Asante MacCon Mod/Radius Color Pivot II/IIsi Cable Build-n-Mate Thread

[Trash80toHP_Mini]

A lot of info on this stuff is scattered about a few convoluted threads, most of it is buried under the incredible heap of crap in the IIsiColorPivotII_PDS_Card_HackProject™ thread so it's high time for a fresh start:

I should be able to dig pics of the MacCon model with the removable ThickNet board up easily enough later on. Probably right in that thread, but that cionversion's a piece of cake!

However someone recently asked about options for modifying a version of the Asante MacCon backplane breakout board PCB lacking that very convenient, removable PCB/ThickNet Connector daughterboard recently so I took some pics of an even earlier model's breakout board:



This one hasn't even got 10baseT/RJ-45 on board, but if I ever get around to sourcing a transceiver for it, I'll definitely go with ThinNet to use with one of my old Hubs anyway. But with insanely inexpensive Pivot Cards available, there's really no other choice!



The ThickNet connector is a special, high profile version of ye olde DA-15 for use with clips on a backplane plate mounted way out in the cheap seats to match the depth of the ThinNet BNC Connector's form factor, this makes things very easy.



Oblique view of the High Profile DA-15 connector next to a standard DA-15 connector with solder cup termination for panel mount applications, which exactly how it will be employed! [] ]'>



Here are the connectors overlapped in top view, we've got plenty of head room for the custom built cable. Gotta love having about 12 lbs. of air dry modeling clay stickumtogethastuffs at hand! [] ]'>



Here's a view of the cable wire/solder cup interface and how easily the wire will bend to an even tighter radius than necessary to clear the PCB behind the wiring harness.

I used a wire about the same gauge as the breadboard jumper wires toledogeek used for his fabulous Cheap, Quick and Dirty flexible SE/30 PDS extender hack. I've got these laying around in the prototyping piles as well. I'll post a current link to them on eBay when the time comes.

I guess that about wraps it up for soldered high profile DA-15 -> Panel Mount DA-15 conversion specific information for any ThickNet connector equipped IIsi or SE/30 NIC.

The rest of the thread will detail building the almost universally missing link, the standard RCPII/IIsi breakout cable with DA-15 connector and the version I'm really interested in building, the RCPII/IIsi breakout cable to HD-15 VGA connector with 16" Resolution conversion done in the cable.

Meanwhile, take a look around for a ratty video cable or any another donor for your Ferrite Ring.

 

[68krazy]

Excellent, thank you so much for this thread!  I just picked up a IIsi Radius Color Pivot a few months ago and would like to get it installed in my SE/30 in the next year or two, this thread will help immensely.

For anyone interested in the IIsi Radius Color Pivot, they are currently available on eBay for around $5.  Grab one while they're cheap!

 

[BadGoldEagle]

Great thread! Unfortunately, I don't have time ATM to do anything Mac-related but that is exactly what I planned on doing.

You still need to cut a piece of that Pivot card though. And do the right angle conversion to the MacCon. But that should be doable.

Didn't know the ThickNet connector was that thick compared to a normal video connector (could have probably figured that out just by looking at the name though...). But that's great. There's enough room now to fit the new connector.

Speaking of which, are the metal stand-offs part of the ThickNet connector, or can they be removed and reused on the new DA-15?

Here are the IIsi Pivot pinouts, in case anyone needs them...

top of the card
A->1
B->2
C->3
D->4
E->5
F->6
G->7
H-> NOT CONNECTED
I->9
J->10
K->11
M->12
L->13
N->14
O->15
(FPU socket side)

Notes:
-A thru O: IIsi pivot connector
-1 thru 15: DA-15 connector
- Pin #8 is not connected to anything.

FYI, here's a picture showing how the pins on the DA-15 connector are numbered
http://q-syshelp.qschome.com/Content/Resources/Images/DA15%20connector.png

Edit: 

Meanwhile, take a look around for a ratty video cable or any another donor for your Ferrite Ring.

Is that really necessary? The original video cable from Radius doesn't have one...

 

[Trash80toHP_Mini]

Thanks so much for jumping in!

I can't find my cable ATM to check, but IIRC every SE/30, IIsi and SE video breakout panel cable in my collection has a ferrite ring shrink wrapped a couple of inches back from the DA-15. Dunno if it's necessary, but it's probably a good idea to follow that general precedent.

