SITE INVESTIGATION 0F CHANGES IN CONVEYED WATER QUALITY AFTER LINING MAINS WITH BLAST FURNACE SLAG CEMENT MORTAR GLASGOW FIELD TRIALS
Report No DWI0285

Jul l989

SUMMARY

I OBJECT

To investigate the effect on water quality of lining two dead end mains with blast furnace slag cement mortar in low and moderately low alkalinity waters.

II REASON

Cement mortar is widely used to internally renovate corroded iron water mains throughout the UK. In some cases the pH levels of the water passing through newly lined mains can be elevated above the EC directive maximum advisable values. To alleviate this, WRc has developed a modified mortar replacing 65% of the ordinary Portland cement with blast furnace slag (BFS). The new BFS mortar, during laboratory and field trials has been found to reduce the pH in the conveyed water by up to one pH unit in comparison to OPC. The trials in Glasgow have been conducted as a final part of a series of field trials to investigate the effect on pH of using BFS mortar in waters of varying alkalinity. Two sites were chosen, one at Millerston where the alkalinity of the water was approximately 14mg/1 as CaCO3, a level not previously examined and at Howwood where the alkalinity was approximately 22mg/l as CaCO3.

III CONCLUSIONS

  1. At a water alkalinity of approximately 19-23mg/l as CaCO3 (Howwood trial) the pH of the conveyed water in contact with the blast furnace slag cement mortar for 1 hour remained above the EC directive MAV of 9.5 for a period of 9 weeks. For a contact time of 6 hours the conveyed water pH remained above 9.5 beyond the duration of the trial 15 weeks after first returning the length to service.

  2. At a water alkalinity of approximately 12-16mg/l as CaCO3 (Millerston trial) the pH of the conveyed water in contact with the blast furnace slag cement mortar for 1 hour remained above the EC directive MAV of 9.5 for a period of 12 weeks. For a contact time of 6 hours the conveyed water pH reached a value of 9.5 units, 15 weeks after returning the length to service.

  3. Initial pH values at Millerston may have been artificially low compared to the Howwood trial as high flows were maintained over the trial length due to consumer supply problems. However on rectifying the flow, pH values stabilised rather than increased.

  4. The concentrations of aluminium, calcium, potassium and sodium within the conveyed water was initially high, with aluminium above its EC directive MAV at both sites and potassium above its respective limit at Millerston. Potassium concentrations had fallen sufficiently by the eighth day but aluminium remained above the EC directive MAV, 9 weeks after lining at Millerston and 5 weeks after lining at Howwood.

The high concentrations of aluminium measured at the Howwood site was aggravated by the high source water levels of aluminium.

IV RECOMMENDATIONS

  1. Although the conveyed water pH values remained above the EC directive MAV for a significant period of time after lining, the use of blast furnace slag cement "Mainline" shows benefits in terms of pH when compared to Ordinary Portland cement. Therefore, if no alternative renovation or renewal technique is available for critical condition mains the use of blast furnace slag mortar is currently the best option and should be continued to be used.

  2. With the high levels of pH and metals occurring within the conveyed water the first few days after lining, a flushing programme could be initiated to alleviate water quality problems during this period.

  3. Cement mortar lining of dead ends or long rural lengths where little or no flow exist should be avoided. However, in certain circumstances flows could be artificially created through the mains by joining overland dead ends or by having controlled wastages.

  4. Source water to proposed relining areas should be monitored and if feasible controlled to reduce potential water quality problems over a short period, especially pH and aluminium.

  5. Further field measurements should be taken at both Howwood and Millerston 4 months after the initial return to service. Values after a 1 year period would also provide useful information.
Copies of this report may be available as an Acrobat pdf download under the 'Pre 2000 Reports' heading on the DWI website.