Follow on Instagram
Verified creator

AMP HAUS ⚡

E-Bikes — Builds · Speed Tests · Battery Work

We measure everything.

E-bike builds, speed tests & battery work. Real numbers. No fantasy specs. If a brand prints it on the box, we test it on the bench — and we publish what we actually find. Now with a full digital lab: a 3D garage, an engineering toolbox, and the investigation files, all in one place.

Illustration of an e-bike on a diagnostic platform in the AMP HAUS lab, surrounded by holographic engineering displays
LAB VIEW — ILLUSTRATIONBENCH 01
NOT 5,500W
Advertised5,500W
By the math~960W
Battery fitted720Wh
▶ Follow on Instagram Enter the 3D Garage ↓ Read The GT50 Files ↓
AMP HAUS // BENCH TELEMETRY — GT50Measured data
0VSystem voltage — the GT50's battery platform
0AStock controller ceiling — the number that caps everything
0WCalculated peak power (48V × 20A) — not 5,500W
0WhBattery actually fitted (15Ah) — 1,200Wh was promised
5,500WAdvertised peak power on the GT50
~960WActual peak by the math — 48V × stock 20A controller
$650.04Paid — bought with our own money. Not sponsored.
Scroll — Command Deck

Command Deck

Pick a system. Everything is connected.

SYS.01

The Lab

Investigations, testing, measurements and technical analysis. The GT50 Files live here.

Open the Lab →
 SYS.02

The Garage

An interactive 3D bench. Pull the bike apart, tap a component, read what we found on it.

Enter the Garage →
 SYS.03

The Intelligence

The AI Copilot plus the Engineering Toolbox — calculators that check any spec claim in seconds.

Run the numbers →
 SYS.04

The Network

The community deck: discussion categories that run live on Instagram, the bench-knowledge challenge, and your explorer badges.

Enter the Network →
 SYS.05

The Market

Deals I stand behind — every partner, every code, disclosed on the card. Search and filter included.

Browse the Market →
 SYS.06

Submission Bay

Put It On The Bench: brands and riders submit products for possible independent testing.

Submit a product →
 SYS.07

The Media Room

The GT50 Tapes podcast on Spotify, the investigation videos, and the discussion threads where questions get answered.

Open the Media Room →
 SYS.08

Ask AMP HAUS

A bike problem or a question about a test? The conversation lives on Instagram — DM is open.

Ask a question →
⚡ BULLETIN
A mystery bike is on the bench.

You've seen this bike before. It's coming back changed — two batteries, one question: how far does it really go now? Full measured exposé coming soon. No fantasy specs.

Follow @amphaus.garage for the drop
You're in. Watch your inbox to confirm.

Email the second it drops — not just Instagram.

SYS.01 · The Lab — Investigation Center

The GT50 Files

Case File 001 · Askgo GT50 · Published

Advertised vs. measured. We bought the Askgo GT50 with our own money and measured everything. Not sponsored.

Bookmarks and reading progress live in the Explorer Mission Log (The Network) — for this visit.
Spec
Advertised
Measured
Peak Power
5,500W
700W nominal motor; about 960W peak by the math (48V × the stock 20A controller) Not 5,500W
Top Speed
40 MPH
About 30 MPH
Battery
25Ah confirmed by the seller in writing (1,200Wh)
15Ah fitted (720Wh) — 60% of what we paid for, runs out about 40% sooner Short-changed
Brakes
“Dual hydraulic”
Rear hydraulic, front mechanical disc (the cable is visible in our video) Claim was false
Steering
Works
Stem clamp metal torn, bolts won’t tighten, not rideable Not rideable
What IS Real
—
The display, and the Shimano Tourney TZ 8-speed kit ✓ Real
5.7×
The headline number is 5.7 times the physical ceiling.5,500W claimed ÷ ~960W possible (48V × 20A) — the same check the Claim Checker in the Toolbox runs on any listing, in seconds.
Run it yourself ↓
Case
001 — ASKGO GT50
Status
● PUBLISHED
Claims on the table
6 TESTED
Evidence logged
EV-01 → EV-09 · 9 ITEMS
Headline gap
5.7× OVERSTATED
Your file progress
0 / 7 SECTIONS
Exploded inspection deck — The Garage ↓

Power Gauge

0W 5,500W ADV.
0W
Calculated peak · 48V × 20A

The needle stops where the physics stops. A 20A controller on a 48V system cannot pass more than about 960W — the advertised 5,500W would need roughly 115A.

Battery — Promised vs Fitted

Promised (25Ah)1,200Wh
Fitted (15Ah)720Wh

60% of the battery we paid for. Watt-hours are the honest unit: volts × amp-hours. Run the same check on any bike in the Toolbox.

Investigation Highlights

· Peak power claim: 5,500W advertised, ~960W by the math
· Top speed: 40 MPH advertised, about 30 MPH measured
· Brakes: “dual hydraulic” advertised — front is mechanical
· Steering: stem clamp torn — not rideable
· Genuinely good: the display and the Shimano Tourney TZ 8-speed kit

Every future investigation follows this same framework: the claim, the measurement, the method, the verdict. Nothing published without numbers behind it.

Peak power — the gap, visualised

Advertised5,500W
By the math (48V × 20A)~960W

Claimed vs Verified — The Whole Gap On One Chart

Every quantified claim in Case 001, drawn to the same rule: the grey bar is what was sold, the lit bar is what the bench found, and the chip is the share of the claim the hardware actually delivered. Hover, tap or tab through the rows for the finding behind each bar — or sort the file by how big the gap is.

