Yes. With 74ls245 socket.
Its possible to make an 245 jumper. With the 245 jumper breakout.
See in the right corner of the pcb

Yeah! I hope this board eliminate all signal problems!

Yes. With 74ls245 socket.
Its possible to make an 245 jumper. With the 245 jumper breakout.
See in the right corner of the pcb

Can I ask what the 245 issue is?

Not really an issue, it is more a characteristic of the Enterprise and the MSX computers.

If you check the Enterprise EXDOS original schematics and even the Hungarian classic expansions, you always will see a 74LS245 protecting the Z80 Data Bus. You already know that the Enterprise is very noisy, with all the buses running from right to left of the PCB.

And on some MSX and all MSX2 and upwards, there is a signal named /BUSDIR that every cartridge has to create from the ports decoding and Ram/Rom allocation, to return to the computer slot, to be passed to a 74LS245 inside, guess for what?, to protect the Z80 Data Bus.

I am very new to interfacing "things" to the Enterprise, if someone had told me only a few years ago what I am doing now I would not have believed it...

May be the issue is me... When I started to try the SE-ONE(MP3 MSX cartridge) from Hans(TMTLogic) on the Enterprise, I connected the Buses and some signals and it started to sound almost immediately. But it was only casualty, or the very good components and construction of such MSX cartridge.

But when I tried to connect the GFX-NINE(Yamaha V9990 graphics) with the same system, I found that the cartridge had a strange behaviour and that often hanged the Enterprise.

Some months passed and then I found the /BUSDIR signal and the 74LS245 chip  on the schematics of most MSX computers. Zozo pointed me that the chip was also used on the Enterprise classic expansions, and then, advised by Hans, I tried to intercalate the 74LS245 on my M-SLOT adapter with total success.

What I learned was that some expansions add noise to the Data Bus, so it is convenient to only communicate with them when the Z80 wants it. But wrongly I also assumed that this was because the V9990 was an under tested and under used chip that was refused by the MSX computer makers on the MSX2+ phase.

When I started to interface the Symbiface3 to the Enterprise, I had forgotten all those problems, SF3 was the totally new all-in-one-card planned for the Amstrad CPC series, so I assumed that it would work straight.

It more or less did it accessing Z80 ports, but I had some strange issues with memory on my 2 CMOS Z80 Enterprises. When Zozo tried to test my hand made interface on some stock computers, he had worse results, probably due to the NMOS Z80...?

Some months after, and thousands of tests later, Hans proposed me again to put a 74LS245 and /BUSDIR signal(I had been rejecting that option several times, unbelieving about it). Hans sent me a new CPLD update and instructions on the point where to take the /BUSDIR signal. It took me an hour to put it aerial with a socket and some leads.... Of course it cured all the errors, and it was the base for Hans to design the SF3 adapter I presented Yesterday.

[ Guests cannot view attachments ]

One lesson I learned (maybe a proper h/w designer would have already known): use a real LS245, not a CMOS HCT245, because CMOS inputs left floating (eg on the tri-state data bus) are bad - can cause the chip to oscillate, and I have seen this. Even on a chip designed to interface to busses.

No because I'm expert here ... But I think proper bus should never float. If in it's high-Z state not driven, passive component (pull-up/down resistors) or even an active solution designed for that purpose should cause to have a proper well defined state. The physical realization of the bus (ie PCB traces) alone can produce problems if it's allowed to float time-to-time. But it's just my opinion, surely there are much more clever people than me in this area ...

@ BruceTanner
how is the automatic increase made with the Chip select?
do you have a circuit diagram of that?

Sadly, yes! Some pics here https://enterpriseforever.com/maintenance/glitch-on-rd/

My problem is in the WIZ W5300 network chip - it has an internal FIFO with a auto-incrementing data pointer. So to read a received frame, for example, you just do lots of reads, and the pointer auto-increments to the next byte. I interfaced it using I/O, so it happened when the I/O address was decoded and there is a /RD cycle. The main glitch happens when /RD is low so the W5300, being a modern fast device, sees 2 read cycles and increments the FIFO pointer twice. Not every cycle, just occasionally.

I have seen the glitch on a EP with no expansion attached - you just need to look at /RD on a scope that can trigger on a short high pulse (basic Rigol scope in my case). My early EP64 seems worse than my later EP128s.

Having got to the bottom of this, Zozo pointed out that a few EXDOS cards have a factory-fitted bodge on the back. When I looked, it was a R and a C on the /RD line going to the EPROM, and it successfully removed the glitch! I could also see it at the WD1770 but it works...so must happen at a slightly different point in the cycle. Or the WD1770 is too slow to notice the glitch. Or a double read doesn't matter to it.

@gflorez sorry, I have accidentally hijacked your thread

B.

In order to make work the SF3 with the new adapter, a simple lead has to be sold:

[ Guests cannot view attachments ]

[ Guests cannot view attachments ]


To fix some memory issues that happened interfacing the SF3, a 74LS245 chip was needed to protect the Data Bus. Actually this chip is in every Enterprise expansion, so it was the logical move.

But Hans found that there was no free pin to implement it on the CPC connector, so he redefined the TP3 point to be /Busdir, a returning signal that activates when memory(RAM/ROM) or ports are accessed(decoded) inside the SF3.

The lead will connect TP3 to pin 1 on the CPC connector.

This is not the perfect solution, because pin 1 on the CPC connector is the Sound output, and being the SF3 made for the 4 SymbOS computers, this would cause noise when connected to a real CPC.

There is planned a revision of the SF3 with the 74LS245 chip included, is for that the Enterprise adapter has a detachable pcb scrap to make a dummy chip for the 245 socket.

[ Guests cannot view attachments ]

One idea I have is to use a flat ribbon cable male to female like this to connect the SF3 to the future SF3 slot adapter, making a Z with the 2 pcbs and the ribbon cable the SF3 can be lowered at max. to the Bus expander surface, but this only can be made on the first slot because this approach can be very thick.


But then the SF3 would need to be fixed in some way to the future SF3 slot adapter.