BeebSCSI 7_7

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
User avatar
danielj
Posts: 7497
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester
Contact:

Re: BeebSCSI 7_7

Post by danielj » Sat Aug 03, 2019 1:42 pm

I usually boot dos, load gem and draw a picture of Ian.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Sat Aug 03, 2019 1:50 pm

danielj wrote:
Sat Aug 03, 2019 1:42 pm
I usually boot dos, load gem and draw a picture of Ian.
Well I have just built the random smallymouse2 board you sent me ages ago with the freefi232 boards. So I could give that a test at the same time. Is it important that it is Ian?

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Sat Aug 03, 2019 1:53 pm

I usually boot dos, load gem and draw a picture of Ian.
The Ian-test is a solid one, I use it for testing SmallyMouse2 boards too :)
ian.jpg

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Sat Aug 03, 2019 2:01 pm

I think you should add this test to the wiki page

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 10:13 am

I wanted to use superform as a standalone without using the pre-made scsi disk (and I can currently pull it from here), Beebmaster mentioned there are several versions of superform kicking about, can anyone point me in a link to a download the correct version to use for BeebSCSI?

Has anyone tried using the gotek adfs disk support with beescsi, I assume so.

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 10:19 am

Just download the quick start LUN image from the following link:

https://www.domesday86.com/?page_id=400

You can either a) use the quick start LUN image, mount it and use the superform image to format other LUNs, or b) copy it to a floppy from there (or Gotek) or c) use ADFS explorer to move superform from the LUN image to another ADFS image (of either a harddrive LUN or floppy drive).

BeebSCSI works perfectly well with Goteks (I have several machines set up that way) - combinations of BeebSCSI, real floppies and Goteks also works - the original ADFS can't tell the difference between BeebSCSI and an original Winchester, so there are no compatibility issues.

The correct version of SuperForm is the one supplied with the original Winchester systems (which is the one in the quick-start LUN).

/Simon

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 10:29 am

I cant use do a) currently that is why I need a standalone copy. Hmm no Linux version of ADFS Explorer? And Mac version is broken as it is old. Will have to hope it runs on a windows 10 VM. What happened to the multi-platform version you were working on last year?

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 10:37 am

Doesn't make much sense... if you have a BeebSCSI and an SD card; you just copy the quick-start LUN to the SD card (as per the doc) and boot the BBC. Once you have loaded SuperForm you can even use it to format the current LUN; or you can load it; *SCSIJUKE, change LUN sets completely; then OLD, RUN to run superform and format a completely uninitialised LUN set.

Copying a file from a zip to an SD card can be done on any OS... what specifically is the issue?

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 10:52 am

I didn't want to post the issue until I had exhausted all the possibilities I could think of. The last of which is a superform format via standalone superform. But basically, the beebscsi cant see the quick-start lun, so I want to create a fresh one to see if that works or fails. This previously worked until an upgrade to 2.5 (which may, or may not be relevant) but even downgrading back to 2.4 doesn't seem to help. But it is a very long post of things tried (involving 3xSD Cards, 3xBeebSCSIs, multiple working cables and 2 BBCs) and output from the serial debug, which I dont have time to do until this evening.

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 10:58 am

What error does the Beeb give when you try to access the LUN?

Edit: If you could also link to the BeebSCSI trace, that would be great.
Last edited by simoni on Thu Aug 15, 2019 11:01 am, edited 1 time in total.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 11:02 am

The usual 2c one when it can't find a lun. It will happily use the same SD Card to up or downgrade the firmware. The serial debug (turned on vai the *fx command) matches that as well. As it happens I did export the output, it repeats a lot but basically says:

Code: Select all

SCSI State: Selected by host ID 1
SCSI State: Command phase
SCSI State: Information transfer phase: Command
SCSI Commands: CDB byte 0 decode: Command group 0 (6 bytes) opcode 3
SCSI Commands: Received command operand bytes: 3 0 0 0 0 0
SCSI Commands: REQUESTSENSE command (0x03) received
SCSI Commands: Target LUN = 0
SCSI Commands: Sense bytes = 4
SCSI State: Information transfer phase: Data in
SCSI Commands: Error flagged - sending to host
SCSI State: Status.  Status byte = 0
SCSI State: Information transfer phase: Status
SCSI State: Message In.  Message byte = 0
SCSI State: Information transfer phase: Message in
SCSI State: Bus Free
SCSI State: Information transfer phase: Data out
SCSI State: Selected by host ID 1
SCSI State: Command phase
SCSI State: Information transfer phase: Command
SCSI Commands: CDB byte 0 decode: Command group 0 (6 bytes) opcode 8
SCSI Commands: Received command operand bytes: 8 0 0 0 2 0
SCSI Commands: READ command (0x08) received
SCSI Commands: Target LUN = 0
SCSI Commands: Attempting to Auto-Start LUN (as it is currently STOPped)
File system: filesystemSetLunStatus(): ERROR: No file system mounted - cannot set LUNs to started!
SCSI Commands: Could not auto-start LUN #0
SCSI State: Status.  Status byte = 2
SCSI State: Information transfer phase: Status
SCSI State: Message In.  Message byte = 0
SCSI State: Information transfer phase: Message in
SCSI State: Bus Free
SCSI State: Information transfer phase: Data out

User avatar
hoglet
Posts: 8529
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: BeebSCSI 7_7

Post by hoglet » Thu Aug 15, 2019 11:25 am

I had a similar problem, which turned out to be hardware in nature.

