality control etc. and there is a chance that you the designer have over specified something, increasing cost yet again. Not all factories accept jobs like this.
This process is more for mass production where minutae detail can be negotiated. However, there is also the other cheaper option for low quantity production. Just accept the vendors settings and be done with it. This is faster, cheaper and involves less micromanaging the producer too. For most products the factories are good to go, as is.
Yes drill wear affects the hole size. Thing is though the factory has already accounted for this. in their specs.
No it does not. PCB manufactures simply can not tolerate the amount of wear that would result in hole size differences. I thought I made that clear already.
For the other things, yes I agree, if you go with the cheap services, then you are just happy with what they deliver, but those services don’t bother with drill tables either. The need for drill tables (if any) is beyond those cheap services, and into more controlled production.
Yes most shops have calculated the cost of doing so and don’t drill excessively on a drill so it wears out. It would be bad for the machine. That does not change that the statement that drill wear affects tolerance isn’t true. Its just a half truth that may not reflect realities on how things are currently made.
So the book is not wrong per see. When I make my own boards in proto settings i do monitor the drill wear manually, because i dont have big volume statistics and dont have a huge inventory of drills around. So if someone specifies a tolerance my current setup does currently do, but could do, then we have to wait for new tools to arrive.