PEAK POWER — 960W DELIVERED OF 5,500W CLAIMED · 17% OF THE CLAIM

Bench Specification — The Machine As It Actually Arrived

The complete fitted specification from the Case 001 bench record — not the listing. Filter it down to just the discrepancies, or just the parts that checked out, and select any row to read why it matters.

System
Bench finding
Status
Peak Power
~960W ceiling by the math (48V × 20A) — 5,500W advertised
Not 5,500W
Motor
700W nominal — the sustained rating on the file
✓ On record
Controller
20A stock — the ceiling-setter for the whole system
✓ On record
System Voltage
48V — the platform all the math runs on
✓ On record
Battery
15Ah fitted · 720Wh — 25Ah · 1,200Wh promised and confirmed in writing
Short-changed
Front Brake
Mechanical disc, cable-actuated — sold as “dual hydraulic”
Claim was false
Rear Brake
Hydraulic — the half of the brake claim that held up
✓ Real
Top Speed
About 30 MPH measured — 40 MPH advertised
Overstated
Steering Stem
Clamp metal torn, bolts won’t tighten — not rideable
Not rideable
Drivetrain
Shimano Tourney TZ 8-speed — as described
✓ Real
Display
Fitted and as described
✓ Real
Purchase
$650.04 paid — our money, not sponsored
✓ On record

12 SYSTEMS ON THE BENCH RECORD — SELECT A ROW FOR THE FINDING BEHIND IT

Evidence Locker — Case 001

Every finding in this file traces back to a piece of evidence — a document, an inspection, or footage you can watch yourself. No finding stands on our say-so alone.

EV-01Document

The listing, as sold

The claims were recorded exactly as advertised before the bike was touched: 5,500W peak, 40 MPH, “dual hydraulic” brakes. That record is the left-hand column of the table above — the benchmark every measurement is judged against.

EV-02In writing

Seller confirmation — 25Ah

The battery size wasn't a guess from a spec sheet: the seller confirmed 25Ah (1,200Wh) in writing. That's the number we paid for — and the number the fitted pack is measured against.

EV-03Receipt

$650.04 — our money

Bought outright, not sponsored, not a review unit. Nobody sent this bike and nobody can pull the findings. The receipt is why the verdict column can say whatever the numbers say.

EV-04Inspection

The fitted pack — 15Ah

The pack actually on the bike is a 15Ah unit. The math is the evidence: 48V × 15Ah = 720Wh, which is 60% of the promised 1,200Wh — and why it runs out about 40% sooner.

EV-05Video

The front brake cable

“Dual hydraulic” was the claim; the mechanical cable on the front brake is visible on camera in Part One. Hydraulic systems have no cable to film. Watch it in Part One →

EV-06Inspection

The torn stem clamp

Hands-on, not on paper: the stem clamp metal is torn and the bolts won't tighten. Steering that can't clamp is why the file's verdict is not rideable — full stop.

EV-07Video

Part Two — the update

The follow-up inspection, published as the file grew: what the deeper look found, and the thread where reader questions get answered on the record. Watch Part Two →

EV-08Verified real

What checked out

Evidence cuts both ways: the display and the Shimano Tourney TZ 8-speed kit were inspected and are exactly what they claim to be. The file credits them in full — that's what makes the rest of the file mean something.

EV-09The math

The ceiling calculation

48V × the stock 20A controller = ~960W maximum. The advertised 5,500W would need roughly 115A — through a 20A controller. Reproduce it yourself in the Toolbox; arithmetic is the one witness that can't be argued with.

Evidence Timeline — How Case 001 Was Built

The case in the order the Lab worked it — claim first, verdict last, nothing published before the evidence was in. The file is kept by evidence rather than by calendar: each stage below is pinned to the evidence items it produced, and the two video exhibits at the end are the footage the findings stand on.

  1. 1
    Stage 01 · Before the bike was touched

    The claims are frozen

    The listing is recorded exactly as sold — 5,500W peak, 40 MPH, “dual hydraulic” brakes — so nothing can quietly move later. That record becomes the left-hand column of the claims table. → EV-01

  2. 2
    Stage 02 · Before purchase

    The seller confirms it — in writing

    The battery is not taken from a spec sheet: the seller confirms 25Ah (1,200Wh) in writing. That is the number that gets paid for — and the benchmark the fitted pack is later measured against. → EV-02

  3. 3
    Stage 03 · Purchase

    Bought with our own money

    $650.04 paid — not sponsored, not a review unit, nobody able to pull the findings. The receipt is what lets the verdict column say whatever the numbers say. → EV-03

  4. 4
    Stage 04 · Bench inspection

    The hardware tells a different story

    Hands on the machine: the fitted pack is a 15Ah unit (720Wh), not 25Ah. A mechanical cable runs to the front brake that was sold as hydraulic — and it is visible on camera. And the stem clamp metal is torn, its bolts refusing to tighten: the bike is not rideable. → EV-04 · EV-05 · EV-06

  5. 5
    Stage 05 · The arithmetic

    The ceiling calculation

    48V × the stock 20A controller = ~960W maximum. The advertised 5,500W would need roughly 115A through a controller built for 20A. The headline number fails before a wheel ever turns. → EV-09

  6. 6
    Stage 06 · Publication

    Published in two parts — credit where due

    Part One publishes the inspection; Part Two follows up as the file grows, with reader questions answered on the record. And the evidence cuts both ways: the display and the Shimano Tourney TZ 8-speed kit checked out exactly as claimed, and the file credits them in full. → EV-07 · EV-08

Video Exhibit A · EV-05

Part One — The Inspection

The primary footage. Watch for the front brake cable: the “dual hydraulic” claim dies on camera, because hydraulic systems have no cable to film.

