hello,
I am working on a tool to build virtual PCBs. Other people or organisations
are probably also working on this. I have tried to use the most simple
models for the dimensional and the electrical analysis. The dimensional
analysis involves weight and thickness, the electrical part concerns
impedances, xtalk, ...
My background is in electronics and EMC engineering. It is only lately that
I have began to explore PCB manufacturing. So forgive me my expressions
for I am not used to the slang used in PCB manufacturing. Also I apologize
for this lengthy mail. I appreciate you going through this text.
I have some first results on the dimensional part. We actually have build 10
6-layer test boards and we have a simulation run using some detailed data
on preg, core laminate and foil. Of these we have used very specific data
of the reinforcement, of the resin and of the copper. Also included is
the copper area/layer, mask data (or coating), plating details, plated
perimeter of vias and the board dimensions.
The preg is an Isola with 1080 reinf., the core is an Isola with 3x7628
reinf. The average finished resin content of the preg is 65+/-3%.
The buildup is described below at the end of this email bewteen
** example ** and *** example done ** and although you do not know the
exact input syntax, you should be able to figure out the data and the
buildup intuitively (note: a comment line starts with ; or *, all
other lines do something, e.g. introduce the resin, do an analysis, ...)
At the bottom of the example you will the results of the weight and thickness.
They are repeated here (the board is renamed to XYZ6):
***************** build summary for XYZ6 *****************
mean max min sdev
preg weight: 5.09E-02 | 5.33E-02 | 4.88E-02 | 6.40E-04
core weight: 1.22E-01 | 1.26E-01 | 1.18E-01 | 1.16E-03
foil weight: 4.33E-02 | 4.76E-02 | 3.90E-02 | 1.28E-03
plate weight: 9.83E-03 | 1.25E-02 | 7.44E-03 | 7.17E-04
-----------------------------------------
total weight: 2.27E-01 | 2.33E-01 | 2.19E-01 | 2.00E-03
preg thick: 5.12E-04 | 5.41E-04 | 4.88E-04 | 7.64E-06
core thick: 1.02E-03 | 1.07E-03 | 9.68E-04 | 1.40E-05
foil thick: 6.26E-05 | 7.01E-05 | 5.57E-05 | 2.16E-06
-----------------------------------------
final thick.: 1.59E-03 | 1.66E-03 | 1.53E-03 | 1.60E-05
The colomns give the mean, the maximum, the minimum and the
standard variation of the weight and thickness of the simulated
buildup (in kg and m). You can also see the contibutions of the
induvidual materials.
The weight of the 10 test boards we build is 227.12g with a stand. dev of 1.79g.
The simulation fits the measured data good with a simulated weight of 227g
and a standard dev of 2g.
These are my questions:
(q1) Is this analysis useful? Or is this just a coincidence that the
match is good?
If the answer to q1 is affirmative:
(q2) Does anyone use these kind of tools?
(q3) Can someone supply me with the specific weight of typical FR4
resins (and also their variations) or give me directions were
I can find data on these things.
(q4) Is it possible to get an estimate of the thickness of a PCB by
just weighing it? If a do another analysis with an overall
finished resin content of 70+/-3% the weight and thickness data
looks like below. The mean weight is 7g larger and the mean
thickness is now 1.7mm. This simulation indicates that it is
feasable! (Is this reasoning correct?)
