Modified Coffin-Manson (Norris-Landzberg)
Modified Coffin-Manson (Norris-Landzberg)
Modified Coffin-Manson (Norris-Landzberg)
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
• <strong>Modified</strong> <strong>Coffin</strong>-<strong>Manson</strong> (<strong>Norris</strong>-<strong>Landzberg</strong>) Equation<br />
Used to Determine Acceleration Factor from ATC to<br />
Field Conditions:<br />
• Where:<br />
<strong>Modified</strong> <strong>Coffin</strong>-<strong>Manson</strong><br />
Equation for AF Calculations<br />
[]<br />
ΔT<br />
l<br />
AF =<br />
ΔT<br />
f<br />
1.9 1/3<br />
(<br />
f<br />
l) f<br />
1414<br />
f<br />
(<br />
{1<br />
exp<br />
T max f<br />
AF = Calculated Acceleration Factor (on a Cycle Basis).<br />
ΔT = Package/Board Temperature Difference Between Ton and Toff (°K).<br />
Tmax = Maximum Solder Joint Temperature (°K).<br />
f = Cyclic Frequency (in Cycles per 24 Hour Day). Minimum Number of Six<br />
Which Assumes Maximum Damage Takes Place Within Four Hours.<br />
f, l = Subscripts to Denote Field and Lab (i.e., ATC) Conditions, Respectively.<br />
m Customer Applications Support 512-933-7289 10 April , 1998<br />
1<br />
- })<br />
T max l
Information and Asssumptions<br />
The key temperature for predicting the board level solder joint reliability is the<br />
temperature of the solder balls at the board. It is not the device junction temperature.<br />
The examples given are for illustration. Individual applications can be addressed<br />
by contacting the Customer Applications Support department.<br />
The <strong>Coffin</strong>-<strong>Manson</strong> predictions are generally recognized as “conservative” or to<br />
pessimistically predict failures under most long term use conditions.<br />
New data and process improvements are an ongoing effort and for the latest<br />
information please contact your SPS Sales contact or the Customer Applications<br />
Support department.<br />
The design of the board that the CBGA is assembled to affects the solder joint reliability as<br />
well as attachment of heat sinks. Most life testing is done on an FR-4 2S,2P board that is<br />
0.062 inches thick and the components are assembled on one side. Other materials and<br />
configurations can be tested by special request through your SPS Sales contact.<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Lifetime in Years<br />
1000.00<br />
100.00<br />
10.00<br />
1.00<br />
10<br />
FR-4 Board<br />
1 cycle per day<br />
0.01% Cumulative Failure Rate Prediction<br />
15<br />
20<br />
25<br />
Delta T in Celsius<br />
Toff = 50C<br />
m Customer Applications Support 512-933-7289 10 April , 1998<br />
30<br />
35<br />
CBGA<br />
Size, Thickness<br />
21 mm, 1.0 mm<br />
25 mm, 1.0 mm<br />
25 mm, 1.2 mm
Lifetime in Years<br />
1000.00<br />
100.00<br />
10.00<br />
1.00<br />
10<br />
Thermount Board<br />
1 cycle per day<br />
0.01% Cumulative Failure Rate Prediction<br />
15<br />
20<br />
25<br />
Delta T in Celsius<br />
Toff = 50C<br />
m Customer Applications Support 512-933-7289 10 April , 1998<br />
30<br />
35<br />
CBGA<br />
Size, Thickness<br />
21 mm, 1.0 mm<br />
25 mm, 1.0 mm<br />
25 mm, 1.2 mm
Time to 0.5% Packaging Failure (Years)<br />
Case A: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
1000<br />
100<br />
10<br />
1<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
Case A Assumptions<br />
• T off = 23°C<br />
• 6 cycles per 10 hour day<br />
• 200 & 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
Case B: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
1000<br />
100<br />
10<br />
1<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
Case B Assumptions<br />
• T off = 40 °C<br />
• 6 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
1000<br />
Case C: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
100<br />
10<br />
1<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
Case C Assumptions<br />
• T off = 23 °C<br />
• 10 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
Case D: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
1000<br />
100<br />
10<br />
1<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
Case D Assumptions<br />
• T off = 40 °C<br />
• 10 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
Case E: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
1000<br />
100<br />
10<br />
1<br />
Case E Assumptions<br />
• T off = 23 °C<br />
• 1 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
Case F: 21 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
1000<br />
100<br />
10<br />
1<br />
Case F Assumptions<br />
• T off = 40 °C<br />
• 1 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from -40 to 125 °C<br />
21 mm CBGA ATC Data<br />
0 10 20 30 40 50 60 70 80<br />
Delta T (°C)<br />
21mm CBGA 200 work day/yr<br />
21mm CBGA 365 work day/yr<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
25 mm, 1mm Thick CBGA Solder Joint Reliability With 4 Balls Off<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
100<br />
10<br />
1<br />
0.1<br />
Assumptions<br />
• T off = 23 °C<br />
• 3 & 6 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from 20-80 °C<br />
25mm ATC Data<br />
6 cpd,365 day/year,Toff=23<br />
6 cpd,200 day/year,Toff=23<br />
3 cpd,365 day/year,Toff=23<br />
3 cpd,200 day/year,Toff=23<br />
0 5 10 15 20 25 30 35<br />
Delta T (°C)<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
25 mm, 1mm Thick CBGA Solder Joint Reliability With 4 Balls Off<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
100<br />
10<br />
1<br />
0.1<br />
Assumptions<br />
• T off = 40 °C<br />
• 3 & 6 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from 20-80 °C<br />
25mm ATC Data<br />
6 cpd,365 day/year,Toff=40<br />
6 cpd,200 day/year,Toff=40<br />
3 cpd,365 day/year,Toff=40<br />
3 cpd,200 cpd,365 day/year,Toff=40<br />
0 5 10 15 20 25 30 35<br />
Delta T (°C)<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
100<br />
10<br />
1<br />
21x25 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
Assumptions<br />
• T off = 23 °C<br />
• 3 & 6 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from 0-100 °C<br />
21x25mm ATC Data<br />
6 cpd,365 day/year,Toff=23<br />
6 cpd,200 day/year,Toff=23<br />
3 cpd,365 day/year,Toff=23<br />
3 cpd,200 day/year,Toff=23<br />
0 5 10 15 20 25 30 35<br />
Delta T (°C)<br />
m Customer Applications Support 512-933-7289 10 April , 1998
Time to 0.5% Packaging Failure (Years)<br />
100<br />
10<br />
1<br />
0.1<br />
21x25 mm, 1mm Thick CBGA Solder Joint Reliability<br />
<strong>Norris</strong>-<strong>Landzberg</strong> Predictions from Accelerated Thermal Cycling Data<br />
Assumptions<br />
• T off = 40 °C<br />
• 3 & 6 cycles per 10 hour day<br />
• 200 and 365 work days per year<br />
• Log Normal 0.5% Failure<br />
extrapolations from 0-100 °C<br />
21x25mm ATC Data<br />
6 cpd,365 day/year,Toff=40<br />
6 cpd,200 day/year,Toff=40<br />
3 cpd,365 day/year,Toff=40<br />
3 cpd,200 cpd,365 day/year,Toff=40<br />
0 5 10 15 20 25 30 35<br />
Delta T (°C)<br />
m Customer Applications Support 512-933-7289 10 April , 1998