BPi-2400 is a methodology for truing up an energy model to match the energy usage calculated from the utility bills. BPI-2400 is a requirement for models used to calculate the % savings for the IRA HOMES rebate threshold. Read more on our blog here.

Make sure you have 12 months of historical utility bill data for electricity and the primary heating fuel. This is required to get a BPI-2400 calibrated model.

At the time of job creation, you can toggle strict BPI-2400 calibration on or off:

Programs can also choose to have the toggle set to on by default. This setting can be turned on by the Snugg Pro support team.

1. Click on the customer name in the job header to open the job info modal

2. Click on the job information at the bottom of the modal

3. Set the toggle "Use Strict BPI-2400 calibration" to Yes or No

3. Click on "Confirm Changes"

While all audit jobs in Snugg Pro attempt BPI-2400 calibration during modeling, the strict BPI-200 calibration flag does a few important things:

It sets the min/max values of modeling inputs to match the min/max range specified by the BPI-2400 standard.

Snugg Pro provides helpful pointers to troubleshoot calibration errors.

The calibration results are displayed in the tech specs section of the audit report

You may have noticed that the Input screen of a job has many fields where a range of information or non-precise data is requested. Take for instance the attic insulation section. The dropdown menu for insulation thickness shows 0, 1-3, 4-6, 7-10, and so on. In the input screen we do not ask for an actual R-Value of the attic insulation to be calculated. We use the range to calculate multiple attic R-values to test in our utility bill calibration process.

Similarly, for wall insulation, we do not ask for a specific R-value of cavity insulation since it's very hard for even the most highly trained building scientist to know what the actual R-value of the wall insulation is. We ask a non-specific qualitative question: How well insulated are the walls? Well, Poorly, Yes, No. We use the answers to these questions to generate a range of R-values for the wall insulation that we then test against the utility bills. Even the efficiency of the HVAC equipment is generated based on the model year and a range of efficiencies based on that model year are used in the calibration process.

The goal here is for the user to enter their best estimate of what they're observing in the home. We want the data collection process to be time-efficient. We don't require the measurement of every surface area down to the nearest 6 inches like some modeling standards. We take advantage of the weather normalized utility bill data, which is the most accurate representation of the energy that the home uses, and test many variations of the energy model against the bills until we find a match that's within the requirements of BPI-2400 — heating, cooling, and baseload should be within 5% of the weather normalized utility bills.

If you are absolutely certain about a data point (for example, the house has continuous rigid foam under the siding), only then should you modify the base R-value on the refine screen. Setting values in the refine screen locks those values out of the calibration process. The model assumes that that value is 100% accurate and it will vary other parts of the model in order to attempt calibration.

After modeling the job, the Calibration Errors section of the BPI-2400 Calibration window points you in the right direction on where to look if the calibration fails.

In this example, you can see that the model is slightly off from the bills and is failing BPI-2400 calibration based on the model's inputs. If you hover over any of the red numbers (tap on the information icon on a mobile device), you'll see specifics about which direction the model needs to change for that item:

If calibration fails — the loads are not within 5% of the bills — then the user will need to make adjustments to the assumptions about the job. In the BPI-2400 calibration panel — you'll find this under the Alerts icon after modeling a job where the BPI-2400 option is turned on — you can see how far off the model is from the bills for heating, cooling, and baseload for each fuel type used in the building. If it's within 5%, the text will show green, meaning you don't need to adjust anything in the model related to that component. If it's 5% or more in either direction, the text will be red. If you hover your mouse over the red text, you'll see more details about what's happening. And we also provide an extensive list of suggestions as to what might be wrong with the model and how to adjust your assumptions.

The most important thing to remember is to be sure you've got correct answers for things that you are certain about. If your model isn't calibrating and you actually went into the attic and observed a quality install of lots of insulation, then you shouldn't adjust the attic inputs further. Instead, try adjusting inputs to the model that you weren't able to gather information for, or that would take a lot of effort to ascertain precisely. Blower door tests always help a lot since they are a fairly precise measurement of how leaky a home is. Once you've made sure your geometry is in the right ballpark, and that the rest of the envelope and equipment data is close to what you've observed, then you can resort to adjusting thermostat setpoints.

For an explanation of the BPI-2400 standard, refer to our blog post on the subject. This article is for Snugg Pro users who want to get the energy model to calibrate according to the BPI-2400 standard.

Snugg Pro’s utility bill calibration methodology has been verified as compliant with the BPI-2400 standard and meets all DOE requirements for data collection within the IRA Homes Program. You can read more about this in the press release. Here