I was trying to test a small LCD screen I wanted to make into a monitor that plugs into the monitor port of the IIc. As I was plugging in the pins, I accidentally got one of the pins wrong and plugged something into pin 15, which apparently in the Color Reference. When I turned the computer on I got the text on this image:
And now no matter how much a reset the computer, the text characters show up the same way. I did the self test and it looks like the RAM is ok, but all text is messed up. Please tell me I didn't fry a chip!
The Apple IIc video expansion port is not compatible with any other Apple DA-15 video port (as found on the IIe RGB card, IIgs, or the Mac II, etc). This is a big problem if an inexperienced user assumes it is the same port.
You plugged "something" into pin 15: a lot depends on that "something".
Pin 15 comes directly from UD4, the TMG HAL. The equations for this chip have been published, so it can be replaced with a new GAL. It also connects to UD6 GLU (which is harder to replace), and UE1, though.
Definitely don't worry about the RAM -- there's no way RAM can distort the shapes of text characters.
Although I can't immediately pinpoint the problem, your picture certainly demonstrates that the character generator is attempting to draw the "Apple //c" text. Either the character generator itself is faulty...or the character generator is receiving bad signals from the IOU. That suggests three likely causes:
Here's a diagram of that part of the circuit with some colored highlights to indicate:
Apple IIc character generator.png
Upon seeing the photo, my first thought was a bad connection in the VC signal, which selects between the top and bottom half of each character.
[EDIT]
Note that I typed this comment before robespierre's comment had been published, so I wasn't disputing his guess.
You're right, it must be the timing generator.
I confess that I didn't even consider the the timing generator because I thought it couldn't possibly affect the character generator. But after reading robespierre's comment #2 I noticed there's a direct connection between SEGB and pin 6 of the timing generator.
And, if pin 6 is internally shorted to either power pin then it would jam SEGB, which would cause characters to be drawn exactly like they appear in your photo: two rows of pixels are correct, then the same two rows of pixels are erroneously repeated.
Apple IIc shorted PAL.png
PS: Keep that Apple //c switched off until you replace the timing PAL at D4. The internal short at pin 6 of the PAL puts an enormouse load on the SEGB output at pin 4 of the IOU, and you really don't want to stress the IOU because replacements are not readily available for it. (Sadly, there's a risk that the IOU has already been harmed.)
Wow, I feel like such an idiot. Thank you all though for your help, this was so fast and insightful.
Where can I get a new TMG GAL? I don't have a way of making that chip myself.
Also, if I take the TMG out and leave that space open, can I check the IOU in any way to make sure it wasnt harmed?
You can order a TMG GAL from ReActive Micro and perhaps other places.
I'm not sure about your other question. It's difficult to test ASICs like the IOU in isolation.
I tested continuity on the TMG between pins 20 (5V) and 6 (SEGB) and it shows around 3k ohms. It's the same from pin 20 of the TMG to the SEGB pin (4) on the IOU. I dont know if that means anything.
On the IOU, there is 3k ohms between 5V and SEGB as well.
Here's some context of what happened, which is very embarrassing, but I'm clearly new to this and so grateful of a community like this that helps each other.
I'm working on a small LCD monitor for my IIc. I had an old car backup camera system with a small screen that runs off of 12V from the car outlet. I found out that one of the pins from the IIc's display port outputs 12V, plus there's a composite output pin as well, making for everything I need to drive this small display without the need for any other wires or PSU.
So I wired it all up to the correct pins, and it worked perfectly. But the color artifacting quality of the composite video has been concerning me so I decided to do another test to see if I could clean up the signal.
Instead of plugging in the composite video pin to my screen, I plugged it into the color reference pin and then started probing around my screen's display board to see if I could get a signal. More than likely, I must have ended up sending 3.3V to that pin on the IIc, which goes straight to the TMG.
I've been trying to find the specs on all these custom chips, but all I see is that they take 5v from the board, so I don't know how far my voltage got through the path. I'm praying it didn't hit the ICU. I'm waiting for more advice before making the purchase for a new TMG GAL. I'd appreciate more feedback.
One more thing to add:
After I tried cycling power a few times, I tried an ADTPro disk to see how it looks, and while I don't have a screenshot, I can tell you it had the same character issues but everything was almost in the right place, meaning the only major thing wrong I'm noticing is the second row of pixels being off. Hopefully that helps.
Yes that does mean something...but it's every bit as confounding as it is helpful.
According to Applelogic.org, the Apple IIc's TMG is based on a PAL16R8, a bipolar device according to its data sheet. Bipolar transistors tend to fail as short-circuits, but your tests suggest pin 6 is not explicitly shorted to the power pins. (But maybe it's shorted to another pin? Or something else internally?) That could be interpreted as good news.
What's confounding is that pin 6 functions exclusively as an input, so it shouldnt be capable of affecting other signals like SEGB if the PAL array was merely corrupted. But maybe it's internally shorted to some other output? Or maybe it's not the source of the fault after all?
I took the TMG out to get ready to replace it, I have the replacement on the way. I decided to hook the old one up to 5V and see what it does. All input pins have about 1.6V on them, the outputs show nothing. An oscilloscope just shows a constant line of DC, no activity, but it's just sitting by itself not calculating anything, so I don't know if that tells me anything helpful. None of the pins seem to be shorted to each other, I tried them all. The chip feels hot to the touch, too.
Well I took the TMG out and replaced it with a new HAL, there's no change to the text on screen, so it must be deeper than that, most likely the IOU. I'm going to order a replacement. I found a seller on eBlueJay that has a few pulled from doner boards:
https://www.ebluejay.com/item/8432212
Is anyone familar with this seller and feel it is low risk?
And if anyone has any other suggestions, please let me know!
IOU chips are unobtanium from any other source than pulls from donor boards. I am not at all familiar with that seller.
I did some digging and found out it's the store for the well-revered Apple2Online.com, it's also recommended by ReActive Micro as another A2 resource. The eBlueJay site seemed questionable at first, but they are using it for their commerce so they don't have to deal with eBay's fees and taxes.
So it looks like I'm going to try putting in a donor IOU! Keep your fingers crossed this fixes it!
I replaced the IOU and we are all good now!
Very big lesson learned: don't mess with things on this computer that I don't fully understand!
I will also totally vouch for Apple2Online.com for being incredibly helpful and fast.
Thanks everyone for your help on here as well! While I was at it swapping chips, I went ahead and upgraded the ROM to 4x.