Test Set cabinet frame using aluminum T-slot extrusions

[PStechPaul]

Some time ago I started a thread discussing a high current transformer using toroid primaries and either tubular or rectangular copper bus bars for the output. It is rated at 2000 amps and 6.9 volts output, and weighs about 200 pounds. The final design that we came up with uses two loops of 1/4" x 2" bus. But we also had to make a cabinet, 12.5" wide, 29.5" long, and about 24" high, and since we wanted to keep the total weight under about 300 pounds, it "was decided" to use 1.5"x1.5" and 1.5"x3" aluminum T-slot extrusions. I had suggested a welded tubular steel construction which I thought would be cheaper and stronger and probably not much heavier, but the mechanical design was given to a consulting ME and a local machinist / fabricator.

We had a previous design, called a PI-2500 and AUX-5000, that used these extrusions because another consulting engineer was familiar with them and claimed they would be better than steel tube, but after they had been in the field awhile, the hardware that was used to bolt the pieces together became loose from the unit bouncing around in the back of a truck and being pushed and pulled around on factory floors. So the local fabricator came up with reinforcing pieces that supposedly would positively prevent this problem in the future (although I remain skeptical).

If you are not familiar with these, here is a catalog page showing standard fasteners, and we had used the "anchor fasteners" as pictured, which required milling a round hole in the ends of the extrusions:

http://www.file-vault.us/pdfs/tslots/catalogSec4.pdf

http://www.automation4less.com/tslots2.htm

So, anyway, here is how this cabinet has been put together. I think it's going to have problems (and I think it will wind up costing a lot more than one made from welded steel tube) but it's "out of my hands" and I am responsible ONLY for the electrical portion of the design (although I had to make a 3-D model showing where all the parts will go because the ME has absolutely no idea):









I'm not sure what sort of T-nuts are being used in the extrusions, but it seems that each piece is only being held with a couple of cap screws on only one edge. Also, having these on the outside makes it rather ugly (IMHO) and also necessitated some extra machining of the phenolic output plate (which originally was going to be sandwiched between the braces and the frame, until I objected and they reluctantly agreed). I also suggested that the corner braces could be made stronger and more cheaply as 45 degree angles but they said it was being made using CNC so it didn't matter and they will use these.

Note that the frame fits into a "dolly" made of steel angle, and there are long handles on each side made of heavy steel pipe and welded brackets that will be used to lift the unit and toss it into the back of a truck, and probably lashed down to keep it from sliding around too much. The smaller square section above the transformer will hold two variable transformers ("Variacs" or "Powerstats" or "Verniers" as we often call them), and they weigh about 60 pounds for the pair. They have decided to mount them on two 1/4" aluminum plates about 11" square rather than directly to the extrusions as I had originally designed it, because my "boss" who is paying me to design this didn't think it would work and he had the frame raised another 4" and added these plates which have considerable weight and won't be cheap. But he made the decision without my approval and I think there will still be problems, but not MY problem because the "experts" have assumed all responsibility for the mechanical design. This project is WAY past several "deadlines" because of the mechanical boondoggles, and when it's finally done then they will probably breathe down my neck to finish the wiring and testing in just a few days.

Sorry for the rant, but I'm just frustrated by this and I know I could have built the whole thing including the transformer much more quickly and better and cheaper (yes, ALL 3!), but my client is insisting on having it done "professionally" including a complete set of production drawings and building TWO of these units without even making a prototype first. So it will be interesting to see how this pans out, but I think one day the customer will take his test set for a ride down a bumpy road and find his $20,000 toy in pieces!

[Punkinhead]

That looks like 40x80mm Item extrusion - we use tons of the stuff for machine bases. I use it to build controls cabinets supporting SCARA robots that run at very high speeds and don't have any problem with joints coming loose. Those external corner braces look horrendous - I'd be embarrassed to send something like that to a customer. Assuming that is Item brand extrusion, you can look at their catalog for various ways to screw them together, including internal corner braces although I don't think they're necessary in your application. They also sell handles, feet, hinges, and other accessories that fit the t-slots so you don't need to fabricate and paint your own. They even have a design service to design and quote stuff like you're doing that would probably end up being cheaper and much nicer than the outside consultant you used. I've used their design service a few times when I got busy and was surprised at how fast, cheap, and nice their designs were. Personally, if I were in your situation I'd send them some pictures and dimensioned sketches and ask them for a design and quote. I bet they'd be able to ship you a completed cabinet cheaper than you can make it in house - that's what we've found on some of our machine bases. http://www.itemamerica.com/catalog-online.html

[Old Hat]

Same concept as quick-gage components, where you knock them around till a CMM says it's good
then dowel it down. NO SPENCIVE TOOLMAKERS NEEDED.

This stuff allows you to build a containment with nothing more than a coldsaw and a drill-press.
NO SPENCIVE FAB-MEN AND WELDERS NEEDED.

