Getting 65A off a PCB to another PCB

Hello,
We need to get 65A off a PCB on to another PCB.
We are thinking of ten 4mm plug jacks in parallel...but is their a better way?
Its just for a test jig, so that we can get the current to a dummy load PCB. Voltage is 1V5.
Its the output of an SMPS.

Not enough information.

Is this DC or high frequency AC.

How fast do you want to change the board.

How many times do you have to change the boards out.

How much room do you have for making the connection on each board.

Can you layout each board anyway you want.

consider, screw terminals, plug in connectors, spring probes.

Yeah there are lots of ways. A simple one would be brass standoffs. You just need holes and pads on each board.

Sixty five amps eh ?

I would use a ruddy great nut and bolt with a copper lug.
Can be removed in seconds, and will never let you down.

Warpspeed said:

Sixty five amps eh ?

I would use a ruddy great nut and bolt with a copper lug.
Can be removed in seconds, and will never let you down.

Click to expand...

Copy that................gold or silver plated nut & bolt

Thanks,
its DC, there is flexibility over the size and layout, hopefully we'll be selling lots and changing the boards many times..theres plenty room for getting the connections on the board.

The board will actually get the 65A off the SMPS PCB via six ERNI 214548 connectors, but then we need to get it off that board to the other board which has the dummy load on.....we cant put the dummy load on the first board as it would be too awkward.

The nut and bolt sounds good and quick, i dont know why, but i always assume such connections are a little "hit and miss".....also, it woul dneed to go to a wire which will need a crimp, and we all know how expensive crimp tools can be...also, the crimp and crimp tools are expensive for the heavy wire that we would need for the 65A.

Try this...
**broken link removed**

Available here...
https://www.digikey.com/product-sea...erconnects/terminals-screw-connectors/1442846


Udhay

The nut and bolt sounds good and quick, i dont know why, but i always assume such connections are a little "hit and miss".....also, it would need to go to a wire which will need a crimp, and we all know how expensive crimp tools can be...also, the crimp and crimp tools are expensive for the heavy wire that we would need for the 65A.

Click to expand...

Bolted lugs have been around for a very long time and are used for aerospace applications.
There are a few simple rules to follow though to do it successfully.

First thing to realise is that we are clamping a copper lug to another copper surface using a through bolt.
The bolt itself carries negligible current, so steel is fine. The bolt only provides a clamping force, the electrical joint is (or should be) copper to copper.
Fibreglass is not rigid, it will compress and flow slightly under high compressive force over time, so caution is advised.

The lugs themselves are pretty fool proof, and there is nothing wrong with solder.

In fact unless you have the correct exact specific tooling, solder is probably preferable to crimping anyway. And with heat shrink tube the lug can be colour coded and look quite neat.

The trick with the bolt is that it has to be elastic and stretch sufficiently to hold the joint in compression, even when your fibreglass compresses slightly over time.

So what we really need is a long skinny high tensile bolt, rather than a short fat coarse threaded bolt which will loosen.
A high tensile fine threaded bolt and two steel washers and maybe even a spacer (to allow for a longer bolt stretch), can be really torqued down and work very well.

Its not difficult, it just needs to be thought through.

Just realise that many of the very high power semiconductor modules use 5mm or 6mm screws on top, and can be rated to carry several hundred amps, and they are perfectly reliable.

There is no reason why the same sized lugs and clamping screws cannot be made to work equally well at the PCB end.



These modules all use a copper strap with a captive nut below. Something we are all very familiar with.

They follow the above simple rules of having a copper to copper joint, with a steel tension bolt. Its as simple as that.

SMPS for computer PCs often deliver more than 65A (5V or 3V3) under routine operations. They are connected with multiple (more than 5) connectors (in parallel). I have never seen the connectors failed.

I like the "big copper screw" idea, but as you know, it will need a ring terminal and connection to a wire…which means buying a big expensive crimp tool, whereas we only need one test unit and this may not be worth it.

- - - Updated - - -

the PC's usually deliver the big current to the same board as the module thats producing the 65A..so no need for board to board conenctors for it.

When the working voltage is low (say less than 5 or 3V; I understand your case is about 1V), it is wise to pay attention to contact resistance losses. Voltage drops, due to contact imperfections, can be a real nuisance because they are usually a constant loss (say 0.1 to 0.3V due to oxide on the surface) that will be a significant part of the total voltage at low supply voltage.

the PC's usually deliver the big current to the same board as the module thats producing the 65A..so no need for board to board conenctors for it.

Click to expand...

Not true in general for most desktop PCs; they have a separate power supply.

Hi,

There are flexible copper wires like these.
You can buy them in different length, isolated or not....