mean max min sdev
preg weight: 5.93E-02 | 6.26E-02 | 5.66E-02 | 8.59E-04
core weight: 1.22E-01 | 1.26E-01 | 1.18E-01 | 1.16E-03
foil weight: 4.33E-02 | 4.76E-02 | 3.90E-02 | 1.28E-03
plate weight: 9.97E-03 | 1.27E-02 | 7.53E-03 | 7.27E-04
-----------------------------------------
total weight: 2.35E-01 | 2.42E-01 | 2.27E-01 | 2.08E-03
preg thick: 6.16E-04 | 6.56E-04 | 5.83E-04 | 1.03E-05
core thick: 1.02E-03 | 1.07E-03 | 9.68E-04 | 1.40E-05
foil thick: 6.26E-05 | 7.01E-05 | 5.57E-05 | 2.16E-06
-----------------------------------------
final thick.: 1.70E-03 | 1.77E-03 | 1.63E-03 | 1.74E-05
***************** example *************************************
***************** example *************************************
BUILD V1.0: build from file 'test.bui', Fri Sep 5 11:43:37 1997
;file test.bui for initial test build code
;example XYZ 6 layer type 1
;model type nstat
model 1 3000
******************** material data ********************
; Isola core 510 micron, 3x7628, class 2 (+/-10%) 510 (561-459) micron
; Isola preg 1080, RC% 65 +/- 3%
; E-glass fabric 1080 46.8 g/m2, level 2 -> 45.4- 48.5 g/m2
; E-glass fabric 7628 204.4 g/m2, level 2 -> 210.0-199.4 g/m2
; E-glass data spw 2540 kg/m3, dk 6.1, variations unknown
; foil 18 micron 153 g/m2, level 2 -> 168.3-137.7 g/m2
; foil 35 micron 305 g/m2, level 2 -> 335.5-274.5 g/m2
; foil spw 9860 kg/m3, conductivity copper 17.8E-9 S/m, variations unknown
; resin data not exactly known, estimated spw 1300 kg/m3 and dk 3.6
; mask data spw 1300 kg/m2, 1350-1250 kg/m3, estimated dk 3.0
; thickness_over_conductor > 8 micron, thickness_over_substrate 40 micron
;plating data copper spw 9860 kg/m3, thicknes 35 micron, (25-50) micron
******************** material data ********************
******************** build data input ********************
;resin name spw,+,-,sdv dk,+,-,sdv
resin EPON12 1300,1300,1300,0 3.6,3.6,3.6,0
;rinf name w,+,-,sdv spw,+,-,sdv dk,+,-,sdv
rinf E_1080 0.0468,0.0485,0.0454,0.0005 2540,2540,2540,0 5.8,5.8,5.8,0
rinf E_7628 0.2044,0.2100,0.1994,0.0025 2540,2540,2540,0 5.8,5.8,5.8,0
;foil name w,+,-,sdv spw,+,-,sdv [conductivity]
foil ELF_17um 0.153,0.1683,0.1377,0.0075 9860,9860,9860,0 17.8E-9
foil ELF_35um 0.305,0.3355,0.2745,0.0150 9860,9860,9860,0 17.8E-9
;preg name rc%,+,-,sdv resin rinf
preg ISOLA_1080 65,68,62,1.0 EPON12 E_1080
;core name t,+,-,sdv resin rinf1 ... rinfn
core ISOLA_3X7628 510E-6,561E-6,459E-6,10E-6 EPON12 E_7628 E_7628 E_7628
;plate name t,+,-,sdv spw,+,-,sdv
plate OUTER 35E-6,50E-6,25E-6,2.5E-6 9860,9860,9860,0
;mask name spw,+,-,sdv dk,+,-,sdv toc,+,-,sdv tos,+,-,sdv
mask POLY 1300,1350,1250,15 3.0,3.0,3.0,0 10E-6,10E-6,10E-6,0 40E-6,40E-6,40E-6,0
******************** build data input done ********************
******************** the actual build ********************
;job number
job 100789
;build name www hhh
build XYZ 0.2945 0.215
;signal name layer_name area
signal ELF_17um L1 0.01077
;preg name
preg ISOLA_1080
preg ISOLA_1080
signal ELF_35um L2 0.0059
;core name
core ISOLA_3X7628
;plane name layer_name area
plane ELF_35um P1 0.05944
preg ISOLA_1080
preg ISOLA_1080
plane ELF_35um P2 0.05916
core ISOLA_3X7628
signal ELF_35um L3 0.00021
preg ISOLA_1080
preg ISOLA_1080
signal ELF_35um L4 0.01183
;mask name area
mask POLY 0.075
mask POLY 0.075
;plate name top_area bottom_area perimeter
plate OUTER 0.01077 0.01183 3.701
;hole area
hole 0.00113
******************** the actual build done ********************
******************** dimensional analysis ********************
;end, start the analysis
*** STARTING BUILDUP for job 100789 ***
***************** build summary for XYZ6 *****************
mean max min sdev
preg weight: 5.09E-02 | 5.33E-02 | 4.88E-02 | 6.40E-04
core weight: 1.22E-01 | 1.26E-01 | 1.18E-01 | 1.16E-03
foil weight: 4.33E-02 | 4.76E-02 | 3.90E-02 | 1.28E-03
plate weight: 9.83E-03 | 1.25E-02 | 7.44E-03 | 7.17E-04
-----------------------------------------
total weight: 2.27E-01 | 2.33E-01 | 2.19E-01 | 2.00E-03
preg thick: 5.12E-04 | 5.41E-04 | 4.88E-04 | 7.64E-06
core thick: 1.02E-03 | 1.07E-03 | 9.68E-04 | 1.40E-05
foil thick: 6.26E-05 | 7.01E-05 | 5.57E-05 | 2.16E-06
-----------------------------------------
final thick.: 1.59E-03 | 1.66E-03 | 1.53E-03 | 1.60E-05
********************* end of example ***********************
##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe: SUBSCRIBE TECHNET <your full name>
To unsubscribe: SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
##############################################################
|