TF536 Dimensions

[Lemaru]

So I have been looking at this today, measuring up and coming up with ways that the A600 case could close with the TF536 installed. So far I have come up with these two ideas. One I couldn't complete even if I wanted to as the free version of Eagle won't allow me to place the 68k DIP where I want (even though the board size is smaller than what the free version of Eagle allows).

PLCC2DIP.JPG (160.14 KiB) Viewed 3775 times

This would fit around the keyboard header, and still leave the onboard IDE header accessible. Three of the HDD mounting holes can be used together with some 3D printed standoffs and M1.6 screws to support the adaptor.

The other version has the 68k DIP mounted partially above the PLCC socket, and again uses two of the HDD mounting holes to support the adaptor. This one would also leave the onboard IDE and keyboard headers accessible. However this one would be trickier to solder, and would involve soldering the SIL strips in place, trim the pins that sit above the PLCC as flush as possible to the board, and then solder the PLCC in place.

PLCC2DIP2.JPG (94.2 KiB) Viewed 3775 times

In both of these you could move the DIP back even further as it would actually hover above the keyboard connector and the membrane ribbon would just tuck around the edge (theres plenty of length to do it).

I can't complete the first one due to free Eagle limitations, but was going to route the second one and give it a go, but found my routing skills aren't that great so would take forever (autorouter doesn't do a great job).

Anyone have any other ideas, or any reason why either of these wouldn't work before I waste any time on it?

I have attached the schematic and board file (based on Kludge's Ice Scraper design), if anyone wants to play with it.

68k PLCC to DIP.zip

(47.37 KiB) Downloaded 177 times

[Aeberbach]

I have thought about this a bit. The position of the CPU doesn't have to be limited by overlaps. I still don't have an A600 - it's been sitting at a French airport for 8 weeks - but I think there is plenty of vertical room. Why not have two boards relocating the CPU, one fitting onto the PLCC and connecting to the 68000 board at a convenient place that does not overlap, then the 68000 footprint on its own board overlapping the PLCC if you like? It doesn't have to be hard to solder and might even get around those Eagle imitation.

(Why not KiCad, no limitations at all)

[Lemaru]

I did consider two boards connected by a cable, but a 64pin cable is just something else to have to fit inside, and another possible point of failure with interference and the like.

As for space, there really isn't much room inside an A600. The actual issue is locating it so that the case can still be closed. The keyboard sits in such a way that you are really limited by height. If the board comes any further toward the front of the case then you simply cannot closed the case. You don't really want to interfere with the keyboard cable too much neither as the A600 membranes can already be pretty fragile as it is. And other locations are not ideal due to the height of other components such as capacitors (you could go over the top of these, but again that results in the case not closing). The right side of the case is pretty much a no go due to the trapdoor slot and floppy drive. The board can be moved further left but I wanted to leave the on board IDE accessible, as other A600 boards do. Using the TF IDE results in not being able to use PCMCIA, so the reason behind this thinking was that maybe a new firmware could be released disabling TF IDE and allowing us to use native IDE and PCMCIA.

Kipper2k also now has a design for one but the location he has usd is pretty much the same as Kludge's adaptor so the TF again won't fit in a way to close the case. He had the new A500 FPGA board in mind for this adaptor though so obviously not interested is making it compatible with others.

I think once you get your machine you will realise how limited you are, particularly by height.

It doesn't affect me a great deal as mine will be going in a Checkmate A1500+ case, so space isn't an issue. But thought if designing a board then it would be better to design one everyone can use.

(As for KiCad, I already know the basics of Eagle so it's just convenient to not have to learn new software, but when time allows I will have a go)

[8 Bit Dreams]

I was the lucky one who's tested @kludge's ice scraper,
I quess the best way to go is to make it on two boards, here is a side view of it:

2020-07-16 12.46.47.jpg (36.37 KiB) Viewed 3718 times

One board is for PLCC socket and the other has 68k socket, these boards may be connected with some pinstripes..

[Lemaru]

I was the lucky one who's tested @kludge's ice scraper,
I quess the best way to go is to make it on two boards, here is a side view of it:
2020-07-16 12.46.47.jpg
One board is for PLCC socket and the other has 68k socket, these boards can be connected with some pinstripes..

This was another option I considered, angling the TF board so that it sits over the connectors and "lip" at the rear of the case (though on one board and not two), a bit like how the Indivision ECS sits at an angle in an A600. This would work for the standard case, but wouldn't fit in a Checkmate case unfortunately.

I guess the ideal is wait for the files for the TF536 to be released, and then mod it with a PLCC socket instead of the DIP, getting rid of the need for an adaptor at all.

[Lemaru]

I have thought about this a bit. The position of the CPU doesn't have to be limited by overlaps. I still don't have an A600 - it's been sitting at a French airport for 8 weeks - but I think there is plenty of vertical room. Why not have two boards relocating the CPU, one fitting onto the PLCC and connecting to the 68000 board at a convenient place that does not overlap, then the 68000 footprint on its own board overlapping the PLCC if you like? It doesn't have to be hard to solder and might even get around those Eagle imitation.

(Why not KiCad, no limitations at all)

Just had a quick look at this briefly again, and the first idea i posted is looking the best route in my opinion. This wouldn't get in the way of any on board connectors, or capacitors, does not overhang at the rear so would fit standard and checkmate cases, gives room above for the keyboard allowing the case to fully close, and importantly would be supported at three points which is pretty much required to take the weight and prevent PLCC connection issues (as this is the common complaint with most A600 cards, or they just use cheap rubbish sockets that are not correctly modified to give a good grip and connection).

When I get some time, I will do it in KiCad, get it routed up and get a batch of 10 made if only for my own use I guess.

[Aeberbach]

The main thing I was thinking is connecting a PCB holding the PLCC to a PCB with the DIP via pins. They wold not be far apart and you may not be able to access all the pins on either socket once they are joined. I'm thinking of a permanent connection with no more than about 5mm extra height rather than sockets because the relocator board is either going to work or not, no need for rework. Very little height gain and overlap can be achieved. Anyway I know it's tight, I had an A600 (I have made mechanical keyboards for them) but could not convince the kind person that loaned it to me to sell it! We will work something out.

[Lemaru]

I have just taught myself the basics of KiCad as coming from Eagle so have one of my designs done. I will try and get it routed over the weekend and get a couple made up.

I can't get my TF536 boards made yet as I am still waiting on the pin strips to arrive (i think they may of got lost), so it is giving me something to do in the meantime.

If it works, I will stick all the files online anyway, so between us all we should come up with a few usable ideas