One of our Title 24 clients, Okamoto Saijo Architecture, recently completed a  $50M retrofit that included creating a 900-kW PV system that is currently one of the largest affordable-housing solar installations in the world. We interviewed one of the principal architects, Eric Saijo, about how the Crescent Park project went from his perspective. He was actually quite happy with the outcome, and after 4+ years of budgeting, negotiating with utilities, the project is completed.

Crescent Park, an affordable-housing solar retrofit by Okamoto Saijo Architecture.

How were you selected?

In the last 12 years we’ve done lots of affordable housing rehabilitation projects. In this day and age people get put into certain categories. We’ve developed a reputation.

Eric Saijo and Paul Okamoto of Okamoto Saijo Architecture, at one of their own project sites

It’s less glamorous, but that’s OK. This client had been working with another architect and the project got put on the back burner for half a year. In 2005, they came to us. We spent 2 years in design and documentation to figure out the project scope.

How was scope determined?

Identifying the budget is always a challenge. They had a wish list of many items, including PV for 100% of electrical needs. We did feasibility studies to analyze whether they had the budget for all the things they wanted to do: update kitchens, flooring, waterproofing. The solar portion was only one aspect.

You did a lot of analysis in addition to design.

We worked hand in hand with our contractor (Brandon Slater of West Coast Contractors) from Day 1 on pricing and budget.

What would you do differently next time?

A better question might be what have we learned? Let’s talk about this instead. We learned more about handling the specific challenges of pulling off a PV installation in a 40-year-old multi-building complex.

Okamoto Saijo Architecture has done "green" private residences as well as sustainable public housing.

We started off with plans for 100% of onsite electrical needs generated by PVs. Our solar engineer did a layout showing where we could put panels on all roofs, taking tilt and orientation into account for each building. We also had an idea of the number of kWh that we needed to generate. Then we could look at all the other constraints.

What were the other constraints?

Not enough roof area on the existing buildings, and existing building systems that were not designed for the structural loads of the panels and the installation process.

Are solar panels really that heavy?

No, but retrofitting existing buildings triggers all sorts of re-analyses, and one of these is to re-analyze for seismic load. Any building that’s 40 years old won’t pass today’s seismic code requirements. And any increase in load over 5% triggers this seismic analysis… it’s a huge limiting factor.

Most PVs get installed in a design-build fashion.

When installing panels, there should be no live load on the roof where the panels are… adding even a minuscule amount of weight can be a problem sometimes.

What was the problem with retrofitting to use existing equipment?

How to make the most of the existing electrical service equipment in a retrofit! The simplest thing to do for an individual building is to install a large PV system and replace all the service equipment and tie in the PV to the entrance panel breaker.

Connecting those beautiful photovoltaics to the public utility's metering and grid system can be a "non-trivial exercise".

The electrical code is written in such a way that PVs are considered a “load” meaning that you might have to up-size the service equipment. The secondary field lines from PG&E, with extra runs to each building, for 24 buildings on 6 acres… this becomes a huge cost.

However, the electrical code does allow for line-side tap between the meter and the main shutoff switch. The equipment is now 40 years old which the code still allows, but it is physically difficult to implement line-side taps. We had to persuade PG&E and the head building official to conceptually approve it.

The larger buildings had enough space in their service equipment to clamp onto existing conductors when we needed to do that. We had to modify the charge condition meter main shut off and route it through a new gutter. We could do new tap here. and then clamp to conductor.

UL certification was another issue. The existing equipment, being 40 years old, wasn’t UL fabricated. Small enough to service meter and switch board were separate pieces of equipment. When the contractor, the electrical building inspector, the electrician, and the solar engineer got together – it was a tense moment!

When architects and building inspectors meet…

The next challenge was getting our systems approved by PG&E. You have to get approval from your local utility in order to submit for solar tax rebates. A 900 kW system like this one must undergo review at PG&E’s engineering department. They analyze their own infrastructure, including their transformers and underground conduits. In this case, PG&E’s equipment was also 40 years old, and perhaps built to prior standards. It took them quite awhile to analyze our proposal.

At first, they rejected it and wanted us to pay to upgrade all the transformers serving the complex. This was due to a loophole in the agreement for rebate systems for PVs, which allow the utilities to charge the client for these upgrades.

But if the power is generated onsite, why do you need those transformers?

In the middle of the day in a residential complex, power will be flowing out towards the grid. Changes to the photovoltaic systems had to be calibrated on the utilties’ side as well as ours. They had to change their meters so they could spin backwards.

Is that a smart meter?

No. A smart meter is one which is read remotely. [Communication is essentially what makes a smart meter more intelligent than a dumb meter]

Does PG&E do enough to support people like you?

There are people in various departments who did. The engineers really got behind us and worked with us to MAKE it work. Then there are other departments. All of them get delayed for a number of months without clear explanation. It could just be under-staffing.

Here’s an example of the type of problem we had to solve together with PG&E engineering… in one area of our project, there were 5 buildings served by one transformer. At first the told us that we had to pay for a new transformer, as well as pay for new primary and secondary feed lines – this would cost $200K.

Naturally our client wasn’t happy. PG&E countered that they were concerned that the kW would bump up the voltage above what they’re legally required to keep it under. Then they said, “But.. if you set the trip point on the inverters down, then we’ll approve.”

Inverters are normally set to 132V and they wanted us to set them at 127V. The PG&E grid is 120-122V but it just happens to be high in this particular location. As we installed systems in those 5 buildings, the inverters started to trip off. We’re still negotiating with PG&E over what to do.

What would you do different?

I wouldn’t accept a trip point! It was a sign that PG&E has real concerns about its own systems, that they were worried about the potential for voltage to increase too high . Our solar engineer had never run into that before. It was a real learning experience!

Even a genius can have a learning experience.

At this stage, our client didn’t know yet how much financing they could get. The budget was still in flux, and they really weren’t willing to accept sudden new costs. Especially in large renovations, you have to hold a large contingency fund; with our project, those funds are now available for post-construction.

What do the residents think about it?

Affordable housing is a very complex thing in our society. These are extremely low-income people. I didn’t have much contact with them but my sense is that they are appreciative of the renovations that included window replacement and other building improvements which improved their comfort and quality of life.

Another affordable housing project from Okamoto Saijo Architecture, PositiveMATCH is an adaptive re-use of a historic building in San Francisco, serving women with HIV and their children.

The buildings were made more airtight, with better insulation, new windows. And… cleaning the duct work after 40 years most likely improved the air quality.

Drainage for the entire site was improved. It’s very close to the Bay, with a high water table, so flooding is a concern. The storm drains were constantly backed up prior to the renovation, and ground floor units had water infiltration. All of these measures made the units more comfortable.

Another private residential design from Okamoto Saijo Architecture, a passive-solar farmhouse in Napa, CA

We also worked with the client to improve the visual appearance of the buildings. New paint schemes, and individual colors for each building. Before that, all 26 buildings on 24 acres had been colored the same. How monotonous!

Of course the resident’s don’t pay their own electric bill. That’s usually the thrilling part for homeowners is seeing their utility bill reduced. In this case, our client financed the PVs because they also pay the utilities.

Financing was through bonds. Our client was not the original developer. The project was originally built by a market-rate developer together with HUD. Our client bought it later, around 20 years ago.