You are not logged in.

#26 2015-11-06 23:30:35

bbraun
Member
Registered: 2014-05-29
Posts: 1,064
Website

Re: Project30

I've been using one of these:
416YmVxj8kL._SY355_.jpg
But it sucks.  I mean, it's better than nothing, but still is not super fun.  Doing a ROM SIMM connector with one of these makes the hand very tired.  Doing an 030 PDS slot or something, probably best to break it into separate sessions.  I'm probably nearing getting a real desoldering iron.

Relevant:
Hakko Fr-300 Integrated De-soldering iron
Inexpensive Integrated De-Soldering Gun

Offline

#27 2015-11-07 00:27:57

mcdermd
Member
From: Corvallis, OR
Registered: 2014-05-12
Posts: 972
Website

Re: Project30

bbraun wrote:

I've been using one of these:

sledgehammer.jpg


Daily Drivers: 27" iMac 2.8 GHz Quad-Core i7 (Late 2009), 21.5" iMac 2.7GHz Quad-Core i5 (Late 2013), 11" Macbook Air 1.6 GHz i5 (Mid-2011)
See the restored heroes here.

Offline

#28 2015-11-07 00:34:55

bbraun
Member
Registered: 2014-05-29
Posts: 1,064
Website

Re: Project30

Yes, that's it exactly!

Offline

#29 2015-11-08 03:34:29

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Offline

#30 2015-11-09 01:01:33

MJ313
Member
Registered: 2014-09-23
Posts: 498

Re: Project30

That's a man on a mission right there.

Offline

#31 2016-11-24 17:29:47

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Heh! wink I've been pretty much out of the scene for quite some time, but the meds were overhauled recently and I'm back to this old hobby. The first step will probably be to get two NuBus slots up and running in the IIsi/Rocket as a test case for switching around slot addressing on the DuoDock's Gemini card in the SE/30.

I picked up a cheap Chinese knockoff of mcd's Hakko 808 vacuum desoldering gun before I dropped off the face of the earth and into hermit mode last winter. That should make things much easier! smile

Offline

#32 2016-11-24 17:47:02

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Related project for the proposed IIci/Accelerator slot at the top of the SE/30 NuBus adapter board:

Cloning the Daystar PDS Adapter for the SE/30 - Take 2

It's boils down to clean room reverse engineering of the chip on the Artmix board.

Offline

#33 2016-11-27 18:20:57

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Correction; text from reboot thread:

Again, this project could mistakenly be described a clean room reverse engineering of the function of the IC Artmix achieved for their SE/30 board from what's likely the same data available for that successful project. It's actually an extension of the Gamba2 project relevant to cloning as many of DayStar DIGITAL's PowerCache adapters as possible, beginning with the IIsi adapter and then modifying for the SE/30.

Change of gears, new title of other thread and link to post rebooting topic: Cloning the Daystar PDS Adapters for the IIsi and SE/30 - Take 2

My bad. neutral

Last edited by jt (2016-11-27 18:34:07)

Offline

#34 2016-11-29 02:23:15

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

I was back at it again today. I got the Gemini Card (Nubus splitter doohicky from the DuoDock's NuBus implementation) set up in the /30 and got its oddball NuBus passthru interboard connector hooked up to a IIsi PDS NuBus adapter with a handy dandy 5" ribbon cable extension. Hopefully that will be short enough to be adequate for experimentation. Hopefully everything is pointed down the line in the proper orientation. Too tired to check tonight. neutral

Now I need to noodle out how to flip-flop slot assignment signals on the Gemini card and in the Ribbon Cable Slot Extender so both "DuoDock" Slots are addressed as IDs available in the IIsi PDS across my cable. Getting this up and running in the IIsi will be the first step and trying to get the kluge adapted to work in the /30 will come if and when. I gave this Two-NuBus-Slots-in-a-IIsi-Hack a wild whirl a while back and it didn't work out because I hadn't made any provision for remapping, just the old hook 'em up, cross fingers and apply power methodology!

Silly me! roll

Offline

#35 2016-11-29 19:37:37

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Update time!

Took a look at the Cable/Gemini Card combo this morning. The ABC rows are flipped to CBA on the Gemini's interboard (almost NuBus) connector, adaptation required. roll   Dug out the PowerCache from the parts lot mcd swapped me for his first Kanga, LOL/IIRC. That helped in working out component layout and clearances. Both connectors on the IIsi NuBus adapter will be replaced with a wire wrap header/socket setup that will coincidentally provide a bit of skootch room for the RA PowerCache (IIci Cache Slot) connector.

Of course this all depends upon getting the Gemini TwinSlot Hack up and running on the IIsi. partial protoboarding for the /30 will be used in that project. So even if it doesn't work out, I will be able to implement more of the prototype setup in the attempt to get a single NuBus slot up and running in the /30. Futura SX Video/10bt Daughtercard combo, it's been a long time coming!