It's beyond my ken, but any issues rectified by the ferrite ring might be especially important when venting output through a NIC's expansion cover plate. That's the aim of this particular thread. Dropping it from the spec we're developing's no problem. Building a simplified cable for a dedicated, single purpose cover plate is a side issue, as is my in-cable 832 x 624 16" resolution, HD-15 VGA connector conversion.

Noodling out an easily built solution to the DA-15 to HD-15 punched opening in the Nic's plate is a side benefit! But if a printed adapter would be robust enough for the application, maybe somebody will jump in with a printed 3-D model. [] ]'>

One of BMOW's posts in that insane mess of a research thread:

Oh snap, I got it to work! I had to connect a real monitor to the Pivot card before I could see anything - simply putting jumpers on the sense lines and then looking in the monitors control panel didn't seem to be enough. I haven't been able to test a dual-head setup yet, since I only have one monitor, but I was able to boot and run at 832 x 624 x 256 colors @ 75 Hz with the Pivot card as the sole video card.

For future reference, pin 1 on the Pivot card is the pin adjacent to the "J2" text on the board's silkscreen. The pins are:

1 RED GND
2 RED
3 C SYNC
4 SENSE[0]
5 GREEN
6 GREEN GND
7 SENSE[1]
8 N.C.
9 BLUE
10 SENSE[2]
11 C & V SYNC GND
12 V SYNC
13 BLUE GND
14 H SYNC GND
15 H SYNC If you want to build a video cable to connect to a Mac-standard DB-15 monitor, then just wire those pins straight to the DB-15, matching them up pin for pin. Pivot card pin 1 connects to DB-15 pin 1, Pivot pin 2 connects to DB-15 pin 2, etc.

If you want to build a combo video cable and VGA adapter with a hard-coded 832 x 624 sense code (which seems to be the highest possible Pivot resolution that's also supported by modern LCD monitors), here's the pin mapping:


PIVOT VGA

1,6,13 4,5,6,7,8 VIDEO GND
2 1 RED VIDEO
4,10 NC PINS 4 & 10 JUMPERED TOGETHER ON PIVOT END OF CABLE
5 2 GREEN VIDEO
9 3 BLUE VIDEO
11 11 C & V SYNC GND
12 14 V SYNC
14 10 H SYNC GND
15 13 H SYNC

In a previous post BMOW composed an excellent list of the tradeoffs inherent in my VGA converting cable build:

I think the diode is only necessary if you're trying to build a cable that simultaneously converts from the Pivot connector to Mac video, *and* from Mac video to VGA video. IMHO that's not a great idea, since it means you could never use it with a vintage Mac monitor that doesn't have a VGA connector. It also means you wouldn't have any switches to change the sense lines, in order to pretend to be different monitor resolutions. Mac to VGA monitor adapters are cheap and plentiful, so why not use one?

But if you want to try it, pretty much any diode should work. Here's one: {C}http://www.jameco.com/webapp/wcs/stores ... 1_36038_-1{C}The little stripe on the diode body shows you which way to orient it. The diode's triangle in the schematic diagram points in the direction of the stripe. I don't really understand the theory behind why there's a diode - it seems to contradict the sense code table below the diagram, on the page you linked to. But I'll assume whoever made the diagram knew what he was talking about.

< snip >

Not too worried about them though. The breadboard cables are available in Male-> Female and resolution switching can likely be done mid-cable. Might need to utilize that version for easily extending cable length anyway?

I HATE having VGA adapters hanging off the @$$ end of my Macs, pushing them farther from the wall and stressing connections.

We'll see how things progress, meanwhile I'm loving the process of harvesting the contributions from that other thread! [] ]'>

 

[Trash80toHP_Mini]

Since the SE/30 has become the target system of choice for this card, I'll add a bit of tangential hacking info due to a recent development detailed below:

This would be another reason I had for having looked into converting the high profile ThickNet connector into a panel mount, solder cup DA-15 for video out on this particular card:



My approach for fitting the RCPII/IIsi to the SE/30 is readily apparent in that new pic:

/monthly_09_2014/post-902-0-15914000-1411110922.jpg">

Compare the NIC breakout boards leaning on the /30, you can see the difference in width of the breakout PCBs. Either will work, but I have that older MacCon and the ThinNet Hub ready at hand. One side of the card clears the Pivot card nicely and the other is itching to have a hole drilled in it.

