Helium Accumulation Chamber Leak Test Equipment 1.0E-6 Mbar.L/S For EV Battery Pack Hood Method Detection

A finished battery pack cannot be evacuated, cannot be wetted and cannot be inspected seam by seam in a takt. The hood method is the only one of the three that keeps the pack dry, gives you a number, and fits a production shift — which is why this is the system we build for pack-level sealing verification.

Why a pack breaks the usual rules

Every leak testing method assumes something about the part. Vacuum chamber testing assumes the part can be evacuated, and a pack cannot — the cells, the structural foam and the seals are not designed for external vacuum, and pulling one down will damage it or, worse, deform it in a way that creates the leak you were testing for.

Water immersion assumes you can dry the part afterwards. On a high-voltage assembly, that assumption is unacceptable, and it leaves no record you can show a customer.

Air decay assumes a small, stiff, temperature-stable volume. A pack is the opposite: tens of litres, a lot of thermal mass, and an acceptance criterion that is usually far below what air can resolve in a shift let alone a takt.

The hood method sidesteps all three. The pack stays at ambient pressure, stays dry, and the measurement is a tracer gas concentration rather than a pressure difference — so the material and the volume stop fighting the test.

How it works, and where the sensitivity comes from

The pack goes into a sealed hood of known volume and is charged with a helium mixture through its service port. Anything that escapes accumulates in the hood. A circulation fan mixes the hood atmosphere so there are no dead zones, and a mass spectrometer samples that atmosphere and measures the rise over a fixed window.

The key point, and the one that decides your cycle time: sensitivity is governed by accumulation time, not by the analyser alone. Doubling the dwell roughly doubles the signal. There is always a trade between the two, and it should be your trade — we will tell you the shortest dwell that honestly meets your criterion instead of quoting a sensitivity number in isolation.

Two practical details make the difference between a stable result and an argument. First, the hood atmosphere is monitored between cycles: a workshop where a cylinder has been venting will otherwise drift the readings all morning. Second, the hood and the pack are both recovered, because on a large volume the pack itself holds most of the gas.

Design choices and the reasons

Hood volume kept small relative to the pack

A hood that is much larger than the pack dilutes the signal and buys you nothing. We size the hood to the pack envelope with enough clearance for loading, not to a round number.

Mixing, not just waiting

Helium is light and will stratify. Without a fan and a baffle arrangement the sample point reads whatever is local to it. Our hoods circulate and were checked for dead zones during commissioning, not assumed.

Background monitoring between cycles

Ambient helium compensation is not a nice-to-have on a pack line; it is the difference between a usable limit and a permanent argument with your quality team.

Recovery from hood and pack

Both are evacuated back through the recovery loop after the measurement. Without that, a large-volume pack vents its whole charge every cycle and the running cost quietly becomes the biggest number in the business case.

A recipe per pack, not a recipe per machine

Pack volume, charge pressure and dwell all change with the product. Recipes are selected by barcode so nobody edits parameters at the station.

How a cycle runs

• Load and seal — pack into the hood, lid closed and sealed, barcode scanned, service port connected.
• Background check — the hood atmosphere is sampled before charging; if ambient helium is too high, the cycle waits rather than producing a false fail.
• Charge the pack — helium mixture to the agreed pressure through the service port.
• Accumulate — fan circulation for the recipe dwell; the analyser logs the concentration rise.
• Measure and convert — the rise is integrated and converted to a leak rate in mbar·L/s, then compared with the limit.
• Recover and release — hood and pack recovered, lid opened, pack released with a verdict and a stored record.

Technical parameters

What it finds

Setting the limit, and tying it to IP

Pack customers usually arrive with an IP67 or IP6K9K requirement and no leak rate, and expect a conversion table. There is no universal one, because it depends on pack volume and on the pressure profile the standard applies. What works is an empirical correlation on your own packs:

• Test known-good packs and record the spread. That establishes what normal looks like on your line.
• Introduce a calibrated leak at a known location and confirm the hood sees it at the dwell you plan to run.
• Set the limit with margin — typically one third of the specified rate — so a marginal pack is caught before the vehicle plant does.
• Revalidate after a design change. A new pack volume changes the conversion; do not inherit the old limit.

Compared with the alternatives

If the pack design allows evacuation, a vacuum chamber is faster and more sensitive and you should buy one. It is the sealed pack with a large internal volume that makes the hood the practical answer — and that describes almost every finished pack we have been asked about.

Delivery, installation and the first month

• Factory acceptance test with a calibrated leak plate, video record, witnessed FAT on request.
• Hood sized to your pack envelope, with the loading concept agreed before fabrication.
• On-site commissioning including correlation runs on your own packs, so the limit comes from data.
• Recipe per pack model with barcode selection; drawings handed over with the machine.
• Operator training on loading, background reading and what to do when the cycle aborts.
• 24-month warranty on mechanics, pneumatics and control; pumps per manufacturer terms.
• Spares in Guangzhou: hood seals, circulation fans, helium valves, pump service kits — 48-hour dispatch.
• Annual calibration with traceable certificate and remote support for recipe tuning.

Frequently asked questions

About Guangzhou Hejin Automation Control Technology

We have built leak detection and electrical test equipment in Guangzhou since 2003, with 50–100 employees and roughly half of our output exported to Europe, North America and Southeast Asia. Our team commissions chamber systems, tracer gas hoods and air decay testers, so the method recommendation comes from production experience across all three rather than from a single product line.

Every system is assembled, leak-checked against a calibrated standard and recorded under its serial number before shipment. That record is still available years later when a customer needs to compare a new result with the original acceptance data.