Here's a shot of the assemblage (KLUGE!) I was messing with last night.

IG2QPS.jpg

I've got more vertical clearance for this PowerCache (Universal?) Accelerator. Dunno, are the rest and the '040 models about the same size?


If this pic is right, my board will have the connectors flip flopped from the setup of the TwinSpark board.

nmW05s.jpg

I always wondered why he made provision for only one PDS passthru, as there are gobs of (at least two?) PDS cards shown  in the PDS slot multiple upgrades project on gamba2? I'm wondering if another PDS slot might work on my board. IIRC, there are three Slot assignments available  on the PDS. If not, a PDS card ought to work if a two way jumper setup is flipped to deactivate one of the NuBus Slots to enable the PDS passthru connector.

Even if I score a /30 PDS PowerCache Adapter, I'm assuming the "read fuses" will be blown on the PALs. Does anyone know if it's feasible to read the input/output combos with a logic analyzer? tongue



edit: caveat: I'll have to lift the power leg of the (soldered) FPU on the IIsi adapter if I have any prayer of adapting it to the /30.

Last edited by jt (2016-11-29 19:54:38)

Offline

#36 2016-11-30 01:23:40

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

SE/30 PDS Revelations! big_smile

pin C2 = /NUBUS = NuBus Space Address
pin C3 = /TM0A = NuBus transfer mode bit 0 


from: DCaDftMF2e p.289

Offline

#37 2016-11-30 01:43:55

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Can't find my FDD cage for the /30. Is it the same as the cages in the Macs II/IIx/IIfx?

Offline

#38 2016-12-02 17:09:55

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Found it and clearances look spot on. The longer NuBus Card in top slot just clears the CRT and there's enough room between the lower card and the top of the FDD cage to mount a 2.5" Savvio HDD.

Better yet, in checking out docs on the NuBus slot ID setup, I discovered it's just four lines and encoded by resistors. I found two neat little rows of four on either side of a larger resistor on the solder side of the Gemini card. Gotta buzz out the connections yet to be sure, but the odds are pretty good. Not only that, they're located right where I can easily get at them! big_smile

IIsi TwinBus feasibility trials, here we go again  .  .  .


.  .  .  eventually.  roll

Offline

#39 2016-12-05 18:06:41

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Spent much of yesterday blissfully engaged in recreating the DuoDock Gemini board's PCB layout via my reverse engineering process in Illustrator. Love this stuff! big_smile

Researched the operational setup/implementation of Slot ID signals for NuBus in DCaDFtMF2e.  /ID0 is the the only pin that shows a visible modification on the solder side of the Gemini Card's Nubus Slot $D, the lower of the two DuoDock slots. I didn't have a card in the top slot of the Dock+ when using profiler to identify the two slots. Working assumption is that it's Slot $C as the Dock itself is a PDS expansion card for the Duo's PDS docking connector located at Slot $E and slots are normally implemented in sequential order.

Need to determine if the top Slot ID's resistor's four bit encoding setup is located on the Dock's logic board. That would make sense from what I've seen so far looking at the locations of the resistors and traces on the Gemini Card. At first blush, it looks like NuBus Slot $D encoding could be a single bit Delta value for /ID0 on pin C26. Gotta build the Slot ID table from the Docs, but bumping that one signal by one bit from the bus traces between Slot $C(?) and Slot $D would do the trick in the most simple method possible.

Relocating the Slot IDs from $C(?) and $D on the Gemini Card to bring them into line with Slots $9, $A a/o $B for the IIsi PDS should be done quite simply.

Same deal for the /30! big_smile

Offline

#40 2016-12-07 05:15:41

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Info from DCaDftMF3e p.54: Card slot identification signals

Identification signals 3 through 0 (/ID3-/ID0) are binary encoded to specify the physical location of each slot. The highest numbered slot ($F) has the four signals wired low. The lowest numbered slot ($0) has all ID signals high.

< snip >

.  .  .  each of the 16 possible NuBus card slots has an address  .  .  .
____________________________________________________________________

I think I've noodled it out, here's the table:
x3o45Y.jpg
This would confirm my guess that "NuBus Slot $D encoding could be a single bit Delta value for /ID0 on pin C26."

Looks like I just need to make sure the top slot of the Gemini Card IDs as the IIsi's native $9 (instead of the Duo's $C) and then just pull /ID1 low on pin A26 to morph the lower slot into the IIsi's $B slot assignment. Same for the /30! big_smile

Does this look right? I always wondered why the Mac's IDs jumped from $0 to $9, it never made sense before.



edit: just thought I may have it backwards. Would the normal states for signals usually be high and pulled low or where "wired low" is pulled pulled high for encoding?