The older MacCon NIC/two flavor/two connector breakout panel PCB offers added flexibility for installation of the antenna/card for ants' wonderful SE/30 wireless hack. With the wrong angle hack for that Asante NIC and same for the RCPII/IIsi, I'm hoping I can set the SE/30 up on the AppleDisplayUnit™ as a wireless bridge to both Ethernet equipped and the PhoneNet limited machines sitting on its rainbow hued shelves. I don't have the wireless conversion goodies on hand for testing, but I'm guessing fitment issues will need to be overcome to get things arranged behind a three connector flavored NIC?

I've got no wired connection available, so this appears to be coming together rather nicely, even without the 50MHz PowerCache accelerator I'd never expected to have up and running.

If the IIsi Pivot card HeaderChopHack doesn't work out, I've got the SE/30 version available to get the SE/30 up and running on the KVM setup if I do the wrong angle hack to the three connector MacCon NIC. Having it set up as a server for the menagerie would be another benefit of getting my unexpected /30 up and running there.

At any rate, ants has opened up some great possibilities for all with his wireless SE/30 conversion.

I've always figured on desoldering the annoying connector Radius used for the breakout cable and wiring the cable fab directly to its thru-holes.So lopping off the card's panhandle for the connector seemed the obvious approach to take for SE/30 fitment to me.

Required mods to the other end of the RCPII/IIsi would be:

Removing/jumper wiring/hot gluing one large cap "box" due to interference with the chassis mounting ear.

Moving the addressing headers/jumpers to the solder side due to a bit of interference with the CRT.

Filing a "V" notch on the front of the card to interface with same on the edge of the return bend of the chassis. The notching would help in replacing the lateral support of the blocked mounting ears of the chassis.

Removing the addressing jumpers/headers from the component and re-installing them on the solder side for CRT clearance.

Back on topic:

Maybe this tangent will get others interested in the RCPII/IIsi, this cable build, wireless SE/30 and possibly wrong angel PDS card hacks. Not a bad application at all for an unaccelerated SE/30! [] ]'>

 

[Bolle]

So I finally managed to hack a cable together to test my card.

I wired up a standard Mac video connector and plugged in an adapter to get VGA. Adapter is set to 13" 640*480. I get an image output to my multisync flat panel that supports all the oddball Mac resolutions - nothing wrong here so far.

Problem is I can not seem to move the mouse cursor now. The keyboard is responsive and mouse button clicks are recognised as well, I just can not move the pointer around.

ROM on my card is version 2.6. Happens with stock SE/30 ROM as well as with IIsi ROM.

So 24 or 32bit addressing does not matter as it seems.

OS does not matter as well as I can not move the mouse even without an OS loaded.

As soon as the card is installed it defaults to having the main monitor on the external screen so I can not check if the mouse would move on the internal screen and only gets stuck when moving it over to the external one.

Any ideas?

 

[Bolle]

Figured that one out... seems like something with the interrupts is getting messed up if you set the jumpers wrong on the Pivot card and that is why the mouse won't move.

Now I got the card working together with the Asante and also modified the Asante panel:



And another picture because colour

 

[Trash80toHP_Mini]

Very neatly done. What termination type connector did you use? I really like your right angle connector twist.  [] ]'>  Looks like you used the female ends of the breadboard jumper wires as your board connector?

Was looking for my pic of the SuperMac(?) cable showing the dedicated RGB ground lines in twisted pair config with the RGB video lines. That and adding the ferrite ring are the only possible improvements I can think of for your build.

While I was looking, I ran across BMOW's observation that there are different pinouts for two different versions of the RCPII/IIsi. I'll have to dig 'em out and post pics to get the differentiation documented.

bigmessowires:

I just noticed you appear to have two different Pivot cards, and they're different!

Look at the area between that rectangular yellow capacitor C4 and the video connector. This post shows a card with CR8 and R14 between the capacitor and the connector, but this post shows a card with only CR6 in that space. Those aren't the same card.

Mine looks like the first example, with CR8 and R14. I don't know if that's an important difference or just a minor board rev. Do both of your cards actually work?

Here's his cable build info and the VGA converter schematic:

Oh snap, I got it to work! I had to connect a real monitor to the Pivot card before I could see anything - simply putting jumpers on the sense lines and then looking in the monitors control panel didn't seem to be enough. I haven't been able to test a dual-head setup yet, since I only have one monitor, but I was able to boot and run at 832 x 624 x 256 colors @ 75 Hz with the Pivot card as the sole video card.

