A4000 tower...help!

[Morley]

Sorry for the lame subject-topic...

Just received my A4000Tower (original) today, and whooooops, it doesn't work as it should...oh no.

Whenever I put a SCSI HDD on the SCSI chain, the machine refuses to boot. I can't access the Early Startup menu, it just stops on a grey screen.

Also my mouse(pointer) only wants to travel up&down, not sideways... the right mouse button is also defunct.
I have tried with 2 different hdd's and mice, but no help...think there is some bad contacts on the cables from the io-board to the mainboard.

And when I fire it up with no hdd's installed, it loads from floppy, but it stops loading workbench half-way...could this bee from the missing workbench.libray not being on the floppy's? I only have WB3.0

What I look for is an explanation of the DIP-switches on the scsi-card, and some tips from existing users...

Any help/tips highly appreciated, this is an absolutely sad moment for me...waiting for it to arrive, only to learn that it is'nt working :-o

[Morley]

well, I just checked the jumpers on the mainboard...what shoyuld they be set to? CLK90 clock-source and CPU clock source are set to external, guess this OK for a 3640 board.
There are no expansion boards here, just plain original with 3640 ver. 3.1 here. Shouldn't this be a ver 3.2 as the entire machine seems to be from '95?

[Morley]

'at least Gloom works and boots! :-D

I think the mouse not wanting to move sideways is caused by bad contact in the cabling between the mainboard and the io-board. Probably just a minor fix.

But what irritates me is that it is unstable, it guru's on rebooting, on different occasions, just without a cause. I think maybe there's something with the clock-jumper settings on the mb...what should i use with the 3640?

[Morley]

Hi!

Thanks for the answer, but unfortunately, my jumpers were set correct. I tried switching them btw, but that led to the machine not wanting to boot at all.

I spoke to the person ho sold it, and he said that he tested it several hours before shipping it off. But he switched the cpu-card, I think he tested it with an accelerator or something similar, and according to what he said, he just took a 3640 from a pile he had of them, but it has been tested.
Anyways, he thinks there could be some combability issues with the different 3640's and the 4000tower. My card is a v3.1, but my machine is stamped'95, so I really think a 3.2 should be what it was shipped with? If you still have your original 3640, have you got a chance to check out the rev. version on it?
I think this could solve the SCSI problem, but the mouse thing is a little more tricky...I have triple-checked the cables, and the joystick works fine, so I really don't think that the cpu-board has anything to do with this?

[Morley]

Well...I have only got one SCSI-device hooked up, and it's id is set correctly. Termination is OK on the drive end, but do I need to add a terminator on the controller? seems strange...

As for the mouse, I got it working now...I just took everything out of the box, and reinserted it again, whohey! the mouse works :-)

[Morley]

@darksun9210

hmm...think that could be the source of some of my problems, I believe I never turned the termination on. I thought it was ment for the external devices only.. but anyway I think I flipped that switch also when I was testing it. I think I tried all combinations of those jumpers :lol:  But shouldn't it just refuse to see the drive if the bus-termination is bad? mine hangs completely, this is different from my blizzard1230IV who doesn't care about termination at all :-) I've got the computer working now btw., booting from ide. When I run hdtoolbox it checks all adresses without crashing, so the controller chip should be OK, wild guess.
I have also got some instability problems on my wb, but I think that is because of the sys: partition being dumped from my 1200 and some libraries tweaked for '030. It goes guru without warning every now and then. But when i boot from floppy everything is 100% (except for the ¤%#!! SCSI controller) :pissed:

[Morley]

Well... latest status update for those interested:

I now have it working from an IDE hdd, everything is 110% from the IDE hdd, I even got to mount a SCSI cd-rom. But the instant i add an SCSI-hdd, - nothing. Just a boring black screen and the drive led on the tower flashes something like 2x per second..huh.
Also when I fire it up from cold-boot with an ide-hdd installed, it hangs before booting anything, and I have to warm-reset it, and then everything works...
Strange, huh? :-?

[Morley]

@Markus

hmm...just tried the power-up without power to the scsi trick. Everything works nice then, also the cd-rom drive on the bus. But after inserting the power cord and waiting for the drive to speed up and intialize, hdtoolbox freezes and the drive led flashes the wellknown 2x/ps. like it usually :-P does. Then I alter hdtoolbox' settings to search 2nd.scsi.device (scsi) only, and it shows no drives at all...not even the cd-rom. But it works. If i then try to access the ide disk, it freezes+led flash. =>this is straaange! with scsi hdd connected, ide access freezes the machine? humm. this also happens if i try without any ide hdd's connected, just by searching the scsi.device in hdtoolbox.

