IIgs no longer seeing 3.5 drives. Bad VGC?

[joethezombie]

So my replacement "dead" mainboard arrived today.  I was going to use its VGC chip first, and failing that, try its IWM.  But looking at the "dead" board, it looked mighty nice, so I decided to try and boot it.  Imagine my surprise when the "dead" replacement board I bought for $15 came right up!  And it's a ROM3 board at that!  Diagnostics check out, everything works,!  Happy me!  :b&w:

I still want to fix my original mainboard, so I guess I'll go looking for parts again.

 

[Gorgonops]

Oh, no, it doesn't actually work. Hooking your drives up to your Mac destroyed them, just you wait and see.

 

[bigmessowires]

Replying to a 4-year-old thread for documentation purposes, in case somebody else has the same problem in the future. For the IIGS to communicate with a A9M0106 Apple 3.5" Drive, or any compatible "dumb" 3.5 inch floppy drive, the disk I/O signals HDSEL and EN35 must both be working. Both of these signals come from the VGC chip, not the IWM chip like the other disk signals do. And neither of these signals is used for communication with 5.25 inch drives or Smartport/Unidisk drives. So if your IIGS suddenly won't work with 3.5 inch drives, but works fine with other drives, you probably have a bad VGC.

There's something very strange about both the HDSEL and EN35 signals: they both can potentially be short-circuited by a connected drive. EN35 is an output from the IIGS, but if you connect a 5.25 inch floppy drive, that pin is connected to ground inside the drive. HDSEL is also an output from the IIGS, but if you connect an intelligent drive like the Unidisk 3.5, that pin is connected to ground inside the drive. Either scenario would create a potential short-circuit between a VGC output and GND. Apple guards against this by putting a 470 ohm inline resistor in the path of each signal on the IIGS motherboard. You can see this on page 4 of the IIGS schematics. Note that EN35 is called 3.5DIK on the schematics.

The resistors limit the short circuit current to about 10 mA. Presumably that's a safe level of current for the VGC. So why are some people seeing this kind of VGC-related 3.5 inch drive failure? I've tried to help a couple of people troubleshoot this in the past few years, but I don't have an answer. Maybe 10 mA is a safe current level for a newly-made VGC, but a 30-year-old chip will have degraded silicon and 10 mA continuous current will eventually burn out the drivers for those pins?

@joethezombie if you still have that busted IIGS motherboard, and an oscilloscope or logic analyzer, I'd be curious to know what you see on HDSEL and EN35. You can measure the direct outputs from the VGC on the motherboard at test point TP153 (HDSEL) and TP111 (EN35). You can measure the voltage after the protection resistor at the external disk port, on pin 16 (HDSEL) and 4 (EN35). During disk activity with a 3.5 inch drive, you should see both signals switching high and low.

 

[RPGCoder]

Replying to a 4-year-old thread for documentation purposes, in case somebody else has the same problem in the future. For the IIGS to communicate with a A9M0106 Apple 3.5" Drive, or any compatible "dumb" 3.5 inch floppy drive, the disk I/O signals HDSEL and EN35 must both be working. Both of these signals come from the VGC chip, not the IWM chip like the other disk signals do. And neither of these signals is used for communication with 5.25 inch drives or Smartport/Unidisk drives. So if your IIGS suddenly won't work with 3.5 inch drives, but works fine with other drives, you probably have a bad VGC.

There's something very strange about both the HDSEL and EN35 signals: they both can potentially be short-circuited by a connected drive. EN35 is an output from the IIGS, but if you connect a 5.25 inch floppy drive, that pin is connected to ground inside the drive. HDSEL is also an output from the IIGS, but if you connect an intelligent drive like the Unidisk 3.5, that pin is connected to ground inside the drive. Either scenario would create a potential short-circuit between a VGC output and GND. Apple guards against this by putting a 470 ohm inline resistor in the path of each signal on the IIGS motherboard. You can see this on page 4 of the IIGS schematics. Note that EN35 is called 3.5DIK on the schematics.

