Page 1 of 3

V2.6 under design (supersedes the V2 .2) - project now abandoned

Posted: Fri Nov 10, 2017 1:07 pm
by exxos
26.JPG (232.04 KiB) Viewed 2907 times

9 hours into manually routing revised version 2.2 layout..

The original 2.2 design used six layers. However as mentioned previously, it is too expensive to produce these in small batches. So I've been spending a lot of time into compacting the design to use two GALs instead of three. Also squashing the PCB routing into four layers instead of six. The auto route failed to get beyond about 82%. It would only be possible if I used less than 0.2 mm tracks in gaps. While this will be possible, it again starts increasing costs of the boards as it is getting a little bit to specialist for general manufacturing.

This new design uses four layers, two layers are dedicated for 5 V and 0 V, however the top of the board and the right side of the board are using those two layers for routing. These are outside the IC area, so it should not affect the ground plane areas. Once I rooted as much as possible on those layers, I then used the remaining two layers squashing the rest of the routing.
gnd.JPG (186.34 KiB) Viewed 2906 times
Obviously I still have a little bit more squashing yet with the capacitors and resistors.. I will likely sell these in kit form first of all, as I simply do not have time to assemble and test them.

Re: V2.6 under design.

Posted: Fri Nov 10, 2017 6:24 pm
by Maeke
even in a kit, i'm still interested.

Re: V2.6 under design.

Posted: Fri Nov 10, 2017 6:44 pm
by exxos
Some more hours later..
642d7eb10e8c81c6e1f42b056fe98fec.png (127.94 KiB) Viewed 2892 times
3b20a2b6f2be86b669b76fdbd78fdc53.png (113.16 KiB) Viewed 2892 times
top.png (113.61 KiB) Viewed 2892 times
bot.png (138.65 KiB) Viewed 2892 times
2.png (109.18 KiB) Viewed 2892 times
1.png (106.04 KiB) Viewed 2892 times
Capacitors had to go on the rear of the board, there was simply no room or no way to route them on the top.

49 Via's, not bad at all. The auto route had over 300..

Big copper areas in the centre of the board for 5V and 0V. Managed to squash it all in :twisted: 8-)
all.JPG (193.24 KiB) Viewed 2887 times

Re: V2.6 under design.

Posted: Fri Nov 10, 2017 8:13 pm
by rpineau
Well ... you rock !


Re: V2.6 under design.

Posted: Fri Nov 10, 2017 9:05 pm
by Anzac
Damn, respect!

Re: V2.6 under design.

Posted: Mon Dec 04, 2017 9:51 pm
by exxos
PCBs arrived today. Though I am having huge issues getting this board to do anything at all. I have been trying for the past several hours.

Booster board when first built, it seems to work stable at 8mhz, and just seems to get worse to the point nothing works at all. At the moment I just have it wired up as a DIP to PLCC adapter, and there are no GALs in there at all. But the board still refuses to work and I cannot figure out why.

This does remind me of the issues I had with the PLCC sockets a few weeks ago. While these are new sockets, and they look okay, obviously something is wrong somewhere.

If anyone has any free time to do a basic wire up one of these PCBs, then please let me know, as I could do with someone else building one of these in case I am doing something incredibly stupid.

IMG_2087.JPG (171.62 KiB) Viewed 2664 times

Incidentally, my original V2 booster works perfectly well in the same machine. So it is definitely something to do with this new board :roll:

Re: V2.6 under design.

Posted: Tue Dec 05, 2017 8:08 am
by DrF
exxos wrote:
Fri Nov 10, 2017 6:44 pm
Some more hours later..
Amazingly a new board is born :D
exxos wrote:
Mon Dec 04, 2017 9:51 pm
, it seems to work stable at 8mhz, and just seems to get worse to the point nothing works at all.
Would that be something burning out?

Re: V2.6 under design.

Posted: Tue Dec 05, 2017 10:38 am
by IngoQ
Reminds me a little of the issues I had with my STE booster adapter. It was wired up correctly, it only added some length to the signal paths, and this seemed to be enough to put my STE over the edge. Regardless of 8Mhz oder 32Mhz operation.

I think we can rule out clock or the other bus signals, since these can apparently be wired over some distance and still be okay.

Maybe there is some capacitance building up somewhere?

Re: V2.6 under design.

Posted: Tue Dec 05, 2017 10:49 am
by exxos
I suspect pulling in swapping out the GAL has weakened something on the circuit board as when I first built it originally did work fine at 8 MHz. I made changes to my code and the thing stop booting. So I thought I did something wrong. Long story short, I bridged over the GAL with links and it did not boot either.

Last night I pushed down on the DIP side of circuit board and it booted. I re-solder the motherboard socket just in case, but the motherboard worked fine with my original booster anyway. I also re-soldered CPU socket.

I lifted up the corner where the CPU socket was last night, and it did tried to boot as I got four bombs. So I re-solder the socket again, and out is not working at all.

I also spent up CPU socket pins a lot thinking it may be that connection on the socket, but this did not change anything either. There is literally nothing on the board other than the socket now.

What I'm going to do is continuity check another new board, and then heat it up in my PCB although, and continue to check it again. I can only think with the board been bent a little bit during testing and changing the CPU etc, that it has flexed the boards enough to break some of the vias. It just seems that it was working originally and seem to progressively get worse. Each time I re-soldered it, seem to not make any difference, or if anything get even worse. So I really think these PCBs are just bad :(

Re: V2.6 under design.

Posted: Tue Dec 05, 2017 1:19 pm
by exxos
I build a third board up just, absolutely nothing on this board other than the plcc socket and a bunch of wire links. Oddly this board worked with a 68000 CPU but not a HC68000. This is exactly what was happening when I had socket issues..

So I have updated my socket page as I believe these sockets are bad also...