As for different hdd's, the two I have tested work great on my 1200 and on my xp machine. huh.

[Morley]

@markus

I have tried with and without fast-scsi, without parity check, termpower and synchronous, and all combinations of these:)

The strange thing though is that the cd-rom only works when the term switch on the disk/controller card is set to off? I thought this turned the internal termination on and off, and now it doesn't make sense at all.
Maybe the drives have a problem with controllers with other id's than 7? I think the Adaptec card in my pecee and the Blizzard1234 scsi-card is set to id7. Just a thought, as exeperimenting with the id-jumpers on the controller doesn't do anything.

As for the termination on the controller card, cn260 and cn261 are empty, but this is for the floppy i guess. The controller seems fully poulated otherwise.

[Morley]

@markus

hm... if the scsi id of the controller is 7, and then fixed, my guess,  - then what are the scsi address jumpers on the back off the controller for? I thought they were for setting the id for the controller

[Morley]

Oh my God! This is embarrassing :-P
I have been using Scsi on my Amiga's since '96 and I can't get a hdd to work in my tower? I'll rather work for 3 months instead of getting another scsi-controller. Anyway, I think I'll borrow some old 50 pin scsi drives tomorrow to test with my tower, if I can find any :-? .

Just tried using the active terminator on the io board instead of the passive on my cd-rom (i also have a looase active terminator which fits on the cable), but this didn't help either. But now I can leave the Term switch on the controller on when i use the cdrom.

[Morley]

Well, I have just ordererd a new disk I/O module, new A/V module and a new C= 68030 cpu-card from Vesalia. If they have it in stock that is, which I believe they have since the items were on some sort of clearing-out sale.
Anybody know how long it takes from ordering to Vesalia sends a order-receipt back?

With a shining new I/O module SCSI will be proper again I hope!

..oh! the price? 30€ total! :-D
 :banana:

[Morley]

AAARGH {bleep}{bleep} silly computer!

NOW! I'm angry.

Just got my new SCSI/floppy module and some 3200/3400 boards from Vesalia. Well, I thought maybe the SCSI module or the 3640 board I'm using could be the problem, so I bought theese.
So. Inserted the new SCSI module, plugged a HDD onto the SCSI bus, power-on -> nothing. Just as before. Then  I threw in another cpu-card.. -> same result. The machine starts up just fine, initializes the SCSI disk, then it hangs and the drive led just flashes once a second. No kickstart screen, just a black screen. It's not completely locked, I can still do soft-resets which hangs the machine again after initializing the disks. Hmpf. I have tried all sorts of combinations on the jumpers at the back, 5(!) different SCSI hdd's, all with same result. The strange thing though is that it works just fine with a SCSI cd-rom and an IDE hdd...huh? Only thing that makes it cough-up is SCSI hdd's...

Somebody have any good suggestions? I really can't figure this problem out...

[Morley]

 :bump:

Well, today I experienced something odd... Itried to plug a 9.1GB Western Digital 80pin SCSI drive through an adaptor in my 4000T, the drive was'nt recognised, but at least the machine booted into workbench for the first time with a SCSI hard-drive connected. I then gave up the 80 pin drive, and connected another 68 pin drive I had lying around, which did'nt cooperate very well in my 4000T. (see previous postings in thread)

Guess what? Now it is found by hdtoolbox and works like a charm. I haven't touched any settings from the last time I tested it, just threw it in...
And now it works. Maybe my 4000T was so glad I gave it a scandoubler that it decided to give me access to my scsi hdd's? :-)

[Morley]

darkcoder wrote:
I don't know if it helps...I suspect the A4000T has a special version of workbench 3.1. I guess this is because of a special version of scsi.device, which should be able to make work both the onboard IDE and SCSI controller.
My 4000T came with WB disks labeled "3.1 4000T".

cheers

Cheers!

Yep, I know the 4000T had a special version of WB. Infact workbench.library was on disk instead of in ROM because of the second scsi.device.

This was not the problem though, my A4000T disliked all scsi hdd's. I had a scsi cd-rom working fine along with two IDE hdd's, but adding a scsi hdd always made the machine choke. But today, wohoo! :-D  it works perfectly?! And that is what is strange, because I haven't touched anything since last time I tried to connect a SCSI hdd.

Humm...haunted computer?