Jump to content
Trash80toHP_Mini

Farallon ETHERMAC LC NSC w/NuBus drivers in the SE/30 PDS?

Recommended Posts

Dunno where this one will lead, but gotta give it a shot. So many folks have been asking about the prospects of doing something like this for so many years that it has to be tried. maceffects has placed two of his Farallon NICs into my grubby little paws, so it's time for a bit of olde schoole wire wrapping fun.

 

The 68030 LCIII PDS version of this diagram has been kicking around here and there in different projects for years. This time it's been dumbed down to the bare essentials of its 68020 LC PDS origins:

 

RosettaCache-01-002.thumb.jpg.5d1a9f3ec4eed6fb85e9e6382c362d05.jpg

RosettaCache-01-002.PDF

 

The signal known as /SLOTIRQ in the LC docs is better described as /SLOTIRQ.E in the expanded 32-bit 68030 PDS LCIII update of theLC PDS. We've been talking on and off about prospects for using it on an available interrupt on the 68030 PDS of the SE/30 and IIsi. That NMRQ line bit in the signal description makes me wonder if there's another pin on the 030 PDS? /NuBus might work where the /IRQ lines may not?

 

Is the 12V line on the LC PDS to power a FDD on a IIe Card's octopus cable or might the NIC need that voltage as well?

 

Any help with proofing the signal map would be much appreciated. I found one boo-boo, but proofreading your own work is just oxymoronic. ::)

 

Share this post


Link to post
Share on other sites

Yeah, the other one I have here has unimplemented components for ThinNet just like like your Farallon NICs. I'll be buzzing for connections for 12V just out of curiosity, but I'm going ahead and hooking up 12V in any case. At first I thought +12V might only be for powering the FDD on the IIe Card's breakout cable. Transceiver power makes sense.

 

 

There's no -12V connection on the LC PDS, so no NIC nor any other LC Slot Card have need of it, but you got me curious:

Ratline-LCIII-030-003.jpg

-12V is not on the extended LCIII slot either. What the heck uses uses that voltage? Is it for differential signalling?

 

I've got a nice piece of perf board here with copper pads on one side and grid lines for 5x5 contacts on the other. It's thin enough for me to wire wrap the protruding soldertails of connectors I had on hand so all the parts are in place. Had it set up in the chassis unencumbered bench setup for the SE/30 and all looks good. I'll probably make a thin plexi carrier board to bolt the RCPII/IIsi, Farallon NIC and perf board up together for a unitized assembly for installation and as a support base for doing the dirty deed with the wire wrap tool. Tested wrapping the short pin protrusions and there's no problem with a single layer wrap. Should be fun! [:)]

Share this post


Link to post
Share on other sites

Here are some pics of the test setup. Thought it best to get it documented before the project crashes and burns. Dunno if this is the first time an LC NIC has been fitted to the SE/30 PDS, but once I've wire wrapped it we'll see what happens. (crosses fingers)

 

WireWrapOne-01.thumb.JPG.feed2878806b41f10f5171546eb2d8e2.JPG

 

WireWrapOne-00.thumb.JPG.e1226b6ad04fa9fae50a66aaf44ce712.JPG

 

WireWrapOne-02.thumb.JPG.66642aebd4455452c064f0dc096b9c0e.JPG

 

I've got a couple of days off to work up the wrap templates. They'll be laser printed, rubber cemented to poster board, cut to fit and then bolted up to to either side of the perf board. One side keyed for wrapping and the other keyed for testing the connections. I'll see if I can get any of the wrapping done before I'm off again next Monday/Tuesday, but not likely. The only component yet to be fitted is the header block on the reverse side so swapping the address jumper can be swapped around without removing the board assembly.

 

 

edit: decided to go ahead and test the rig in the SE/30 right off the bat to avoid any complications the faster clock of the IIsi might induce. KISS, the less variables the better.

 

One additional question for you boffins: should I keep all the wires the same length (meandering) or is it safe to keep the wrap neat by using length appropriate wires? The card should be running off C16M in the IIsi, but I'm wondering there might be a problem with a IIsi clocked to 25MHz?

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

Finished off the physical setup for the attempt, realized the shots above are a bit cluttered so:

 

Assembly-Outside-Vies-2P.thumb.JPG.ff34d1e98a483b7eeee53f697c40194b.JPG

Outside the box.