Watch on Instagram →
Video Exhibit B · EV-07

Part Two — The Update

The follow-up inspection as the file grew — and the thread where reader questions get answered on the record, in the comments.

Watch on Instagram →
Testing methodology

Every AMP HAUS investigation runs the same way:

  • Record the claim exactly as sold — listing text, seller messages, box copy.
  • Do the ceiling math first — volts × controller amps caps the watts; volts × amp-hours gives the real battery size in watt-hours.
  • Inspect the hardware — what's actually fitted, not what the spec sheet says: brake type, controller rating, battery label, fasteners and clamps.
  • Publish both columns — advertised and measured side by side, with the honest parts credited too.
Findings & conclusions

The GT50's problem was never that it's a weak bike — a 700W nominal motor moving at about 30 MPH is a normal machine. The problem is the gap between the box and the bench: 5,500W that physics won't allow, a 25Ah battery that arrived as 15Ah, “dual hydraulic” brakes with a cable on the front, and a steering stem that tore before it could be ridden.

Verdict: judge this bike by the measured column — and judge every listing the same way. That's the whole point of The Lab.

Technical note — nominal vs peak vs fantasy

Three different numbers get printed as if they were one:

  • Nominal (700W here) — what the motor can sustain all day without cooking itself. This is the honest working number.
  • Peak (~960W here) — the short-burst maximum, and it is not chosen by the motor or the marketing department: the controller's current limit sets it. 48V × 20A ≈ 960W, full stop.
  • Fantasy (5,500W here) — a number with no hardware path to exist. Reaching it at 48V would take ~115A through a controller built for 20A. When a listing's watts exceed volts × controller amps, you're not looking at a fast bike — you're looking at a typo with a price tag.
Technical note — how the Lab checks a battery

Amp-hours on a listing mean nothing until volts join them. The check the Lab runs on every pack:

  • Read what's fitted, not what's promised — the pack's own label and configuration, on the bike, before any ride.
  • Convert to watt-hours — Wh = V × Ah. The GT50 promised 48V × 25Ah = 1,200Wh; the fitted pack is 48V × 15Ah = 720Wh.
  • Judge range from watt-hours — at a typical 20–30 Wh per mile, a 480Wh shortfall is range you paid for and never received. That's why this file reports batteries in Wh first.

Battery Comparison — Promised, Fitted, and the Parallel Question

The two bars on record, plus the question the Mystery Bike exposé exists to answer: what does a second pack in parallel actually add? Set the second pack's size below — the combined bar, the watt-hours and the range estimate update with your numbers, using the same 25 Wh-per-mile planning rate as the Range Estimator. The exposé will publish the Mystery Bike's real fitted packs and measured range; this panel is the math, not the result.

Promised — 25Ah1,200Wh
Fitted — 15Ah (verified)720Wh
Fitted + second pack in parallel1,440Wh

720Wh fitted + 720Wh second pack = 1,440Wh · est. range ~58 miles at 25 Wh/mile (estimate, not a measurement)

Safety, always: parallel packs must match in voltage and charge state and be joined with proper fused connectors — mismatched packs can dump current into each other. The configuration math lives in the Battery Configuration Explorer; pack building belongs to people with the right gear.

Battery Analysis — What The Missing 480Wh Costs On The Road

The shortfall, translated into the only unit a rider feels: miles. These are planning estimates, not measured range — both bars divide the pack's watt-hours by a consumption rate you set, inside the same 20–30 Wh-per-mile band the Lab's battery method uses. The point that never moves: the fitted pack holds 60% of the promised energy, so at any riding style it delivers about 60% of the promised miles.

Promised — 1,200Wh~48 miles
Fitted — 720Wh (verified)~29 miles

At 25 Wh/mile: promised ~48 miles · fitted ~29 miles · shortfall ~19 miles of range paid for and never received (estimates, not measurements)

Technical note — why the Lab judges batteries in watt-hours

Amp-hours are half a number: the same 15Ah means 540Wh at 36V, 720Wh at 48V and 780Wh at 52V. Watt-hours (V × Ah) are the energy actually on board, and range is just energy divided by appetite. Run the endpoints of the planning band on this case and the shape of the loss stays identical:

  • Gentle riding, 20 Wh/mile — promised ~60 miles, fitted ~36 miles.
  • Hard riding, 30 Wh/mile — promised ~40 miles, fitted ~24 miles.
  • Either way, 40% of the range is missing — because 40% of the energy is. That is why this file reports batteries in Wh first, and why the exposé on the Mystery Bike's dual-battery setup will be judged in watt-hours and measured miles, not brochure numbers.

Case Progress — Read The File Like The Lab Filed It

Work through all seven parts of Case 001 and the case closes — the sections tick off as you actually read them (the two case notes count when you open them). Progress is a learning trail for this visit only.

