Any of the expansion anchors on the page below are simple to install. The same size hole is used through the ledger into teh masonry. This eliminates trying to spot holes from the wood to the masonry.

I use lots of galv. 1/2"x4" lag bolts into lag shields...expansion anchors. My decks aren't collapsing.

steve

Not allowed by code uin many locations.
Lag screws and lead shields have very poor withdrawal reistence.
The decks can pull away from the houses and collapse.
They dio not move part way, but typically fail all at once when a large number of people are present loading the deck.

The link below is a typical set of deck requirements.
Lag screws for ledgers are only allowed for attachment to band joists. A lesson learned after hard experience.

steve

A ½ inch Rawl Power Stud embedded 3 inches deep in 2000 PSI concrete has a tension load rating of 5100 pounds. With derating 510 pounds.

A 208 pound withdrawal every 18 inches on a deck is pitifully low and can easily be exceeded under normal conditions. As soon as the deck has any pitch the loads change from shear to tension.

Lag shields are not a safe method of attaching ledgers to concrete for deck support.

"Of course nobody has mentioned how much torque was applied to the bolt. "

Cause it actually does not matter once you apply enough to obtain the maximum rating of the fastener. More torque can actually decrease the strength. If the shield starts to rotate in the hole the withdrawal resistance has been seriously compromised. It can turn at the start of tightening a little, but once it starts to expand it must stop rotating and not start again.

You have not provided adequate resistance to the fasteners pulling out of the wall (tension loading).
The numbers I used are for Rawl shield anchors at nominal installed strength.
Normal derating is to 10% of the ultimate strength so a 2000 pound strength is rated as only 200 pounds in actual use.
200 pounds if allowable tension load over 18 inches is only 133 pounds per foot of ledger. That is not a large number at all, and grossly inadequate.
500 pounds is 333 pounds per foot of ledger. Still a little low but could be acceptable if the deck is not very high off the ground (less than 4 feet or so) and not very large (say less than 8 feet perpendicular to the ledger).

Higher or larger the fastener spacing needs to be tightened up based on the actual area.

Positioning the anchors in a perfectly straight line in the wood also reduces the strength. The wood can simply split along the line of anchors and the deck pulls away.
The arrangement is normally for the anchors to alternate being 1/3 down from the top of the ledger, and the next one third up from the bottom.

As a PE I testify at trials over deck collapses.
Many times I just present the findings to the insurance company and they settle.
The Fairfax code is just one of many that have been promulgated over the past 10-15 years following deck collapses, many with serious injuries and even a few deaths.
A more recent problem has even been the overloading of band joists in frame houses and the entire side of the house (band joist, sheathing, etc.) all pulling away. Most of the nailing fastening the band joist is in tension (either through the joist into the ends of the floor joists, or through a joist hanger and into the band joist) and tension loaded nails have reduced strength. In the case of no joist hangers the nailing is into end grain and is normally rated at zero strength in withdrawal for structural purposes.

You didn't define "derating", you only applied percentages.
It sounds like youre saying a fastener system can only be counted on for 10% of it's maximum capacity. Is that so?

I know many posters here are not deck builders, but at least one above is and appears proud he does not know what he is doing.
I hope he has good liability insurance.

I like to stay out of these debates, especially with know-it-all number crunchers...but you seem to want to drag me back in. You like to use terms like derating, and throw numbers and %s at us. you know what they say..."figures don't lie, but liars figure"

let's look at your data...you say those 1/2" bolts with 3/4"
shields will support 208 lbs (with derating). So if I use 2 every 18"...typically 20-30 per ledger...what will that support? what if I always fasten to brick, not concrete? those bolts are not gonna shear, so we're just talking about pulling away from the house. Add to that, a series of posts (4x4 or 4x6) supporting the deck about 8' out. those posts are set in concrete, at least 2.5' deep.

you're an engineer, and you testify at trials over deck collapses. how many trials have you been to where a deck collapsed because of lag shields? I'd like to read some supporting articles. Even if you can site a few examples, who knows who the contractor was, or who he hired to do the installation. Was it his first deck? was it a friday afternoon or monday morning? into brick? 1/2" bolts?

you can throw your numbers around, but you're not a deck builder...and you're certainly not on my site to witness the way I set a ledger.

Most decks collapse because the builder used nails to fasten a ledger into a rim joist. I've read articles about them. I never even deal with rim joists or nails. here in brownstone Brooklyn, my clients all have brick buildings. I've built plenty of parlor floor decks in the last 15 years, and I am confident that they will stay up at least until the wood deteriorates (nothing lasts forever).

Proud? yeah I'm proud of my work. check out my website if you want to see what I do. and don't worry about my liability insurance.

steve

It's as I thought, I would state it as "being we don't know who built it, we're allowing for most things to go wrong."
And that's OK.
A factor I haven't heard mentioned was a deck that isn't level. You get a bunch of people on a deck that has a slope and that increases the tension tremendously (IMO).
On the other hand... a large, "overbuilt" deck weighs ALOT...in my calculations like 2000-3000 pounds. If the deck is level that's alot of weight going straight down. If good piers...no problemo. As a non-engineer, with enough mass and excellent piers there might even be a resistance to pulling away from the house, or in any direction. The height would have alot to do with it. Are you with me on that concept.

Because the wedge grips the concrete more and more as you torque the nut, you do want to make sure you apply the proper torque during installation. Also, make sure you drill the proper diameter hole. I have included a link to the specs below and as you can see, even the little guys can develop quite a bit of strength (scroll down to the strength tables using the link) - much more than a couple hundred pounds quoted by brickeye.

IMO, the lag shield type anchors are very much inferior to the wedge type shown in brickeyee's link and the alternate link I provided. These anchors can be purchased at a reasonable price at the big box stores and they work well so I don't see a good reason to use the lag shields.

Lag shileds are all made from softer metals (lead or zinc being common) and do not have the strength of steel expansion anchors. Red head, Rawl exapansion anchors, they all work about the same.
The expanding steel will not slip the same way a shield type anchor does and the pressures that can be developed against the sides of the hole are nothing compared to steel anchors.
The guard rails on highways are attached using steel expansion anchors for a reason.
The strength and repeatability of steel anchors os far superior to soft metal anchors.

"If the deck is level that's alot of weight going straight down."

Decks are rarely level ebnough to eliminate tension loads, and are very unlikely to stay level for very long after construction. Wood moves.
Even the sagging of joists affects the loads on the anchors of the ledger.

"Even the sagging of joists affects the loads on the anchors of the ledger."

The sagging of the joists causes a rotation of the ledger board and due to the constraint of momvement caused by the concrete wall, the ledger board does not rotate about the anchor bolts but rather the lower edge of the ledger that is in contact with the concrete.

This results in a tensile load on the anchor bolts rather than just a shear loading. As such, the pullout performance of the anchor bolts is important.

"Allowable values are based upon a 4 to 1 safety factor. Divide by 4 for allowable load values."

That is vastly different from the 10-1 safety factor talked about earlier in this thread. Using the #s provided by The Smith Fastener Company website for shear and tension loads, you could fasten your ledger to your foundation so that even several full size pickup trucks yanking on it couldn't pull it out.