Last edited by jt (2016-12-07 12:45:06)

Offline

#41 2016-12-09 16:56:10

jt
Member
From: Bermuda Triangle, NC USA
Registered: 2014-05-21
Posts: 1,408

Re: Project30

Well, Two setbacks and two major breakthroughs!

My hopes of easily changing slot IDs on the Gemini Card almost came to a screeching halt because of the insane complexity of the blue circuitry that changes the state of the bottom slot ID's encoding from high to low on the /ID0 pin.
ut6LuT.jpg
There's no way in the world that I'm tackling the reverse engineering of that f*****g blue mess! I might add that the yellow mess shouldn't even be there at all! The affected yellow pins along with any other pins labeled "t" are documented as "unused" in DCaDftMFH3e for standard Mac NuBus as well as the later NuBus 90 implementation! roll

Ruminating on that and then deciding to concentrate on the Gemini Card I'd already cut down for the IIsi version of this project was a major breakthrough (one can hope anyway) for this project. Removing the Gemini card's interboard connector section that sticks up, protruding from the bounds of the case of the IIsi had been a necessity. I'd soldered machine pins to the stubby soldertail pins of the abbreviated board's top Slot $C and then plugged them directly into the IIsi NuBus adapter's connector located at Slot $9 for testing.

I'll have to edit in the "lost" pics from this thread for clarity of that indecipherable verbiage/explanation:

TWO SLOT IIsi Nubus Card hack . . . (SE/30 hacks) https://68kmla.org/forums/index.php?/to … e30-hacks/

Quote from MLA thread:
________________________________________________________

I chopped the top off the Gemini Card, so the assembly will fit inside the IIsi . . .

Nothing changed, behavior is exactly as before. There are three possible states when the hacked Gemini Card is inserted into the IIsi NuBus Adapter:

A - Radius Card installed in top slot - Unserviceable Slot Interrupt
B - No NuBus Card installed - boots normally
C - Radius Card installed in bottom slot - boots normally

In Case C: If the Radius card were hooked up to the monitor, it would probably have the splash screen up from the ROM, but no comms with the MoBo, I did a screen shot and there's nothing in the Video Dump but the 12" RGB Display's window, as expected.

________________________________________________________

Editing in the pics (if/when I find the originals or recreate them as a worst case scenario) will help a lot! I was told they did over there anyway.

Second breakthrough was realization that installing this hack in Slot $E of the IIcx would be the best possible test for the modified Gemini Card!

on2fPf.jpg

Slot ID encoding /ID0 - /ID3 for both target systems (IIsi and SE/30) are all present on my IIcx MoBo!

I can use 96pin wire wrap EuroDin connectors to replace both soldertail versions on the Gemini Card. I'll have to source new ones, just discovered that the connectors I have on hand are missing connections for center Row B and I need all three rows, AB&C. Matters not, I can test my signal hack procedure on these connectors in the meantime:

Remove enough of the connector's plastic to enable me to bend the four long wire wrap legs for /ID0 - /ID3 causing them to protrude from the sides of the connectors for easy jumpering of their signals. Clearing the MoBo and any components on it will be key.

This will allow me to encode the Gemini Card's top Slot $C to match the IIsi NuBus adapter's Slot $E independent of circuitry on the Gemini Card and that of the IIsi NuBus adapter. This should (hopefully will) alleviate the Unserviceable Slot Interrupt failure of my original IIsi test procedure. Addding the circuitry for manually encoding the values on the legs of Slot IDs /ID0 - /ID3 should be quite simple to implement.

Doing the same bent pin hack/independent encoding procedure to the Gemini Card's Slot $D to match NuBus Slot $B of the IIsi and SE/30, decouples the incredible blue circuit kluge on the stock Gemini Card and relieves me from dealing with that horrible mess entirely! wink

Plugging the hacked Gemini Card into Slot $E of the IIcx will test that slot's ID mod when a NuBus Card fills it. Plugging a NuBus card into the bottom slot (hacked to Slot ID $B) should (hopefully will) be the same as plugging it directly intoSlot $B on the IIcx MoBo.

We shall see. At least I have a clear path for experimentation independent of any foibles of slot assignments in the target systems! big_smile



Note: maybe I should encode the second slot to ID $A to keep them contiguous. Choosing $B involves a change of only one Slot ID signal, but changing two for Slot $A should be just as easy. I'll be cribbing from the Mac II series standard of the IIcx Slot ID setup circuits. Dunno, probably won't matter.

Last edited by jt (2016-12-09 17:28:24)

Offline

Board footer

About ThinkClassic

ThinkClassic specialises in the maintenance, repair, restoration and modification of Vintage Apple and Macintosh computers. Ask questions and find answers about classic Apple desktops, laptops, accessories and peripherals.