PCB costs
 

Hello,

I am working on a PCB-based antenna, with copper on both sides.
Believe it or not, the hardest part is figuring out how much the PCB
material would cost. Does anyone have any PCB costing numbers for
FR4, FR408, Nelco 40013, Rogers 4003, any other material? I
understand that thickness plays a major part in the cost so if
possible please include the board thickness with the cost number (I am
currently thinking of using 20mils thick PCB). It would probably be
easiest if the cost was in units of US$ per sq. inch or sq. cm. Many
thanks.

From my days in the Silicon Valley in California, the biggest thing I
remember about PCB costs weren't related directly to the size or
thickness, they were quantity....

I've had quite a number of double sided PCBs fabricated, along with
numerous multilayer ones, and the overriding issue with cost was the
"non-recurring costs " of generation of the masks, tooling and setup
needed to convert the supplied circuit trace films to finished boards.
Once the fixed costs are covered, the individual costs of each board
weren't usually all that high, if you had enough volume. As a simple
example, a small jumper board of about 2.5" square, double sided 60 mil
thick and only about 75 holes, with soldermask cost about a buck each at a
quantity of 200. At a quantity of 500, the price was something like sixty
cents each... Of course, this was the sort of thing that a hundred pieces
would be a near lifetime supply, and at a quantity 100, the price was
about 1.75 each.... Such is the problem of amortizing the fixed and setup
costs over a relatively small quantity.

So, any price that you get quoted on a per square inch basis will have to
carry some serious disclaimers or additional explanation of the fixed and
setup costs incurred in making the first board, along with price breaks
for quantity....

The best thing to do is to contact some places that do engineering
prototype quantity work, and see what they say. You can usually get a
decent price quote by providing the outside dimensions, basic construction
(thickness, number of layers and copper weight (thickness) along with
things like number of holes, how many different sizes of holes involved,
whether they are plated through, and whether you want a bare board, solder
plating on the exposed copper and such things as solder mask and legend
(the exact term for legend, the lettering on one or more sides of the
board doesn't come to me at this moment,)) and, of course, the quantity of
the run you want made. Of course, tolerance is also important.... If you
need ultra-precise registration between layers, that will tend to cost
more than an alternate board with a looser tolerance...

It seems like a lot of information, but once you get used to it, it's not
so bad.

Of course, if the relative amount of circuitry on opposite sides of the
board isn't somewhat equal, (called copper-balancing) the board may have a
tendency to warp during manufacture, or even warp or curl with changes in
temperature, particularly if one side uses a thick (heavy) copper and the
other lighter weight copper. This is especially important if you intend
to wave solder the PCB during assembly or other manufacturing
processes....

Of course, yet another variable in the cost issue is where the PCB
manufacturer is and what sort of volume they deal with. If you're looking
for quick turn shops, there are (or at least used to be) a lot of them
around the Silicon Valley area. However, if you're expecting to find
someone in Podunk WA, KS, OH, FL or similar locations away from high-tech
development, you may be looking at higher prices unless they are "hungry".

Good Luck
-_Rick AH7H

basketball_jones wrote:

Hello,

I am working on a PCB-based antenna, with copper on both sides.
Believe it or not, the hardest part is figuring out how much the PCB
material would cost. Does anyone have any PCB costing numbers for
FR4, FR408, Nelco 40013, Rogers 4003, any other material? I
understand that thickness plays a major part in the cost so if
possible please include the board thickness with the cost number (I am
currently thinking of using 20mils thick PCB). It would probably be
easiest if the cost was in units of US$ per sq. inch or sq. cm. Many
thanks.

Two more things:
Assume very high, production volume so setup, tooling etc. are not a factor.
There is only one hole.

Many thanks.

(basketball_jones) wrote in message . com...
Hello,

I am working on a PCB-based antenna, with copper on both sides.
Believe it or not, the hardest part is figuring out how much the PCB
material would cost. Does anyone have any PCB costing numbers for
FR4, FR408, Nelco 40013, Rogers 4003, any other material? I
understand that thickness plays a major part in the cost so if
possible please include the board thickness with the cost number (I am
currently thinking of using 20mils thick PCB). It would probably be
easiest if the cost was in units of US$ per sq. inch or sq. cm. Many
thanks.

