Questions

Yes, and most of what we do is publicly procured. Send the invitation to bid or point us at the advertisement. We would rather look at the documents early than on the last day of the bid period, because if there is a problem with the sequence that is when it can still be raised as a question to the engineer rather than as a claim after award.

Yes, in writing, during the bid period, as a formal question. That sometimes makes us unpopular and it is a great deal cheaper for everybody than the alternative. The specific thing we look for on a retrofit is whether the sequence of construction can actually be built in the outage window the drawings assume, because that is the failure that is most expensive to discover late and easiest to see early.

The prices on the unit price page are planning ranges for this class of work in this market, published so a district can size a project before spending money on design. A bid is a fixed price against defined documents, a schedule and a specification, and it is a commitment. If you need a commitment you need a bid. If you need a number for a board meeting two budget cycles out, the price page is the more useful document and almost nobody else publishes one.

Yes, and it costs nothing. Telephone with what the facility has to do, what it does now, what the permit requires, and what can be taken out of service. We will tell you what class of project it is, which of the four cost variables is going to decide your number, what has to be defined before anybody can price it, and roughly what order of magnitude to plan for.

It depends on how well-defined the scope is and how much risk you want to hold, and it is worth deciding deliberately rather than by default. Hard bid is right for a well-defined scope with few unknowns. CMAR is usually right for a retrofit on a live plant, because it puts a contractor in the room while the sequence is still a drawing. Progressive design-build suits performance-driven and schedule-driven projects. The three are compared on delivery methods, and we have no interest in pushing one.

Regularly. Where a building scope dominates a project, a building general contractor should hold the contract and we are content to be their civil, utility and site-work subcontractor. That is frequently the right structure and we will say so rather than trying to hold a contract we are not the natural party for.

More than 75% of a typical project, with our own workforce and our own equipment: excavation and trench support, cast-in-place concrete including water-retaining structures, pipe in open cut, manholes and structures, selective demolition inside operating facilities, bypass pumping, process and yard piping, mechanical setting, and site civil. Electrical and instrumentation, trenchless crossings, immersion coatings and paving are subcontracted to specialists. Full detail, including how to verify the figure: what we self-perform.

On a well-planned project, yes — that is what the bypass and the sequence are for. What makes it fail is a bypass sized on average flow instead of peak wet-weather, a tie-in scheduled without a contingency, or a sequence that leaves the plant with no fallback for longer than a storm. All three are planning failures rather than construction failures, and all three are visible in the submitted sequence if somebody reads it properly at bid time.

Work stops on that element, the owner and the engineer are told in writing the same day, and nothing is worked around until there is a decision. On a plant built in stages over decades the drawings are frequently wrong, and the ways they are wrong are not random: abandoned lines shown as live, live lines not shown at all, inverts off by a foot. Improvising around an unexpected buried condition on a live plant is how a retrofit becomes a spill.

Three: the bypass is proven under real flow before anything is demolished; the tie-in procedure is written, rehearsed and resourced before the window; and every test result is in hand and accepted before anything is buried or accepted. Each sits immediately before a moment where the plant is exposed or something disappears underground. Passing one without proof does not remove the risk, it moves the discovery to a worse moment. See the process page.

Where the work requires it, routinely. Tie-ins are done at the lowest flow the system sees, which in most collection systems means the early hours of a weekday morning. Arterial road work is often worked at night because the traffic control is impossible in the day. Night and weekend work is a priced item and it is in the bid, not a surprise afterwards.

Our crew and superintendent, and the owner's own operators, who have the authority to stop the work. Not as observers — plant staff know things about the plant that are not in any document, and the person who will live with the consequence should be able to halt it. The engineer attends where the contract provides for it.

Because that is the document our customers actually need and almost nobody publishes. A district cannot ask its board for money it cannot name, and the usual industry answer — call us for a quote — is useless at the point where the decision gets made. Publishing ranges with a stated basis costs us nothing we were not going to tell you in the first meeting anyway.

They are the APWA uniform colour code: blue is potable water, green is sanitary sewer. It is the same code that gets painted on a street before anyone digs, and on this site it identifies what is in a pipe and nothing else. It is never used to mean important, or new, or good. Each coloured band also carries its APWA code as text, so colour is never the only way to know what a row is about.

Field photography belongs to the project it was taken on. Record drawings, plant layouts and site access arrangements on a water or wastewater facility are commercially sensitive and frequently security-sensitive, and using them as marketing is a poor trade for a client's confidentiality. The drawings on this site are worked illustrations instead.

In its present state it makes zero external network requests: no analytics, no pixel, no CDN, no embedded map, no chat widget, no cookies of its own, and therefore no consent banner, because a banner asking permission to set nothing is theatre. It is built to load a Microsoft Advertising conversion tag when we run paid search, and the cookie policy is generated from the same setting that switches it on, so it cannot say the wrong thing.