Jump to content

SCSI to Ethernet Adapter on New Hardware


Recommended Posts

  • Replies 240
  • Created
  • Last Reply

Top Posters In This Topic

I've been running this Daynaport firmware variant for almost a week I think so far and I'm just posting to confirm that it runs 100% fine on my end with my se/30 Mac's. its a bit slower then the nuvolink firmware with transfer speeds in fetch and Appleshare but superjer2000 has told me he thinks its likely the daynaport driver on the Mac that is the bottleneck. ive actually started redesigning the pcb for an internal version withe IDC-50 pin connector should anyone be interested in that.. I'll be putting the design files up on GitHub when that's done so anyone can go build it, I have some future ideas regarding the scuznet that ive been talking to superjer2000 about and I certainly don't want to take the spotlight away from saybur and the incredible work his has done here with the scuznet, but its been great project to fiddle around with and I want to give something back that hopefully some of you can find useful

Link to post
Share on other sites
  • 68kMLA Supporter
Posted (edited)
On 4/7/2021 at 8:53 AM, Chopsticks said:

I've been running this Daynaport firmware variant for almost a week I think so far and I'm just posting to confirm that it runs 100% fine on my end with my se/30 Mac's. its a bit slower then the nuvolink firmware with transfer speeds in fetch and Appleshare but superjer2000 has told me he thinks its likely the daynaport driver on the Mac that is the bottleneck. ive actually started redesigning the pcb for an internal version withe IDC-50 pin connector should anyone be interested in that.. I'll be putting the design files up on GitHub when that's done so anyone can go build it, I have some future ideas regarding the scuznet that ive been talking to superjer2000 about and I certainly don't want to take the spotlight away from saybur and the incredible work his has done here with the scuznet, but its been great project to fiddle around with and I want to give something back that hopefully some of you can find useful

 

It turns out I was wrong about the reason for the lower transfer speeds.

 

There is a marked difference in transfer speeds using the Daynaport firmware (based on GETing a file via FTP with Fetch) depending on whether I write the file to a ScuzNet volume or a separate hard disk.

 

On my SE/30, using a 6MB file, transfer rates stabilize at about 55kB/s to a ScuzNet disk and closer to 90kB/s on my spinning hard disk (on a larger file I topped out closer to 98kB/s with Daynaport over FTP).  AppleTalk (AppleShare) performance shows a similar difference in percentage terms.

 

I ran MacBench and on the Mix test, my Scuznet disk scored fairly well (i.e. 128 or so versus Pmac 6100/60 at 100) but there is a marked difference in copy times with my internal hard disk and Scuznet:  Duplicating a 6MB file with my ScuzNet disk takes about 35 seconds whereas it takes around 10 seconds with my internal disk.

 

What's really weird is other than for these network tests and Finder copying tests, Scuznet's disk performance feels quite snappy in regular usage.  Boot times are completely fine and I don't feel like app load times are really any different so I don't know what to make of this.  Overall, I'm happy with the speed and stability I'm getting so I'm not sure if I'll pursue this much more.  I think I will try running Fetch via ScuzNet on my SE/30 with a SCSI2SD and see if there is any incremental improvement.  One other item that's quite odd - when I was running the original Nuvolink firmware, my transfer speeds were better than with the Daynaport firmware so I'm not sure if there is some kind of odd interaction there.  I did go back to the Nuvolink firmware to test the disk performance of the Scuznet and it seemed about the same as the Daynaport firmware).

 

I have been getting the odd freeze error on shutdown if, and only if, I don't use AppleTalk (i.e. don't mount any shares) and only run a TCP program and then proceed to shutdown.  Every so often the system freezes doing something on the network.  I can force quit the finder, open the AppleTalk control panel (AppleTalk gets disabled by the system prior to this), click on "Don't make Appletalk active when closing the Control Panel", force quitting the finder again and then shutting down.  It seems like the Daynaport driver is sending something out in this particular circumstance that I haven't been able to trap yet but I'm working on that.

Edited by superjer2000
Link to post
Share on other sites
On 1/22/2020 at 6:15 PM, aperezbios said:

I have a batch of Printed Circuit Boards on the way, if anyone would like one.

Late to the party! But at least I found it..

 

if there are any boards left I’d really appreciate one :)

 

supercool project.. can finally get my Mac classic to earn its keep ;)

Link to post
Share on other sites

I finally got some more time to devote to this project and implemented version 2 PCB changes. I've attached a rendered image of the board.

 

One significant change was switching from microSD to full SD. This should simplify soldering of the annoying little connector and make it easier to swap cards with the case put together.