Hello,

I am working on a PCB-based antenna, with copper on both sides.
Believe it or not, the hardest part is figuring out how much the PCB
material would cost. Does anyone have any PCB costing numbers for
FR4, FR408, Nelco 40013, Rogers 4003, any other material? I
understand that thickness plays a major part in the cost so if
possible please include the board thickness with the cost number (I am
currently thinking of using 20mils thick PCB). It would probably be
easiest if the cost was in units of US$ per sq. inch or sq. cm. Many
thanks.


4003 is very good, but there are many less expensive alternatives, depending
on the app and freq required. I am puzzled why you haven't priced this with
manufacturer's reps or even board fab houses. They are not shrinking violets,
and happy to give you info

Also, it would be helpful to know who you are. Basketball and antennas don't
often get placed together.

73,
Chip N1IR

Don't know about that Chip!
Every player on my sons team had an earpiece and tranceiver so the
coach could chew them out while playing.

TTUL
Gary



WIreless basketball. That's a new direction!

;-

73,
Chip N1IR

My name is Kevin, Mr. Fractenna. The reason I don't call the reps is
because I am not nearly as interested in the numbers they tell me as I
am in the numbers that folks have already negotiated. In this way
when I eventually call them I have an idea of how much I am getting
ripped off. Last time I called a rep he wanted me to pay 60 cents for
a nickel part. So how much I am overpaying is the answer I am after.

You indicate that you are aware of less expensive alternatives to
4003. Would you kindly let me know what those are. Are we talking
materials other than FR408? BTW the frequency is in the 3-4GHz band,
so loss tangent and other stuff only somewhat matters. There is a
price/efficiency tradeoff in play. I would like to 'see the whole
field' before I decide which way to run.

Thanks.
Kevin Smith

....and that's exactly why you need the reps in.

We do thousands of square feet of PCB a year, and we never had to use a
newsgroup to figure out a good price, or ask anyone in the way you suggest.
Call the reps. If you want to play games, then you can invite them over on the
same day and ask for their best price.

The substrate houses are eager for business and you owe it to good
vendor-customer relationships to play it straight. That's my opinion. But hey,
what do I know:-)?

BTW, putting FR4 and 4003 in the same sentence indicates you may really need
the vendors to fill you in: No one uses FR4 efficiently at those freqs--lousy
loss tangent. OTH, I, personally, do not choose 4003 much due to cost, save
for very specific high margin needs. But others, of course, may choose these
great product according to their needs.

73,
Chip N1IR

basketball_jones wrote:
BTW You have NOT answered one question I asked so far. It would be
helpful if you either answered the questions or stopped wasting my
time.

It's a little ironic that you're trying to save a penny or two on your
design and think that bitching about free advice is somehow going to help
you achieve that goal...

Look, call up a dozen or so vendors and get some price quotes -- it'll give
you some idea of what a 'reasonable' price should be. Figure out what you
can charge for your product... if there's enough money left for you to make
profit, just go for it and leave the rest of your learning/profit
maximization until you actually have an income stream and some track record
of the difficulties in building your products.

Anyone worrying about $.01/sq. in. either has such a huge volume that they
should be able to afford an expect to get the best prices and redesign the
product (this oftens makes a huge difference!) specifically to be easy to
manufacture or else is probably never going to get their product out the
door in the first place. I've seen many people who seemed to have viable
product ideas never get anywhere because they spent far too much time
worrying about how to squeeze the last penny out of manufacturing costs and
ran out of enthusiasm to simply design and document the product properly.

---Joel Kolstad

It's a little ironic that you're trying to save a penny or two on your
design and think that bitching about free advice is somehow going to help
you achieve that goal...

Look, call up a dozen or so vendors and get some price quotes -- it'll give
you some idea of what a 'reasonable' price should be. Figure out what you
can charge for your product... if there's enough money left for you to make
profit, just go for it and leave the rest of your learning/profit
maximization until you actually have an income stream and some track record
of the difficulties in building your products.

