/diy/ - Do It Yourself

>>1603798

Looks good to me, anon. First day, you say?

>>1603801
Um ground is not in the sky. Last time I checked

Ground doesn't have to be on the ground anon.

>>1603798
>the only things wrong that I think I see is that might be wrong are

needs a flexible connection for the larger pipe

the rating of the angle iron and the anchors used might not be rated to hold this

possible working space clearances issues with the sprinkler pipe in the way

since they used angle iron instead of strut I doubt they got engineer approval to see if the rated roof can handle the added load... not that anything would actually happen.

lack of isolation pads under the transformer feet

that 3/4" pipe in the background isnt supported within 3' or even using the exception for unbroken pipe at 5'

-not wrong but lb's are kind of an expensive way to change over to flex, would be cheaper to just buy proper gofroms

OP is that a 30, 45 or a 75 kva?

>>1603815
The support is fine, if it gets hit by a crane or jlg it may fall but that would honestly be better than bringing the roof down.

>>1603817
I couldnt find the strength ratings for angle iron.. here we just use unistrut and all-thread because the strength ratings are listed in the caddy catalog we order from so its on site and can easily be proven to an inspector.

>>1603819
>here we just use unistrut and all-thread
A fellow patrician, I see.

>>1603820
I tip my hat to you sir

>>1603819
We primarily use unistrut because it doesn't require welding(tons of flammable shit around). It's quick, easy and modifiable. That's angle steel, it's incredibally resilient. Like you mentioned the fastners will fail first, the steel will just deform. It's not brittle like iron.

>>1603822
>We primarily use unistrut because it doesn't require welding(tons of flammable shit around). It's quick, easy and modifiable

yeah its great stuff, when ever I see someone build racks or like ops transformer out of anything else I start to question their knowledge of the industry.

were you OP, and if so where did you find this?

>>1603824
Not op, just seen tons of stupid shit over the years so I decided to check in.

It's stupid because it's not fucking grounded!!

>>1603859

>>1603859
Is the roof not connected to the ground?

>>1603910
Five bucks he's never changed a taillight bulb

>>1603910
It's in the air

>>1603815
>since they used angle iron instead of strut I doubt they got engineer approval to see if the rated roof can handle the added load... not that anything would actually happen.
I can tell you right now that that roof can handle it. That I beam is pretty thick. The angle irons are more than enough to support that transformer.

>>1604011
We gotta genius over here! Yeah, the thing isn't in freefall so it's being supported.

None of the parts really move so that's going for it. The transformer is being supported by the thin membrane of the roof, that potentially could introduce repetitive stresses and cause something to fatigue. The frame holding the whole mess together is clearly not square so good chance it's some combination of not very rigid and has some stress built in to its members. There's not even any guarantee the beam is a useful ground either, and using it as a ground is an excellent vector for corrosion to begin prematurely in a structural member, yay! The absolute worst offense and 100% this is a sign it was never signed off is that, of this fails, it's smack into that sprinkler pipe, which will be super fun.

Almost certain failure within 10 years, very likely within 5, you'd be surprised how much a flat membrane responding to wind pressure causes failure by fatigue. The i beam is a write off.

>>1603817
Please post the strength ratings...

No seriously you get engineers to sign off on things like this simply to defer liability to them anything ever happens...it's not so much the I beam but more of the z pearling that could be an issue. I agree though that nothing would likely happen.

>>1604032
Meant for >>1604011

>>1604018
It's pretty standard to take it to the structural steel here. We are required to bond the building steel at the service anyways. Typically we are required to show that the price we choose for this ground is welded to other structural pieces and not just simply bolted together.

>>1604034
If it's actually grounded it'd be fine, but I have my doubts. On an additional note if there isn't a good route to ground it coukd be a fun day if anyone ever needs to get on top of that roof.

>>1604009
>>1603859

>>1603798
>transformer "grounded" to the ceiling
Look up 'bonding'. Pic is not wrong as long as there are proper grounding conductors inside conduit as well.

>>1604018
>The transformer is being supported by the thin membrane of the roof

It's _very clearly_ attached to the heavy I-beam supporting the roof and the smaller beams holding up the sheets making up the roof.

>>1603859

A good thread died for this shit.

>>1603859
Ur right it's bonded. Ground on a transformer that size usually takes an ngr.

>>1604561
you know... since its a separately derived system one could argue that it is grounded

NEC 2014 250.30A(4) considers the steel to be a grounding electrode so essentially you are grounding it,
and 250.30A(5) states you size the grounding electrode conductor by 250.66.
But then you get to 250.30A(8) and they start talking about bonding the neutral to structural steel..

so since we are tying all the grounds and neutrals together it can really go either way depending on what exactly you are talking about.

yes my autism is in overdrive right now

>this thread

>>1604814

yes I like it to.

on the other hand, goddamn I can't wait for firefox to fix their shit so the flashing crap can be banished from my internet experience.

>>1604814
more like this board desu

>>1604011
Do you know the span of the beam? Noone can say from that pic if the beam can hold. There are other concerns as the other poster replying to you. The only thing WE know is that it has held untill know. So stop talking out your ass.

>>1603808
"ground" == huge source/sink of charge (e.g. a big metal beam, a block of metal, a copper rail in the literal Earth).

"Ground" does not imply a literal connection to the Earth

>>1603798