smd pad count is usually wanted (for quote purposes) for test reasons
(SMD pad count can be reckoned easily from BOARD INFORMATION report)

drill hole count these days is usually not a factor unless holes per 
square inch exceed some very high densisty which is not typically 
encountered (possibly in very high volume it could be different)

the reason is that the production throughput bottleneck is in the 
plating tanks not the drilling so nowdays i have found the 
quotes are almost strictly based on square inches

BTW, a friend of mine recently visited a shop in China 
it was 100K+ square feet with 2 rows of drill machines extending 
"as far as the eye could see"
the machines were running 24/7 for the last 2 years and were just 
starting to approach payoff of the drills

re the SMD/TH part count ...
i can't imagine that it would take more than a few minutes cruising 
through even a rough BOM to come up with numbers adequate for the

Dennis Saputelli

Abd ul-Rahman Lomax wrote:
> At 05:35 AM 11/19/2003, Leo Potjewijd wrote:
> >At 19/11/2003 01:09, Abd ul-Rahman Lomax wrote (among other things):
> >[a list of assumptions]
> >
> >I must say that these statements are not entirely correct: many SMD
> >connectors need holes in the PCB for positioning or letting the mating
> >part through.
> >I regularly use a <>FFC connector
> >that has pegs for positioning and strain relief. You will need two
> >non-plated holes in the PCB to accommodate that thing.....
> First, what I wrote were presented as simplifying assumptions, not as
> necessary truths. Secondly, you need the holes, but you don't necessarily
> need a pad. Protel will allow you to use a via. However, there is no
> provision for an "unplated" via, so one might need to manually ensure that
> a via with hole size exceeding the diameter is not plated.
> This isn't very satisfactory, however, because it would involve modifying
> footprints just so that report generation would be a little easier. Better
> to generate reports for all three logical types (pure SMT, pure
> through-hole, mixed).
> Note, however, that there is an additional complication, which is that
> there can be a "component" which is neither SMT nor through-hole, it is a
> non-assembled device for accomplishing some purpose other than providing a
> place for assembly. It might be a test point, it might be a component with
> no holes whatever, representing something not fastened to the board but to,
> for example, an adjacent board, and one wants to place a dummy footprint
> and lock it in place to keep the space clear.
> >The most likely reason the fab house asked for the pincounts is
> >calculation of the bare PCB price (at least, my boardhouse does use this
> >info for that purpose): holes need to be drilled, many different sizes
> >and/or really small ones result in greater setup/machine/handling time.
> >In electrical testing, SMD pads most likely need a different test setup:
> >thru-hole pads can be accessed on both sides of te PCB while SMD pads can
> >not, SMD on both sides will thus result in higher cost.
> The question was directly posed by Mr. Malik and it was explicitly about
> "an add-on utility or a process script that can separately count the number
> of SMD and Through-hole components."
> It's not likely Mr. Malik was asking in response to a fabricator query,
> because fabricators have little need to know the number of surface pads;
> surface pads are merely copper features and openings in the solder mask;
> however, if the question was about pads, he asked the wrong question. It's
> easy to get a pad report, use Edit/Export to Spread.
> I do notice that Advanced Circuits asks about SMD pads in their instant
> quote form. Oddly enough, they *don't* ask about holes and the variety of
> hole sizes, which would have a larger effect on price, I'd think. However,
> I do recall a conversation in which a fabricator stated that holes have
> become so cheap to drill that they are not a substantial contributor to
> cost....
> Mr. Malik, on the other hand, made another inquiry seeking recommendations
> for assembly houses, and I'd expect an assembler to need the information
> for a quotation; it's common to see figures of so many cents per SMT
> component and so many per through-hole component....
> For this, the information would not need to be absolutely accurate, and the
> existence of a few ambiguous components would have little or no effect on
> the cost.

Dennis Saputelli

  ========= send only plain text please! - no HTML ==========
Integrated Controls, Inc.   
2851 21st Street                    tel: 415-647-0480
San Francisco, CA 94110             fax: 415-647-3003

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
* To leave this list visit:
* Contact the list manager:
* Forum Guidelines Rules:
* Browse or Search previous postings:
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to