Great stuff if you like helping put people out of work. I won't use it.
I also refuse to use self-check-out lanes, and I walk into the bank
that still has tellers with eye-contact and a friendly greeting.

Folks complain about megga-rich corperatists, and then willingly participate
in destroying another's occupation. SAD.

[Toolguy]

I agree with you on all points Paul. I have been in the same situation too many time in the tool and die shops. They outsource to "professionals" when their own REAL professionals could have done a much better job faster and cheaper.

[Weston Bye]

...Great stuff if you like helping put people out of work...

Yep, just like anything better than a hammer and anvil.

I am no particular fan of the extruded aluminum, but it does have its place. Indeed, I just finished a side-job where it was appropriate. The slots in the sides were perfect for capturing the necessary aluminum and plexiglass panels that served as guarding, and no screws through the panels were necessary. the extrusions were sized and mounted for good structural integrity for the application. (Photos at 6 if anyone is interested)
The assembly method allowed me to make an efficient product that allowed me to make the sale.
Sad to say, most people expect Lego assembly from the extrusions and ready-made fittings rather than a good substantial weldment, but a weldment of aluminum would have been out of the question as would machining from solid. Nobody would have got the job.

[Rosco-P]

They bought into 80/20 extrusion stuff, but didn't understand the whole picture. Loos like it's sitting on a furniture dolly, couldn't casters have been bolted directly to the frame? The unfinished steel handles and brackets, the angle iron holding the toroids in the base....., whole thing looks like ass. Did Farmer Fred build this?

[Old Hat]

Yep, just like anything better than a hammer and anvil.

Sad to say, most people expect Lego assembly from the extrusions and ready-made fittings rather than a good substantial weldment, but a weldment of aluminum would have been out of the question as would machining from solid. Nobody would have got the job.

If we are willing to go Mr Spock, then why not just eliminate human labor at every opportunity?
I've not seen it, but it's said there is now ultra-high-end software for plastic injection molds.
You drop in your sugar bowl with lid, in the morning, you got drawings by lunch.

In the 20's a film called Technocracy red-flagged the folly of taking the human out of the industry.
It must be a timeless temptation for the greedy and the short-sighted.
>> Obviously this stuff has a roll in prototyping an unstable design.<< . . . but!

As for the outcome of the OP's thread, I can see weekness in this thing, but I think that was part of his point too.

..... sad face ..... in the title block.

[PStechPaul]

The ME contacted me today, requesting me to give him the box of components I had received from ETI (my client who has hired the ME and FAB, and who is building this unit to his customer's rather loose specifications. The customer just wanted a unit that would do about the same, or better, than our competitor's unit (Megger DDA-1600), which is actually based on a test set I had designed for EIL around 1985.

http://www.megger.com/us/products/ProductDetails.php?ID=54&Description=

The customer requested that the unit be removable from the dolly, so that is why it was made in two pieces.

Here is my concept of the frame, without supporting brackets:



Here is what resulted from the combined efforts of the ME, FAB, and ETI:



Note that the top section has been extended vertically with the intention of being able to assemble the "verniers" in the upper right corner as a separate assembly so they can be properly aligned with a common shaft through the centers. ETI actually "yelled at" the ME for using my original design and now they are mounted on the 1/4" thick aluminum plates and spacers through the four mounting holes of the flanges, and the result is that they are dangerously close to each other and it will be necessary to remove at least one of the top rails to install it as an assembly. And good luck holding it in place and getting to all the mounting screws to the inside surfaces of the extrusions.

My concept shows a simple "U" bracket in the left side for mounting the electrical components. But the ME and FAB came up with a much more complex piece that requires a "T" weld and also a rectangular opening to mount the control panel as shown below. And the panel has studs which involve using nuts on the inside which will be difficult to access:



Here is my latest concept drawing before the design was taken from my control:

[Bob Fisher]

Contrary to public opinion so far, I love the stuff! Ordered $600+ today in fact. Built a simulated pick up box in about 4Hrs. Would have taken longer with steel, cutting, welding, cleaning, and painting. And when it is no longer needed, scrap! The 80/20 stuff is reusable until the pieces get too short. Even then it's still brings a good price as scrap. Agreed the pictured fixture look pretty ugly, there are different ways of fastening this stuff that are more aesthetic. With proper torque, these fittings should never loosen. Bob.

[PStechPaul]

I agree that it is quite versatile and well-engineered. I have no doubt that, properly assembled, this would be a good-looking and durable product. But since ETI apparently doesn't want his own technicians to do any mechanical assembly other than mounting and wiring components, I think it would be (or would have been) better for the FAB to have welded a frame. This is the same machinist who said he really could not make my chuck screws using his manual machines because he uses CNC for almost everything and hadn't done single point manual threading for many years. I like the guy, and I feel a bit sorry for him because he has what seems to be an extreme case of "essential tremor", and although he had once enjoyed playing guitar and has perhaps a dozen high-end and unique instruments, he has not been able to play them for years. That might also explain his inability to do the manual threading job.