CellNet Comments on Statistical Load Profile Report

To: Toby Tyler, PG&E
       Carl Silsbee, Edison
       Bob Hansen, SDG&E

CellNet has three comments of the draft Statistical Load Profile (SLP)
Workshop Report.

1. ELIGIBILITY - The Report Should Have a Method that Faithfully Implements
the CPUC Decision

The Report does not have a consistent and equitable formula for determining
which customers exceed the CPUC-established 20 kW threshold. Section 5.1 of
the Report presents methods that vary by UDC, customer class, and some
within customer class. Some formula is obviously needed for those customers
who are not clearly above or below the threshold and who do not have hourly
or demand meters.

A simple, fair, consistent, and easy to implement approach is to use the
same SLPs that are used for billing customers to test whether those
customers are exceeding the 20 kW threshold. Any other method would seem
inherently unfair, because it would use one data source to bill customers
and a different data source to determine the threshold.

The proposed method is straightforward. In applying SLPs to a customer's
total consumption to bill for Direct Access, the billing agent (or Meter
Data Management Agent) must estimate consumption for each hour in the
billing period. By definition, if a customer's estimated hourly
consumption, used for billing purposes, for a particular hour exceeds 20
kWh, the customer's peak demand exceeds 20 kW. Also, because this
calculation must be done for billing anyway, there is also no additional
work or cost associated with implementing this method.

The UDCs have proposed 20 kW tests that have problems that could be avoided
through use of this proposed method (CellNet's desire is not to criticize
the UDC proposals, but to show additional benefits of the proposed method).
For example, SDG&E proposes a customer be characterized as above 20 kW if
their usage exceeds 12,000 kWh for a billing month. This test translates
into a load factor of 83.3 percent (12,000 kWh divided by 20 kW times 720
hours - 14,400 kWh - equals 0.833), far in excess of SDG&E's system average
load factor of 59 percent (according to Enova's 10-K report for 1996).
SDG&E's average customer - 59 percent load factor - with a peak load of 20
kW would have usage in that month of only 8,500 kWh (0.59 times 14,400
kWh), not 12,000 kWh. The 12,000 kWh customer, if their usage has the
average load profile, has a peak load of 28 kW, not 20 kW.

In another example, PG&E proposes 35 hp as being equivalent to 20 kW.
According to the Edison Electric Institute's Handbook for Electricity
Metering, 35 hp is actually 26 kW (Ninth Edition, 1992, page 13: "One
horsepower is approximately equal to 0.746 kW").

For both examples, there would be many Direct Access customers with demands
over 20 kW that were directed by the CPUC to install hourly meters but were
deemed, via the proposed UDC formulas, to have demands below 20 kW.

2. METER AVAILABILITY - Availability is Not an Issue, Provided Sufficient
Lead Time is Given

The Report notes that Parties at the Workshop expressed concerns about the
availability of hourly meters (at page 16). Availability will not be a
problem, assuming the CPUC acts in a timely fashion on the adoption of
meter standards and assuming that the CPUC adopts reasonable standards that
allow for flexible implementation of hourly metering in a competitive and
innovative meter market (i.e., the standards are not overly prescriptive
and restrictive). Given these assumptions, hourly metering will be
available from a variety of retrofit device providers (CellNet,
International Teldata, and others) as well as a variety of solid state
meter manufacturers (Schlumberger, General Electric, ABB, Landis & Gyr,
Process Systems, Inc., and others). These manufacturers currently deliver
well over 1 million hourly-capable meters nationwide each year, and that
delivery rate could be doubled given a lead time of approximately six
months. Many, if not all, are expected to, and are planning to, supply the
California market, so even significant uptake of Direct Access by 20 kW
customers would not be expected to result in shortages.

3. METER AFFORDABILITY - The Competitive Metering Market Created by the
Commission Will Result in Much Lower Metering Costs

The Report notes that Parties at the Workshop expressed concerns about the
affordability of hourly meters (at page 16). Many Parties in the Direct
Access proceedings testified that metering costs will decline significantly
as a result of competition. Most significantly, Customer Choice for Energy
Services said: "Competitive meter manufacturers and metering, billing and
information services providers will have financial incentives to negotiate
with customers to provide them with meters at little or no charge"
(Rebuttal Comments by Customer Choice for Energy Services, February 7,
1997, at page 7). Similarly, in D.97-05-039, the Commission said, "Where
there is customer choice, there may be... a greater likelihood that
competitive forces will keep prices low" (at page 28). CellNet shares these
views.

SUGGESTED REPORT MODIFICATIONS

Insert at the end of Section 5.1:
One party suggested an alternative to the UDC approach. This alternative
would specify that the statistical load profiles used to bill Direct Access
customers without hourly meters also be used to determine whether those
customers exceed the 20 kW threshold for required hourly metering. This
alternative would result in a consistent methodology for all three UDC
service areas and for all customers.

Add a third sentence to the second paragraph of Section 5.2:
One party commented that that meter availability and affordability were
discussed in the Commission's deliberations on the Unbundling of Revenue
Cycle Services, and that various parties, including the Commission,
expressed the belief that competition in metering will help alleviate
affordability and availability issues (D.97-05-039 at page 28).



CellNet appreciates the opportunity to comment.


.