Help using/fixing Apple // europlus

You need that one RAM chip. The machine will not work without it. When you put a language card, this chip is simply moved to the language card and you connect the ribbon cable in its socket.

If you don't have a RAM chip to put in the empty socket, you can still test the machine, but you have to remove the entire row. Then the computer will have only 32K of RAM instead of 48K, but it will boot.

I can also see that you have a 16K RAM card. You can take one of the 8 identical RAM chips on the right side of the card and put it in the empty RAM socket on the motherboard. Then you should have the full 48K.

With no cards, you simply get a beep and a Basic prompt:

                         APPLE ][

]■

Try to find the missing Language card. It probably has the missing RAM chip.

The language card adds another 16KB of RAM to your system, so it has 64KB. And if it's an original Apple Language Card (not a clone) then it will also contain a newer version of the boot ROM with the "autostart" feature. Another reason to be looking for the missing card... :)

Having a missing RAM chip will not prevent the computer from working unless it's in the bottom row. You're missing chip will only disable the top row and leave you with a 32K machine.  

As for the ROM, you might want to take a look at the ROMX  .

jeffmazur wrote:

Having a missing RAM chip will not prevent the computer from working unless it's in the bottom row. You're missing chip will only disable the top row and leave you with a 32K machine.  

As for the ROM, you might want to take a look at the ROMX  .

I don't think the system is smart enough to realize it has only 32K complete if just one chip is missing unless I'm mistaken, which is possible.

I think, but I'd have to try it to verify that with no disk (boot device) installed it will go to Applesoft (or IntBASIC) but any memory access in the 3rd bank will be bad because one bit will be missing of every byte.  Since the Apple II doesn't do any kind of memory checks to speak of it will work up to a point. 

A missing chip in the 3rd row will cause DOS loading to fail unless the DOS is INITed for 32K or a MASTER.

I think the reason why you have to INIT for a particular size is that's the only way it knows where to try to load DOS.  I'm not sure how a MASTER disk knows what the memory size is 32K or 48K (since DOS 3.x doesn't know how to load to a 16K language card without additional software).  I suspect it does some kind of memory test in software loaded with the DOS boot process.  Like trying to write bytes and read them back maybe?  I don't remember if I knew at one time.

softwarejanitor wrote:

I don't think the system is smart enough to realize it has only 32K complete if just one chip is missing unless I'm mistaken, which is possible.

BASIC (both flavors) will do a memory check and set HIMEM accordingly. That's the only way the computer can detect how much memory it has. And a single bad/missing RAM chip will disable the entire row.

tony359 wrote:

Is there a writable IC I could use to replace the ROMs, possibly with no or little modification?

 Unfortunately, if you replace the ROMs with EPROMs you are going to need adapter sockets (with inverters) for each chip. That will cosrt way more than the ROMX. You could also butcher the motherboard with your own EPROM modification but that's a bit more difficult.

jeffmazur wrote:

 

softwarejanitor wrote:

I don't think the system is smart enough to realize it has only 32K complete if just one chip is missing unless I'm mistaken, which is possible.

BASIC (both flavors) will do a memory check and set HIMEM accordingly. That's the only way the computer can detect how much memory it has. And a single

I stand corrected on that.  It's been too long, I didn't remember all of that.  I still think the part about DOS 3.3 boot failing on a 32K matchine if it was INITed on a 48K machine is true unless you used MASTER CREATE.  Somehow I have a weird recollection of that being needed.  Let me know if that is wrong also.

tony359 wrote:

Ok I need to understand this a bit more.

The original ICs are PROMs if I understand correctly? Particularly 2316B. My TL866II+ doesn't even read them so I sold it and ordered the T48. 

Once I have it, I can at least check the ROMs against known good images. 

 

If one ROM is defective, which IC can I use and will it require an adaptor regardless? 

<

The TL866 won't read some of the weird old PROMs like a 2316 w/o an adapter. but that's true of a lot of them.  I don't think that just to read a 2316 chip you need an adapter with an actual inverter though, the simplistic ones made with two sockets should probably work.

The T48 programmer looks interesting though.  I will have to see if there is Linux support for it.

Not sure if it will read a 2316B either though.  It seems to be touted as the next iteration of TL866.  It looks like it may support some newer chips that the older versions don't.  But FWIW, support for extremely old chips, the original TL866CS that I have will program a lot of 27xx chips that the newer TL866-II that I have won't.  However, I've found the TL866-II seems to do a better job on some GALs.  So I keep both.

tony359 wrote:

Ok I need to understand this a bit more.

The original ICs are PROMs if I understand correctly?

