Jump to content

Arduino SCSI device - Work in Progress


Recommended Posts

  • Replies 237
  • Created
  • Last Reply

Top Posters In This Topic

  • 68kMLA Supporter
On 10/28/2020 at 7:53 AM, Chopsticks said:

also might be a handy idea to add a standard HDD power connector to your pcb

 

On 10/28/2020 at 2:03 PM, erichelgeson said:

There is no need, it is powered by the SCSI bus.

 

I have had issues with some of "the competition" (as it were) on some machines where the SCSI bus doesn't seem to provide quite enough juice, even though it ought; there are a couple of models that seem particularly prone to this.  So it may be worth putting one on if room can be found, perhaps even a small floppy-style one.

Link to post
Share on other sites
5 minutes ago, cheesestraws said:

 

 

I have had issues with some of "the competition" (as it were) on some machines where the SCSI bus doesn't seem to provide quite enough juice, even though it ought; there are a couple of models that seem particularly prone to this.  So it may be worth putting one on if room can be found, perhaps even a small floppy-style one.

interesting, i should probably, check how much current the se/30 can provided on the scsi 5v termination line compared to the current draw of this hdd emulator, as well as the current draw of the microsdcard i was using.

perhaps its not a problem on more 'modern' vintage macs?

Link to post
Share on other sites
  • 68kMLA Supporter
Just now, Chopsticks said:

perhaps its not a problem on more 'modern' vintage macs?

 

I don't think it's even that; my suspicion is that it's just slightly borderline design on a couple of models.  Earlier ones and later ones both seem to work.  Though I haven't looked into this scientifically other than sighing and finding workarounds :-), so take this with a pinch of salt.

Link to post
Share on other sites
2 minutes ago, cheesestraws said:

 

I don't think it's even that; my suspicion is that it's just slightly borderline design on a couple of models.  Earlier ones and later ones both seem to work.  Though I haven't looked into this scientifically other than sighing and finding workarounds :-), so take this with a pinch of salt.

thats interesting to know, i guess maybe the se/30 is one of those borderline designs?
 

i gave up on this device awhile ago anyways, it was to slow compared to the hdd i have in my mac currently, still considering how cheap it is to buy the micro and the pcb im sure this is a great option for many people

Link to post
Share on other sites
3 hours ago, cheesestraws said:

 

I don't think it's even that; my suspicion is that it's just slightly borderline design on a couple of models.  Earlier ones and later ones both seem to work.  Though I haven't looked into this scientifically other than sighing and finding workarounds :-), so take this with a pinch of salt.

It may not even be a matter of current. I noticed while poking around with my SE/30 that there appears to be a Schottky diode drop on the SCSI termination power rail, resulting in a voltage of about 4.6V iirc. This makes sense, as you don't want another device on the SCSI chain also providing term power (like my CD-ROM does) "fighting" with the power supply. This might just be marginal enough even with the low dropout of the regulator on the Blue Pill that the processor isn't getting enough voltage. And I imagine some Blue Pills have better LDOs than others, considering everybody makes them.

Link to post
Share on other sites
On 10/26/2020 at 10:16 PM, erichelgeson said:

Got the mouser order tonight - BlueSCSI 1.0-a fully assembled - tested termination - works!

Sweet! Is this considered "ready to go" and should I order some boards from your plans or are you going to do another rev or two before you call it finished?

Link to post
Share on other sites
2 hours ago, nyankat said:

It may not even be a matter of current. I noticed while poking around with my SE/30 that there appears to be a Schottky diode drop on the SCSI termination power rail, resulting in a voltage of about 4.6V iirc. This makes sense, as you don't want another device on the SCSI chain also providing term power (like my CD-ROM does) "fighting" with the power supply. This might just be marginal enough even with the low dropout of the regulator on the Blue Pill that the processor isn't getting enough voltage. And I imagine some Blue Pills have better LDOs than others, considering everybody makes them.

hmm that's likely it then, as 4.6v is on quite the low end of the scale and it probably means that even if the stm chip turns on it would be triggering brown out detection and constantly reseting itself

 

does anyone know just how many Macs are effected by this? as if its only a few models it probably doesn't matter but if it is common to many models it would probably be a good idea to add a power connected to the board with a. 3 pin header to allow people to select either termination power or power from an external source through the power connector (hope that makes sense), the small increase if PCB footprint size would barely have any effect on the overall cost of getting boards manufactured so its probably something worth exploring?

also these bluepill boards can be overclocked a bit and I wonder if that would improve the overall speed of scsi transfers?