I definitely had this in the logs:

Code: Select all

File system: filesystemSetLunStatus(): ERROR: No file system mounted - cannot set LUNs to started!
My memory is a little hazy on the details, but if I recall there was a dodgy joint on the card sense pin of the SD Card connector, which meant the firmware never detected the card, and never even applied power. At least, that's what I think was the problem.....

Dave
Last edited by hoglet on Thu Aug 15, 2019 11:26 am, edited 1 time in total.

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 11:42 am

It's almost certainly hardware related; only the snippet of the log file misses the important bit (which is from when the SD card is detected).

I'm almost certain that the card detect switch isn't soldered properly; however, if you could post the log including the bit when the BeebSCSI starts (or just do a CTRL-break) there will be a section where it looks for the card - that's the important part here and it clearly states if a card is present. If you are unsure what to post; just post everything the BeebSCSI trace gives during and after a CTRL-break.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 12:18 pm

hoglet wrote:
Thu Aug 15, 2019 11:25 am
I had a similar problem, which turned out to be hardware in nature.

I definitely had this in the logs:

Code: Select all

File system: filesystemSetLunStatus(): ERROR: No file system mounted - cannot set LUNs to started!
My memory is a little hazy on the details, but if I recall there was a dodgy joint on the card sense pin of the SD Card connector, which meant the firmware never detected the card, and never even applied power. At least, that's what I think was the problem.....

Dave
Is that SDCARD_CD/DET a pin 10 going to 15 on the Atmel?

https://www.domesday86.com/wp-content/u ... Page_5.jpg

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 12:28 pm

simoni wrote:
Thu Aug 15, 2019 11:42 am
It's almost certainly hardware related; only the snippet of the log file misses the important bit (which is from when the SD card is detected).

I'm almost certain that the card detect switch isn't soldered properly; however, if you could post the log including the bit when the BeebSCSI starts (or just do a CTRL-break) there will be a section where it looks for the card - that's the important part here and it clearly states if a card is present. If you are unsure what to post; just post everything the BeebSCSI trace gives during and after a CTRL-break.
That is all I wrote out to the capture file at the time. Odd as that card was working when I last used it, dry joint maybe.

So this would be G3/G4, no surprise as they are pains to test.

Edit: on the Molex diagram they call them g1-g4 but the beebscsi schematic has pins 9-11. I assume 11 is g1 (and maybe g2 as well), but not sure how 9 & 10 are mapped to g2/g3.
Edit2: Interestingly (maybe) on most SD CARD designs they seems to ground G1-G4. I am guessing in this case everything is grounded except for the single g pin (either g3 or g4), which goes to pin 15 on ATMEL
Attachments
molex_sdcard.png
Last edited by Elminster on Thu Aug 15, 2019 12:57 pm, edited 2 times in total.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 1:09 pm

Pin 15 on ATMEL buzzes out to what Molex call G4 (which I assume is pin 10 on Beebscsi schematic). Neither is shorted to ground.

Testing G2 & G3 are a bit more tricky, I get a ground but hard to tell if one of them is perhaps not grounded as they are underneath

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 1:24 pm

simoni wrote:
Thu Aug 15, 2019 11:42 am

I'm almost certain that the card detect switch isn't soldered properly; however, if you could post the log including the bit when the BeebSCSI starts (or just do a CTRL-break) there will be a section where it looks for the card - that's the important part here and it clearly states if a card is present. If you are unsure what to post; just post everything the BeebSCSI trace gives during and after a CTRL-break.
I know the bit you mean, can't remember what it said, will need to redo a capture this evening.

I checked continuity and shorts of all pins and components before programming the board, and this one was working. Not impossible there is a cold joint, but pin 10<->15 okay at the moment.
Last edited by Elminster on Thu Aug 15, 2019 1:24 pm, edited 1 time in total.

User avatar
danielj
Posts: 7497
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester
Contact:

Re: BeebSCSI 7_7

Post by danielj » Thu Aug 15, 2019 1:30 pm

Just check that the relevant pin on the atmel is grounded when you insert a card. If not then the card holder isn't soldered properly. The connection between the holder and atmel being present doesn't rule it out.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 1:33 pm

danielj wrote:
Thu Aug 15, 2019 1:30 pm
Just check that the relevant pin on the atmel is grounded when you insert a card. If not then the card holder isn't soldered properly. The connection between the holder and atmel being present doesn't rule it out.
Which if that was the case would suggest that G3 was not grounded, which also happens to be the hardest pad on the board to test.
Last edited by Elminster on Thu Aug 15, 2019 1:36 pm, edited 2 times in total.