 

There was a discussion a page or two back about how to power the board, whether through USB, termination power, or the current approach. After doing some more research, I landed on keeping power through the barrel jack only. My thinking was:

  • Adding USB requires relying on a "dumb" USB supply or adding proper enumeration support, along with some mechanism to turn off the termination regulator and/or other components to avoid drawing too much power when connected to a computer. I wanted to avoid having to implement this if I could get away with it.
  • USB-C has alternative power detection options through the cable resistors. The analog ports on the microcontroller are blocked and it would have required significant rework to expose them, and I didn't want to add more parts to the board to do it with external circuitry.

I did add a micro USB plug, but the board cannot be powered through it. It is present mainly to help program blank chips with the factory-supplied DFU bootloader, and for future expansion.

 

Other changes:

  • Base schematic updated from Kicad 4 to 5. Footprints were switched to the ones in the new packages.
  • The SI-60062-F Ethernet connector is now quote-required on Digikey. Switched to the more common RJMG1BD3B8K1ANR.
  • The board depth was increased by half a millimeter to account for the case-wiggle in the Hammond 1455L801. Mounting holes were also added per request.
  • The packaging on the power regulators was changed. The 3.3V regulator was oversized, it is now down to SOT-223. The 2.63V termination regulator was switched to TO-220 to simplify soldering and save some board space.
  • A second LED was added to help with diagnostics.
  • Numerous other minor changes, like switching all resistors to 0603 packaging and tweaking the BOM.

I'd love to hear if anyone has suggestions for other things to modify. I'm planning on sending off for prototypes sometime in the next few days, and if things work I'll post the files on Github. If you want them in advance let me know, but be aware they are untested.

scuznet_v2.jpg

Link to post
Share on other sites

Hello. 
Great work being done here!
I’d love to see kind of an hybrid board that can be used both externally and internally. 
It would be great having both:

- internal 50p connection

- possibility of having the network port in a separate small board that could be fitted on the rear expansion port of the SE and SE/30 

 

Link to post
Share on other sites
  • 68kMLA Supporter
21 hours ago, pfuentes69 said:

- possibility of having the network port in a separate small board that could be fitted on the rear expansion port of the SE and SE/30 

 

@PotatoFi 3D printed a small carrier for an RaSCSI that would fit in the rear expansion port of the SE(/30); I'm sure something similar could be made for this too (not volunteering them, just saying it's probably possible!)

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

 

@PotatoFi 3D printed a small carrier for an RaSCSI that would fit in the rear expansion port of the SE(/30); I'm sure something similar could be made for this too (not volunteering them, just saying it's probably possible!)

That could simplify it a lot, for having external access to the RJ45 port. But if that's not possible (because the SCSI connector, maybe) a "daughterboard" should be easy too.

For the rest, I think it should be doable to put the place for the 50p SCSI internal connector and a floppy style power connector. Like this people can choose if soldering the DB25 or the internal port.

 

scuznet_v2.thumb.jpg.98258b9241103640a79397b2706174a9.png

Edited by pfuentes69
Link to post
Share on other sites
On 4/25/2021 at 6:08 AM, saybur said:

I finally got some more time to devote to this project and implemented version 2 PCB changes. I've attached a rendered image of the board.

 

One significant change was switching from microSD to full SD. This should simplify soldering of the annoying little connector and make it easier to swap cards with the case put together.

 

There was a discussion a page or two back about how to power the board, whether through USB, termination power, or the current approach. After doing some more research, I landed on keeping power through the barrel jack only. My thinking was:

  • Adding USB requires relying on a "dumb" USB supply or adding proper enumeration support, along with some mechanism to turn off the termination regulator and/or other components to avoid drawing too much power when connected to a computer. I wanted to avoid having to implement this if I could get away with it.
  • USB-C has alternative power detection options through the cable resistors. The analog ports on the microcontroller are blocked and it would have required significant rework to expose them, and I didn't want to add more parts to the board to do it with external circuitry.

I did add a micro USB plug, but the board cannot be powered through it. It is present mainly to help program blank chips with the factory-supplied DFU bootloader, and for future expansion.

 

Other changes:

  • Base schematic updated from Kicad 4 to 5. Footprints were switched to the ones in the new packages.
  • The SI-60062-F Ethernet connector is now quote-required on Digikey. Switched to the more common RJMG1BD3B8K1ANR.
  • The board depth was increased by half a millimeter to account for the case-wiggle in the Hammond 1455L801. Mounting holes were also added per request.
  • The packaging on the power regulators was changed. The 3.3V regulator was oversized, it is now down to SOT-223. The 2.63V termination regulator was switched to TO-220 to simplify soldering and save some board space.
  • A second LED was added to help with diagnostics.
  • Numerous other minor changes, like switching all resistors to 0603 packaging and tweaking the BOM.