Anyone worrying about $.01/sq. in. either has such a huge volume that they
should be able to afford an expect to get the best prices and redesign the
product (this oftens makes a huge difference!) specifically to be easy to
manufacture or else is probably never going to get their product out the
door in the first place. I've seen many people who seemed to have viable
product ideas never get anywhere because they spent far too much time
worrying about how to squeeze the last penny out of manufacturing costs and
ran out of enthusiasm to simply design and document the product properly.

---Joel Kolstad

Finally a thoughtful response. BTW I have called up the vendors,
that's just common sense, but the prices they quote are rediculous
even if you promise them volume. They don't believe you about the
volume (rightfully so, that's why I am asking someone who has the
volume) and at those prices there is no business. I have done all
that analysis.

Yes, I am worried about $.01/sq in. That's a buck on a 100 sq in
board. That's A LOT of money to me. BTW, I have done as much antenna
footprint minimization as possible (that's a different topic). Now to
your point about never getting anywhere due to too much time squeezing
out pennies. I have experienced a lot more of the opposite, where
people build stuff without paying enough attention to the costs and
then figure out that its too expensive and try to cost reduce. At
that point their previous decisions limit their options and the stuff
can never be built cheap enough. So I am trying to strike a balance.
And if I can get a quick answer on what those costs are, that would
leave me more time to optimize the design. BTW I am NOT asking who
you are, I am just asking how much you pay. That's the beauty of
these boards. I don't see the foul.

ps Free advice is the goal of these boards, I am not bitching about
the advice, I am bitching about lack thereof.

basketball_jones wrote:

It's a little ironic that you're trying to save a penny or two on your
design and think that bitching about free advice is somehow going to help
you achieve that goal...

Look, call up a dozen or so vendors and get some price quotes -- it'll give
you some idea of what a 'reasonable' price should be. Figure out what you
can charge for your product... if there's enough money left for you to make
profit, just go for it and leave the rest of your learning/profit
maximization until you actually have an income stream and some track record
of the difficulties in building your products.

Anyone worrying about $.01/sq. in. either has such a huge volume that they
should be able to afford an expect to get the best prices and redesign the
product (this oftens makes a huge difference!) specifically to be easy to
manufacture or else is probably never going to get their product out the
door in the first place. I've seen many people who seemed to have viable
product ideas never get anywhere because they spent far too much time
worrying about how to squeeze the last penny out of manufacturing costs and
ran out of enthusiasm to simply design and document the product properly.

---Joel Kolstad

Finally a thoughtful response. BTW I have called up the vendors,
that's just common sense, but the prices they quote are rediculous
even if you promise them volume. They don't believe you about the
volume (rightfully so, that's why I am asking someone who has the
volume) and at those prices there is no business. I have done all
that analysis.

Yes, I am worried about $.01/sq in. That's a buck on a 100 sq in
board. That's A LOT of money to me. BTW, I have done as much antenna
footprint minimization as possible (that's a different topic). Now to
your point about never getting anywhere due to too much time squeezing
out pennies. I have experienced a lot more of the opposite, where
people build stuff without paying enough attention to the costs and
then figure out that its too expensive and try to cost reduce. At
that point their previous decisions limit their options and the stuff
can never be built cheap enough. So I am trying to strike a balance.
And if I can get a quick answer on what those costs are, that would
leave me more time to optimize the design. BTW I am NOT asking who
you are, I am just asking how much you pay. That's the beauty of
these boards. I don't see the foul.

ps Free advice is the goal of these boards, I am not bitching about
the advice, I am bitching about lack thereof.

I gotta think you must be taking the wrong approach with these PC houses
if you can't get an answer from them. Besides, the answers you get from
us don't amount to much if you can't get anybody to make your boards for
the prices "quoted" here.
Bottom line is you won't get the low digit price unless you're going to
spend some big money. If you let the PC houses know how big a project
you have, and it's big, then they'll be plenty anxious to quote you a
good price.

73, Jim AC6XG

Jim Kelley wrote in message ...
basketball_jones wrote:

It's a little ironic that you're trying to save a penny or two on your
design and think that bitching about free advice is somehow going to help
you achieve that goal...