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

check how much current the se/30 can provided on the scsi 5v termination line

Please do and let me know - I guess all of my machines provide a good 5v - but if that's not common (and probably with 30 year old machines it varies) it would be a good idea to have a 5v line. An option is they do make molex to usb to provide 5v, but they're a bit hard to source. I'll look at the cost and layout.

 

10 hours ago, Chopsticks said:

also these bluepill boards can be overclocked a bit and I wonder if that would improve the overall speed of scsi transfers?

This is something I'll try this weekend, I've noticed the option, just never selected it.

 

12 hours ago, nyankat said:

Is this considered "ready to go"

The scsi termination works well. Would just adding niceties to it now - eg: 5v Power, activity LED pin out, fix placement of debug pins (not enough room), likely remove fuse, remove 2nd hole as there is not enough room to use it, 3d print a bracket. But all that said it works well and I'm testing/using this rev (and ztto's rev) daily. If you're comfortable figuring out any issues you run into yourself 1.0-a is ok for you. If you want a set it and forget it - let me think on a few of these items - likely a 1.0-b will be happening soon.

 

Since there is quite a bit of demand I'm considering doing an order of 100 (shipping really kills the deal for small batches) I do have 2 1.0-a's and 3 ztto v2's if anyone wants a blank now - with the expectation of feedback.

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

Please do and let me know - I guess all of my machines provide a good 5v - but if that's not common (and probably with 30 year old machines it varies) it would be a good idea to have a 5v line

 

FWIW, on a couple of IIsi boards I have here, putting any actual load on the power seems to drop the voltage to about 2.5v.  I really think a 5v in is a good idea if there's room for it.

Link to post
Share on other sites
  • 68kMLA Supporter

Re power, might it be best to add a floppy power connector to the PCB design, but leave it unpopulated? That would take little space on the board and wouldn’t change the BOM, but if anyone had issues using term power it would be trivial to solder in the connector.

 

also, I’m definitely in for a few PCBs.

Edited by Michael_b
Link to post
Share on other sites

well with a passive terminator the the stm bluepill l uses 140ma of current on a 5v volt rail. but I don't seem to need to use scsi termination when its connected to my se/30. i wouldn't be surprised if the bluepill has built in internal pull-ups on the lines used for the scsi interface so perhaps its not needed, or maybe just not needed on the se/30? 

 

also I decided to overclock the bluepill and was able to get a subtle improvement in transfer speeds as shown below. its about 100KB faster on read and about 40kB faster on write speeds. everything is the same but with the overclock option enabled before flashing so its going from a stock clock speed of 72mhz to 128mhz

 

[edit] forgot to mention to that when I don't power the bluepill external and try to use the 5v termination power from the se/30 that 5v coming into the bluepill shows 1.3v

 

IMG_0325.thumb.jpg.8e891c9d0582869aa29066c107a2d624.jpg

Edited by Chopsticks
Link to post
Share on other sites
12 hours ago, erichelgeson said:

The scsi termination works well. Would just adding niceties to it now - eg: 5v Power, activity LED pin out, fix placement of debug pins (not enough room), likely remove fuse, remove 2nd hole as there is not enough room to use it, 3d print a bracket. But all that said it works well and I'm testing/using this rev (and ztto's rev) daily. If you're comfortable figuring out any issues you run into yourself 1.0-a is ok for you. If you want a set it and forget it - let me think on a few of these items - likely a 1.0-b will be happening soon.

 

Since there is quite a bit of demand I'm considering doing an order of 100 (shipping really kills the deal for small batches) I do have 2 1.0-a's and 3 ztto v2's if anyone wants a blank now - with the expectation of feedback.

Remember I built one of these myself without any circuit board and before I found this thread, I'm quite comfortable figuring out any issues that come up. :) Put me down for at least three boards, please.

Link to post
Share on other sites
On 10/30/2020 at 9:51 PM, Chopsticks said:

i wouldn't be surprised if the bluepill has built in internal pull-ups on the lines used for the scsi interface so perhaps its not needed, or maybe just not needed on the se/30?