Any IC that has a customer part number and copyright (e.g. "APPLE") is not a programmable anything. It was manufactured with data permanently wired at the fab as a Mask ROM.

These were sometimes called "mask *programmable" chips, but the term is confusing, because there is no way for a device programmer to write to them, and they never existed as a blank device.

You will not find bipolar PROMs used in any Apple II. They simply hold too few bits and are too expensive and laborious to program.

tony359 wrote:

In that case, wouldn't I be able to replace a faulty 2316 with a 2716?

Yes you can replace the 2316 with a 2716 EPROM. But you will need to use an adapter to re-map a couple of the pins. At the bare minimum you can do this with two sockets (see Sather, "Understanding the Apple II" page 6-15). Without an inverter however, there will be a conflict with the Apple ROM and Language Cards as well as any other board that uses the INH signal. So you should always use a proper adapter.

If you want to read a 2316 ROM in your programmer, ReActiveMicro does make an adapter.

robespierre wrote:

You will not find bipolar PROMs used in any Apple II.

 

 

 Except for the DISK II controller board...

softwarejanitor wrote:

...DOS 3.3 boot failing on a 32K matchine if it was INITed on a 48K machine... Let me know if that is wrong also.

 

You are correct on this. The INIT command creates a "slave" diskette which needs at least as much memory as the system that created it. And won't use any excess memory if booted on a larger system.

tony359 wrote:

... not all 2716s are the same. Am I correct?

Just stay away from the TI 2716; it is not compatible. But a TI 2516 will work.

There are plenty of RAM testers like the Arduino one you mentioned but they don't always find every bad chip. Some may pass the test but still fail when used in the Apple II.

As for Diagnostic ROMs, there is an Apple Diagnostic Card but they're hard to find. Or the MultiROM card. But honestly, the ROMX includes a dead RAM Memory test, a full, burn-in quality RAM tester, as well as the Apple Diagnostics Program in one of its ROM banks. I know the shipping cost is an issue - maybe you can find a used one locally.

If anyone want a Apple II Diagnostic Card I have a couple available for €55 plus €15 postage doesn't matter where you live. Alternatively you can just use the eBay link 

https://www.ebay.com/itm/125980776422?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=BxWC1D9HRTK&sssrc=4429486&ssuid=BxWC1D9HRTK&var=&widget_ver=artemis&media=COPY

jeffmazur wrote:

robespierre wrote:

You will not find bipolar PROMs used in any Apple II.

 

 Except for the DISK II controller board...

There's also the original Centronics Interface card and their original Parallel Printer card.  (These were two distinct cards.)

Plus Apple's original serial interface card, the version before the "super" serial card.

And don't forget Apple's 1982 combo Printer/Centronics card, which employed an LS472 to double the quantity of bipolar PROM in order to include the firmware from both the Centronics Interface and Parallel Printer cards.

(In short, there are a lot of Apple branded cards that use bipolar PROMs.  A lot of third-party printer cards do too.)

You can also boot without any of the ROM chips if you have a MultiROM card. I have one for sale that I don't need any more. I bought it when I also thought that something is wrong with my ROMs. They turned out to be fine and I wouldn’t be surprised if yours are fine too. PM me if you are interested. Shipping it from Bulgaria to the UK will not be more than $7.

MultiROM.png

MultiROM card.JPG

tony359 wrote:

- How do I run "inspector" or "Watson" which I put on the ROM?

To run Inspector or Watson, just jump to the base address of the ROM socket it's in.  If you have both Inspector and Waton installed, it will start as Watson (the sector viewer/editor).  I regularly use Watson, but I don't have any experience with the Inspector.

So if you installed your Watson ROM in the  D0  socket, run Watson with the command   D000G  in the Monitor.

The web page doesn't give a schematic diagram, but it's straightforward to follow the circuit of the ROM Replacement Board from the PCB.

The circuit interfaces a 27256 by lifting three address lines (A11, A12, A13) from the motherboard, and controlling A14 with the bank-selection jumper.

So, here's how the contents of the EPROM appear to be mapped to the motherboard sockets D0, D8, E0, E8, F0, F8.

The bank-selection jumper chooses which set of ROMs will be present.

When the bank-selection jumper is placed in the first position, then the EPROM adapter behaves as though the sockets contain ROMs for Watson, The Inspector, Integer BASIC, and the original Apple Monitor.  When you switch-on the computer, it will start with a Monitor prompt.  To start Integer BASIC, press CTRL-B and press return.  To run Watson, type D000G and press return.  (The Applesoft ROMs and Autostart Monitor are not available in this configuration.)

