214 F. Supp. 3d 478
D.S.C.2016Background
- This MDL involves owners of Pella Architect and Designer Series windows (1997–2007) who allege a common defect causing water intrusion and damage to windows and adjacent walls.
- Plaintiffs proffered three SGH experts (Louis, Clark, Faulkner) who inspected hundreds of windows, performed destructive testing, conducted spray-rack and nozzle water tests, reviewed Pella documents, and visited Pella plants.
- SGH’s core opinions: the windows have defective "water management" with leakage paths (sash/frame gasket compression, glazing pocket sealant, frame corners) and that wood preservative treatment/application was inadequate.
- Pella moved to exclude the SGH experts under Rule 702/Daubert, arguing testing was methodologically flawed, samples were non-representative/biased, the experts lacked qualifications on wood treatment, and evidence was insufficient to generalize to all ~7.5 million windows.
- The court held a Daubert hearing, examined ASTM/AAMA standards (notably ASTM E2128, E1105; AAMA 502, 511, 501.2), and found SGH’s diagnostic testing and sampling methods failed to simulate or tie to actual exposure conditions and were not reliably generalized.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Reliability of water testing (spray-rack & nozzle) | SGH followed ASTM E2128/E1105 and industry practice; tests recreated leakage paths. | Tests used unrealistic pressures/volumes, ignored AAMA 511 diagnostic guidance, did not simulate actual exposure, so results are unreliable. | Excluded: court found tests inconsistent with ASTM E2128/AAMA 511, testing conditions were unrealistic and created an analytical gap under Daubert/Joiner. |
| Consideration of alternative causes | SGH inspected and observed stains/deterioration that justified focusing on leakage; testing corroborates. | SGH failed to adequately address other causes (installation, condensation, construction defects); did not rule out alternatives. | Court: SGH did not sufficiently investigate or rule out alternatives; tests did not close the gap to show leakage was the probable cause. |
| Sample representativeness / generalization to all windows | SGH inspected hundreds of windows and tested over 100 across many residences; industry standards do not require quantitative sampling. | Sample biased toward plaintiff-owned/problem windows; no statistical method to generalize to ~7.5M units; possible selection bias. | Excluded for generalizability: court found sampling nonrandom, potentially biased, and unsupported by statistical or industry-accepted methods to prove product-wide defect. |
| Qualifications to opine on wood treatment effectiveness | Louis has decades in waterproofing/engineering and familiarity with IS-4; SGH relied on observations, documents, and plant visits. | Experts lack formal training/experience in wood science/wood preservative testing and haven’t conducted IS-4 testing; conclusions venture into biology/chemistry beyond expertise. | Excluded for qualifications and analytical gap: court found Louis not shown to have requisite specialized experience with IS-4/tests; wood-treatment opinions insufficiently tied to reliable evidence. |
Key Cases Cited
- Daubert v. Merrell Dow Pharm., 509 U.S. 579 (court must ensure expert testimony is reliable and relevant)
- Kumho Tire Co. v. Carmichael, 526 U.S. 137 (Daubert gatekeeping applies to all expert testimony; factors depend on context)
- Gen. Elec. Co. v. Joiner, 522 U.S. 136 (district court may exclude expert opinion if there is too great an analytical gap between data and opinion)
- Westberry v. Gislaved Gummi AB, 178 F.3d 257 (expert must address alternative causes or explain discounting them)