from this post:

http://forums.tupvians.com/index.php?showtopic=2617&st=50

ang akon ya is on the aspect of thermomechanical. mismatch in the coefficient of thermal expansion (CTE) between mold compound and the leadframe (or substrate) may cause bending stress. More bending means more tensile stress sa mold compound and as a result, will induced more compressive stress sa die. one factor that diebond process contributes to the cracking of die is the bond line thickness (BLT) sg epoxy. higher BLT almost always have higher fillet height, especially when bonding force is mataas. higher fillet height means lesser degree of freedom sa die. lesser degree of freedom can prevent die from expanding and when the die is overly contrained, it is more likely to absorb all the stress. and since silicon dice are brittle in nature, minute surface flaws can be a crack starters.

pamangkot: what happen to the adhesion between die-glue or glue-leadframe interface when to much CTE mismatch occurs?

*no .... indi ko kabalo...

i guess.. since indi match and CTE nila, it will cause either undersized or oversized product... or broken product.... ( :o ... nosebleed) :lol: :lol: :lol:

hehehe.. mayo gid migo rostan ya.

this problem is typical to epoxy mold compound (EMC) and leadframe (LF) interface. with EMC having thermal expansion coeff (TCE) usually at 8ppm/K (<Tg) and 32ppm/K (>Tg) versus LF at around 17ppm/K.

above Tg (TCE: EMC>LF), EMC tends to pull the LF upward which causes the IC package to bend on a smiling manner :). whereas below Tg, it's the opposite kay mas daku ang TCE sg LF, so frowning behavior ya :( .

with various change in temperatures tungod sg assembly (+ reliability) process like molding (175) to RT (25) to reflow (260) to RT (25) to TMCL low (-55) to TMCL high (150), ang behavior sg package is daw laminated ID nga gina-lungilungi. eventually, the adhesion b/n EMC and LF interface weakens. kag with the presence of moisture since ang polymer is hygroscopic in nature, it will delaminate.

later try ko mag-upload image. rockon

theory lang ang akun migo baltix kay indi ako involve sa amu ni nga mga bagay sa ubra.. though i have various job encounters about TCE sa facilities installations..... will try to share also paf naka luwag luwag....

in principle sakto na imo mig ah. application-wise, pwde man japon sa facilities mig. mga expansion sg mga valve niyo da ah.

mainvolve lang kami kung macroscopic ang expansion nga involve pero kung mm to cm lang.... may factor of ignorance da kami nga ginausar.... :lol:

ari na daw maayo na gid nga discussion ni. So, knowing the effect of CTE mismatch, how do we correct it in the assembly production floor if we are seeing this issue in our chips? Based on physics, how are we to correct or remedy this kind of process problem? :)

idol raymie,

sa assy manufacturing point of view, daw ka unusual gid to alter the material property. pero i think, we can lessen the impact or improve the adhesion performance by making sure nga ang surface sg LF is clean enough. one can use plasma cleaning to make sure nga limpyo ang surface sg LF. by removing oxides and other foreign material can definitely improve the reliability of an interface.

another novel idea is by roughening the surface of the leadframe through chemical etching. with the surface being rough, it improves the adhesion by providing better linking b/n two materials.

my 2-cents.

improving the LF design is another thing...

baltz.... change materials ka...... GO GREEN! :lol: :lol: :lol: :lol:

mig,
usual suggestion na sg process engineer. material issue. hehehe..

you control the environment... imho...

pero dalagku sweldo sang mga process engr..... gusto ko man tani ma gprocess galing kay sa construction ang akun san kapalaran nagkadto... :lol:


ice, anu meaning sang imho man?

in my humble opinion... :lol: :lol: :lol:

abi ko hinubog nga version sg "imo environment" :bleh:

given na ina guro migo ice ah. kay in cleanroom, maintained ang environment sina. sigurado na dira si migo rostan kay siya na ang ga-design. hehe..

may ara man factor of ignorance ah.... :lol: :lol: :lol:

Hmmm I think i would agree with this Migo Balts... PLasma Cleaning... ( it could be 2x or 3x plasma or mas taas pa) to get the desired cleanliness of the surface.

my 3 cents.... :lol: :lol: :lol:

may pamangkot ko gali ah. this is an actual problem. any suggestion or idea is very welcome.