Klaus

"
The trick with the bolt is that it has to be elastic and stretch sufficiently to hold the joint in compression, even when your fibreglass compresses slightly over time.
"
NO!, N0!, NO!
I have worked for a long time in high power broadcasting and am familiar with high current circuits. Using the "elasticity" of ANY insulation material (GRP, Paxolin. . .) will lead to a short life. Mount your stud to the board then take your wire to the stud where it needs to be fixed with a nut and locknut.
Many years ago I did a course on high precision components and a bit I seem to have remembered is about very good switch contacts. You are talking sub milliohm here. The real problem is not the contacts but the tarnish that gets in between them. So the contacts were a lot of thin flat ended blades, that literally scraped the tarnish off as they rotatated.
I have some "old fashioned" 4mm type test plugs where the ends are a bundlde of wires, so on insertion they scrape the tarnish off. Shame I can't find them on the net!
Frank

Chuckey;
I remember, many moons ago on a power load board, I saw exactly what you describe.

A bolt was soldered and with a nut fully secured to the board.
Then the crimped wire was attached to the bolt and tightened in place with a second nut and a star washer.

In this instance, both the bolt and the first nut carried current, so they were made of tin-plated copper.

Very reliable.

I like the "big copper screw" idea, but as you know, it will need a ring terminal and connection to a wire…which means buying a big expensive crimp tool, whereas we only need one test unit and this may not be worth it.

Click to expand...

The screw really needs to be high tensile steel. A copper screw would just shear off before you could get enough clamping force. Cheap hardware store mild steel bolts are almost as bad.
What we need for low resistance, low temperature rise, and low voltage drop is a large cross sectional copper area. And a bolted lug, copper to copper will give you that.
It will very likely occupy less board space than a multi pin monster plug that has insulation between individual pins for a given copper conduction area.

You do not need to crimp for prototyping or limited production. The wire will slip neatly into the lug, and with a large soldering iron, soldering will work perfectly well.

THIS is a soldering iron.

Warpspeed said:

The screw really needs to be high tensile steel. A copper screw would just shear off before you could get enough clamping force. Cheap hardware store mild steel bolts are almost as bad.

Click to expand...

This is an excellent solution except that if you try to connect -disconnect more than 2-3 times, you will need to change everything (the hardware).

After a while, you will see that the copper lugs has developed a fine (but visible) rectifying oxide layer- unless the contact area is large and flat.

Ran across these, when I was just looking around to see what was out there. They look pretty awesome. No need for bolts, threads, wrenches, torque, etc
https://www.te.com/usa-en/about-te/news-center/snap-lug.html

Probably cost you an arm, leg, kidney, eye, ear, lung, and indentured servitude of the next 3 generations of your family to buy one. ;-)

How far are one board apart from another ? For short distances, the use of interconnections directly with male-female connectors has the added benefit of serving as a robust support bracket, when placed in pairs. To avoid the risk of unbalanced distribution of current - which would be totally expected - would be necessary to use an amount of terminals in a baggy factor exeeding the average current rated per pin.

c_mitra said:

This is an excellent solution except that if you try to connect -disconnect more than 2-3 times, you will need to change everything (the hardware).

After a while, you will see that the copper lugs has developed a fine (but visible) rectifying oxide layer- unless the contact area is large and flat.

Click to expand...

The bolts and washers should be fine to reuse even if rusted, they will still clamp up tight, only deliberate abuse can really damage a good quality bolt.

A positively clamped high pressure joint has to be better than any sliding pin, and a really tightly clamped joint will work just as well as a crimp, and for exactly the same reasons.

Sufficient clamping pressure clamps the parts rigidly together in intimate contact, excluding air, and the oxidation process. If you have ever dismantled ancient equipment using properly bolted up lugs, the exposed outer surfaces can certainly be rough and pitted and absolutely filthy.
But the underside of the washers, and the lug surfaces, will be clean and pristine.
There is also special electrically conductive grease containing silver particles if you are really desperate.
This is routinely used for crimping aluminium conductors into ordinary copper lugs to prevent moisture ingress and ionic corrosion from developing from the dissimilar metals.
It is really unnecessary for copper to copper, but it would absolutely exclude any air or moisture ingress from a clamped joint.

Only extreme circumstances like occasional exposure to sea water can cause problems, but your expensive slide action plugs and sockets will likely fare even worse in similar circumstances.

You should check out some high power car audio amplifiers, they have a few different ways of getting a big power wire into the PCB. Google something like car audio power amplifier terminals - do you have to put terminals on the wires? Many of the larger power amplifiers don't use terminals on the wires - you just strip the wire, stick it in the hole and tighten the set screw (great if you don't have to repeatedly make connections).