A601 1/2 meg expansion question.

[Zetr0]

This gets more and more fascinating all the time,

The expansion bus LCAS and UCAS connectivity for extra 1MB ram

LCAS [pin 35] (_BCASL(1))
UCAS [pin 36] (_BCASU(1))

------------------------------

Where as Pins 8 and 9 on the expansion port are

DRD 2 (pin 8)
DRD 3 (pin 9)

the plot thickens and thickens

[Zetr0]

Originally Posted by FOL

Yeh, but what does it all mean, . Spill the beans, also are the new high res pics ok? or do you want more, done a mixture of both flash and non flash.

I just sent you a pm (maybe I should have just replied here, sorry).

LOL no worries FOL, i have to back up my eab PM's on a near enough daily basis anyway... i just wish bippy would understand that its over.... meh....

okay... now looking at new pics

[Zetr0]

Now this is very interesting.

I think i have noticed some intended damage to a couple of the signal lines. not sure where they go to, have a lookie at the attached.

[FOL]

Originally Posted by Zetr0

Now this is very interesting.

I think i have noticed some intended damage to a couple of the signal lines. not sure where they go to, have a lookie at the attached.

Someone didnt read my post, already point this out, . 1 end is going to the green wire, ill check where other end goes soon, just kicking horaces ar*e in renagade squadron, .

It does look as though they may have not intentionally damaged the track next to it.

[FOL]

Right Zetr0, heres my findings.

Correction first. Sorry zetr0, I misunderstood the edge connector. I now see that after looking at manual.

U17 pin 28 goes to pin 1 & 2 (VCC) of edge connector
U17 pin 29 goes to pin 35 (_BCASL(1)) of edge connector


The track that is cut, goes to pin14 of U17, it then goes to pin 38 (_BRAS(1)) of edge connector. It is also connected directly to pin14 of U19.

Now the track next to it, that looks broken but isnt. This is connected to pin14 of U18, which goes to pin 37 (_BRAS(0)) of edge connector and no where else from what I can see.

Now with the track cut, that wire is telling U17 (pin14) to use pin 37 (_BRAS(0)) of the edge connector instead of pin 38 (_BRAS(1)). Making it a single chip configuration?

So from the looks of it, if the track wasnt cut and that wire not there. U17 and U19 would use pin 38 (_BRAS(1)) of the edge connector and U18 would use pin 37 (_BRAS(0)) of the edge connector on its own.

I have also uploaded 2 more highres clear pics. You can see where I have scrapped lacker from the tracks for testing.
Pics are in NEW folder, under Trackcut1.jpg and Trackcut2.jpg.

[FOL]

Zetr0, I added 2 x 256kb (60ns) i.c.'s to that A601. It seems to be working fine, . It only seemed to work the way they modded it, all 3 i.c.'s are using _BRAS(1).

Filled the full 768kb and read from it, with no issues. So I shall leave it at that.

Turns out, that it was shipped with this A601 preinstalled as 256kb. As it states in the Service Manual.

Been looking at another schematic of the A600, 1's posted here, seem to small. Found aload of Commodore schematics where you can actually read all the writing.

[TheCorfiot]

@FOL

Glad you got the card working,
What is even odder though is that it was only equipped with 256K in the first place, 512K I can understand to a point but 256K, thats half as much as the original A501 (A500 expansion).
What use would a 256K expansion add to an A600 even if the A600 was factory shipped with only 512K, remember the A1200 was planned to originally ship with 1MB, so a 512K A600 may have been planned.
Commodore really had some strange Marketing ideas & concepts...


TC

[coze]

FOL, that chip on the A601 is a 256K, 16 bit chip. which makes 512Kbytes. Are you sure you read 1.250 mbs ? if you're referring to RAM which is reported at workbench bar, that is the amount you have after workbench use, so you have more than that (1.5 mb).

Also the showconfig chip ram address range (no RAM $FFFFF, with RAM $17FFFF) suggests 512Kbytes have been addressed.
So probably you didn't need the third chip in there.

[Zetr0]

@FOL

This is all clearly fascinating, and i really need to answer your PM too

indeed that is a 512KB chip, i think you will find that U18 is addressed in the same space as U17... really really cool bit of kit

if you get 5 mins have a screen shot of the show config would round this puppy up

a huge thanks for the pics, hires hardware pron... always good for a desktop

[FOL]

Originally Posted by coze

FOL, that chip on the A601 is a 256K, 16 bit chip. which makes 512Kbytes. Are you sure you read 1.250 mbs ? if you're referring to RAM which is reported at workbench bar, that is the amount you have after workbench use, so you have more than that (1.5 mb).