amo ni ang scenario ho: in die shear test (using Dage equipt), may data ako in Kg-f. but this data represents the amount of force req'd to peel off the die from the substrate (or leadframe).

now, what i wanted to know is the stress intensity factor,K, for the initiation of crack or crack-point at the start of the interface (die-lf). how can i use the current F data (Kg-f) that i have to solve for K?

damo ka pwede mahimo kapin pa if the product is in its infancy and the process window in still open for optimization. Since LF, epoxy, die, and emc ang considerations, and CTE mismatch ang topic, you can:

1. find a molding compound with the nearest CTE to your other materials.
2. if the device does not need msl level 1, don't use a very high adhesive molding compound. The more it adheres, the more constraints there is and more failure is expected in the other interfaces.
2. evaluate epoxy bond line thickness against TCT. A very thin BLT would be disastrous in this case.
3. identify the lowest possible process temperatures for elevated temp processes like oven curing, wb, mold, etc.
4. design mechanical stress reliefs inside the package
5. make the package thinner
6. make the LF thinner
7. make sure there is 100% epoxy coverage in DA. Fillet height is not important in this scenario but coverage is.

haay, kapoy type. dugangi nyo pa gid bi.

If my understanding is correct, you are trying to determine the stress-raiser factor for this exercise. Since you tool is normally flat on the side of the die, there should be no factor for stress concentration. Since your equipment is normally inducing the force slowly and without impact, then factor for impact loading remains unity. If this is the case then, I would say that the epoxy ultimate stress is what your reading gives you and the factor is unity.

idol raymie,

close ko lang ni anay nga topic ah. daw medyo complicated ang gusto ko himuon. will try to share kon tapos na ang research. thanks gid!

Sige, share mo pagkatapos ha. But since this topic is quite wide in application, let us check sa iban nga applications abi. Since si Toto Rostan sagad sa construction, let us try to evaluate ang application sang thermally induced stresses sa reinforced concrete beams. Considering a beam of reinforced concrete, how can we prevent the concrete from cracking (by design ha) due to tensile loading from the steel reinforcement bars? Again, we know that the CTE of steel is higher than the CTE of concrete. We also know that concrete has virtually zero tensile strength but a very high compressive strength. So guys, let us grind our thinking caps again. ;) ;)

agree.... CTE of steel is a lot larger than the CTE of concrete BUT in construction application, specially on reinforced concrete steel, the delta T in the area where steel and concrete is having contact is approximately Zero...

aside from that, the CTE of steel is only about 0.00000645 /degF .... which means that for every deltaF, an inch of steel will only expand about 0.00000645 inch....

the critical factor in determining steel lenghts is the span distance (distance between supports/joints) which is limited to only about 6 to 20 meters, why?... its not the temperature that majorly counts but as engineers, we also need to balance the cost and the loading capacity, the longer the span, the bigger the beam, the higher the cost is...see moment of inertia on strenght of materials sa computation sa pag determine sang type of beam (am talking of I beam generally)...

assuming a 20m beam is induced to a 2degF deltaT = it will only expand to about 0.258mm...

while for concrete, there are two factors sa expansion.... Thermal (caused by temperature) and Hygric (caused by moisture)... lets only talk about thermal here....take for example a certain concete mixture has a CTE of 0.0000135/degF considering a change in temp above of 2degF, the mixture will expand to by 0.54mm

comparing the expansion of both 20meter concrete reinforced with steel I beam, the change in lenghts are 0.54 for concrete and 0.258 for steel....

THAT IS KUNG MAY 2DegF ka nga delta T.... in most cases, there is no delta T....

info: most of the steel span on I Beam ranges only from 6 to 12 meters...

on some steel applications, expansion joints are provided if the material is exposed to sunlight....

hope this helps..