For future reference, pin 1 on the Pivot card is the pin adjacent to the "J2" text on the board's silkscreen. The pins are:
 

1 RED GND

2 RED

3 C SYNC

4 SENSE[0]

5 GREEN

6 GREEN GND

7 SENSE[1]

8 N.C.

9 BLUE

10 SENSE[2]

11 C & V SYNC GND

12 V SYNC

13 BLUE GND

14 H SYNC GND

15 H SYNC
If you want to build a video cable to connect to a Mac-standard DB-15 monitor, then just wire those pins straight to the DB-15, matching them up pin for pin. Pivot card pin 1 connects to DB-15 pin 1, Pivot pin 2 connects to DB-15 pin 2, etc.

If you want to build a combo video cable and VGA adapter with a hard-coded 832 x 624 sense code (which seems to be the highest possible Pivot resolution that's also supported by modern LCD monitors), here's the pin mapping:


PIVOT VGA



1,6,13 4,5,6,7,8 VIDEO GND

2 1 RED VIDEO

4,10 NC PINS 4 & 10 JUMPERED TOGETHER ON PIVOT END OF CABLE

5 2 GREEN VIDEO

9 3 BLUE VIDEO

11 11 C & V SYNC GND

12 14 V SYNC

14 10 H SYNC GND

15 13 H SYNC

http://pinouts.ru/Vi...GA_pinout.shtml

I remain buried under the reorganization/project condensation/hack abatement rubble, but I'm slowly getting there. :

May have decided on taking a different approach to modifying the ThickNet/ThinNet breakout board in the IP. Instead of using ThinNet, I'll likely be removing both connectors. I'll incorporate ants' fabulous WiFi in Compact Mac hack. The BNC port is perfect for the antenna and I'll use a solder cup DA-15 to make a hardwired cable for the required transceiver to the vacated ThickNet connector's thruholes. Should be pretty slick if I ever actually get around to doing it. [] ]'>

 

[Bolle]

What termination type connector did you use?

It actually is a AUI port I desoldered from a AUI/AAUI adapter box. It was a 90° PCB mount type connector.

Looks like you used the female ends of the breadboard jumper wires as your board connector?

 

Indeed. After removing the crimp contacts from their plastic housing on one side they did fit perfectly onto the solder legs of the DA15 connector where a drop of solder holds them in place.

 

[Trash80toHP_Mini]

Interesting, did you have a problem with using the plastic housings as a removable connector? The housings on my cable fit perfectly, but a bit firmly into the card's connector shroud. Is there a problem with the pin to crimp contact connection electrically?

I'm almost certain that using MEK to weld the housings together to form a unit will be problematic for use as a removable connector. The unitized assembly may go in, but it probably won't come back out unless the outsides of the resulting rectangle are filed (maybe sanded) to a bit smaller profile.

I've been looking for the pic I have of the SuperMac(?) VidCard cable showing the ferrite ring and the twisted RGB/dedicated ground pairs used in place of the Radius cable's micro co-ax cable shielding. I ran across the zombie's mention of his gallery pic of the Xceed cable assembly in your Xceed thread:

No joy yet on the pic or in finding my borked Radius cable. :   However I did find a pic illustrating the only instance of a ferrite free VidCard cable assembly I've ever collected. I wish a ferrite juju master would chime in to explain why it may or may not be advisable to implement it for this cable build. I'm curious to learn how size and shape matter in terms of tuning its effectiveness? It's curious to me that the Xceed cable uses tubular form factor "rings" without the wire loopage seen in most "ring" configurations.

I also found the full set of VGA sense coding pages from which the diagram above was taken. I'll need to post those here as collecting tangential sense coding/cable/adapter building info here seems the thing to do.

 

[Trash80toHP_Mini]

.  .  .   are the metal stand-offs part of the ThickNet connector, or can they be removed and reused on the new DA-15?

Missed your question. The standoffs are part of the card assembly, not integral to the connector. One appears to have been soldered to the ground plane of the PCB. Its Philips head has been soldered over anyway, I'll have to check that one.

 

[Bolle]

The stand-offs have to be reused as they hold the bracket to the PCB.