 

Component-Jumper-Cards-2P.thumb.JPG.6ae31ca1e4009bb3eeadb6f08c55917a.JPG

Closeup view of cards w/address jumper detail.

 

Assembly-Inside-View-2P.thumb.JPG.8af4984290de85accde3c83bee182d60.JPG

Inside the box.

 

WrapHeaders-Cards-2P.thumb.JPG.488e1489b4c31976fc71a157147aca5c.JPG

Closeup of cards

 

Connectors-HotGlueMess-2P.thumb.JPG.a3e08aacbefdf7e9954ebe3f1e272c5c.JPG

Connectors-HotGlueMess/FUGLY Side up!

 

You can see the joint between the two sections of EuroDIN-96 I hacked together to make a EuroDIN-120-M for the 030 PDS passthru just above the vertical blob that covers the near side of the joint. I found enough 4-40 Nylon to bolt everything up and make standoffs for card support and base for the wire wrapping process. Now that the headers and standoffs are in place I can do my wrap and test diagrams. The only soldering I need to do is to tack the base of the headers to the copper side of the perf board. At the same time I'll loop the wire for /SLOTIRQ.E on the NIC around the headers and pass it thru the board to be wrapped to that pin.

 

 

@Bolle haven't heard from you. Should I keep all the wires of equal length or is it OK to go for a much neater olde schoole wrap like they did when they didn't have to worry about latching higher frequency signals? 25MHz is as fast a system (overclocked IIsi) as this thing could be installed in, but I don't think even that should matter? The NIC itself will be running off C16M in the IIsi, no? What clock do these things run off in an LCIII or Quadra 630? I'll have to look it up, but can't imagine it's not off the 16MHz reference clock. :huh:

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

It's a passive RA Adapter for using the SE/30 Version of one of my VidCards in a IIsi. I've switched over to it from the Pivot Card to comply with the KISS principal now that I'm tilting straight at the SE/30. I don't want to get ahead of myself, but this it does work out the carrier card will look a lot like a wrong Angle Hacked ASANTE NIC, an artistic interpretation with a neat twist or three. [:)]

 

Just finished my wrap map and I'll double check everything later after I get something to eat.

PERF-00-01.PDF

 

Gotta figure out the longest wire and start stripping matching color coded lengths. I'll just bolt a clear plexi cover plate to the thing to contain/protect the mangy KynarCritter at some point.

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

It was exciting yesterday, today it was mostly a PITA finding errors in graphics right after I posted, correcting them and trying to edit the corrected versions in within the time allotted. Didn't make it, but the only one that really matters is that last .PDF so I can give it a good going over at work. I'll check it again in the morning before work. Things sometimes look very different there for some reason. They almost always look different just by posting images here for study as opposed to trying to proof my work while in AI. Whatever works I guess. :/

 

 

edit: forgot to mention I cut all the wires to length and got the orange ones stripped for the data bus. I'll be doing that during downtime at work as well. Less than eighty to go  .  .  .

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

Got tired of thinking and curled up at the Illustrator Workstation for a bit of fantasy downtime:

 

ADAPTER-GAMMA.thumb.JPG.56d3cffc82b9981aaad2f70fde531854.JPG

ADAPTER-GAMMA.PDF

 

This is the only way I can figure out to get all my goodies in the box with an LC NIC. Radius Color Pivot SE/30 goes on top, it's the same max board form factor as that grayscale unicorn thing and plenty good enough for me. There's maybe just enough clearance for the Video Out Connector behind the LC NIC. Dunno, the ferrite ring on that sucker is BIG. Conversion to an IDC DA-15 Connector/Ribbon Cable setup for any VidCard should be no problem if spacing is too tight. I think it's a nice touch to have the LED activity light hoisted up the flagstaff next to a WiFi Antenna inside a clear case if it all works out.

 

Can anyone think of a Card for the SE/30 that has a huge breakout board besides a NIC? NIC/Accelerator/VidCard is the only SE/30 expansion upgrade path I've seen since the gamba day. Is there anything else to consider?