When the bank-selection jumper is placed in the second position, then the EPROM adapter behaves as though the sockets contain ROMs for Applesoft BASIC and the Autostart Monitor.  When you switch-on the computer, it will automatically start BASIC or boot a disk drive if one is installed.  (Watson, Integer BASIC, and the original Monitor are not available in this configuration.)

Integer BASIC was the first ROM BASIC for the Apple ][, adapted by Woz from a prior BASIC he wrote for the Apple I. By its name, you can see how it is special: the only numbers it supports are integers from -32768 to 32767.

Applesoft BASIC was based on Microsoft's BASIC and supports floating-point numbers (9 decimal digits with a two-digit exponent). There are a few other syntax differences.

The principal difference between an original Apple II and an Apple II+ (plus) is the former came with Integer BASIC ROMs and the original Monitor ROM; the latter came with Applesoft BASIC ROMs and the Autostart Monitor ROM. The hardware of a II and a II+ are identical.

Either a II or a II+ could run the "other" version of BASIC by installing a Firmware Card or a Language Card in Slot 0. The Firmware Card has sockets for the alternate BASIC in ROMs, and a switch to select whether the card's ROMs or the motherboard ROMs are used. A Language Card has 16KB of RAM into which the alternate BASIC can be loaded by a "Master" boot floppy. Once booted from DOS, the "INT" and "FP" commands switch between the two versions of BASIC.

The two BASICs used different prompts: the Integer BASIC prompt is a greater-than sign, ">"

The Applesoft BASIC has a right-bracket sign, "]"

Some software requires either Integer or Applesoft BASIC, so having a Firmware or Language Card was important, besides the possibility of using the additional 16KB of RAM (on the Language Card) for machine-code programs.

The Apple IIe and GS don't have a Slot 0, so can't use either of these cards. Their AUX slots are used for different memory expansion options (such as the 80 column-64KB card for the IIe).

robespierre wrote:

Integer BASIC was the first ROM BASIC for the Apple ][, adapted by Woz from a prior BASIC he wrote for the Apple I. By its name, you can see how it is special: the only numbers it supports are integers from -32768 to 32767.

...

Actually it's from -32767 to +32767. Integer Basic doesn't support -32768.

robespierre wrote:

Integer BASIC was the first ROM BASIC for the Apple ][, adapted by Woz from a prior BASIC he wrote for the Apple I. By its name, you can see how it is special: the only numbers it supports are integers from -32768 to 32767.

Applesoft BASIC was based on Microsoft's BASIC and supports floating-point numbers (9 decimal digits with a two-digit exponent). There are a few other syntax differenc

Actually you can use an INTBASIC ROM Card in a //e if you do a little mod.  Because of where the switch is on the board it won't fit into a //e so you have to unsolder the swittch and put it on some wire you can feed out the back.  The ROM card won't automatically be enabled like it is on a ][+ in slot 0, but you can put it into another slot and then flip the switch and hit CTRL-RESET and get into INTBASIC.  It will also work when modified like this in slots other than those where the switch would normally like up in a ][+.  I used to use a ROM Card like that for copy protection cracking.

Forgive me for nit-picking, but some of these details are worth straightening out because some useful features are being clouded in confusion: 

• The Apple Firmware Card (aka "ROM card")  is not limited to slot 0.  Its hardware works in any slot, but DOS 3.3 is hard-coded to use slot 0 so you can't use DOS's FP and INT commands if the Firmware Card is installed in a different slot -- you'll just need to use POKEs or Monitor commands to activate the Firmware Card.  (See below for addresses.)
• The Apple Firmware Card will fit inside the Apple //e and Apple IIGS without modification.  Apple shifted the slots away from the backplane to make room for cards that formerly poked through the cutouts in the old Apple ][ case.
• The Apple Firmware Card is especially suitable for slot 3 in the Apple //e because it doesn't use IOSEL, the signal that's pre-empted by the Apple //e 80-column card.  (IOSEL is also missing from slot 0.)
• The Apple Firmware Card can accept 2716 EPROMs by modifying the wires labeled "2716" in the silkscreen.  It can further be modified for two ROM banks by using 2732 EPROMs, disconnecting the Vpp resistor and routing it to the 7474's spare flip-flop for bank-selection.
• As noted by robespierre and softwarejanitor, a ROM card provides quick access to hacking tools like Watson, a ROM-based sector editor from Omega Microware.  (But Watson requires RWTS16 to be loaded in RAM, so it's not a complete ROM solution.)

Here are the IO addresses and POKE commands for activating the Apple Firmware Card.  It's best to use the Monitor or machine-code to switch ROM banks, as the POKE commands will crash if you activate a ROM configuration that doesn't include the same BASIC interpreter. 