  • ○The claims table — advertised vs measured, all six rows
  • ○The 5.7× verdict — why the headline number fails
  • ○The evidence locker — EV-01 through EV-09
  • ○The evidence timeline — how the case was built, stage by stage
  • ○Testing methodology — open the case note
  • ○Findings & conclusions — open the case note
  • ○Battery analysis — what the missing 480Wh costs on the road

0 / 7 sections read — the case is open.

Investigation Report — Take The File With You

The complete Case 001 report — claims, measurements, evidence index, method and verdict — generated in your browser, from the findings on this page, as a plain-text file. Nothing is uploaded and nothing is added: if it isn't in the file above, it isn't in the report.

Got evidence of your own? Public evidence uploads arrive with the Lab's backend phase — they are not faked here in the meantime. Until then, evidence reaches the bench the working way: through the Submission Bay and an Instagram DM.

Test Queue

What's on the bench, what's published, what's next. Only items approved for public display appear here.

3 Public entries 1 Published 1 On the bench 1 Open

Showing 3 of 3 entries

Published
The GT50 Files — Case 001 Askgo GT50: full advertised-vs-measured investigation. You're reading it.
LifecycleOpenOn the benchPublished

Filed above on this page: the claims table, Evidence Locker EV-01 → EV-09 and the verdict. Watch and listen in The Media Room.

On the bench
Mystery Bike — Dual Battery Exposé A familiar bike, back with two batteries. One question: how far does it really go now? Full measured exposé coming soon.
LifecycleOpenOn the benchPublished

Announced in the Bulletin at the top of this page, with a teaser card in The Media Room. No results are published until the exposé is.

Open
Next case — could be yours Brands, sellers and riders can put a product on the bench via the Submission Bay. Submission never guarantees testing or a good word — that's the deal.
LifecycleOpenOn the benchPublished

The slot fills through the Submission Bay, where the six-step track — Received → Published — is set out in full.

Published

The investigation is complete and live. Advertised vs measured is public — claims, evidence and verdict — whatever the numbers said. Case 001, above, is the standing example.

On the bench

Approved for public display and being worked now: measured, filmed and documented on the bench. Nothing from it is published yet, and no result is claimed until the exposé goes live.

Open

A public slot with no case assigned yet. It fills only through the Submission Bay — and a submission never guarantees testing, coverage, or a positive review.

How this board works: AMP HAUS controls what reaches public display — nothing appears here just because it was submitted. Live queue management — submission tracking, accounts and status lookups — arrives with the backend phase; until then, this board is updated by AMP HAUS when a case's public status changes.

▶ Watch Part One ▶ Watch Part Two — the update

The comments are where the questions get answered:

Join the discussion on Part Two Part One discussion

SYS.02 · The Garage

The 3D Bench

WebGL · Drag to orbit · Scroll to zoom

A detailed 3D bench model of the GT50 — real proportions, real materials, rendered live in a working lab — built for one job: tap a component and read what we actually found on it. Drag to orbit, use the exploded slider to pull it apart, and on a phone you can enable motion controls and tilt to look around. Sensor data is used locally, on your device only — nothing is collected or sent anywhere.

BENCH CAM 01 — GT50 · REALTIME 3DDRAG · ORBIT   SCROLL · ZOOM   TAP · INSPECT
Select a component

The Bench Model

Ten inspectable systems carry the GT50 story — the five headline findings plus wiring, drivetrain, suspension, frame and display electronics. Pick one — on the model, on a floating label, or from the list — and the findings panel fills in, with verified findings and general knowledge labelled as what they are.

Full investigation in The Lab →

Pull the bike apart along its guide lines — every system travels on a clean axis with its finding labelled, the way a real engineering exploded diagram reads.

Tilt your phone to look around the bench. On iPhone, tapping Enable asks for motion permission — that's the browser's rule, and the data never leaves your phone.

Keyboard works too: arrow keys orbit the model when the bench is on screen. Reduced-motion settings are respected automatically.

The camera glides to each workstation angle — the same smooth move a diagnostic rig makes.

Mute

Sound never starts on its own — it only exists after you switch it on here or in the Console. Ambience is a low lab hum generated live in your browser; nothing is recorded or sent anywhere.

Exploded View · Component Inspector

Nine systems. Pull every one apart.

A professional-style inspection deck for the whole machine. Select a system and the schematic separates into its layers — drag the explode slider, read the engineering explanation, and check exactly which claims AMP HAUS verified on the GT50 and which panels are general engineering knowledge. Nothing here invents a measurement: verified findings are labelled verified, everything else is labelled general knowledge.

Select a system

Component Inspector

Nine systems, one machine. Pick a layer set above and this panel fills with how it works, what it does, and — where the GT50 file has one — what the bench actually found.

    INSPECTED 0 / 9

    Inspection progress is for this visit only — a learning trail, not a profile. Finish all nine and you have read the whole machine the way the Lab does.

    SYS.03 · The Intelligence

    Engineering Toolbox

    Real math · Runs in your browser

    The same arithmetic behind every AMP HAUS verdict. Every calculator is live — change a number and the answer updates. And when a question needs more than math, the AMP HAUS Copilot (bottom-right, always on) answers from the investigation files — it separates verified AMP HAUS findings, manufacturer claims, and general knowledge, and it never invents a measurement. You can even attach a spec sheet (.txt) in the chat and it will check the claims against the math.