 

Hacks territory there, the internal pull-ups are not low resistance like the SCSI bus terminators so technically the bus isn't "in spec" if you go that route, but I'd vouch I personally prefer that hack if it works because it also negates the need for transceivers with high-current sinking capability.  And, therefore, you cut most of the cost of RaSCSI.

 

And actually, that's a good point to mention, looks like this is using "direct connection" to the SCSI bus and I wouldn't think the STM32 is rated for 48mA on each I/O pin.  And it isn't, I checked the datasheet and it says 25mA as the "absolute maximum."

 

Note that 48mA is the maximum rating in the SCSI spec assuming a bus has two passive terminators on each end, which supply 23mA each.  Of course our typical Macintosh SCSI buses fare well with only a single passive terminator.

Link to post
Share on other sites
53 minutes ago, quorten said:

 

Hacks territory there, the internal pull-ups are not low resistance like the SCSI bus terminators so technically the bus isn't "in spec" if you go that route, but I'd vouch I personally prefer that hack if it works because it also negates the need for transceivers with high-current sinking capability.  And, therefore, you cut most of the cost of RaSCSI.

 

And actually, that's a good point to mention, looks like this is using "direct connection" to the SCSI bus and I wouldn't think the STM32 is rated for 48mA on each I/O pin.  And it isn't, I checked the datasheet and it says 25mA as the "absolute maximum."

 

Note that 48mA is the maximum rating in the SCSI spec assuming a bus has two passive terminators on each end, which supply 23mA each.  Of course our typical Macintosh SCSI buses fare well with only a single passive terminator.

that's really interesting, yeah I haven't looked at any specs on current sinking required on the scsi bus for termination. though if the stm32 is rated at a max of 25ma then I would imagine that some 245's or something on the IO lines is probably needed for long term reliably. again though I haven't really looked into it. like even the overclocking of the stm32 chip while listed as an option in the Arduino IDE may or may not be a good thing. Im not really sure what the max operating specs are of these devices tbh

Link to post
Share on other sites
On 10/28/2020 at 2:27 PM, Chopsticks said:

there must be something wrong with my se/30 then as i couldn't get it to run using the +5v from the scsi bus

I have been testing a MacSD card and found it wouldn't turn on connected to an SE/30 due to a blown 1A fuse (F3). Replacing the fuse fixed the issue. I wonder if the term powered devices strain that fuse since they are aged and this is another component that may need replacement like capacitors.

Link to post
Share on other sites
6 hours ago, tt said:

I have been testing a MacSD card and found it wouldn't turn on connected to an SE/30 due to a blown 1A fuse (F3). Replacing the fuse fixed the issue. I wonder if the term powered devices strain that fuse since they are aged and this is another component that may need replacement like capacitors.

I don't think that's my issue, I recently did a 100% full restore of my se/30 replacing all the caps on the psu/analogue board/logic board, as well ass replacing the SCC chip and repairing a couple wires. I also replaced the Simm ROM socket so I've been over my Mac extremely thoroughly and if it was as blown fuse then I wouldn't get 5v on the termination pin (25 of IDC header I think from memory, I may be wrong on that atm though)  when nothing is plugged in. and I also wouldn't the the 1.3v or so when the device is plugged in. im not sure what the current limit is on thermination power but yes is the macSD uses more then it can provide it would blow the fuse or at the very least drag down the voltage like mine is doing

Link to post
Share on other sites
8 hours ago, erichelgeson said:

My test machine is an LC and it works great. Likely kit (with pre flashed arduino) and fully assembled.

how much it will cost? Is it possible shipping to Poland? I am interested kit and fully assembled prize.

 

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

how much it will cost? Is it possible shipping to Poland?

Working on that now - international shipping will be more of course - I just haven't shipped a lot internationally. Cost will be determined on how I can source all the parts - have a spread sheet and working on it now.

 

This weekend I added a floppy power port (though seems to be some conflicting info on if the pitch is 2.54 or 2.5 - though both would probably work), added activity LED pin out, moved the debug port, re-worked the mounting, made the board 1mm wider to not cut off silk screen. Will be testing a few of the power options before actually placing the orders here on non-TERMPOWR macs (like the plus.)

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...