IMG_1203.JPG

IMG_1205.JPG

S.Elliot, do you have the schematic or photo handy for the Apple ROM Card mod?

Also, what is the difference with Watson 4.0 and Watson 6.0?  (I thought there was only one release of Watson)

It will barely fit, but it more or less requires having the top cover off and even then is often difficult to flip the switch in a //e unless you relocate the switch on some wire like I did for mine.   It's a simple mod and although possibly not technically necessary for a //e, it is extremely convenient to have the switch outside the case.

macnoyd wrote:

S.Elliot, do you have the schematic or photo handy for the Apple ROM Card mod?

I don't have a schematic.  Any paper notes of the modification would have been discarded long ago.

But here's a quick summary and photo of it: 

• Remove resistor R2.  This disconnects the pullup resistor at pin 21 of the ROM/EPROM sockets, an unused chip-select for Apple's 2516 ROMs.  (Or the Vpp programming voltage for 2716 EPROMs.)
• Attach a bodge-wire from pin 11 of the 7474 socket to pin 3 of the 7474 socket.  This attaches CLK2 to CLK1 so that both flip-flops are now activated by DEVSEL.  (The original design only used flip-flop #2, for the purpose of turning the ROM card on or off.)  This enables the spare flip-flop to be used, as described below.
• Attach a bodge-wire from pin 21 of any ROM socket to pin 5 of the 7474 socket.  A 2732 EPROM uses pin21 as address bit A11, so this attachment enables the flip-flop to switch A11 on or off -- effectively dividing the 4K address space of a 2732 into two 2K banks, chosen by the flip-flop.
• Attach a bodge-wire from pin 7 of any ROM socket to pin 2 of the 7474 socket.  This attaches address bit A1 to the flip-flop input D1, so this attachment enables the flip-flop to capture a address bit A1 from the bus when DEVSEL is strobed, thereby adding bank-selection to the card's existing IO.  (The ROM card already uses the other flip-flop to capture address bit A0 when DEVSEL is strobed, which toggles the card on and off.)
• Attach a bodge-wire from pin 14 of the 7474 socket to pin 4 of the 7474 socket.  This attaches Vcc to PR1, which serves no apparent purpose.  It would have been more sensible to attach CLR1 to CLR2.

IMG_1211.JPG

macnoyd wrote:

Also, what is the difference with Watson 4.0 and Watson 6.0?  (I thought there was only one release of Watson)

I don't know what's different.  I don't actually know where I got a second version of Watson -- I had saved Watson 4.0 onto a floppy disk so I could use it in the computer lab, and I modified most of my disks to load Watson into the Language Card along with Integer BASIC.  Some time around 1987, I noticed that some of my copies started with "VER 6.0" instead of "VER 4.0".  (second line, right edge)  A binary comparison shows they're totally different, suggesting they really are distinct versions -- not just a patched/hacked version string.

capture354.png

Thank you S.Elliot!  The mod is now added to my archive. 

Maybe I will create a complete schematic for the mod in the near future based upon the pic.

Do you by chance have a source for the 2 Watson verions?  (4 & 6)  Thank you in advance!

The F8 Monitor ROM in an original Apple II does not autostart; the F8 Autostart ROM was part of the package on an Apple II+.

The Language Card also has a socket for a substitute F8 Autostart ROM that overrides the motherboard F8 socket.

tony359 wrote:

At present, it sits on "Apple ][". If I press RESET I end up to Basic prompt ]

 

If I type PR#6, it boots from the floppy and runs programs.

 

With no controller card in it, when I power up I immediately end up on Basic prompt ]

Taken together, those three behaviors might indicate a faulty 74LS138 at position H12 on the motherboard.  That's the 3-to-8 Line Decoder that sends IOSEL to each slot, which activates the bootstrap ROMs for each card.   If the 74LS138 has developed a fault that causes it to erroneously activate IOSEL on slot 6 when the Autostart ROM searches for a boot ROM in slot 7, then it would erroneously try to boot from slot 7.  Since slot 7 doesn't actually contain a disk controller, it would get stuck waiting at the Apple ][ message until you press RESET.

But that's just a guess, a shot in the dark.

Because the disk controller can operate in any slot between 1 and 7, you can test it in another slot and make note of how it behaves.  For example, with the controller installed in slot 7 does Autostart boot it automatically?  With the controller installed in slot 7, can you boot it manually with PR#7?

Ooo, you're a quick study!

