Jump to content

Arduino SCSI device - Work in Progress


Recommended Posts

Scratching my head over it, I had imagined that i could source low profile pass-through female headers in a 2.54mm pitch, but i sure don't see them for sale through the usual sources. 

 

Maybe more realistic to lay out a version that uses a 20 lead ribbon cable to connect to a pin strip soldered directly to the scsi chip, or perhaps a short chunk of old 40 pin PATA cable, with half the pins grounded at the 40 pin header. 

Link to post
Share on other sites
  • Replies 273
  • Created
  • Last Reply

Top Posters In This Topic

I have built a poor man blue SCSI and have problems with it. When I connect it to my SE/30, the drive is not found. Lido does not show the drive. The bluepill blinks 5 times, which probably means that it does not see the SD card. I changed the speed to 18 MHz, with the same result. What can I do?

Edited by dochilli
Link to post
Share on other sites
  • 68kMLA Supporter
21 hours ago, dochilli said:

The bluepill blinks 5 times, which probably means that it does not see the SD card.

 

Try some different sd cards. Not all sd cards support the interface used by bluescsi. If they all fail, check the soldering on the sd card slot.

Link to post
Share on other sites
  • 68kMLA Supporter
Posted (edited)

FWIW to get my cards to work I had to format them with diskpart—quick Explorer formatting would give me blinking on the BlueSCSI. I managed just fine with exFAT, though I know some have said it wouldn't work for them.

Edited by jessenator
Link to post
Share on other sites

Ok, I used two different SD cards but had no success. I formated FAT32 and exFat without success. I used different filenames (HD20_512, HD31_512), but nothing worked. I double checked the cables and all are connected correct. Then I changed the code of the program because I wanted to see if the bluepill shows the change. I inserted a 3s delay before the SD card is checked and the LED blinks and the bluepill shows this delay. So the bluepill and the programming seems to work.

Then I measured the voltage at pin 25 (TPWR) of the DB25 and got 4,0 to 4,4 V. I tried this at two different macs (SE/30, Classic) and no one showed 5V. Is this voltage to low? The SD card reader is connected with the 5V pin. Should I use the 3.3V? But perhaps 4.4V may damage the reader.

Edited by dochilli
Link to post
Share on other sites
15 hours ago, dochilli said:

The SD card reader is connected with the 5V pin. Should I use the 3.3V? But perhaps 4.4V may damage the reader.

 

What SD card adapter are you using?

 

SD cards are supposed to run from 3.3V, so unless your adapter has an onboard regulator you may damage the card.

 

If it does have a regulator then your 4.0-4.4V may be below the dropout voltage. Also it will have either voltage dividers or level shifters on the data pins, either way this would be messing with the levels between the uC and SD card.

 

I recommend soldering pins to a spare MicroSD to SD adapter and using that directly on 3.3V from the BluePill.

I'd also suggest powering the BluePill from USB and see if the LED stops blinking and if there is a log file generated. You should be able to verify R/W card access before plugging it in to the SCSI.

Link to post
Share on other sites

I use the same (blue) SD card adapter as shown by Dav before. It has voltage regulators on the board. He got it to work with the 5V pin.

I have many SD to micro SD adapters, but I do not know which pins I have to use. Can someone tell me the pins for a connection to the poor mans blue Scsi? Then I will try this way.

When I use the USB for powering the blue pill blinks too.

 

Edited by dochilli
Link to post
Share on other sites

Success! With the SD to micro SD Adapter the blinking stops. The problem was the blue SD Card reader. I have a logfile on the SD! My SE/30 mounts the drive and everything works! Thank you for your help!

Link to post
Share on other sites
  • 68kMLA Supporter
1 hour ago, dochilli said:

Here is a photo of my „quick and dirty“ design...

 

Oh man, that's awesome.  I'm guessing the cables are short enough termination isn't needed?

Link to post
Share on other sites
1 minute ago, Scott Squires said:

You're just here to show off your 128k/512k running System 7.5+. :)

 

It's a Plus actually, with 4mb of ram. And someone hacked in what i am reasonably sure is a composite video output. 

 

Machine was university surplus, maybe it was an instructor machine hooked up to a projector. 