Look, call up a dozen or so vendors and get some price quotes -- it'll give
you some idea of what a 'reasonable' price should be. Figure out what you
can charge for your product... if there's enough money left for you to make
profit, just go for it and leave the rest of your learning/profit
maximization until you actually have an income stream and some track record
of the difficulties in building your products.

Anyone worrying about $.01/sq. in. either has such a huge volume that they
should be able to afford an expect to get the best prices and redesign the
product (this oftens makes a huge difference!) specifically to be easy to
manufacture or else is probably never going to get their product out the
door in the first place. I've seen many people who seemed to have viable
product ideas never get anywhere because they spent far too much time
worrying about how to squeeze the last penny out of manufacturing costs and
ran out of enthusiasm to simply design and document the product properly.

---Joel Kolstad

Finally a thoughtful response. BTW I have called up the vendors,
that's just common sense, but the prices they quote are rediculous
even if you promise them volume. They don't believe you about the
volume (rightfully so, that's why I am asking someone who has the
volume) and at those prices there is no business. I have done all
that analysis.

Yes, I am worried about $.01/sq in. That's a buck on a 100 sq in
board. That's A LOT of money to me. BTW, I have done as much antenna
footprint minimization as possible (that's a different topic). Now to
your point about never getting anywhere due to too much time squeezing
out pennies. I have experienced a lot more of the opposite, where
people build stuff without paying enough attention to the costs and
then figure out that its too expensive and try to cost reduce. At
that point their previous decisions limit their options and the stuff
can never be built cheap enough. So I am trying to strike a balance.
And if I can get a quick answer on what those costs are, that would
leave me more time to optimize the design. BTW I am NOT asking who
you are, I am just asking how much you pay. That's the beauty of
these boards. I don't see the foul.

ps Free advice is the goal of these boards, I am not bitching about
the advice, I am bitching about lack thereof.

I gotta think you must be taking the wrong approach with these PC houses
if you can't get an answer from them. Besides, the answers you get from
us don't amount to much if you can't get anybody to make your boards for
the prices "quoted" here.
Bottom line is you won't get the low digit price unless you're going to
spend some big money. If you let the PC houses know how big a project
you have, and it's big, then they'll be plenty anxious to quote you a
good price.

73, Jim AC6XG

Folks we are missing the point here. I can certainly get a quote from
them. You can safely assume I already have it. Ok, that tells me the
MOST I am going to pay. Now I want the LEAST I can possibly pay. The
only way for me to get that is to go through board manufacturers' cost
structure and assume their take. I am not TI or HP, I don't have the
resources to do that, and even if I did I wouldn't get it right. Now
why would I want to know the LEAST even though I can't get that price
at this point. Well, the situation is likely that if I have to pay
the MOST, I lose money. Do I still lose money when I get to pay the
LEAST? (As far as getting the houses to quote a big project) I can't
get the big project funded, much less quoted until I understand that.

Look people. Determining the 'right' price is one of the 'killer aps'
of the internet. Use it.

basketball_jones wrote:
Do I still lose money when I get to pay the
LEAST?

I think what you need is a book on the business of manufacturing. The
price is basically going to fit a line y = mx + b. Think of b as a
tooling charge. The unit cost is m. x is the number of units, and y is
what you pay per lot.

73, Jim AC6XG

BTW You have NOT answered one question I asked so far. It would be
helpful if you either answered the questions or stopped wasting my
time. You, my friend are the one that's playing games.


Please Kevin,

Don't be so righteous. I told you what to do on a private e-mail, way before
you sent this. If you couldn't get pricing from the reps--REAL pricing--then I
would provide it.

I, too, have priced out boards for microwave apps, and I certainly didn't need
to go around the reps.

We need to keep a viable substrate industry continuing in this country, and
anything that could be construed as undermining them is not in my vocabulary.
They've bled way too much already.

73,
Chip N1IR

Jim Kelley wrote in message ...
basketball_jones wrote:
Do I still lose money when I get to pay the
LEAST?

I think what you need is a book on the business of manufacturing. The
price is basically going to fit a line y = mx + b. Think of b as a
tooling charge. The unit cost is m. x is the number of units, and y is
what you pay per lot.

73, Jim AC6XG

Excellent, thanks, now, what is the rough equation for 'm' vs volume for PCBs?

basketball_jones wrote:
Excellent, thanks, now, what is the rough equation for 'm' vs volume for PCBs?

