Smartport disk emulators and daisy-chaining

There is already a schematic for a daisy-chain adapter, and it was made by Apple in 1985: the daisy-chain board inside the external Apple 3.5 Drive, with its DSIC chip. Remember that the 19-pin port on the IIgs can drive two Apple 3.5 drives, plus two Apple 5.25 drives; or a SmartPort device like the Unidisk 3.5.

The only pins wired "in parallel" on the daisy-chain board are RDDATA, WRDATA, /WRREQ, and WRPROT, plus the power and ground pins. These go straight through to the female connector on the back of the external drive where downstream devices can be connected. When passing data thru from/to downstream devices, these pins must be high-impedance on the drive. The DSIC ensures this using hardware, which was prototyped using TTL logic chips, then designed onto a Sony CMOS gate array for production.

https://www.applefritter.com/files/2025/12/04/Apple_3.5_Drive_Schematic.pdf

In post #1, 'colinleroy' wrote:

" In the process, I somehow fried the WRPROT-related part of my IIc's IWM chip, so now I have a full read-only Apple IIc, and I'll have to wait to see a friend who is, contrary to me, good with electronics to desolder the chip and replace it. "

Uncle Bernie comments:

Ouch, but your mishap (and 'fessing up to it) gives us more insight how 'IWM's can be damaged / killed. So other than bad DRAMs leading to the dreaded "check disk drive" message leading to tinkering with the floppy disk drive, we now have seen that the IWM can also be killed by trying to plug daisy chainers into the backside floppy disk port of an Apple II.

I would recommend you to use an ultra low profile IC socket for the replacement of the IWM. Just in case your "mishap" happens again. It has been my experience that the aged Apple IIc PCBs do not like repeated desoldering and soldering of IWMs (or DRAMs) much (telltale bubbling when resoldering tells me it's already decomposing and has compromized through holes).

I happen to have three 'IWMless' beta test units with ultra low profile sockets available. These will be the last ones with regular pins and a socketing capability, as they are very tedious and expensive to make.

- Uncle Bernie

I would repeat myself -- such experimenting should be done (first) with a built with non-ASICs Apple ][+ motherboard or clone and a DISK II controller or clone. 

In post #6, 'colinleroy'wrote:

"I had seen the IWMLess thread but thought you had distributed all the beta units."

Uncle Bernie answers:

Oh, I can make as may beta test units as I want. The problem is the ongoing trade war between the U.S. and the rest of the world. But as long as you are located in the USA I still can send you an 'IWMless' and I even would make you (if you want) a special version that is not as easily "fried" by tinkering on the Apple IIc external floppy disk drive connector - no warranty to be indestructible, though, so if you fry it, no warranty and no refund. All I can do is to put in some series resistors on the SENSE and RDDATA lines. If despite of that you fry it, you lose your money, sorry, but this is life, and life is tough - I only offer these "extras" to you because I think your daisy chaining work may be valuable to the Apple II user community. You still need to pay the full costs (of mine) for the beta test unit. I ain't no U.S. Government welfare office handing out freebies to parasites. But I am volunteering to put these two extra resistors in (worth 2 cents plus my work) free of charge just because I think your daisy chaining work is worthy to get support.

Send me a PM for more details (sorry, if you are not in the USA I can't help you, not my fault or policy, but when extra customs paperwork is involved, sorry, I can't spend my time on any such nonsense).

- Uncle Bernie

In post #8, 'colinleroy' wrote:

" Sadly, I'm in France, so that's not gonna be an option. "

Uncle Bernie comments:

Oh, this is bad luck. I wish I could have helped you with your project. But please understand that my refusal to send electronic stuff to the EU is not due to lazyness ("filling out the customs form" ;-) but due to massive problems I had with EU customs in the past. I wrote about this topic here on Applefritter a while ago. But in a nutshell, it seems to me that since Y2024 or so,  EU customs started to crack down hard on import attempts of vintage electronic components ("not RoHS conforming") and now it also seems they want a special EU id number for each and every component and certification it was not made by "Child Slave Labor", as if children ever had worked in the semiconductor industry, never happend, they could not do this kind of highly specialized work anyways.

