Macintosh Classic stalls booting indefinitely when anyting is plugged into scsi bus

23 posts / 0 new
Last post
Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
Macintosh Classic stalls booting indefinitely when anyting is plugged into scsi bus

I have a MAC classic board which will boot and workfine from a floppy drive but the second I connect a hard drive via internal ribbon or external connector the system just sits forever without progressing to even the floppy drive. Is the SCSI chip toast or could it be something else? I did try the ROM from another Classic board and it made no difference.

Tom Owad's picture
Offline
Last seen: 2 days 4 hours ago
Joined: Dec 16 2003 - 15:14
Posts: 3146
Make sure you have SCSI

Make sure you have SCSI terminators (resistors) on the drive. If you already do, then play around with different termination and drive combinations. 

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
Yeah Tom I already tried that

Yeah Tom I already tried that. Lots of cables and terminators and The same hardware and cables works fine on another machine. While its scanning the SCSI bus its supposed to move on or timeout eventuallty right? Its not meant to stay there for an hours and a half?!

Tom Owad's picture
Offline
Last seen: 2 days 4 hours ago
Joined: Dec 16 2003 - 15:14
Posts: 3146
No, it sounds like it's

No, it sounds like it's freezing. Does it freeze even if there's no power to the drive?

I'd try probing the bus with a tool like SCSIprobe or similar. Can it probe the bus and does it see the Mac on it?

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
Well its not really freezing

Well its not really freezing as the mouse cursor still can move. ITs more like stalling. Again, if nothing is plugged into the SCSI bus the machines bypasses and moved onto the floppy drive. Even unpowered devices still plugged into the scsi buss cause the hang.

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
Termination
Verault wrote:

Even unpowered devices still plugged into the scsi buss cause the hang.

Unpowered devices should have a high impedance to the SCSI signal lines (not affect them). It sounds as if there is a problem with termination even though you used a terminator, because there is no power for the terminator to work. Is the termpower fuse blown? I think that's F1 on the Classic. Do you have any of the active terminators with a barrel jack for power?

With no functioning termination, there can be ringing such that when the initiator probes for devices on the bus, it receives noise and goes into an invalid state. 

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
NO I dont own a powered

No I dont own a powered terminator. Do we have a schematic of the Classic to show me exactly which fuse is related to termination? 

i believe there are 3 fuses on the board.

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
What about an active

What about an active terminator with a LED? The LED is an indicator that the terminator is getting powered from (some device on) the bus.

F1 is just the fuse closest to the SCSI port, I didn't check a schematic.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
No Im afraid I dont even have

No Im afraid I dont even have a terminator with an LED either. So I think maybe I should test these board fuses. IS there a schematic?

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
BOMARC

There are cleaned scans of the BOMARC schematics here: https://archive.org/download/Macintosh68kSchematics

The Classic is one of them. It shows that termpower is supplied through diode D3, and bypass caps C7 and C52, and the external port protected by fuse F1. The internal port is unfused.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
if the internal port in

if the internal port in unfused that negates the whole issue of the termination being culprit then right?

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
A fault could also be in the

A fault could also be in the diode or capacitors

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
the board has been recapped.

the board has been recapped. I can test the diodes, thats easy enough. What diode(s) is associated with the SCSI bus?

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
The diode is D3, directly

The diode is D3, directly next to fuse F1.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
thanks Robespierre, I tested

thanks Robespierre, I tested F1 and F2 ( I dont know why I thought there were 3?), both fuses are fine. Diode D3 is also behaving properly.

 

I found a seller on ebay with a SCSI terminator with diag leds, I am trying to come to an agreed price with them.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
I bought this to help

I bought this to help diagnose the SCSI issue: https://www.ebay.com/itm/373673811451

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
Power Sniffer

I have not seen that device before, but it looks like what you need.

The (one sheet) instructions are at SCSI Power Sniffer

To check termination power, the TRM light should be on. When the Classic probes the bus you should see SEL blink.

Note that although the instructions talk about hot-plugging devices, you still need to use extreme caution.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
Glad its the right thing. I

Glad its the right thing. I will post back when I get it. Thank you for the link to the pdf.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
Ok so I am at a loss. I just

Ok so I am at a loss. I just go back to this today and using the power sniffer all I see is the term power LED and BSY LED are on and thats it. The Term pwr led is a tad les bright thatn the bsy LED but its not that dim. Im no better off than I was before. What can I do now? This is really disheartening.

 

Im going to take a break from this since I really dont know what to try next besides replacing the SCSI IC. Please anyone with any advice on why the SCSI bus is not working let me know.

 

Im going to work on a very burned Mac II board now.... at least that ones has.. well "Potential", I have had nothing but bad luck with The MAC classics'. I have repaired horribly broken MAC II boards,, but the Classic... seems they just dont want to live. :(

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
The TRM and BSY lights  show

The TRM and BSY lights  show that the Classic is supplying termination power to the bus (good) and attempting arbitration (good). What happens next should be that the host wins arbitration and asserts SEL to probe for target devices. At that point BSY should go out.

What seems to be happening instead is that BSY stays lit? That could mean that there is a short between data lines either in the cables or on the board, or a bad SCSI chip. The host chip asserts BSY and its own ID (data bit 7) and waits reads the bus back to see if any bit is set other than 7, meaning that there was contention and it didn't win arbitration. At which point the cycle repeats.

I should say that the doc for the Sniffer doesn't say how much stretching it does of the signal so I don't know if you can see brief phase transitions that only normally last a microsecond.

Also you want the switch on "safe", that turns on the FPT termination inside the plug.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
I have tried 3 different

I have tried 3 different cables and 3 different terminators and two known good external hard drives and an internal blueSCSI. They all do the same thing. Term led and bsy LED are always on. IF there are any resistors (resistor packs) on the mainboard which could possibly affect this let me know. I have a dead classic board I can salvage a SCSI chip from but I like to try all avenues first. I just dont get it this board is super clean! I cant believe the scsi bus has an issue.

Offline
Last seen: 3 hours 53 min ago
Joined: Feb 27 2021 - 18:59
Posts: 65
There aren't any resistors

There aren't any resistors relevant to the SCSI bus on the Classic board, in fact there is nothing protecting the SCSI chip at all; it connects directly to the internal and external SCSI ports and they are tied together.

So if there is ever any ESD on the SCSI pins (from hot-plugging a cable or device, or other possible causes), the SCSI chip takes the full brunt of it. It's not a robust design.

It also shares the property of other early Mac SCSI ports that there is no on-board termination; the only way both ends of the SCSI bus would be terminated like it's supposed to be is if there are terminators at the end of both internal and external SCSI chains. When only the internal disk is connected, the other end just reflects the signal. This was "normal" and people lived with it despite not being compliant with the spec. When the IIfx came out and required an external terminator, they acted like that was all weird.

Verault's picture
Offline
Last seen: 5 hours 25 min ago
Joined: Dec 21 2017 - 14:19
Posts: 254
so its a fair assumption that

so its a fair assumption that the SCSI chip has failed then if thats how poorly implemented the SCSI bus is on this unit. Well I have needed to buy a hot air rework system. 

 

So the General consensus is bad SCSI chip at this point? Cant be anything else i suspect?

Log in or register to post comments