 

@Bolle will the overhang of my P33 clear a RA connector for the LC NIC as positioned above? The drawing is based on your interpretation of the DiiMO adapter. The existing outline is there in black.

 

 

Playtime over. :mellow:

 

Share this post


Link to post
Share on other sites

Gonna build something from cardboard to verify if an LC card will fit there. Can’t tell from the top of my head.

Share this post


Link to post
Share on other sites

Cool! Thanks for the assist, knocked together a low profile display version. 50MHz puts out a lotta RFI anyway, but there's still a passthru if needed.

 

.ClearBucketAdapter.thumb.jpg.3fa10e298d0cebd2671ab59b7a24107d.jpg

 

 

Share this post


Link to post
Share on other sites

I love the drawing, really makes this feel like it can work!  I actually like this layout, I assume one pass-through is for video and another for cache/processor upgrade?  Then we add conversions for the LC PDS on the rear?  This would be nice as it really maximizes internal space. 

Share this post


Link to post
Share on other sites
9 hours ago, maceffects said:

I assume one pass-through is for video and another for cache/processor upgrade?

Exactly, that's Bolle's DiiMO style PowerCache Adapter with a fourth connector grafted onto it ass. IF we can get this up and running that will provide networking for those without Asante NICs or the like, but are lucky enough to have an accelerator or are planning to source one. Might as well get all that done on one card. Especially for folks uncomfortable with the Idea of doing the wrong angle hack on an expensive SE/30/IIsi NIC.

 

I'm torn on what I'd want to do with with a clear case. It would be very cool to display the leaning tower of Grayscale if that's your thing. But seems better to me to keep the big toys in a standard case with its shielding for the bits toiling away at transmitting RFI in the 50MHz range. I'm thinking it would be better to use the clear case for a more stock setup where PSU, CRT, cabling etc are clearly seen, not blocked by PCBs wallpapering the entire side of the case.

 

That's why I tend more toward the passive display adapter design. With only the passthru connector (held in reserve) peeking up from behind the chassis and the shark fin/LC NIC/WiFi setup lending some current tech/hack cool factor that's clearly displayed (with activity LED even) at the back with a clear view into the goodness that made the SE/30 such a great machine it seems like a winning combo.

 

ClearBucketAdapter-CLR.thumb.jpg.fa4a1d96163c32b937b4bec8c94f4545.jpg

 

Didn't have time to post anything better before work today. Here's a clear chassis view of the passive board with the three possible clearance issues/limitations indicated. At the bottom, the NIC has to be far enough back to clear that horrible, oversized stock HDD cage's overhang. At the top, it seems like it will need to be tilted back to clear the neck board? Those constraints drum up the problem in the middle, which would be clearance for anything like a video output connector.

 

Anyway, that's the one I like most for the SE/30, but mine's not slated to emerge from the bench setup anytime soon. A clear bucket would be much better for my SE with its Radius 16 under the floorboards and the Radius TPD card displayed on the back wall with a clear view of the insides. The LC NIC will look better as its component side will be facing forward. TPD Card's components face the back. :/ 

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

Looks like I have to re-do the CRT layer of my graphics. The neck/neck board are farther back than the diagram would indicate, but matters not. There's about a half inch of play vertically for the NIC between the deck and the neck board. Better yet with the NIC raised to the level of the neck board there appears to be almost a half inch of clearance between the two when set in the same plane.

 

Now if only my P33 will clear the socket and NIC mounted on the tail end of your adapter card, Bolle. [:)]

Share this post


Link to post
Share on other sites

OCD AI session most of the day so I'm beat, crosseyed and my back hurts  .  .  .  but my wrists are fine! [:P]

 

716301532_DisplayAdapter-BETA.thumb.jpg.e5b84fa282b42b2d34172a3f133c7f3a.jpg

 

Scaling errors  almost got the best of me, so I gave up and tried to underlay a photo that was distorted every which way and I'm knocking this crap off until I can make a Physical mockup of the adapter and take my own pics of the real deal and finally let it go. I have real life issues for brushing up my OCD/ADD skills for a week or so. The mockup will take a less than an hour to do. :lol:

 

648528721_DisplayAdapter-Beta-2.thumb.jpg.e79757f2a406dda76c1d50769ce4004f.jpg

Display Adapter-Beta-2.PDF

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×