Listing every air compressor brand certified for the US market from the DOE certification database with Python

I built this actor; it's a paid tool on Apify with a free trial credit.

If a company sells air compressors, pumps, distribution transformers, commercial fridges, water heaters or central air conditioners in the US, it has to certify each model to the Department of Energy first. Those filings sit in the DOE Compliance Certification Database (CCMS) at regulations.doe.gov. On 8 November 2026 it held 4,714,625 certified records in 73 product groups.

That makes CCMS a quiet but complete answer to "who actually sells this kind of equipment in the US?". The catch is the site: one search page per product group, results in pages, and field names like Full_Load_Actual_Volume_Flow_Rate__CFM__d.

I built DOE Certified Appliances & Equipment Scraper to get those records as clean JSON rows. This post shows two jobs with real data from compressors: a brand list, and a spec table you can sort.

Limits first: CCMS has no prices, no addresses, no e-mails and no phone numbers, and it stores the brand, not the legal manufacturer. It is a certification record, not a catalogue.

Job 1: who certifies compressors in the US?

Turn on brand mode and you get one row per brand with its number of certified records. For the Compressors group on 8 November 2026 (11,431 records):

brand	certified records
Atlas Copco	9,645
Kaeser	342
Ingersoll Rand	294
Gardner Denver	280
Sullair	278
Sullivan-Palatek Inc.	144
Champion	96
Pneutech	83
Saylor-Beall Manufacturing Company	73
DV Systems	60
Quincy Compressor	51
K-Series	44
Schulz Compressores ltda	31
Bobcat	9
Bauer	1

Fifteen brands, and one of them holds 84% of the records. That does not mean Atlas Copco sells 84% of the compressors. It certifies every option combination (cooling type, dryer, heat recovery, voltage) as its own basic model, so its count is large. Record count measures catalogue breadth, not market share. Keep that in mind before you put it in a slide.

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")

run = client.actor("rel8ble/doe-certified-equipment-scraper").call(run_input={
    "productGroups": ["Compressors"],
    "uniqueBrands": True,
    "maxResultsPerSearch": 0,
})

brands = client.dataset(run["defaultDatasetId"]).list_items().items
for b in sorted(brands, key=lambda r: -r["brandModelCount"]):
    print(f'{b["brandModelCount"]:>6}  {b["brand"]}')

The same call with "productGroups": ["Pumps - General Pumps"] gave 75 pump brands, led by Grundfos (17,900), Armstrong (1,316) and EBARA (1,166). That brand list is the useful part for sourcing or for selling into an industry: it is short, complete and filed with a regulator.

Job 2: a spec table for one brand

Model mode returns one row per certified model with every rating in specs, under readable names. Here is the prefilled demo (20 Atlas Copco compressors) turned into a table:

import pandas as pd

run = client.actor("rel8ble/doe-certified-equipment-scraper").call(run_input={
    "productGroups": ["Compressors"],
    "brands": ["atlas copco"],
    "searchTerms": ["GA90"],
    "maxResultsPerSearch": 50,
})
rows = client.dataset(run["defaultDatasetId"]).list_items().items

df = pd.DataFrame([{
    "model": r["basicModelNumber"],
    "cooling": "liquid" if "Liquid" in r["productClass"] else "air",
    "cfm": r["specs"].get("Full Load Actual Volume Flow Rate (CFM)"),
    "hp": r["specs"].get("Compressor Motor Nominal Horsepower (hp)"),
    "efficiency": r["specs"].get("Full Load Package Isentropic Efficiency If Applicable"),
} for r in rows])
df["cfm_per_hp"] = (df["cfm"] / df["hp"]).round(2)
print(df.sort_values("efficiency", ascending=False).to_string(index=False))

The six GA90 (121 hp) variants in the 8 November sample:

basic model (trimmed)	cooling	CFM	CFM/hp	isentropic efficiency
GA90PA3 WATER 60HZ HAT	liquid	576.1	4.76	0.862
GA90PA3 AIR 60HZ HAT	air	575.6	4.76	0.853
GA90PA3 WATER FFDRY	liquid	653.6	5.40	0.816
GA90PA3 WATER ENREC FFDRY	liquid	653.8	5.40	0.812
GA90PA3 AIR ENREC FFDRY	air	653.8	5.40	0.802
GA90PA3 AIR FFDRY	air	652.3	5.39	0.791

Two things stand out. Same motor, same 75 psig, but the HAT variants deliver about 12% less air and report the highest efficiency. So "most efficient" and "most air per horsepower" pick different models. And the energy-recovery (ENREC) option barely moves either number. You only see trade-offs like this when every variant sits in one table. The vendor's brochure won't line them up for you.

The specs keys change per product group. Pumps report flow and head, transformers kVA and efficiency, water heaters capacity and energy factor. Read the keys from the first row instead of hard-coding them.

How it works

CCMS runs on a public Solr search service behind those product pages. The actor sends plain HTTP requests to it, with no browser and no login: one request at a time, 200 records each, with a declared user agent. Three things I learned while building it:

• Brand names need exact spelling. The brand filter is an exact match, so the actor first reads the brand facet for the chosen groups and keeps every brand that contains your word. xylem resolves to "Goulds Water Technology - Xylem" and "Bell & Gossett - Xylem".
• Some groups are huge. Central air conditioners and heat pumps hold 3,764,692 records, because every indoor and outdoor pairing is certified separately. Give a model prefix or a brand there, and set maxResultsPerSearch.
• Field names are DOE template names. Full_Load_Actual_Volume_Flow_Rate__CFM__d becomes "Full Load Actual Volume Flow Rate (CFM)". Units stay in the name.

The demo run takes 1.9 s locally and 4.5 s in an Apify cloud run.

Price and caveats

$0.50 per 1,000 rows, no start fee. Searches that find nothing, duplicates and failed requests are not charged. The 15-row compressor brand list above costs under one cent.

DOE says the records are the manufacturers' own certified representations, provided as is, and a listing is not a DOE finding that the model complies. The actor is not affiliated with DOE.

Actor page: https://apify.com/rel8ble/doe-certified-equipment-scraper

Need a custom scraper or a ready dataset? Hire me on Fiverr: https://www.fiverr.com/rel8ble