    Power Ceiling Calculator

    WATTS = VOLTS × AMPS

    The most lied-about number in e-bikes. A controller can only pass so many amps — that caps the watts, whatever the box says.

    Peak power ceiling960 W

    Pre-loaded with the GT50's real numbers: 48V × 20A ≈ 960W — against a 5,500W advertised claim.

    Claim Checker

    ADVERTISED W vs V × A

    Paste any listing's headline wattage next to its battery and controller numbers. If the claim beats the ceiling, the listing is fantasy.

    Verdict—

    The math: 48V × 20A = 960W ceiling. The claim is 5.7× the ceiling.

    Battery Energy Calculator

    WATT-HOURS = VOLTS × AMP-HOURS

    Amp-hours alone mean nothing without volts. Watt-hours are the honest size of a battery — and the number range is built from.

    Battery energy720 Wh

    Pre-loaded with the battery actually fitted to the GT50: 48V × 15Ah = 720Wh. The promised 25Ah would have been 1,200Wh.

    Range Estimator

    RANGE ≈ Wh ÷ Wh-PER-MILE

    An estimate, honestly labelled. Real range swings with rider weight, hills, wind, tyre pressure and how hard the motor works.

    Estimated range—

    Typical e-bike use runs roughly 20–30 Wh per mile. This is a planning estimate — not a measurement, and not a promise.

    Controller Current Calculator

    AMPS = WATTS ÷ VOLTS

    Working the claim backwards: how many amps would a controller actually need to deliver the advertised watts?

    Current required—

    The GT50's stock controller is rated 20A. If a claim needs far more current than the fitted controller, the claim fails on paper before anyone rides.

    Battery Configuration Explorer

    SERIES = +VOLTS · PARALLEL = +AMP-HOURS

    How packs are built. Series strings raise voltage; parallel groups raise capacity. Total energy multiplies either way. Safety: never mix cells of different age, brand or charge state, and leave pack building to people with the right gear — a shorted lithium pack is a fire, not a project.

    Diagram shows each pack as a cell block: S across, P deep.

    Combined pack48V · 15Ah · 720Wh

    This is the math behind dual-battery setups like the Mystery Bike's: a second pack in parallel adds amp-hours (and watt-hours), not voltage.

    The Claim Graph — power vs current

    One graph that ends most spec arguments. The line is physics: watts = volts × amps at the pack voltage below. The amber marker is where the fitted controller stops. The red marker is where an advertised claim would have to live — hover or tap the graph to read any point on the line.

    Pre-loaded with the GT50's real hardware: a 48V pack and a stock 20A controller ceiling out at ~960W, while the 5,500W claim needs ~115A. Change any number — the graph, and the verdict, move with the math.

    SYS.04 · The Network

    The Community Deck

    Live on Instagram · @amphaus.garage

    AMP HAUS doesn't run a separate sign-up or store member accounts — the community lives where it already thrives: on Instagram. Every category below is a real discussion lane; tap one and it takes you straight to the posts and comments where that conversation happens. No new password, no new profile, nothing to leak.

    General E-Bike Discussion

    Anything on two wheels and a battery. Start where the whole garage starts.

    Join on Instagram →

    Technical Help

    Stuck on a fix? DM the question — brakes, batteries, weird noises and all.

    Ask via DM →

    Battery Technology

    Watt-hours, parallel packs, and why the GT50's 15Ah battery matters.

    Open the thread →

    Motors & Controllers

    The 20A ceiling, nominal vs peak, and what controllers really decide.

    Open Part One →

    Performance & Speed

    Advertised 40 MPH, measured about 30. Speed talk with receipts.

    See the evidence →

    Builds & Modifications

    Dual-battery conversions and bench builds — the Mystery Bike is coming.

    Follow the build →

    Product Reviews

    Measured-first reviews and the deals list — every affiliate link disclosed.

    Browse the Market →

    Manufacturer Claims

    5,500W that physics won't allow. Learn to spot the gap in any listing.

    Read the case file →

    AMP HAUS Investigations

    The GT50 Files and every case that follows — claim, measurement, verdict.

    Open The Lab →

    Deals & Recommendations

    ENGWE, Magician (GLEN100), Kingbull, TideWe — the short list, with #ad on it.

    See the deals →

    Community Projects

    Got a bike or product that deserves the bench treatment? Put it forward.

    Submission Bay →

    Ask AMP HAUS — Direct

    The fastest lane of all: a direct message. Questions get answered where the conversation lives.

    Open DM →

    Bench Knowledge Challenge

    Four questions, all answerable from the GT50 Files on this page. No trick questions — just the numbers.

    Explorer Mission Log

    Badges unlock as you work the systems on this page — explore the Lab, pull the 3D bike apart, bust a claim, finish the challenge, close the case, build the pack. Genuine exploration, no popups, no fake urgency.

    Lab Analyst Bench Tech Claim Buster Numbers Don't Lie Case Closed Pack Builder Full Sweep

    Weekly Challenge

    Loading this week's challenge…

    The challenge rotates on a fixed four-week cycle set by the calendar (ISO) week — the same assignment for every visitor that week, no random draw and no countdown pressure. It completes only when the page itself records the matching badge above: there is no claim button, and clicking this text earns nothing.

    Bookmarks

    • Nothing bookmarked yet this visit — the bookmark button sits on the GT50 case file and the Media Room cards.