I'd love to hear if anyone has suggestions for other things to modify. I'm planning on sending off for prototypes sometime in the next few days, and if things work I'll post the files on Github. If you want them in advance let me know, but be aware they are untested.

scuznet_v2.jpg

That looks great, I was part way through learning kicad and reworking the board myself but I might scape that and wait till you get some boards made/uploaded to GitHub and start again. I mainly wanted to replace the power with a molex connector and add an internal scsi idc-50 connector as well as make a daughter board for the se/30’s pds slot on the back of the case for the Ethernet jack ( not enough room on the pcb for a either net jack with the internal scsi connector I want)

 

btw I see you have change the Ethernet ports footprint, is there a reason for that I.e. cost or availability?

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

 

 

btw I see you have change the Ethernet ports footprint, is there a reason for that I.e. cost or availability?

 

On 4/24/2021 at 4:08 PM, saybur said:
  • The SI-60062-F Ethernet connector is now quote-required on Digikey. Switched to the more common RJMG1BD3B8K1ANR.

 

Link to post
Share on other sites
On 4/24/2021 at 6:40 PM, pfuentes69 said:

For the rest, I think it should be doable to put the place for the 50p SCSI internal connector and a floppy style power connector. Like this people can choose if soldering the DB25 or the internal port.

 

I think that's a good idea, but unfortunately it doesn't quite fit on the board. I think @Chopsticks is on the right track with a separate all-internal design, moving the ENC28J60 and associated hardware to a daughtercard.

scuznet_idc.JPG

Link to post
Share on other sites
21 hours ago, saybur said:

 

I think that's a good idea, but unfortunately it doesn't quite fit on the board. I think @Chopsticks is on the right track with a separate all-internal design, moving the ENC28J60 and associated hardware to a daughtercard.

scuznet_idc.JPG

From my tests the enc28j60 doesn’t even need to be on a daughter board, I ran a cable from the Ethernet jack socket to a board with the Ethernet connector that’s about 20cm long and noticed no change in reliability or network speeds, using a shielded cable is likely a good idea though

Link to post
Share on other sites
22 hours ago, Chopsticks said:

From my tests the enc28j60 doesn’t even need to be on a daughter board, I ran a cable from the Ethernet jack socket to a board with the Ethernet connector that’s about 20cm long and noticed no change in reliability or network speeds, using a shielded cable is likely a good idea though


Indeed I have a couple of these extensions, like this one...
https://a.aliexpress.com/_mqir4Vx

 

I guess that this extension, plus something like this: https://github.com/Stephen-Arsenault/OverEasy/

could do the trick to mount it internally... if there’s no enough demand to have one done for the SE family. 
 

BTW... it would be great if someone puts this on sale preassembled, at least with the tiny SMD components. 
 

 

 

 

Link to post
Share on other sites
20 hours ago, pfuentes69 said:


Indeed I have a couple of these extensions, like this one...
https://a.aliexpress.com/_mqir4Vx

 

I guess that this extension, plus something like this: https://github.com/Stephen-Arsenault/OverEasy/

could do the trick to mount it internally... if there’s no enough demand to have one done for the SE family. 
 

BTW... it would be great if someone puts this on sale preassembled, at least with the tiny SMD components. 
 

 

 

 

I probably didn’t explain well, what I meant it that i soldered a ribbon cable to the connection points on the scuznet where the lan port would normally go and then I soldered the lan port on another pcb and then soldered the ribbon cable to that. My plan for an internal version for the daughterboard to also have space for a vga connector too etc

Edited by Chopsticks
Link to post
Share on other sites
  • 68kMLA Supporter
On 4/24/2021 at 5:40 PM, pfuentes69 said:

That could simplify it a lot, for having external access to the RJ45 port. But if that's not possible (because the SCSI connector, maybe) a "daughterboard" should be easy too.

For the rest, I think it should be doable to put the place for the 50p SCSI internal connector and a floppy style power connector. Like this people can choose if soldering the DB25 or the internal port.

 

scuznet_v2.thumb.jpg.98258b9241103640a79397b2706174a9.png

 

This is the modification I would love to see.  It looks like it would be possible to move the Ethernet jack over to the right a little bit to allow for space for the 50 pin IDC connector to clear the barrel jack.  I'm OK with a barrel jack as I would just build a molex to barrel jack adapter for internal use.  I'm OK with the MicroSD card and am OK with programing through PDI but having a dual purpose (internal and external) Scuznet would be awesome.

