August 2019

Confined Space Programs That Actually Work

OSHA 1910.146 has been around since 1993. It's well-established. The requirements are clear. And yet confined space programs are one of the most consistently broken things I encounter at manufacturing and industrial facilities.

Not because people don't care. Because there's a gap between what's written in the program document and what happens when someone actually needs to enter a tank, a vault, a silo, or a pit.

The Inventory Problem

Most programs start with a confined space inventory. Good. But in my experience, about half of the spaces on the list shouldn't be there, and there are spaces on the site that aren't on the list at all.

People add every space that looks remotely enclosed to the inventory because they're being cautious. Meanwhile, a utility vault that gets entered twice a year and has a genuine atmospheric hazard never got added because nobody thought of it as a "confined space." It's below grade, has limited egress, and sits next to a chemical storage area. It absolutely qualifies.

The inventory needs to be done by someone who understands the criteria — not just "is it enclosed" but the full definition: large enough to enter, limited means of entry/exit, and not designed for continuous occupancy. Walk every space. Evaluate each one individually. And revisit the inventory when the facility changes.

Permit-Required vs. Non-Permit

This is where things get messy. A lot of facilities classify everything as permit-required because it feels safer. But that creates a problem: when every entry requires a full permit process — atmospheric testing, attendant, rescue plan, entry supervisor sign-off — people start cutting corners because the overhead is too high for a routine task.

The standard allows you to reclassify a permit-required space to non-permit if you can demonstrate that the hazards can be eliminated. If a space has no atmospheric hazard, no engulfment hazard, and no configuration hazard, and you can document that, then a full permit entry isn't required. That frees up your resources for the entries that genuinely need the full treatment.

But the reclassification has to be documented. Written. Signed by someone competent. And re-evaluated if conditions change.

Rescue Is Where Programs Fall Apart

I can't count the number of times I've asked "what's your rescue plan for this space?" and gotten a blank stare followed by "we'd call 911."

That's not a rescue plan. 911 gets you a fire department that may or may not have confined space rescue capability, and even if they do, response time is measured in minutes. OSHA expects you to have rescue capability that can reach the entrant in a timely manner. For many spaces, that means on-site retrieval equipment and trained personnel.

At minimum, every permit-required confined space entry should have:

The retrieval system is the first line. It's what gets someone out in the first two minutes. Everything after that — fire department, EMS, hospital — is backup.

Training That Sticks

Annual classroom training checks a box but doesn't build competency. The people who actually enter confined spaces need hands-on practice. Set up a training scenario. Have them don the harness, clip into the retrieval system, practice atmospheric monitoring, and run through the permit process start to finish.

Do this once a year at minimum. Do it more often if you have high turnover or infrequent entries. The worst time to learn how the tripod works is when someone is unconscious at the bottom of a tank.

The Real Test

Pull a random permit from last month. Check whether atmospheric testing was done at the right times. Check whether the attendant log shows continuous monitoring. Check whether the rescue equipment was actually staged. Check whether the entry supervisor signed off before entry began, not after.

If everything checks out, your program works. If it doesn't, fix the gaps. That's the whole game.

← Back to blog