    Recently Viewed

    • Systems you open this visit stack up here, newest first.

    Badges, bookmarks, recents and challenge status live for this visit only — this page doesn't store profiles or track you between visits. The full achievement list and this visit's counted progress sit directly below, in Achievements & Session Progress.

    Achievements & Session Progress

    Every badge this page offers, what genuinely earns it, and where you stand right now. There are no points, no levels, no prizes and no streaks — progression here rewards exploration and learning only, and a locked badge simply means the action hasn't happened yet this visit.

    0 / 7 EARNED
    Systems visited0 / 8Sections that scrolled into view this visit — the same count Mission Control shows.
    Lab case progress0 / 7Parts of the GT50 case file the Case Progress panel has actually marked read.
    Systems inspected0 / 9Component Inspector systems opened in the Garage — read from its own counter.
    Calculators used0 / 7None yet — change any Toolbox number and it counts here.
    Badges earned0 / 7Of the seven achievements listed below — counted from the badge row itself.

    Lab Analyst

    LOCKED

    Open The Lab — Case 001's section registers as visited when it scrolls into view, and this badge lights with it.

    NOT YET — THE LAB HASN'T REGISTERED AS VISITED

    Bench Tech

    LOCKED

    Inspect all nine systems in the Garage's Component Inspector — battery through display — until its counter reads INSPECTED 9 / 9.

    0 / 9 SYSTEMS INSPECTED

    Claim Buster

    LOCKED

    Run the Claim Checker in the Engineering Toolbox to a verdict — change its advertised watts, volts or amps yourself and let V × A rule on the claim.

    NOT YET — NO CLAIM CHECKER VERDICT RUN BY YOU

    Numbers Don't Lie

    LOCKED

    Score a perfect 4 / 4 in the Bench Knowledge Challenge, right here in the Network deck. Every answer is in the GT50 Files — it's open-book.

    NOT YET — A PERFECT 4 / 4 IS THE BAR

    Case Closed

    LOCKED

    Read all seven tracked parts of the GT50 case file — claims table, verdict, evidence locker, evidence timeline, methodology, findings and battery analysis — to 7 / 7 on the Case Progress panel.

    0 / 7 CASE SECTIONS READ

    Pack Builder

    LOCKED

    Get all three wiring calls right in Build The Pack, the engineering puzzle below — series adds volts, parallel adds amp-hours.

    NOT YET — 3 / 3 CORRECT CALLS REQUIRED

    Full Sweep

    LOCKED

    Visit all seven core systems: The Lab, The Garage, The Toolbox, The Network, The Market, the Submission Bay and the Media Room. Mission Control lights each one as it registers.

    0 / 7 CORE SYSTEMS VISITED

    Session only, stated plainly: every count on this panel is computed live from actions taken on this page during this visit. Reload or close the page and badges, progress, bookmarks and challenge status all reset to zero — this static page keeps no account, no saved profile and no cross-visit progress, and nothing here converts into rewards, discounts or prizes.

    Engineering Puzzle — Build The Pack

    Three wiring calls, all real math — the same series/parallel decisions behind every dual-battery setup, including the one on the bench right now. Get all three right to earn the Pack Builder badge.

    SYS.05 · The Market

    Deals I Stand Behind

    #ad · Disclosed on every card

    The short list. Every card below is an affiliate link — disclosed on the card, every time. Search it, filter it, copy the code — the partners and the terms haven't changed.

    #ad

    Read this first — how the Market works

    Every link in the Market is an affiliate link. If you buy through one, I may earn a commission at no extra cost to you — the same disclosure printed on every card below, and it covers the comparison table and the Linktree list too. A card here means a disclosed partnership, not a bench measurement: the bike with a full AMP HAUS measurement file is the GT50, and its file lives in the Lab, not the shop window. Where a partner hasn't been ridden yet, the card says so in plain words — see Magician.

    Find your lane

    Four partners, four different jobs. Pick the one that sounds like you and the list below filters itself to the match — every reason given is a fact printed on the cards.

    Showing 4 of 4 partners

    ENGWE

    Folding e-bikes, the travel lane.

    E-BikesFoldingTravel

    Best for: riders who need the bike to fold — the travel lane is the whole point of this card.

    Shop ENGWE →

    #ad — affiliate link, I may earn a commission if you buy, at no extra cost to you

    Magician E-Bikes

    Only code on the list

    Code GLEN100 takes $100 off.

    E-BikesCode · GLEN100

    Best for: shoppers who want the code first — GLEN100, $100 off, ready to copy right below.

    GLEN100

    Straight talk: I have not ridden one yet, so this is a code, not a review.

    Shop Magician E-Bikes →

    #ad — affiliate link, I may earn a commission if you buy, at no extra cost to you

    Kingbull Bike

    E-bikes.

    E-Bikes

    Best for: riders browsing e-bikes — line it up against ENGWE and Magician in the side-by-side under the cards.

    Shop Kingbull Bike →

    #ad — affiliate link, I may earn a commission if you buy, at no extra cost to you

    TideWe

    Heated gear for cold rides.

    Riding GearHeatedCold rides

    Best for: cold rides — this is the gear card, not a bike. Already have a bike and winter coming? Start here.

    Shop TideWe →

    #ad — affiliate link, I may earn a commission if you buy, at no extra cost to you

    Side by side — only what's on the cards

    No invented specs, no prices, no ratings. This table repeats exactly what each card above states, so the four partners can be compared at a glance.