To first order, it's a constant which is proportional to the size of
your PC board.

73, AC6XG

basketball_jones wrote:

Jim Kelley wrote in message ...

basketball_jones wrote:

Do I still lose money when I get to pay the
LEAST?

I think what you need is a book on the business of manufacturing. The
price is basically going to fit a line y = mx + b. Think of b as a
tooling charge. The unit cost is m. x is the number of units, and y is
what you pay per lot.

73, Jim AC6XG


Excellent, thanks, now, what is the rough equation for 'm' vs volume for PCBs?

I hope you are kidding !!!!

Contact the vendors and get a price schedule showing price breaks as a
function of layers, area and quantity of boards.

Do you want finished boards, or will you cut boards from a master sheet,
etc.?

What level of quality control is required? If you are selling into the
NASA or Military market there are VERY SIGNIFICANT documentation,
traceability requirements, and QC requirements and certifications
required. Even to controlling the exact chemistry of the board material.

Depending on your customer your unit material cost may be buried in
recurring paper costs. I'm an expert in this area {Program Chief
Engineer, USAF WS-118, RS/RV ... MX Missile]

basketball_jones wrote:
Folks we are missing the point here. I can certainly get a quote from
them. You can safely assume I already have it. Ok, that tells me the
MOST I am going to pay. Now I want the LEAST I can possibly pay.

In many companies that are large enough to have the kind of volumes you seem
to be after, I suspect that purchasing is done by, uh, the purchasing
department and not the design engineer and said purchasing folks are
probably rather unlikely to be reading this newsgroup. :-) Although I'd
grant you this is about the best place to ask that I could think of...

Once again, I CAN'T get that
price until I am ready to sign and I can't be ready to sign until I
get that price.

K.

Well, I did my duty.

Nope; I won't resend the e-mail. It was labelled ' PCB' costs so it
shouldn't have been counted as spam. I tried to help. That's enough.

If you start getting spam with PCB titles, you really do have some problems I
can't assist with.

Best of luck.

74,
Chip N1IR

basketball_jones wrote:

SNIP

Once again, I CAN'T get that
price until I am ready to sign and I can't be ready to sign until I
get that price.

K.

Isn't that the definition of Catch-22??

Seriously, you need a sit down negotiation session with your potential
suppliers, AND HAVE YOUR PURCHASING AGENT/LAWYER PRESENT. Part of your
strategy is to let your potential suppliers know right up front that it
is a competitive procurement based on either lowest price or best value
[whatever your criteria may be][The stated criteria is common to and
binding on all potential suppliers [you can't change the rules without
inviting a rebid from all suppliers]]. In my experience, once we moved
from the technical issues I yielded all related business risk issues to
the Purchasing Agent or Lawyers. Engineering should NEVER sign a
purchasing contract. The US DOD and my company used legal professionals
to definitize the purchasing agreements/contracts.

Review your potential supplier's quality programs, manufacturing
capacity, inventory levels, their supplier vulnerabilities, [you don't
want your long term sales jeopardized by a sub-supplier's strike etc.,],
cash flow requirements [do you make a partial payment up front or is it
post delivery billing], backlog [you don't want your schedules held
hostage by a supplier's backlog issues]; these are business issues that
exceed 'lowest price' criteria.

Negotiation is more a part of business than creative design. In many MBA
programs there are courses in negotiation. Remember, your supplier NEEDs
business; and, wants to MAXIMIZE their profit also.

Deacon Dave, W1MCE

Best of luck to you. Both you and I know that you have NOT sent that
email in the first place.



That is a false statement.

Kevin, acting this way is no way to win friends in this industry.

Cordially,

Chip N1IR

ps Mr. Fractenna, your motives are transparent ...


Gee, I sure hope so!

Transparency has been a driving factor for me for the last several years. I
know that may be a bad pun to some:-)

Anyway, good luck Kevin.

73,
Chip N1IR

On 6 May 2004 10:22:49 -0700,
(basketball_jones) wrote:
the games you are playing are obvious to everybody.
From your initial post it was clear that you were going
to be less than helpful.

Some things never change....