 

I already bought short panel mount Ethernet jack extension cables for this purpose- my intention is to just 3d print a slot cover for the back of my SE/30 and have the panel mount ethernet port screwed into that.

 

I don't have the skills to make the mod above, but if somebody wanted to do it, esp. on the V01 board, I'd be ecstatic (I still have more parts to build more V1 boards).

 

As an aside, attached is the 3d case I made for my external Scuznets. 

ScuzNet Case Lid.stl Scuznet Case.stl

Link to post
Share on other sites
  • 68kMLA Supporter

While we're on the subject of Scuznet modifications...  the other big one I would love to investigate is to make a Powerbook Wifi edition.  The foot print seems pretty close for a 2.5" hard disk.

 

For connectivity, I've been googling like mad but coming up empty handed to see if there is an option to replace the ENC28J60 with a WIFI option.  Unfortunately it doesn't seem as though just pushing 802.3 frames into 802.11 works.  This is obviously possible as routers are doing just that and the device another member here (Ants?) used (a Vonnet?) to bridge his internal Ethernet card in his SE/30 WiFi shows it can be done.

 

Has anybody had any experience with this?  If the right hardware solution exists that can take standard Ethernet frames and pump them out to WiFi and do the reverse, it would be pretty straightforward to adapt the firmware.

 

Also, for anybody is running the Daynaport firmware, any issues with freezes at shutdown?  I'm getting them every so often but I'm not sure if it's my install, the changes I've been making to try to improve performance, or something with the driver.

Link to post
Share on other sites
  • 68kMLA Supporter
On 4/26/2021 at 8:53 AM, superjer2000 said:

Unfortunately it doesn't seem as though just pushing 802.3 frames into 802.11 works.  This is obviously possible as routers are doing just that and the device another member here (Ants?) used (a Vonnet?) to bridge his internal Ethernet card in his SE/30 WiFi shows it can be done.

 

Yes, the frame format is a bit different (it can have more addresses than just two, for starters, and there's no ethertype).  So you'd need an embedded bridge component.  Those do exist; the usual large suspects (broadcom and friends) seem to have suitable ICs, and I think there are more integrated "just slap it in and go" type hybrid devices as well, but I am coming up at a loss with the google terms I used to find them before.

 

I was wondering if just generating the 802.11 frames on the microcontroller rather than 802.3 frames might help, but I know there's a lot of housekeeping involved in talking 802.11 and I don't know how much of that the controllers do for you...

 

Sorry, on looking back on that, this post is of extremely limited usefulness, really...

Link to post
Share on other sites

 

On 4/25/2021 at 7:26 AM, saybur said:

 

I think that's a good idea, but unfortunately it doesn't quite fit on the board. I think @Chopsticks is on the right track with a separate all-internal design, moving the ENC28J60 and associated hardware to a daughtercard.

 

In my humble opinion, being a newcomer, I think is better having a single version of the PCB... I keep my thinking that a hybrid design would enable this to become a kit (ideally with some preassembled SMD stuff, like the RASCSI) that can be made available for most potential users and therefore lower costs.

Just my two cents! 

Link to post
Share on other sites
  • 68kMLA Supporter
On 4/26/2021 at 6:16 AM, cheesestraws said:

 

Yes, the frame format is a bit different (it can have more addresses than just two, for starters, and there's no ethertype).  So you'd need an embedded bridge component.  Those do exist; the usual large suspects (broadcom and friends) seem to have suitable ICs, and I think there are more integrated "just slap it in and go" type hybrid devices as well, but I am coming up at a loss with the google terms I used to find them before.

 

I was wondering if just generating the 802.11 frames on the microcontroller rather than 802.3 frames might help, but I know there's a lot of housekeeping involved in talking 802.11 and I don't know how much of that the controllers do for you...

 

Sorry, on looking back on that, this post is of extremely limited usefulness, really...

 
hahaha. Any thoughts help!  I was thinking about generating the 802.11 frames but I think it gets complicated on receipt. From what I read 802.11 may have frames with a payload larger than 1500 bytes but the Ethernet drivers on the Mac wouldn’t know how to deal with that. For AppleTalk it shouldn’t be an issue as the frame sizes are always less than that but I don’t know enough about TCP (well I don’t know anything about it actually) but my thought was you might have a tcp packet that might need to be spread across two 1500 byte frames so the Mac driver can deal with it. I’m sure I’m using the terms frame and packet incorrectly so I hope my point is still getting across. 
 

like you noted there must be some bridging chipsets that can do this. But all I found so far is the wiznet630s that has an embedded router with openwrt that might be able to do it. 

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...