SDX FDC05 PAL14L4 dumps

[gilmo]

Hello everyone,

Anyone has the equation PAL dumps for the floppy disk controller MEMOTECH SDX FDC05 classic disk controller?!

those are IC10 and IC11 "PAL14L4" logic equations .jdec/jed files

the controller schematic I am referring to, is here :

http://www.primrosebank.net/computers/m ... 05_1_1.pdf

it's a reproduction obviously .. the equations on the schematic appear to be incomplete or .. I'm not sure 100% if they are accurate .. for example; there are two lines A12 & /RD connected to IC11 but not mentioned in the equations ?! .. and that can't be right .. needless to say the PCB is not working so far (disk read error) .. have done a lot of troubleshooting on this PCB build, and all that is left is to test with an actual PAL-DUMPS from an original working FDC05..


If anyone has the equations or the "actual IC dumps" or can read them from a PAL reader , please provide .jed files here or to my email below, I've been trying to get this to work correctly.

Thank you so much in advance .. your help is appreciated

Gilmo
gilmod714@kakao.com

[Dave]

The schematic came from my website. It, and the PAL equations, were created and checked in conjunction with the original designer (Tony Brewer ). I’d be very surprised if they were wrong, though it is possible. (A12 and /RD are used in the ROM logic, it is possible that Memotech design development removed them from the PAL logic after the PCB was designed).

What makes you think that your board is not working? Are you seeing a “ DISC ERROR” message from BASIC?

If so, are you using a copy of an original Memotech disk? You can not boot any disk in an SDX that has not previously been formatted and loaded with a copy of the CP/M system tracks!

See this page for details http://www.primrosebank.net/computers/ ... sksSDX.htm

It is currently unfinished, but read the bit about system disks

[gilmo]

Hello Dave,

Yes, I am seeing a "disc error" from basic .. I tried different versions of the SDXBASIC 03,07

I don't have an original "system disk" .. but I have created disks from disk images, using many different methods in DOS & linux utilities, one of them was Andy's mfloppy utility

http://www.nyangau.org/mfloppy/mfloppy.htm

I have read and followed all these pages you are referring to on your site and Andy's too ... yes, I know about the first 2 tracks and ram patching .. I have tried many combination of things, which led me to eventually suspect the PAL equations not being correct and the 2 missing lines were a red flag..

Dave, would it be possible to double check the schematic against the original FDC05 you own ?! ..

A PAL dump of the original IC10,IC11 would be awesome too .. this way we can verify if the equations are accurate or not ..

Thank you so much !

Gilmo

[Dave]

The PAL dump comes out with the same equations as on the PCB. Unless there is a problem with the PAL reader hardware or analysis of the output, there shouldn't be an issue.

I'm afraid that I don't have time at the moment to do a detailed comparison of the SDX PCB and the schematic.

The best that I can offer is that if you send me a copy of your system disk I can check whether it can boot my SDX. If not, I can create a good system disk and return it to you

regards
Dave

[gilmo]

Hmm .. ok .. I didn't expect the PAL equations to be identical !

As for the system disks.. mfloppy does actually create perfect disks with CPM header bytes and a hex editor shows the 2 track bytes being present.

"andy_sys.mfloppy-07" (is a good example of that) and mfloppy can write it in linux32 to a floppy disk just fine. I have even used "Greaseweazle" to read that same disk back as an image again, and tested again with Mess just fine. (identical copy - R/W) .. so I would comfortably rule out the floppy disks created from the IMG files.

Dave, could you please email those PAL dumps (.jed) and I will try them out "anyway" in my PCB ?! .. just to eliminate any doubts in my mind, since they come from a working PCB?!... my chip burner is TL866-II (so it only takes .jed files) and I usually burn them to GAL16V8D

If that doesn't work though, then I suspect there might be another issue with the schematic not being 100% identical to the original FDC05 (I'm suspecting the data bus latch section too).. I know that you put a lot of time & effort into reproducing it with KiCad and I applaud you for that.. but I might be the first beta tester of that schematic - unless you know of someone else who actually built it and it worked for them

Just some additional fyi's :
I have tested all the logic chips outside of the PCB, used fresh MB8877 (brand new ones) , scoped out the signal here and there and verified the data separator signals as they should be going in & out while trying to read the disk (wd9216) - 1Mhz & 4Mhz are there too , have tried different version ROM chips "03,07 BASIC", tried different switch combinations etc .. all the same result. Even my 3.5" floppy disk drive is identical to the one you used in your setup on your website "Sony MPF920" modified for DS0.. (works fine on my MSX & Amiga btw)

http://www.primrosebank.net/computers/m ... dx/sdx.htm

Thanking you in advance for your great efforts on creating the schematic and being very helpful here to get this PCB working

Best regards,
Gilmo

[Dave]

"Dave, could you please email those PAL dumps (.jed) "

No, I'm afraid that I can't, all I have are the (reconstituted) equations.