Slow down, though...we need to go back one step:

tony359 wrote:

So if I understand correctly, pin 9 of that 138 is probably stuck low. That way slot 6 is always selected. So at boot up the ROM checks for the drive in slot 7 - but the IC is selecting 6. The ROM finds the controller and thinks it's in slot 7, hence it tries to boot the drive in slot 7, which is not there.

Almost, but not quite.  Yes, pin 9 must be going low at the wrong times...but it can't be stuck low.  The Disk II controller connects IOSEL directly to its 6309 ROM; if pin 9 was stuck low then IOSEL would continually drive the controller's ROM onto the data bus, causing a conflict with the motherboard's RAM and ROM.  But the motherboard's RAM and ROM are working fine, so we can infer that IOSEL is not being asserted...and therefore pin 9 is not always low.

Disk II controller ROM on data bus.png

So, here are a few alternative guesses how the LS138 might malfunctioning:

• A fault in the enable logic or decode logic might be causing it to set all 8 outputs to low, ignoring the whole 3-bit input.  eg: IO to any individual slot might activate IOSEL on all slots simultaneously
• A fault in the decode logic might be causing it to activate 2 adjacent outputs at a time, ignoring just one bit of the 3-bit input.  eg: IO to slot 7 or slot 6 might activate IOSEL on both slots together as a pair
• A faulty NOT-gate in its decode logic might be causing it to activate even-numbered outputs when an odd-number is selected on the 3-bit input.  eg: IO to slot 7 might activate IOSEL on both slot 6 and slot 7, but IO to slot 6 might activate IOSEL on only slot 6.
• Or maybe there's some other mode of failure that I didn't think of...

Luckily there's a simple test to distinguish among all of those cases: install the controller in any slot, then use Monitor commands to list the ROM at each slot's assigned address.  Slot 7 is assigned ROM address $C700-C7FF, while slot 6 is assigned $C600-C6FF, slot 5 is assigned $C500-C5FF, etc.  When the controller is installed in slot 7, you should see a listing of the boot firmware (pictured below) when you enter the command C700L in the Monitor.  But it will just list random byts if you attempt to list the address for some other slot; eg: if you enter the command C500L it shouldn't match the listing below, and it will probably give different results each time.

Disk II firmware listing.png

Bonus Trivia: the 6309 ROM is equivalent to 74LS471.  Some [rare] Disk II controllers have Texas Instruments ROM chips marked "74LS471" at locations B3 and D3.  Does yours?  (Other part numbers are more typical.)

tony359 wrote:

I did check with the scope before you posted and noticed that slots 7,6,5,4 have pulsing happening all the time - slots 3-2-1-0 show pulsing but at a slower pace.

Ooo, that means none of my guesses were entirely right.  And I totally missed a clue!

So the motherboard's built-in IO occupies addresses $C000-C0FF. That's why slot 0 doesn't support devices with ROM -- those addresses are taken by the motherboard IO.  So the 138's  Z0 output (pin 15) is used to activate on-board IO, instead of being the IO SELECT for slot 0.

Here's the schematic of the decoding circuitry for IO SELECT and DEVICE SELECT, with labels for the address wires.  Z0 is used for all motherboard IO and for the DEVICE SELECT for all slots.  Z1, Z2, and Z3 are used to activate device ROMs in slot 1, slot 2, and slot 3.  Your faulty LS138 at H12 always activates all four outputs Z0, Z1, Z2, Z3 at the same time -- so the ROM for any card in slot 1, slot 2, or slot 3 would cause a conflict with all motherboard IO and all slot IO!

So that's why your computer wouldn't work when you installed the Disk Controller in slots 0-3.

IO SELECT decode stage.png

Let's follow the signals to the ON BOARD I/O DECODE stage...

Ordinarily, the motherboard would only activate this decoding if you attempted to access motherboard IO and soft-switches.  But your faulty LS138 will also activate these circuits if you try to read the ROMs for devices in slot 1, slot 2, or slot 3.  Conversely, any device in slot 1, slot 2, or slot 3 will assert its ROM onto the data bus if you try to access any motherboard IO ports.

IO onboard decode stage.png

So, here's a weird experiment...

With the faulty LS138 installed, and slots 0-3 empty, try using the Monitor command  C100.C13F  to print the first 64 bytes of the ROM in slot 1.  I'm guessing it will behave like this:

• The first 16 bytes will contain 0D, the hexadecimal value you generated by pressing the RETURN key.
• The next 16 bytes will just cause the motherboard to acknowledge the RETURN key.  The printed values will be random-ish.
• The next 16 bytes will activate the tape port, generating a brief signal.
• The next 16 bytes will activate the Apple's speaker, generating a burst of noise.

IO with faulty LS138.png

(Note: this picture is just a mockup, my best-guess at how it will behave)