;) :D

In actual application, there is always a delta T, kapin pa kung ambient conditions. In the areas nga may snow or deserts, extremes in temperature can be experienced by materials. Assume naton abi nga there is significant cross sectional area of steel involved vs concrete. The reason for this discussion is for us to review stresses associated with temperature changes. Also, we would want to find appropriate design considerations for compositing materials with different strengths (not only CTE). Ano-ano nga mga equations govern this interplay of stresses and strengths? :)

owkie owkie.... :lol:

may utang ako .... review ko ang topic kay structural applications ini....currently concentrated ako sa mech;l kag electrical... no excuses ofcourse... :lol:

related to the topic, though not structural, sa chilled water piping where in delta T always occur, we are installing expansion joints or flexible couplings....

i saw a structural beam nga indi continuous, means beams are not joint from span to span...

pero ang pamangkut is sa point wherein concrete and steel made contact... may idea na ako what we did sa trusses sang isa ka bldg nga namanage ko dati pero i want to get the full technical details :thumb:

deltaT poses two type of change in lenghts... contraction and expansion.... these are being dealth with specific to the application....

some of the ways engrs opted to install on construction are expansion joint and slip joint ... the name expansion and slip speaks for itself.. right?... so no need to explain in details....

as on previous posts, i had said that the changes in temperature are more often than not, being considered Zero and thus have very minimal change in lenght.. lookng back to my example, a 20meter I beam will expand by 0.258mm, in a macroscopic point of view, is negligible.... which is same is true with concrete which ias 0.54mm...

for the porpuses of discussion, since concrete and steel differs by 0.2++ mm, the method that we are doing is called "caulking" wherein a 3rd material is introduced to absorb the thermal expansion....

lets now talk about "caulking"

"Caulking" is a process and activity to close up joints and also serve as a noise absorption materials.... "acoustics!!".. These are flexible materials such as silicone, Polyurethane, polysulfide and others...

unlike on the application of die in a semicon product, construction deals in macro-perspective...

and by the way, most (if not all) buildings, are designed to move and be flexible..... like the commercial, "skysrapers sway up to 3ft on high winds"....

So one idea based from Toto Rostan is to provide stress relief to the members, allowing them to expand or contract with environmental changes. That is true. But internally (within the composite beam), there are no stress reliefs. The materials inside will try to expand or contract based on their own characteristics and they will be stressed based on the constraints put upon them. What fundamental equations would govern these stresses? And what practical measures can we do to take advantage of material characteristics to make our application successful?

hmmmm..... iban naman sabat bi.... :P :thumb:

:lol: :lol: :lol:

lets take a few steps backward anay...

refresh ko lang ang "stress" since and topic is strenght of materials, the "basic" dimension of stress is expressed in weight / square units... such as, Pa (Pascal or N/sqm), PSI, etc)... basic types of stresses that a material can undergo are: compressive, tensile, shearing, torsional, flexural(or bending) and other variations....

sa una nga pamangkot... fundamental formula for stress is S = W/A; S = stress, W =weight; A = area where the force is applied....


while sa ikaduha nga pamangkot And what practical measures can we do to take advantage of material characteristics to make our application successful?

before answering this, lets review also the characteristics of each material, sample, common two materials are CHB (Concrete Hollow Block) and Black iron Round Bar....

Concrete has a high compressive strenght but low in tensile strenght.... concrete is mainly used to cater compressive stresses and is based on its characteristics....

Black Iron Round Bar when induced to a tensile stress is good since daku ang capacity nya for a tensile stress... sa structural application, may ara gani ginausar nga tension rod on roofing application.... if induced by a force along its axis, it will somewhat resist the compressive stress BUT will fail on the flexural/bending stress, or MABAHUD sya....

to answer the second question materials are being selected based on strenght of its characteristics and application

did i answerit right idol? ;) :D


next fewdays processes naman ro reinforce the basic characteristics.. :naughty:

For the basic stress equations, tama gid ato. For stresses involving thermals and expansion, the governing equation is:

d = La(t2 - t1)

where

d = total expansion in linear units
L = length of the specimen
a = coefficient of thermal expansion or CTE
(t2 - t1) = the change in temperature experienced by the material

So from the equation, we can see that if the temp change is positive or getting hotter, the material will expand because d will be positive. If however the material is being cooled, (t2-t1) will be negative, causing d to be negative, meaning the material shrunk.

If the material is constrained, meaning not allowed to expand or contract in its natural way, there will be a resulting internal stress. What is this internal stress and how is it quantified? :)

dali lang....

daw nag balhin ang topic sa physics haw?... hmmmm :P

Strength of materials na ya. Well actually the whole subject of material strength is still under Physics. Pero specific treatment, how much will the material be stressed if it is constrained from expanding or contracting?