Thunderstorm Generator: what was measured, and what is only claimed.
Measured: a 9.67% lower fuel-use rate on a small generator and 34% less carbon monoxide (CO) at idle; bigger claims unverified.
Malcolm Bendall’s Thunderstorm Generator is an engine add-on that is designed to use water to form and guide plasmoids (tiny, self-organized bits of plasma). This site explains it in plain English, with inventor claims labeled.
Independent education. Not Strike Foundation, Strike Energy, or a licensee.
The problem — and a different path
Scroll through a short story: what ordinary engines do wrong, then how this path reframes the energy process.
Chapter 1 · Problem
Explosive “spend” combustion
Ordinary engines burn gasoline inefficiently. A lot of the fuel’s energy is lost as heat and incomplete burn. Emissions — carbon monoxide (CO) and others — hurt air quality, and the waste shows up on your wallet.
Chapter 2 · Inventor model
Organize energy with plasmoids
Inventor model: the device is designed to form and guide plasmoids so more of the energy is organized — closer to a “store” than a pure blast. That is a claim, not a measurement. What was measured is smaller: a modest fuel gain and lower carbon monoxide at idle. The deeper model lives on Study; the hardware path is on Device.
What was measured
One clear distinction: measured results versus inventor claims. Full compare on Sources and Study.
Embry-Riddle Aeronautical University
Master’s (M.S.) thesis · 2026 · manufacturer supplied road-test hardware
About 9–10% lower fuel use on a small generator, and carbon monoxide (CO) down about 34% in an idle test (carbon dioxide, oxygen and hydrocarbons did not change significantly). Measured — lead with this.
Inventor claims
Labeled as such
Large fuel cuts (often framed as 90% or more), near-zero carbon monoxide (CO) and hydrocarbons (HC), and high chamber temps are inventor claims. This site does not lead with those figures. Use the meter first, then the model for why results might scale — and push for fuller dyno and isotope testing.
Measured vs claimed
Solid bars are measured in the university thesis. Hatched bars are numbers the seller reports on Thunderstorm.tech; we have not seen the full reports. Pick a result:
Change in carbon monoxide (CO) in the exhaust
Bars show the size of the drop on a 0–100% scale.
Embry-Riddle thesis, idle test (2026)
MeasuredBar: % lower, 0–100% scale
Measured: about 34% lower (average)
Test conditions and source
Troy-Bilt 5550 generator at idle inside a trailer, five-gas analyzer, 10 runs each way.
Hall, W. (2026), Embry-Riddle Aeronautical University master’s thesis
Dartford, UK (June 28, 2024)
Seller claimBar: % lower, 0–100% scale
Seller claim: 97.1% lower, as reported
Test conditions and source
Run by Element, a testing company, per Thunderstorm.tech. Engine type not given. We have not seen the full report or who paid.
Clearwater, Florida (June 24, 2024)
Seller claimBar: % lower, 0–100% scale
Seller claim: 98.00% lower, as reported
Test conditions and source
Base-load generator at 60 hertz (Hz) under a 2,400 watt (W) load. Tester not named. Units are unclear in the page’s table.
Star Cement, United Arab Emirates (UAE) (December 5–6, 2024)
Seller claimBar: % lower, 0–100% scale
Seller claim: unchanged at 0.10%, as reported
Test conditions and source
Honda 5 kilowatt (kW) gasoline engine. Described as run in partnership with Strike Foundation (UK).
What you’ll find here
The landing is the hook. Teaching, walkthroughs, and sources stay on their own pages — no textbook dump up front.
- Study
Short lessons and diagrams — start with the basics, go deeper when you want.
- Device
A plain walkthrough of the Thunderstorm Generator path: air, water, plasmoids, engine.
- Sources
What was measured vs what inventors claim — with links you can check yourself.
- Services
Optional install help for kit owners — inquiry only; kits not sold here.
- Connect
Questions, corrections, or just say hello. Independent education site.
Continue learning
Returning? Pick up where you left off. First visit? Skim the problem and solution above, then start Lesson 1 when you’re ready.
Get new explainers by email
About one short email when a new explainer or correction goes up. No ads, no selling your address. Unsubscribe anytime with the link in every email.