For my GS harness I used clip on ferrite thingies and I do not see any difference in the picture if I leave them out. They are probably needed if you want to do this by the book though

 

[Trash80toHP_Mini]

Yep, the model I'm using for this build includes both ferrite ring loopage and twisted pair signal/ground implementations. Anecdotal evidence for both is overwhelming, I've been collecting every picture of every card/cable combination for the Compacts that I've run across for well over fifteen years now and exceptions to either "design rule" amount to one single instance IIRC, though in the auction pics of one combination evidence for or against presence of the ferrite ring is inconveniently hidden from view.

 

[Trash80toHP_Mini]

Cross one off the list: figured out the confusion about my having two versions of the card and I do., but all four examples of the RCPII/IIsi are identical. The first pic I had posted in the other thread was of the Pivot SE/30 cable in its socket on the SE/30 version of the Pivot Card. In the second pic it was shown in the socket of the IIsi version. No mystery there.

Interesting that the two cards are so nearly identical, maybe I can use clues from visible traces on the SE/30 version to figure out where the connector lines wind up on the IIsi card for the header-chop investigation?

Meanwhile, in looking over ants' innovative SE/30 WiFi hack I cribbed his trick for preserving reversibility of this breakout board video connector hack and a most convenient solution for mounting the ThickNet transceiver for mine!

/monthly_08_2017/post-9266-0-36009100-1503310538.jpg">View attachment 13975

If we use the original backplane plate where ants substituted aluminum angle everything is simplified. Mounting small "L-brackets" behind the DA-15 solder cup panel mount connector's standoffs allows for mounting the breakout board to them aligned nicely from behind its ThickNet connector board mount standoffs. There is then no need to desolder anything at all from the breakout board!

Not only that, my ThickNet Transceiver will mount vertically on its intended connector right next to the neck board and the Antenna for the WiFi board mounts on the backplane plate with the video connector with no need for removal of the ThinNet BNC connector in doing so. [] ]'>

I'll have to translate/condense that mess of .TXT into pictures. :

 

[Trash80toHP_Mini]

Due Diligence: Wikipedia on ferrite beads/rings:

Ferrite beads prevent interference in two directions: from a device or to a device.[1] A conductive cable acts as an antenna – if the device produces radio frequency energy, this can be transmitted through the cable, which acts as an unintentional radiator. In this case the bead is required for regulatory compliance, to reduce EMI. Conversely, if there are other sources of EMI, such as household appliances, the bead prevents the cable from acting as an antenna and receiving interference from these other devices. This is particularly common on data cables and on medical equipment.[1]

May not need it, but I'll definitely be putting ferrite rings on my cables. [] ]'>

I ordered up the ThickNet->10baseT transceiver for my SE/30 build and placed an offer on the VONETS WiFi card.

 

[Trash80toHP_Mini]

Meanwhile, take a look around for a ratty video cable or any another donor for your Ferrite Ring.

Is that really necessary? The original video cable from Radius doesn't have one...

The import of that just hit me. Has anyone got a pic of Radius' cable for the IIsi version of the card?

Re-read the ferrite bead article just now and realized why there would be differentiation in the cable implementation between models of the Pivot card as well as the presence of three ferrite rings on Xceed's internal/external cable harness.

The SE/30 has a "Television" inside the case whereas the IIsi does not. Alleviating RFI between components, CRT and external cable wiring for the internal/external monitor setup would have been important for the Xceed GS card setup as it was in bottling up RFI produced by the SE/30's CRT and picked up on the internal wiring "antenna" of the RCPII cable for the SE/30.

There would have been no absolute necessity for the bead's shielding in the case of the IIsi cable because there was no CRT induced RFI to be picked up on the internal cable and the ferrite bead present on the card's dedicated Pivot video cable would have sufficed.

Upshot: while the bead may not be as necessary for an SE/30 cable build in this age of digital TV, it's probably a good idea to include it as it was necessary for regulatory compliance back in that day of analog TV and rabbit ears.

edit: while likely not quite as WA as most of my Guesses, I am guessing here nonetheless. [] ]'>

 

[joethezombie]

Found this one, jt. Ring hiding in the heat shrink.

 

[joethezombie]

I think this part from DigiKey might be the connector to fit into the Pivot's socket.  Order placed, and we'll see in a few.

 

[Trash80toHP_Mini]

erm  .  .  .  might you narrow that down a tad there partner?  Got datasheet linkage?

Very timely resurrection of this thread. Just got the new IIsi board in yesterday and it pained me a might to plug a cable into that life sucking video port. Gotta build some cables  .  .  .  I lost misplaced the SE/30 Pivot Card Cable I finally managed to bork. :