Also the showconfig chip ram address range (no RAM $FFFFF, with RAM $17FFFF) suggests 512Kbytes have been addressed.
So probably you didn't need the third chip in there.

Well, I think I may have mashed something up now, as im getting odd gfx errors when board is in, will investigate further tommorow.

Yes its deffo only reporting 256, so I added another 256 (it wouldnt boot), I added another 256 and it booted fine and gave me 768kb. Another thing is, the jumpers 512kb and 1mb are disabled. You can only have it in the 1mb config.

It does address $17FFFF, with 256kb and 768kb. I cant tell with 512kb as I dont know what components are need on the board to make it work. It just does what I said above, refuses to boot with 512kb. I can use the full 768kb, as I say, I filled it full and read from it with out issue.

[FOL]

Right sorted gfx problem, bit of broken print. In the module is inserted it always displays $17FFFF in showconfig.

@coze you are right, that chip is indeed 512kb. This is where it gets interesting.

I filled the ram again to do some more tests, and checked its capacity each time it said it was full.

Now after filling it once, it changed to 1900kb ram (from 1760kb). I then filled it again, and checked. The ram had gone up to 1990kb.

I tested the stuff copied to ram: and it all worked fine, maybe the auto config cant allocate the 512kb chip(as its maybe faulty), and then detects it after filling it?.

Its turning out to be very odd. I also removed U18 and left u17 and u19 in, it reported 1768kb but then would hang when filling the ram. Im guessing cause it knew U19 was there, it assumed U18 was there also and when trying to access it, it hung.

[Zetr0]

I believe that the A600 chip sense is based on a LOW pass looking briefly at the schematics, by having the adapter plugged in thats enough to fool the controller into thinking there is RAM there, and hence the issues its come up with.

I think from reviewing more and more of the pics, U17 shares its addressing space with U18.

I would attempt removing U17 and populate U18 and U19

[FOL]

Originally Posted by Zetr0

I believe that the A600 chip sense is based on a LOW pass looking briefly at the schematics, by having the adapter plugged in thats enough to fool the controller into thinking there is RAM there, and hence the issues its come up with.

I think from reviewing more and more of the pics, U17 shares its addressing space with U18.

I would attempt removing U17 and populate U18 and U19

Well, I have decided to leave as is. Its stable and doesnt hang, plus the more I mess with removing the i.c.'s increases the chances of damaging more print or the i.c.'s.

I shouldnt need it after doing the Piggy back mod, .

[FOL]

Zetr0, I gave into temptation and removed U17. Result was it didnt accept the A601 was there.

Very odd indeed, I even tried adding another 256x16 IC and it still wouldnt accept the A601 was there.

So will have to leave it as 768kb increasing to 994kb as its used, very odd indeed.

[FOL]

Hi, sorry to drag up and old thread. Seems my 1mb upgrade is working now with the other 2 x 256kb chips.

The problem lies with my A600, it seems to not be allocating the full 1mb. I assumed it was the 1 mb expasion, but its definately my A600 main board.

So im guessing this is why the piggy back mod died. Think Ill leave it as I dont have spares, and dont fancy messing with it while its working.

[squidbass]

I have a rev 2.0 A601 which is also reporting 512K and not 1Mb despite having the 2 ram chips - what was your fix? The rev 2.0 has a different layout but the same jumper config. Can't see anything on the A601 such as broken tracks etc. Although there are a couple of pads that look like a resistor has been removed probably in production...maybe I'll post some pics.

[squidbass]

*bump* I need my A600 fix :-)

[FOL]

Originally Posted by squidbass

*bump* I need my A600 fix :-)

Zetr0 worked out that some boards were different and needed connecting differently. So it was pure chance mine worked for a little bit.

Im sure zetr0 could hook you up with the thread where he explains it all.

[squidbass]

Thanks FOL...

Zetro - if you're around, can you shed any light?

The only thread i can find dealing with this subject is this one.....I'll take some pics of this a601 because it's different to the one in this thread and I'll continue digging...

squid

[Zetr0]

@squidbass

hello there,

hmm, I am inclinded to believe that this is likely a chip fault / failure.

when you disconnect the A601 how much ram do you have available