Transformer Sizing Tool

Sourced values + labelled assumptions · Verify before designVerify assumptions
Back to EngineeringBack to Engineering

Inputs

Homes
kWh/mo assumed

Home type
Level 2 charger · kW at 240 V

Design demand method

Transformer

Impedance %Z

No-load voltage · 120 V base assumed

Limits

Peak limit set
CLPU factor
CLPU short-time limit % assumed

Secondary layout

Click a span, drop or house to edit

Assumptions for Shane to verify

Every value below is a default that is not taken from a cited source, or is a source value applied outside its original setting. Each is marked ☐ verify; the live value is shown. Change them in the Inputs panel or the layout editor.

What this tool does

  1. Design load. The standard method is the NEC 220.82 optional dwelling calculation, applied to each home and summed: 100% of the first 10 kVA of general load plus 40% of the rest, plus the larger of the A/C load (100%) or the electric heat (65%). EV Level 2 chargers are added to the general load. Two alternatives can be selected. Kersting DF: each home's 15-min peak is kW = 0.2 + 0.008 × kWh/month (Kersting Ex 2.1), the sum is divided by the diversity factor for the number of homes (Kersting Table 2.2, N = 1–20), converted to kVA at PF 0.9, and EV chargers are added at full kW. NEC 220.84: for 3 or more units, the connected load of all units times the Table 220.84 demand factor. Commercial entries are kVA, summed.
  2. Transformer. The recommended size is the smallest standard size whose summer peak loading is at or under your limit. 1-phase %Z and angle default to the TDEC guide; 3-phase pad %Z is a range (IEEE C57.12.34 via Eaton) and the default pick is an assumption.
  3. Secondary voltage. Balanced 120/240 V, on a 120 V base. Each span and drop carries the demand of the customers downstream of it by the selected method (with Kersting DF, the diversified demand of Ex 2.1 part 2), so loads accumulate back toward the transformer. Drop = I (R cos θ + X sin θ) per leg, with R at 75 °C and X from Southwire, PF 0.9. The transformer drop uses Zt = %Z × 0.144 / kVA on the 120 V base (TDEC App II).
  4. A/C start flicker. Locked-rotor current 32 A per ton at 240 V and PF 0.6, through the transformer, the span and the drop to that customer. Compared with the 5% transformer + service design limit, with the 4.5–5% GE-curve band for 2–3 starts per hour shown as a warning and the 1.75% visibility line for reference. This treats the start current as constant at rated voltage, so it reads a little high (conservative) against TDEC's App II method, which includes the motor impedance.
  5. Checks. ANSI C84.1 service voltage: Range A 114–126 V pass, Range B 110–127 V warn, outside B fail. Conductors: pass at or under the 75 °C ampacity, warn up to the 90 °C ampacity, fail above. Transformer: pass at or under 100% of nameplate, warn up to your summer peak limit, fail above it.
  6. Spans. The default layout is the transformer pole alone, with every drop from it. Add span left / right adds a span and a new pole (with one new drop and customer) beyond the outermost pole on that side, up to two spans each side. A span carries every customer at its pole and beyond, and a customer's voltage drop and A/C dip include every span between it and the transformer. Removing a span moves the drops on its pole one pole toward the transformer.
  7. Service configurations. Besides single-phase 120/240 V, the tool sizes open wye–open delta banks (120/240 V high-leg, 240 V, 480 V), wye banks or 3-phase pads (120/208Y, 277/480Y) and closed delta banks (120/240 V high-leg; 240 V and 480 V, bank or pad). These use the load list (kVA entries); each load is three-phase (balanced) or single-phase on a chosen phase or winding. Unit loading: wye, S3/3 per unit plus the single-phase load on that phase; closed delta, S3/3 per unit and a single-phase load split between its winding and the other two in series in inverse proportion to impedance (2/3 and 1/3 for equal units); open delta, S3/√3 per unit (bank capacity 86.6% of the two units, 57.7% of a closed bank) plus single-phase load on its unit. High-leg banks size the lighting unit and the power unit(s) separately. Voltage drop and A/C dip use TDEC's 120 V phase-to-neutral base (App IV): Zt = %Z × 0.144 / kVA per unit, three-phase current S × 1000 / 360 through one-way conductor impedance × 3 × 120² / VLL², single-phase current S × 1000 / 120 through the round trip × 120² / V². On open delta each unit's current is taken 30° off the load angle (the worse of ±30°). Shares are added arithmetically, so results are conservative. ANSI C84.1 ranges are applied on the 120 V base (the same percentages at 208, 240, 277 and 480 V).
  8. Growth headroom. Homes are added one at a time on new drops copied from the worst customer's drop and pole; EVs are added to existing homes starting with the lowest voltage. The count stops at the first failed check: transformer over the limit (upsize), voltage under Range A, flicker over 5%, or a conductor over its 90 °C ampacity. The installed size is held fixed.

Transformer loading (IEEE C57.91 concepts)

C57.91 is the IEEE guide for loading oil-filled transformers. It is paywalled, so only its concepts are used here. A 65 °C-rise transformer is designed around a 110 °C winding hottest-spot temperature at rated load and 30 °C ambient. Loading above nameplate is allowed when the ambient is cooler or the peak is short, because the oil and windings take time to heat up; the cost is extra insulation aging (loss of life). A residential summer peak lasts about two hours, so many utilities accept peaks above nameplate. The limit you set (default 120%) stands in for your utility's loading guide. The tool does not run the C57.91 thermal model.

Cold load pickup

After a long outage, thermostatic loads all restart together and diversity is lost. The IEEE PSRC working group report gives about 2.2× the normal peak at 15 minutes, with diversity regained in about an hour; a TVA summer test plateaued near 2.8×. The tool multiplies the design load by your factor and compares it with a short-time limit that you set (an assumption).

Validation

The same calculation file is checked by tx_validate.js against Kersting Ex 2.1 (30.3 kW for 5 customers), Kersting Ex 2.3 (232.15 / 233.1 / 232.7 V) and the TDEC App II sample (7.9 V transformer and 3.8 V service dips). The multi-phase math is checked against the TDEC App IV three-phase sample (3 × 37.5 kVA, 480 V, 100 hp motor: 13.4% total dip) and App III example (three 15 kVA units, 50 hp 460 V motor), the open-delta 86.6% / 57.7% capacity rule, the closed-delta 2/3–1/3 split, and the multi-span layout against hand calculations.

Sources

    Conductor data, diversity factors and every source value are in data/tx-data.js; see data/README.md. The data file has empty slots for Entergy, Xcel Energy and FPL diversity tables.

    Disclaimer

    For illustration and preliminary design only. This is not a substitute for your utility's design standards, load data and transformer loading guide, or the judgment of a qualified engineer. Defaults marked as assumptions must be checked. No warranty is given.

    For illustration and preliminary design only. Load, impedance, conductor, flicker and voltage values are from cited public sources; values marked as assumptions must be verified against your utility's standards.

    Method & assumptions