Dragon 200E Lower Case Daughterboard ROM question

[sixxie]

Just a little note: if you re-spin, can you ensure you take the FS# signal from that pin of the VDG? I think we learnt that your board uses the fact that FS# and MS# are connected on the motherboard.

And the way we learnt this was that earlier versions of my D64 motherboard remake didn't connect these - I instead pinned MS# high so that DA0 would never float, and a HCT part could be legitimately used for IC13. It's a solder jumper on later revisions, but still...

With apologies if it was actually an older lowercase board someone had a problem with!

Edit: yep here we are: viewtopic.php?f=5&t=11061&start=17

[dublevay]

Just a little note: if you re-spin, can you ensure you take the FS# signal from that pin of the VDG? I think we learnt that your board uses the fact that FS# and MS# are connected on the motherboard.

And the way we learnt this was that earlier versions of my D64 motherboard remake didn't connect these - I instead pinned MS# high so that DA0 would never float, and a HCT part could be legitimately used for IC13. It's a solder jumper on later revisions, but still...

With apologies if it was actually an older lowercase board someone had a problem with!

Edit: yep here we are: viewtopic.php?f=5&t=11061&start=17

~MS & ~FS are supposed to be connected together on the D64 motherboard. The schematics had a problem whereby the original was creased right where that connection was. On my lower case board, and probably as per the 200E board, ~MS on the 6847 is connected to the socket, whilst ~FS is connected to ~PE on the HCT161.

[sixxie]

On my lower case board, and probably as per the 200E board, ~MS on the 6847 is connected to the socket, whilst ~FS is connected to ~PE on the HCT161.

Oh, you do tap off FS#? Interesting, I wonder why the person in the link above had issues then.

~MS & ~FS are supposed to be connected together on the D64 motherboard. The schematics had a problem whereby the original was creased right where that connection was.

Yes, but I think in the Dragon 64, it's a bit cargo culty - the purpose is to tristate DA0 during FS: on a D32 (IIRC) that means the input to the SAM probably naturally pulls high while DA0 floats. On a D64, DA0 is passed through an or gate with (effectively) FS#, which takes it high more explicitly. And that means it's possible to just tie MS# high meaning DA0 is never tristated, meaning that OR gate can be replaced with a CMOS equivalent (all its other inputs are "safe" if you replaced other logic too).

..ciaran

[Go4Retro]

On the original lowercase PCB, pin 37 of the header (FS) is not connected to anything. You can see this here:

Screenshot 2024-11-20 224806.png (779.42 KiB) Viewed 5326 times

This is the original PCB photos superimposed, the bottom reversed, and the top is set to 50% opacity.

Pin 37 on the header is the 4th from the right top. Notice no trace either side...

Notice that pin 12 (MS) has a trace heading to the left, and then a via right to the left of the 6847's right set of pins. THat via has a trace on the bottom that goes up to where pin 37 is, and then vias again.

Thus, I theorize that the header *MS and *FS pins are tied together and the signal is then tied to the header *MS pin. Since the '161 needs to get the *FS from somewhere (it's the signal going above the IC5 in the pic, to the left of the one with the via that is connected to pin 31 (INT/EXT). Since it lines up with the upper via sitting to the left of *FS, I think IC 12 and 37 and the '161 are tied together with that trace, and the trace then heads over to the header pin 12.

I updated my v1 schematic to tie FS of the IC to FS and the 161, and MS header to MS IC, no bridging.

[Go4Retro]

~MS & ~FS are supposed to be connected together on the D64 motherboard. The schematics had a problem whereby the original was creased right where that connection was. On my lower case board, and probably as per the 200E board, ~MS on the 6847 is connected to the socket, whilst ~FS is connected to ~PE on the HCT161.

According to this post:

http://archive.worldofdragon.org/phpBB3 ... 630#p16630

"The /MS ping of the 6847 connects to the /MS pin on the socket and to the /PE pin of the 74LS161A."

Jim

[dublevay]

This is exactly what is on my board...I think.

I have the below connections

Pin 12 Socket = Pin 12 6847 = Pin 37 6847 = Pin 9 LS161
Pin 37 Socket = N/C

~MS~FS.jpg (539.2 KiB) Viewed 4981 times

So yes, it is reliant on the motherboard having ~FS and ~MS connected.

I don't see the issue here. If we're saying change the circuit just because we can because we're using CMOS, I'd need a better reason than that.

You may, however, have given me a clue as to why the Atom modded version I made performed so poorly!

[sixxie]

So yes, it is reliant on the motherboard having ~FS and ~MS connected.

Your pic makes it look like you connect the two together yourself as well as them being connected on the motherboard!

I don't see the issue here. If we're saying change the circuit just because we can because we're using CMOS, I'd need a better reason than that.

Not an issue per se. As a daughterboard that's designed to plug into a specific motherboard it's obviously fine - the original motherboard is what it is, your daughterboard will of course work fine with it.

Just that if on your daughterboard you only connected to FS# - which is after all the only signal you actually need - it opens up the possibility of more CMOS use on the motherboard! I wouldn't even have mentioned it if you hadn't suggested you'd be doing a bit of redesigning anyway

(In particular, my D64 board has the different ways of connecting MS# as a solder bridge selection - I've tied it high so that DA0 can feed into an HCT chip "safely")

You may, however, have given me a clue as to why the Atom modded version I made performed so poorly!

Kinda interested in the why of that

[dublevay]

Your pic makes it look like you connect the two together yourself as well as them being connected on the motherboard!

Yes - as it was originally a replica of what the 200E card was, and isn't ~FS and ~MS also tied on there? It is on the traces I got from (I think) Rob originally?

You may, however, have given me a clue as to why the Atom modded version I made performed so poorly!

Kinda interested in the why of that

Just something Phill said a long time ago about the different way that ~FS or ~MS is assigned on the Atom, and that if I have taken across a Dragonism, that could well be the issue. It basically screws up the keyboard entry, making it somewhat 'hair trigger' from memory.