The resistors limit the short circuit current to about 10 mA. Presumably that's a safe level of current for the VGC. So why are some people seeing this kind of VGC-related 3.5 inch drive failure? I've tried to help a couple of people troubleshoot this in the past few years, but I don't have an answer. Maybe 10 mA is a safe current level for a newly-made VGC, but a 30-year-old chip will have degraded silicon and 10 mA continuous current will eventually burn out the drivers for those pins?

@joethezombie if you still have that busted IIGS motherboard, and an oscilloscope or logic analyzer, I'd be curious to know what you see on HDSEL and EN35. You can measure the direct outputs from the VGC on the motherboard at test point TP153 (HDSEL) and TP111 (EN35). You can measure the voltage after the protection resistor at the external disk port, on pin 16 (HDSEL) and 4 (EN35). During disk activity with a 3.5 inch drive, you should see both signals switching high and low.

@bigmessowires, I recently purchased a ROM01 IIGS and I'm seeing this exact problem. I have an Apple 3.5" drive, a Disk IIc 5.25" drive, and one of your excellent floppy emulators. The system reads 5.25" physical disks and images from the emulator, but neither the 3.5" drive nor the emulator in 3.5 mode work. The system won't boot/read from anything off of Slot 5, it appears. I do have an oscilloscope (Siglent SDS1202X-E), so I'd be happy to try to investigate this with you. I may need a little coaching, but I think I have the equipment that will be needed. I don't have access to any other IIGS machines, so I won't be able to try a replacement VGC yet.
 

 

[Verault]

Well for anyone interested I looked in my A9M0106 manual and found this, so I think this speaks for itself in regards to using the A9M0106. You can see the manual says for IIGs and MAcintosh, and the pages inside tell you which mackintosh models exactly. I have no idea why the pictures are sideways, They arent this way on my computer and I cant find a way on this forum to change the orientation.

 

[tetatdo]

Replying to a 4-year-old thread for documentation purposes, in case somebody else has the same problem in the future. For the IIGS to communicate with a A9M0106 Apple 3.5" Drive, or any compatible "dumb" 3.5 inch floppy drive, the disk I/O signals HDSEL and EN35 must both be working. Both of these signals come from the VGC chip, not the IWM chip like the other disk signals do. And neither of these signals is used for communication with 5.25 inch drives or Smartport/Unidisk drives. So if your IIGS suddenly won't work with 3.5 inch drives, but works fine with other drives, you probably have a bad VGC.

There's something very strange about both the HDSEL and EN35 signals: they both can potentially be short-circuited by a connected drive. EN35 is an output from the IIGS, but if you connect a 5.25 inch floppy drive, that pin is connected to ground inside the drive. HDSEL is also an output from the IIGS, but if you connect an intelligent drive like the Unidisk 3.5, that pin is connected to ground inside the drive. Either scenario would create a potential short-circuit between a VGC output and GND. Apple guards against this by putting a 470 ohm inline resistor in the path of each signal on the IIGS motherboard. You can see this on page 4 of the IIGS schematics. Note that EN35 is called 3.5DIK on the schematics.

The resistors limit the short circuit current to about 10 mA. Presumably that's a safe level of current for the VGC. So why are some people seeing this kind of VGC-related 3.5 inch drive failure? I've tried to help a couple of people troubleshoot this in the past few years, but I don't have an answer. Maybe 10 mA is a safe current level for a newly-made VGC, but a 30-year-old chip will have degraded silicon and 10 mA continuous current will eventually burn out the drivers for those pins?

@joethezombie if you still have that busted IIGS motherboard, and an oscilloscope or logic analyzer, I'd be curious to know what you see on HDSEL and EN35. You can measure the direct outputs from the VGC on the motherboard at test point TP153 (HDSEL) and TP111 (EN35). You can measure the voltage after the protection resistor at the external disk port, on pin 16 (HDSEL) and 4 (EN35). During disk activity with a 3.5 inch drive, you should see both signals switching high and low.

@bigmessowires, I could use your help if you have any updated info on this issue....

