VDH needs to get out of the product approval business. VDH's efforts to affect a rational and statistically robust testing program has failed. The entire concept of General Approval and Treatment Level (TL-3) needs to be rationally evaluated.

Puraflo failed its testing program yet was granted General Approval. Both Advantex and Ecoflo used sampling lysimeters that filtered out the bacterial contaminants used as the pass/fail criteria – yet were granted General Approval.

The statistical model used in GMP-147 is fundamentally flawed. Dr. David Edwards of VCU found that the use of standard error and confidence intervals for the mean were not appropriate when interest lies in where treatment unit performance will fall. He further added, the tolerance intervals computed by VDH are too low as a large percentage of treatment units will fail the criterion.

The raw data shows that each of the three units above had average BOD levels ranging from 6.9 to 8.3 mg/l but had a 99% confidence limit range of 28.5 to 43.2 mg/l. This data demonstrates that the units work well most of the time, but cannot achieve the treatment levels of 10/10 more than about 50% of the time. Similar data exists for TSS.

Treatment Level 2 Effluent should be replaced with – “Secondary Effluent” means effluent that has been treated to produce BOD5 and TSS concentrations equal to or less than 30 mg/l each.

Treatment Level 3 Effluent – should be removed for small systems. There are no manufactured treatment units that can achieve a BOD of 10mg/l consistently. This standard is superfluous – TL-2 (Secondary) is achievable and allows for disinfection and meets the EPA/DEQ discharge quality standard. TL-3 is a contrived standard.