None of the low cost programmers available today can directly read the contents of a PAL, that was the whole purpose behind Tony designing the PAL Reader.

http://primrosebank.net/computers/mtx/p ... reader.htm

As far as I am aware, any PAL reading tools that you might come across on the web work in the same way, they only work with non-registered PALs (as Memotech used) and work by generating every possible input combination, reading the corresponding output states, then minimising the resultant output to generate the corresponding logic which can be used to create a new .jed file for a GAL.

The hardware and software tools were verified by testing known devices, but it is possible that the SDX PALs do indeed have an error. I will have to blow some GALs with the equations and try them in my SDX but this is not something that I will be able to do quickly I'm afraid.

To generate a new jed file, you'd need to use something like CUPL./WinCUPL to compile the equations to a jed file for your programmer. (My TL866A used to have problems blowing GALs until I loaded a later release of firmware but your TL866-II might be OK I guess.) If you want to try this, you could try modifying the /CE equations for IC11.

I will try to get to this before too long, but can't commit to a timescale I'm afraid.

regards
Dave

[gilmo]

Thanks Dave, I understand what you're saying ...

Which brings us back to the initial suspicion of the Equations -- If the schematic was 100% traced accurately (and I do trust your work ) .. then we still need clean JED files to make this PCB work as it should.

I have attached my .jed files below , so you could just easily blow them to GALs and test on your original FDC05 - when and IF you have the time.

(see attached)

for compiling jeds in general I use OPALjr under dos , it works great .. that's how I created those jeds. if anything was done incorrectly in the process, the MTX wouldn't even turn on (black screen and buzz) .. having my MTX on and running, responding and seeking the floppy means the burned jeds were done correctly actually - just the equations are questionable still, or there's a mistake in the schematic..

I have indeed modified some equations for testing on /CE in the past .. the results were horrible .. that's when my suspicion started regarding (A12 /RD) lines ... unfortunately, I am not an expert in Z80 bus communications to understand fully how those equations are supposed to look like.

Thank you for your kind help in this matter.. and if you get a chance to test new GALs please let me know .. together, we will get to the bottom of this.. for nostalgia sake:)


Thanks again Dave

Best regards,
Gilmo

[Dave]

OK, not that it matters, but I think that it's unlikely that the MTX would "black screen" if an SDX GAL was not programmed correctly

regards
Dave

[Dave]

I have been in touch with Tony who did the original design who has kindly looked at this for us.

First of all, Tony has confirmed that A12 and /RD are NOT needed in the logic for IC11. The GALs that you have created should be correct.

However, Tony has identified a couple of issues . . . .

Although the layout drawing and Bill of Materials shows decoupling capacitors for all of the ICs, I did not specifically note them on the schematic and so you have not included them on the board. Good practice would dictate that you should fit 47 or 100 nF capacitors across the power lines as close as possible to each IC. I would hope that the board would work without them though.

Tony also queried the power distribution on your PCB, suggesting that the traces may be a little light.

The biggest issue though is that Tony has detected a mistake in my drawing. On IC6B there are a couple of reversed connections. The correct connections should be :-
Pin 5 - INDISK
Pin 12 - NC

Can you try creating a patch to connect INDISK (pin 18 of IC11) to Pin 5 of IC6B and break the current connection to Pin12 of IC6B

regards
Dave

[gilmo]

Wonderful news! Thank you Dave & Tony as well

Yes, I did notice the missing decoupling caps from the beginning and also thought that it wouldn't be a big deal either (just good practice to have them as you said)

yes, the lines are light on this PCB (I should have thickened some pwr traces) but there are no issues with power distribution so far, and I also routed mtx +5 to power the PCB only , and using external 5vdc-1A for the drive via USB cable.

I will correct the connections to IC6B and report back ..

Thank you guys!!!