    Partner
    What its card says
    Code
    Lane
    Link
    ENGWE
    Folding e-bikes, the travel lane.
    None on its card
    E-Bikes
    Shop →
    Magician E-Bikes
    E-bikes with a code — and the honest note on its card: not ridden yet, so the code is not a review.
    GLEN100 — $100 off
    E-Bikes
    Shop →
    Kingbull Bike
    E-bikes.
    None on its card
    E-Bikes
    Shop →
    TideWe
    Heated gear for cold rides.
    None on its card
    Riding Gear
    Shop →

    #ad — every link in this table is the same affiliate link as on the cards above, with the same disclosure: I may earn a commission if you buy, at no extra cost to you. Current prices and terms live on the partners' own sites — nothing here quotes them, so nothing here can go stale.

    Before you buy anything — the AMP HAUS way

    The Market is four cards long on purpose. Run these three steps and you'll land on the right one — or catch a bad claim before it costs you.

    01Pick the lane first

    Folding, a code, e-bikes in general, or cold-weather gear — use Find your lane above. Category first, brand second.

    02Run the claim math

    See a headline wattage number on any listing? Check it before you pay: the real ceiling is volts × controller amps. Open the Claim Checker →

    03Take the code with you

    If Magician is your pick, copy GLEN100 from its card above — the copy button puts it on your clipboard, $100 off at their checkout.

    All deals in one place — Linktree

    Every live deal, code and link — kept up to date in one list.

    Open Linktree →

    SYS.06 · The Submission Bay

    Put It On The Bench

    For brands, sellers & riders

    Manufacturers, brands, sellers and riders: submit a product for possible independent AMP HAUS testing. Fill in the form and it builds your submission packet — copy it and send it to AMP HAUS in an Instagram DM, which is where submissions are received and reviewed. Submission does not guarantee testing, coverage, or a positive review. What gets published is what gets measured. Editorial independence is not negotiable. One more thing, stated plainly: this page stores nothing. The form assembles your packet in this tab only — nothing is sent, saved, or queued until you paste it into the DM yourself.

    STEP 1

    Build the packet here

    Fill in the form below. Only name, email and product name are required to build — everything else you add is one less question at review. The completeness meter under the form tracks you as you go.

    STEP 2

    You send it — by DM

    Copy the packet, open the Instagram DM, paste it in, and attach your images and documents there — the form only notes their file names. Until you paste it, no submission exists.

    STEP 3

    The track takes over

    From the moment it lands in the DM, your packet runs the six-step track in the panel beside the form — Received → Published — reviewed in DM order, with a straight Accepted / Declined answer at step 04.

    Who AMP HAUS is dealing with — the name that goes at the top of the packet.
    Where the step-04 answer — Accepted / Declined — and any step-03 questions go. Double-check it; a typo here means silence.
    Riders: leave it blank or put “independent” — brands and sellers are held to exactly the same track.
    The exact model name as it's sold — that's the name any published result would carry.
    The single strongest thing you can add. At step 02 the claims get the ceiling-math check — volts × controller amps — before anything is agreed. Published specs that survive that math are what a strong submission is built on. No specs usually means a step-03 round-trip.
    Copy them word for word from your own listing — don't soften them here. These are the numbers the bench checks first, and if testing goes ahead, they're the left-hand column of the published advertised-vs-measured table.
    What it is, who it's for, and what it costs — plain facts beat marketing copy at review.
    One specific, measurable question — “does it really hold 30 MPH on a 10% grade?” — is worth more than a page of general praise requests. Vague asks are the ones that stall at review.
    Selecting files here only lists their names in the packet — nothing uploads from this page. Attach the actual images in the Instagram DM with the packet.
    Same rule: names are noted here, files are attached in the DM. A real spec sheet or manual is strong review evidence — it pins the claims to paper.
    Loaner vs keeper unit, shipping, timing — anything practical. Note: a loaner changes nothing about the result. Bought or loaned, it's measured either way and published either way.
    Packet completeness0 OF 12 FIELDS
    Name — required Email — required Product — required Specs — review fuel Claims — checked first

    Three fields build a packet. Specs and claims are what review runs on — add them and your packet skips the most common step-03 delay.

    Or start with a DM →

    Brand Collaboration Center

    Work With AMP HAUS

    Brands: DM open on Instagram. Builds, speed tests and battery work — with real numbers published, every time. That's the deal.

    Independent Product Reviews

    Your product, our money-where-our-mouth-is method: bought or loaned, measured either way, published either way.

    Performance Testing

    Power ceilings, battery verification, brake and component checks — the full Lab framework on your spec sheet.

    Technical Content

    Investigation videos and the GT50 Tapes treatment: claims on screen, measurements on the bench.

    Product Demonstrations

    Builds, dual-battery conversions and real-range tests, filmed as they happen — results included, good or bad.

    Two routes in — same track

    There is no separate brand back door. Both routes land in the same place and run the same review:

    1. A
      The fast route — a DM.
      Message @amphaus.garage on Instagram. Best for a first conversation about any of the four formats above.
    2. B
      The structured route — the Submission Bay.
      The form above builds a packet with your published specs and advertised claims — the exact material review runs on — which you then send by DM.
    3. →
      Then the pipeline decides.
      Claims get the ceiling-math check, missing information gets asked for once and clearly, and the answer at step 04 is a straight Accepted / Declined — for brands exactly as for riders.