This week, all of the sudden my ROM3 IIGS no longer sees my 3.5” drive or the floppyEMU. I thought the drive was bad, but I put it on my IIC plus and it works fine. I read this thread that it's the VGC chip. I probed pin 61 for the 3.5DSK line. It's stuck high, so I tracked down a ROM1 IIGS with a working VGC chip that's socketed.
.
I de-soldered the VGC from my ROM3 IIGS and placed it in the ROM1 IIGS and connected the 3.5" drive with the VGC I suspected is bad. Guess what. It works perfectly. So I took the VGC from the Rom1 IIGS and soldered it into my ROM3 that's not working. SAME ISSSUE! The ROM3 cannot see the 3.5" drive.

I decided to check the resistors on the ROM3 IIGS that's not working. R120 and R190. I see 472ohms of resistance between pin 61 on the VGC and pin 4 on the floppy port. And I see 472ohms on pin 51 for the HDSEL line to pin 16 on the floppy drive.

Lastly, I decided to desolder R120 for the 3.5DSK line and probe the VGC chip to see if it would still be stuck high, and it was. At no point does it go low.

Im at a loss for what could be causing this? It seems like the VGC chip is fine. So are the resistors. I'd love to have some help to try to fix my ROM3 board. Its a shame to see one non functional.

Could it be bit rot on the roms causing some sort of logical error? Im at a complete loss.

HELP!

 

[Arbee]

Quick sanity check: slots 5 and 6 are set to the internal ports, right?

 

[tetatdo]

Yes, it’s set to the defaults.

 

[bigmessowires]

Unfortunately I don't have anything to add beyond what I wrote here in 2021. I would strongly suspect you have a hardware problem somewhere, but your test results are puzzling.

Lastly, I decided to desolder R120 for the 3.5DSK line and probe the VGC chip to see if it would still be stuck high, and it was. At no point does it go low.

The TLDNR of it all is that the IIgs must be capable of driving a logical 0 (close to 0 volts) on both HDSEL and EN35, or else all 3.5 inch drives will be unusable. For unknown reasons, I suspect some IIgs systems lose this ability and one of these signals is stuck at a logical 1 or left with a floating voltage. If you connect a scope or logic analyzer to the IIgs, and assuming slot 5 is enabled and no other drives are present, you should see both signals high after a system reset, except they will go low for a moment (maybe a few microseconds) when the IIgs checks to see if a 3.5 inch drive is present.

 

[tetatdo]

Unfortunately I don't have anything to add beyond what I wrote here in 2021. I would strongly suspect you have a hardware problem somewhere, but your test results are puzzling.



The TLDNR of it all is that the IIgs must be capable of driving a logical 0 (close to 0 volts) on both HDSEL and EN35, or else all 3.5 inch drives will be unusable. For unknown reasons, I suspect some IIgs systems lose this ability and one of these signals is stuck at a logical 1 or left with a floating voltage. If you connect a scope or logic analyzer to the IIgs, and assuming slot 5 is enabled and no other drives are present, you should see both signals high after a system reset, except they will go low for a moment (maybe a few microseconds) when the IIgs checks to see if a 3.5 inch drive is present.

Yes! That's exactly what I see. But its so strange because the VGC is known good. And the original VGC I pulled works in a different system.

It seems like the EN35 and HDSEL lines originate at the VGC. Unless some stuck address is setting the bits for those to stay high, I have no idea what might be causing it. I don't know where to go about from there.

 

[bigmessowires]

Maybe something else is shorted to one of those signals and is forcing it high? I don't remember if the IIgs has some kind of RC filter mechanism on those I/O signals, but those have been known to fail in compact Mac disk signals.

 

[tetatdo]

Maybe something else is shorted to one of those signals and is forcing it high? I don't remember if the IIgs has some kind of RC filter mechanism on those I/O signals, but those have been known to fail in compact Mac disk signals.

I did desolder the two resistors in line and noticed that the signals go low on the port. But they stay high on the VGC. It's almost something on the soft switches is driving it high and keeping it that way.

 

[bigmessowires]

How are you testing? Are you watching the signal with a scope while you reboot the computer? Are there any disk drives connected to the system?

 

[bigmessowires]

You also may need to add a small bias resistor to VCC or GND in order to tell the difference between a true high/low output signal and a signal that's merely floating. For example after you desolder the two resistors, add a 10K resistor between the VGC output and GND. If the signal is still high, then it's actually the VGC driving a high signal. But if the signal is now low, then the VGC is not driving anything.