Link to post
Share on other sites

This is just crazy...thank you so much (erichelgeson and all of you) for figuring this out and posting. I ordered  a board from androda and can verify it works on a Powerbook 180c. I had to change to a different sd card other than that it appears fine. I'm using a PNY 16GB/Go micro SD on which I put 2 disk images (250 MB at SCSI 0 and 500 MB at SCSI 1). Used an external SCSI drive to boot to System 7.5 (no particular reason 7.5, its just what I had ready to go) to install System 7 Pro (7.1.1) on the 250 MB internal disk). The sound of silence...the mechanical drive was grinding.  Thank you again.

Link to post
Share on other sites
  • 68kMLA Supporter

@dalek

I tried a black pill board, and it's a bit of a mess.  I have the code ported and it's working but the black pill at 100MHz is *slower* than the blue pill at 72.  At full 100MHz clock, the black pill gets 720k read and 700k write, where the blue pill gets 900k read and about 600k write.  It's ridiculous that the newer chip with higher clock speed is slower.  Black pill modules are also about 50% more than the blue pill modules and the global chip shortage just keeps driving prices up.

 

Talked with Bill Greiman, the maintainer of SDFat (the SD library we use) and he said it's no surprise that the black pill is slower.  In the switch to running on the F411 I had to use the official ST framework because the Maple framework (used on blue pill) was throwing a huge fit and not compiling.  Bill says that the ST framework's SPI implementation is trash.  The Maple implementation is far superior, as we can clearly see.

 

I have also spent some time trying to get maple working on the F411, might be close to something there.  Although the code compiles and runs, it won't see the SD card.  So many layers to debug.

 

Another option I'm considering is to order some really high-clock (170mhz +) STM chips and see what kind of speed those get.  If Maple won't play nice and the ST framework sucks then "fix it with raw clock speed".  I can do the full board design to use the bare chip, but that drives prices up too because we then have to also supply the 3.3v regulator.

Link to post
Share on other sites
  • 5 weeks later...
  • 68kMLA Supporter

While we dont have CD emulation yet a user on discord discovered you can make and ISO into a HDD with a few simple steps - I installed A/UX 3 on my SE/30 this way. A workable interim solution to get something like the OS Legacy CD or other CD installers on your mac and install. https://github.com/erichelgeson/BlueSCSI/wiki/Mac-CD-to-HDD-(Hack!)

Link to post
Share on other sites
  • 68kMLA Supporter
14 hours ago, erichelgeson said:

While we dont have CD emulation yet a user on discord discovered you can make and ISO into a HDD with a few simple steps - I installed A/UX 3 on my SE/30 this way. A workable interim solution to get something like the OS Legacy CD or other CD installers on your mac and install. https://github.com/erichelgeson/BlueSCSI/wiki/Mac-CD-to-HDD-(Hack!)

 

Woohoo!  Yeah, this approach works well; that's how I install A/UX (not using a bluescsi, but another option) here, too.  I've been procrastinating writing this up for well over two years at this point and now someone else has done it, and better than I would have :D.  Score one for procrastination.

Link to post
Share on other sites

I made a PowerBook version BlueSCSI PCB. I will update on it, if everything works out and it tests fine you can buy an assembled unit from me.

I had no idea androda already made one, well, too late, already ordered the PCBs :shocked:.

I'm going to open source everything once I have tested it.

scsiPB.png

Link to post
Share on other sites
  • 68kMLA Supporter
Posted (edited)
11 hours ago, dehij said:

I had no idea androda already made one

That's generally the reaction I get after bringing it up.

It's listed on https://scsi.blue/ (at the bottom), my site (https://androda.work/shop/), and my github repo (https://github.com/androda/PCBDesigns/tree/main/BlueSCSI).

Edited by androda
Link to post
Share on other sites
  • 68kMLA Supporter

Had the day off so did some testing on PowerPC - what I/we though the whole time was it was some software incompatibility - well turns out it was power! The TERMPOWER on my 6100 was low and would work if I put it on a very short cable, but wasnt enough for the pullup termination. Just plugged in a molex to berg connector and it works on the long cable. Gah! I wish I would have tried this sooner. Sometimes you get an idea in your head the bug lies in software when it's really something else. I'll be updating the wiki with more info.

 

Thanks to @androda and @xunker for debugging with me on discord this AM.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...