User avatar
danielj
Posts: 7497
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester
Contact:

Re: BeebSCSI 7_7

Post by danielj » Thu Aug 15, 2019 1:34 pm

Don't speculate, test! (that ground is hard to solder, I usually use hot air on the entire thing).
Last edited by danielj on Thu Aug 15, 2019 1:35 pm, edited 1 time in total.

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 1:34 pm

I am not speculating. I already know the answer as I tested that earlier.

User avatar
danielj
Posts: 7497
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester
Contact:

Re: BeebSCSI 7_7

Post by danielj » Thu Aug 15, 2019 1:36 pm

So the pin on the atmel does ground when you insert a card?

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 1:38 pm

It does not. Hence suggesting that for whatever reason G3 has gone cold. And it is a pain to test as it is hard to touch it without touching the shield, which is definitely grounded. Will try to reflow it and report back.

It isn't that it not soldered that is odd, it is that it was working. But it must have gone cold. This is the one which I built last year, not the new ones. I blame it on the record-breaking hot weather we had a week or two ago.
Last edited by Elminster on Thu Aug 15, 2019 1:49 pm, edited 5 times in total.

User avatar
danielj
Posts: 7497
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester
Contact:

Re: BeebSCSI 7_7

Post by danielj » Thu Aug 15, 2019 2:09 pm

All it takes is for it to be marginal to begin with and then any slight movement on the joint will break it (e.g. inserting a card).

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 2:12 pm

It does looks that way. Time to get out the moon tape (well this evening anyway).

Ironically I said way back in the thread

viewtopic.php?f=3&t=15139&start=30#p215362

when I originally got the card working that the sdcard was the hardest thing on the whole board to do, looks like it is still causing me issues.

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 2:14 pm

As I already mentioned; you can simply CTRL+break the Beeb with the trace on. The trace will state if the card is detected. If it's not; there's your hardware fault. You don't even need a multimeter to know (although it's a pretty easy test as you have a schematic for the board and it shows that the CD pad is connected to pin 15 of the AVR - just test there and insert/remove the card to see if the line grounds and disconnects).

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 2:19 pm

simoni wrote:
Thu Aug 15, 2019 2:14 pm
As I already mentioned; you can simply CTRL+break the Beeb with the trace on. The trace will state if the card is detected. If it's not; there's your hardware fault. You don't even need a multimeter to know (although it's a pretty easy test as you have a schematic for the board and it shows that the CD pad is connected to pin 15 of the AVR - just test there and insert/remove the card to see if the line grounds and disconnects).
I think you missed a few posts. Pin 15 is correctly connected to the SD card, but it isn't going to ground with a card in, so looks like G3 that should be grounded needs to be reflowed.

Doing the trace isn't an issue but I can't do it ill this evening. Although probably not worth it as looks like a reflow due to a cold joint is required.

Thanks all.

User avatar
BeebMaster
Posts: 2867
Joined: Sun Aug 02, 2009 4:59 pm
Location: Lost in the BeebVault!
Contact:

Re: BeebSCSI 7_7

Post by BeebMaster » Thu Aug 15, 2019 2:35 pm

I wonder if a dry joint or similar hardware fault is the cause of the problem I've been having with &2C errors. Unfortunately I don't have the facility to the do the SCSI trace.
Image

User avatar
Elminster
Posts: 3891
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: BeebSCSI 7_7

Post by Elminster » Thu Aug 15, 2019 2:37 pm

BeebMaster wrote:
Thu Aug 15, 2019 2:35 pm
I wonder if a dry joint or similar hardware fault is the cause of the problem I've been having with &2C errors. Unfortunately I don't have the facility to the do the SCSI trace.
If same issue as me you can test with multimeter.

User avatar
simoni
Posts: 484
Joined: Wed May 25, 2016 6:18 pm
Contact:

Re: BeebSCSI 7_7

Post by simoni » Thu Aug 15, 2019 2:41 pm

Didn't miss anything; I was just explaining that the test was simple to do (and that; if you have issues - reading the trace will tell you what they are; saving a whole bunch of speculation and time).

BeebMaster - There are two common construction faults - one is the soldering around the SD card (which I've seen a few times) and the other is the soldering around the resistor networks on the 1MHz bus termination. A bad SD usually causes an permanent fault where you simply can't mount a LUN. Soldering issues around the resistors usually result in intermittent communication failures over the bus (it will work fine except under load (which is what the speed test BASIC program can help with)). Just a couple of pointers :) But, if you are building boards, then a serial TTL adaptor is a wise investment of a couple of pounds to speed things up.

Post Reply