PCBONEZ wrote:

There are "spacers" aka Standoffs that have a pad instead of a pin that goes through the mobo pan. They just sit on the pan, th […]
Show full quote

candle_86 wrote:

Well alot of newer cases don't have the holes for a BabyAT board anymore, and with the cutouts some of them have plastic spacers dont work either.

There are "spacers" aka Standoffs that have a pad instead of a pin that goes through the mobo pan.
They just sit on the pan, they don't go through it. No hole required.

They don't hold-down but they do limit board flex when installing RAM or add-in cards.
For hold-down the standoffs that do line up are sufficient.
Use metal standoffs and screws in those locations so the board is properly grounded to the case.

In a pinch you can cut the pins off the more common nylon standoffs which turns them into the pad type.
Been doing this since the 90's. It works great.

Standoff.jpg

Also works for parts testing boards that will never see a case.
Keeps all the solder joints on the back of the board off your work surface.
.

I'm talking about where the case is missing metal on the tray for heatsink removal in modern cases, seen some rather large cutouts for that

candle_86 wrote:

I'm talking about where the case is missing metal on the tray for heatsink removal in modern cases, seen some rather large cutouts for that

OIC
Those can be patched by soldering in a patch panel. Or use pop-rivets.
Personally I'm too lazy, ah, I mean busy, for that sort of project anymore and I'd just find a better case.
.
Just saying if you really really just gotta have THAT case you can make it happen.
.

tayyare wrote:

Sorry, I jumped before reading the whole tread 😊

Coming late to the party is okay.
[passes tayyare a margarita]
.

Plasma wrote:

You don't need a hydraulic punch or multiple hole saws. A step drill bit will cleanly cut through sheet metal.

The io shields are way thinner than normal sheet metal. I could give a step drill a try though I guess. If I remember correctly it didn't go to well the last time I tried it on something that thin.

PCBONEZ wrote:

Coming late to the party is okay. [passes tayyare a margarita] . […]
Show full quote

tayyare wrote:

Sorry, I jumped before reading the whole tread 😊

Coming late to the party is okay.
[passes tayyare a margarita]
.

Thanks! 🤣

cyclone3d wrote:

Plasma wrote:

You don't need a hydraulic punch or multiple hole saws. A step drill bit will cleanly cut through sheet metal.

The io shields are way thinner than normal sheet metal. I could give a step drill a try though I guess. If I remember correctly it didn't go to well the last time I tried it on something that thin.

As an engineer/manager that spends 12 years (of his 25 years work life) manufacturing parts from sheet metal, I would say for a sheet metal that thin and flimsy, it (i.e. step drill) will not end in a good way.

cyclone3d wrote:

Plasma wrote:

You don't need a hydraulic punch or multiple hole saws. A step drill bit will cleanly cut through sheet metal.

The io shields are way thinner than normal sheet metal. I could give a step drill a try though I guess. If I remember correctly it didn't go to well the last time I tried it on something that thin.

I suggested the two wood blocks and hole saw method partly because you are planning on doing a number of them.
If you set it up right the blocks will serve as templates so you won't have to spend so much time fussing with centering and aligning each one individually.
.

cyclone3d wrote:

I can get blank I/O shields for ATX cases for cheap. […]
Show full quote

I can get blank I/O shields for ATX cases for cheap.

All I would need to make AT style shields out of them would be a punch for the correct size hole. I would also probably have to make a holder to hold the I/O shield in the correct place when punching the hole with a hydraulic knockout punch.

If enough people are interested, I wouldn't mind ordering a bunch of plates and making AT plates out of them.

I think with shipping, I could send them out for around $7 a piece (maybe less) and they would be significantly cheaper per shield if more than 1 was ordered.

I would just be looking to cover the price for the punch as well as the shield and shipping.

The more that are wanted, the cheaper I can make them for.

To break even at the ~$7 a piece shipped to the USA, I would need to make about 20 of them.

I'm interested in one or two of those.

Use PM.
Site admin frowns on trading.

PCBONEZ wrote:

Both the PCI and ISA in the shared slots lineup to the same slot in the back but you can not use an ISA and PCI at the same time in the shared slots.

I wonder if it is just a mechanical "sharing" (squeezing in one more slot taking advantage of PCI cards being inverted when compared to ISA) or there is some circuitry on the board shared making it impossible to use those two slots together (assuming that one manages to put a card into both).

I once managed to squeeze a really thin USB card on the shared slot (of course all connectors were put on cables and capacitors relocated to other side of the PCB).

Azarien wrote:

PCBONEZ wrote:

Both the PCI and ISA in the shared slots lineup to the same slot in the back but you can not use an ISA and PCI at the same time in the shared slots.

I wonder if it is just a mechanical "sharing" (squeezing in one more slot taking advantage of PCI cards being inverted when compared to ISA) or there is some circuitry on the board shared making it impossible to use those two slots together (assuming that one manages to put a card into both).

So far as the signals and data they can't be sharing. Different bus, different tech.

So far as the -power- supplied to shared slots, I don't think there is any official standard that covers that situation.
That would be up to the discretion of the manufacturer.
Never thought to look, this is speculation.
Most likely they do share the power paths to reduce the number of traces on the board.
That would only be a problem if the traces are too thin at some point to support the current to two cards.
Again, never thought to look on vintage boards, this is speculation.
I imagine -ALL- the slots on a board share power traces (to reduce the board space needed) and they would have enough copper in the traces to support a card in each shared slot.
--- I'm not going to say that all boards do that because as soon as I do a board will show up that doesn't.

I have occasionally seen people use both slots and for them it worked fine.
.