So in the end, it is impossible to send any electronic components to the EU, unless you are a big distributor like Mouser and Digikey, who automatically print out lots and lots of paperwork including the bar codes for all the EU ID numbers and whatnot. Does the EU customs process ever scan in all these barcodes ? I don't know but it sure would be a way to create many unproductive government jobs for useless people who are unfit to do any productive work.

The bottom line is that due to these "paperwork" hurdles demanded by EU customs it became all but impossible to send electronic components or electronics based hobby projects to the EU. This is why so many US based small businesses who do have a good stock of vintage ICs etc. refuse to send these to recipients outside the USA. The EU may be the worst offender but other countries are not far behind and have their own hurdles against imports. It's a trade war, after all.

- Uncle Bernie

In post #10, 'colinleroy' wrote:

" How would you calculate the required resistor values for the chip to still see high/low ? "

Uncle Bernie comments:

Alas, there is nothing here to "calculate" ... the IWM is a NMOS chip and hence has inputs with (almost) no current draw, except for pullup "resistors" which may be on these pins or not ... at the moment I have no time to look it up in the IWM specifications that can be found on the web. The problem here is that even when looked up, the value they give you is typical. And these old processes had huge tolerances.

So this is what I would do: with the IWM plugged in, rig a 10 kOhm resistor to GND (O V) from each pin. Unplug the internal disk drive cable. Turn the Apple IIc on. Measure the voltage on the SENSE or RDDATA pins to GND. If it is lower than 1V, you should be good. Otherwise calculate which current the pin's pullup circuit  sources and calculate the new  resistor to have less than 1V on the pin. Verify the result with this resistor to GND and then use it as a series resistor to protect this pin.

You can improve the protection of you solder any small signal diode (such as the 1N4148)  to the pin (anode to the pin, cathode to VCC/+5V) and this will clamp the voltage on the pin to safe levels while the series resistor limits the current (Note that is is NOT sufficient for CMOS ICs, which would need Schottky diodes such as 1N5711 or BAT42/43 etc. and two of them, the added one with cathode to the pin and anode to GND, which clamps negative voltages on the pin to safe levels, and note that no, you can't use PN junction diodes to protect CMOS this way).

Lab work typically involves the hazard of mishaps and it's very easy to blow up ICs, especially CMOS ones which can and will "latch up" if a pins gets a voltage more than ~0.6V above VCC or below GND. Almost all Schottky diodes have forward voltages of less than that and so they will prevent parasitic BJTs within the CMOS ICs to turn on, which may lead to self destruction of the IC even if you limit the pin current.

NMOS ICs are much more robust than CMOS but still can be damaged if you force enough current into or out of a pin. Some CMOS ICs are worse than others ... some examples are known as "latchup machines": the slightest issue with power supply turn on / turn off sequencing may blow them up. If your CMOS ICs tend to get hot and die, this was "latch up". The industry has made great progress in making modern CMOS ICs more robust against latch up but as long as bulk CMOS processes are used, the danger will never go away. SOI CMOS processes do not have any latchup issue, this is why military and space applications prefer SOI ICs ... they are more rad hard, too.  But if I want to be snarky, I could make a joke that our modern electronics may be too modern to fly to the moon again. The 1960s technology was just right to do it. Not too primitive to do the job (you could not fly to the moon with a vacuum tube based computer), but primitive enough to still be robust against the cosmic background radiation and sun flares. This is no joke ... the recent sun flares have upset the onboard electronics of an Airbus airliner and the are now scrambling to upgrade the software, and if necessary, the hardware, too. Make no mistake ... Boeing avionics ain't no better, they just were not hit by the effects of a CME yet, or they did not make headlines. And with the weakening of Earth's magnettic field it will only get worse. Until we may need to shut down air travel ...

- Uncle Bernie