Very large Steel Milling Machine on existing wood dock piles

[dreber]

So I've got a bit of a tricky situation.

I've got a client with an industrial factory that was built in 1909 in a tidal zone on piles. The structure is essentially a dock, with wooden piles of unknown depth or quality into the soils below, the soils are also of unknown properties, so the existing foundation is impossible to justify via calculation.

Elsewhere in the same building, 8 years ago there was a load test to verify capacity for a 100kip machine.

The proposed machine they want me to design for is a 200kip machine, but it loads a few more piles than the existing 100kip machine. Obviously these piles are different than the ones that had been load tested.

They want to skip load testing for the new 200kip machine. What would you do? Should I design a slab and anchorage to the building and put on the drawings "EXISTING PILE CAPACITY TO BE VERIFIED BY OTHERS"?

I suppose in all of this, where does my responsibility end? I don't foresee any safety concerns if the piles were damaged by the weight of the machine, but it could be enormously expensive to repair any damage.

Should I walk away from the job?

 

[oldestguy]

"Should I walk away from the job?" Looks to me to be a good choice, but a threat like that may put some sense in them. I'd want some form of release from liability if they violate any of your recommendations.
I'd look at driving or jacking some new piles to get a handle on what might be a safe load per pile. Perhaps re-drive an old one also.

 

[dreber]

That's part of what I have been thinking. All the time doctors advise patients of their options, let the patient decide, and then the doctor proceeds, based on some ethical considerations of "first do no harm, etc".

Since I told them this is a dumb idea, if they release liability, and since nobody would get hurt, it's hard to see why not to do it.

On the other hand, I've never heard of an engineer doing something like this, so it's hard to see me doing it.

¯\_(ツ)_/¯

 

[oldestguy]

I have walked away from at least one construction job that I recall due to safety issues with no problems later because of that.

 

[WARose]

Sounds like you don't have drawings and you don't have load tests for the piles you are going on......so that makes the decision pretty easy: either tell the client we need to do [X, Y, etc] or walk away. You can't green light something with info that bad.

 

[SlideRuleEra]

...it's hard to see why not to do it.
I've never heard of an engineer doing something like this, so it's hard to see me doing it.

I did things like this frequently (in-house) for my employer.
Don't make the decision now, based on "guesses". Offer to investigate, gather information, and prepare a report on what you find... then decide what you will do.
Obtain the results from the load test 8 years ago.
Take measurement and elevations at the location of the 100 kip machine, to see if you can get an idea of what has happened there (even though you will not have base-line data from 8 years ago to make a comparison).
Review how the 100 kip machine loads each pile. Compare with how the 200 kip machine will load each pile.
See if soil reports, pile driving logs, etc. are available for other nearby projects.
For certain inspect some of the existing piling.
Once you start gathering information, the path for what else to look for will probably reveal itself.

Make it clear that you may decline the job based on what you find.
If the client is unwilling to pay for the investigation... walk away right now.

http://www.SlideRuleEra.net

 

[dreber]

I do have the original structural drawings, but they are silent on design loads, soil values or pile embedments (this was 100+ years ago mind you).

I also have data about the original load test including its location and the weights included. Based on the previous load test, the piles were worth around 4kip each (additional load beyond what they had previously been subjected to). However, it is impossible to say for sure the allowable loads of the previously tested piles because the concrete beam structure between the machine and the piles would have spread the load around to some unknown extent, based on the stiffnesses of the concrete beams which I am loath to try and account for.

Even if I did know the allowable loads on the other piles, its difficult to be sure that after 100 years, the piles at the new location are still just as good. The only way to be sure is to put 400k of steel there and wait a few days.

The more I think about it, the more I think that if it was my building and my machine, I would perform the load test. I think I'm gonna tell them that I can't help them if they won't load test.

 

[NorthCivil]

why not bring in another engineer who can review the piles and the sub-sea earth and provide guidance on upgrading/approving that portion?

it doesnt have to be sign off on everything or walk away - bring in someone who is knowledgeable on that aspect of the job.

 

[SlideRuleEra]

...it is impossible to say for sure the allowable loads of the previously tested piles because the concrete beam structure between the machine and the piles would have spread the load around to some unknown extent, based on the stiffnesses of the concrete beams which I am loath to try and account for.

This is where the "art" of engineering comes in. Forget about trying to get an exact answer. Do what amounts to a DIY Monte Carlo simulation using various assumptions to determine how sensitive the "answer" is to variables.

The load test you suggest is excellent... if it "passes"... but the results may be inconclusive. Also, if the load test "fails" (collapses the structure), maybe you should have performed the investigation to see if even performing a load test was reasonable.

http://www.SlideRuleEra.net

 

[dreber]

How could I do a DIY Monte Carlo? What would I use as a range of spring stiffness's for the axial load on the piles? For all I know, they are 10 ft deep and bearing directly on bedrock, or 40 feet deep in some crummy clay-mud, or 10 feet deep in crummy clay-mud (this last one is the most likely

 

[Retrograde]

What would I use as a range of spring stiffness's for the axial load on the piles

Estimate the original working load on the pile and assume it deflected 10-15mm (half an inch) under that load. That would be a good starting point for a spring stiffness.

 

[SlideRuleEra]

dreber - The problem you expressed is the uncertainty of how the the previously tested piles were loaded by the "concrete beam structure" since "stiffness of the concrete beams" could not be accounted for... not about soil properties supporting the piling.

The DIY Monte Carlo simulation concerns looking at various assumed stiffness of the "concrete beam structure" to see how that changes test load distribution on the pile.

Concerning the piling themselves, I assume they are timber. Until an investigation proves otherwise, I would be more concerned about structural condition of the piles themselves than soil properties of the soil, rock, "pluff mud", etc. that supports them.

Consider this: As I understand, your (valid) concern is that (for whatever reasons) the existing structure is not capable of safely supporting a 200 k load. With that uncertainty, I would sure not blindly apply a 400 k test load to a questionable structure to see what happens.

http://www.SlideRuleEra.net

 

[dreber]

Thanks, Retrograde. I took that approach and came out to ~160 psf allowable live load. The machine will be ~200psf dead and ~200 psf live.

The load test makes things look a lot more do-able. The results of the load test suggest the allowable load (of the piles directly below that load test) was almost (about 80%) of the demand load from the machine.

 

[racookpe1978]

Look at it from a slightly different viewpoint:

If the machine foundations begin moving, so the whole machine twists by 1/2 inch down on one end or one side, what is the maximum loss? ("Whole machine and assembly line becomes unuseable", or "nobody notices")
How "stable" is the heavier load? Impact (like a classic drop press)? Or slow inertia-loaded like a long, flexible conveyor belt or oven? Static like a tank or rolling mill with low changes in weight per second?

Is the heavier machine on the same "style" piers as the older, lighter weight machine? Or is it configured so a "pounds per sq ft" resistance is near-meaningless?