    The terms, in plain words

    Everything a collaboration runs on is printed here — nothing is negotiated away in the DM:

    • §Real numbers, published either way. What gets published is what gets measured — good result or bad, the finding goes live.
    • §Bought or loaned, measured either way. A loaner unit, a keeper unit, or a unit AMP HAUS buys itself — the method doesn't change with the arrangement.
    • §No pay-to-play, ever. Money does not buy testing, coverage, or a positive review — and no collaboration implies one.
    • §Editorial independence is not negotiable. Brands see results when readers do: advertised vs measured, in The Lab.
    • §Safety before power. Battery and electrical products are inspected before they're powered; if it isn't safe to run, it isn't run — and that gets published too.

    What AMP HAUS measures

    The same Lab framework behind Case 001 — the GT50 investigation — applied to your spec sheet. This is the full list of what a collaboration is actually buying into: measurement, not endorsement.

    PowerPower ceilingsVolts × controller amps — the physical ceiling every advertised watt figure is checked against.
    BatteryBattery verificationVolts × amp-hours = watt-hours, fitted pack against promised pack — capacity is counted, not quoted.
    ComponentsBrake & component checksWhat was sold against what's fitted — the check that found a cable where “dual hydraulic” was claimed.
    On the roadSpeed tests & real rangeBuilds, speed tests and battery work — including dual-battery conversions and real-range tests, filmed as they happen.
    DM on Instagram → Use the Submission Bay ↑

    Reading the fine print that isn't fine print: a collaboration conversation is not a partnership, and being featured in the Market is a disclosed affiliate arrangement — every deal there carries its #ad disclosure. The only results AMP HAUS publishes are measured ones, and the only file with a full measurement record today is Case 001 in The Lab.

    SYS.07 · The Media Room

    Watch & Listen

    Everything AMP HAUS publishes, in one room: the podcast that tells the full GT50 story, the two investigation videos the evidence stands on, and the exposé that's on the bench next. Search it, filter it by format, or take the story route under the cards and run the whole investigation in order.

    Showing 4 of 4 items

    The Podcast

    AMP HAUS — We Measure Everything

    Audio · SpotifySeries — The GT50 TapesThe full story

    The GT50 Tapes. The full story behind the numbers on this page — what was advertised, what we measured, and what happened next.

    🎧 Listen on Spotify
    The audio companion to the written file — every number in it is filed in Case 001 in The Lab ↑

    Investigation Video

    The GT50 Files — Part One

    Video · InstagramEvidence EV-05The inspection

    The investigation opens: the claims on the box meet the bench for the first time.

    Watch for: the front brake cable. It was sold as “dual hydraulic” — hydraulic systems have no cable to film.

    ▶ Watch Part One
    Filed as Evidence EV-05 in The Lab ↑ · the discussion runs on the post itself

    Investigation Video

    The GT50 Files — Part Two

    Video · InstagramEvidence EV-07The update

    The update: what the deeper inspection found — and the thread where the questions get answered.

    Watch for: the follow-up inspection as the file grew — then bring your questions. The comments are where they get answered, on the record.

    ▶ Watch Part Two — the update
    Filed as Evidence EV-07 in The Lab ↑ · Stage 06 of the Evidence Timeline — publication
    Illustration of an e-bike frame fitted with two battery packs under a workshop lamp — teaser artwork for the dual-battery exposé

    Coming Soon

    Mystery Bike — Dual Battery Exposé

    Coming soonDual batteryOn the bench

    A familiar bike, back with two batteries. The full measured exposé is on the bench now — follow along so you don't miss the drop.

    The question it answers: how far does it really go now? When the exposé publishes it will be judged in watt-hours and measured miles — not brochure numbers.

    Follow for the drop →
    Tracked as On the bench in the Test Queue · first teased in the Bulletin ↑

    The GT50 Story — Start To Finish

    New to the case? This is the order the Lab would hand it to you: the written file first, the video evidence it stands on next, and the podcast for the full story — every stop below is a link that already lives on this page.

    01 · READ

    The File

    Case 001 in The Lab: six claims, advertised vs measured, the 5.7× verdict, and the evidence locker — EV-01 through EV-09.

    Open Case 001 — The Lab ↑
    02 · WATCH

    Part One — The Inspection

    The primary footage (EV-05). The front brake cable is visible on camera — the “dual hydraulic” claim doesn't survive it.

    Watch Part One →
    03 · WATCH

    Part Two — The Update

    The follow-up inspection (EV-07), published as the file grew — and the thread where reader questions get answered on the record.

    Watch Part Two →
    04 · LISTEN

    The GT50 Tapes

    The podcast: what was advertised, what we measured, and what happened next — the full story behind the numbers, on Spotify.

    Listen on Spotify →
    And next: the Mystery Bike dual-battery exposé is on the bench in the Test Queue — a familiar bike, back with two batteries, and one question: how far does it really go now? Its teaser card is in the grid above, and the parallel-pack math it will be judged by is in the Lab's Battery Comparison panel.

    Talk To Us

    Ask AMP HAUS

    Got a bike problem or a question about a test? Ask where the conversation lives:

    💬 Ask on Instagram — DM open Follow @amphaus.garage

    Free public AI — answers can be wrong; for anything safety-critical, ask a mechanic or message Glen.