MSX FLEX project

Pagina 2/4
1 | | 3 | 4

Van Alexey

Guardian (2238)

afbeelding van Alexey

13-04-2019, 01:20

Printed the cases. They will look better in white plastic, but I keep the transparent one as an option.

Van ericb59

Paladin (847)

afbeelding van ericb59

13-04-2019, 06:52

Great Idea !
This is something I was thinking ... I need at least 1 of them...
Please keep me informed once I can buy one or two :-)

Van hamlet

Scribe (2356)

afbeelding van hamlet

13-04-2019, 08:42

I'm in for one.

Van Alexey

Guardian (2238)

afbeelding van Alexey

05-05-2019, 23:37

The boards, cable and connectors arrived. Testing the 60-pin extender/converter with 30cm cable. Seems to work fine. Tested with SFG module, network module from Russian MSXs and with Gotek adapter.

The cabling is done so that the adapter part (where the cartridges are inserted) is located behind MSX.

Van rderooy

Hero (559)

afbeelding van rderooy

06-05-2019, 13:29

Nice work. Just wanted to point out that there may also be some people interested in a version they can use with the MSX Expansion bus on certain machines
https://www.msx.org/wiki/Expansion_bus_slots
Looking at the pictures, I think your existing version may just fit to these machines in any case, as it seems you have an 50-pin IDC connector on the top of the MSX Cartridge? So the only issue would be for those systems with two polarizing keys, which will have to remove the single key to fit it.

Or something to substitute the Yamaha CA-01, which also is hard to get
https://www.msx.org/wiki/Yamaha_Expansion_Bus_Connector

Van Alexey

Guardian (2238)

afbeelding van Alexey

06-05-2019, 15:01

The way that the 50-pin cable is made in FLEX should be compatible with the male bus extension connector. The key of the connector is the only problem. I only found the connectors with one key. But if the key is slightly trimmed or the bus connector is modified to accept a single key, then the existing FLEX extensions could be usable with the default expansion bus.

As for CA-01 - this could become a separate project if there is a demand.

Van Alexey

Guardian (2238)

afbeelding van Alexey

09-05-2019, 16:33

Seems that the 50-pin extender module has a problem. It looks like a few cartridges start to glitch when connected to the extender. So far I identified 4 cartridges (out of 20+ that I have) that have problems: MMM cartridge (RAM expansion), our IDE-FDD controller and 2 home-made flash cartridges with Konami mapper (similar to what Eric sells). Everything else, including Carnivore2, SFG, home-made cartridges with ASCII mapper, original game cartridges, simple cartridges (up to 64kb) work without problems.

Today I was testing the 50-pin extender with the SVI-728's expansion bus. I had to install an extra 50-pin connector on the cable because on my main connector the key was reversed.

Interestingly enough that MMM and IDE-FDD cartridges worked correctly with the extender on this computer. But the Konami cartridges still had glitches. If anyone has thoughts on why the glitches with some cartridges happen, please share them here, thanks.

Another interesting fact is that the "brother" of those 2 Konami cartridges works without a problem. Also the glitches are different when the extender is used in different slots of my Yamaha computer. I tried pulling the data bus to +5v via resistors (as it's done in some Yamaha's external devices), but that didn't have any effect. I also tried to isolate the CLK signal from the data signals, but it didn't work either.

Van bore

Expert (115)

afbeelding van bore

09-05-2019, 18:18

Have you measured the clock to see what it looks like?
It could be worth putting a schmitt trigger on the clock just to be sure.
If the cartridge signal drivers are too strong it might be able to "self-clock" by causing enough ripple on the clock.

Maybe also try some extra capacitor on the power at the cartridge end?

Van Alexey

Guardian (2238)

afbeelding van Alexey

09-05-2019, 22:19

Tried 2 capacitors (100nf + 10uF) between GND and +5V on both ends. Nothing changed.
Went through all signals with 2-channel oscilloscope, measured the signals on the slot and on the end of the extender - almost identical readings.
Next I will try pulling the data bus to GND with 10k and 1k resistors like it was done on some slot extenders.

Van Alexey

Guardian (2238)

afbeelding van Alexey

09-05-2019, 23:32

Looks like pulling data bus to ground with 1k resistors works, but not in all cases. Will try to pull the address lines as well...

Pagina 2/4
1 | | 3 | 4