27 replies to this topic

[Yoder]

Hey guys,

I just got an untested Apple IIc in the mail today for a pretty decent price. It's ugly as hell, but I'll retrobright it to address that :-)

Anyways, it didn't have a power supply so I made one from an 18V 1.1A switching power supply from an HP laptop and a 7 pin DIN female connector I bought a few months back (since I've been hunting for one of these). Specs for this are15V/1.2A so I'm pretty close. I've read that they can take up to 23V. Maybe it has a 12V voltage regulator inside?

Anyways, it seems to work fine and it loads disks, but when I pull up DOS, any key I press just repeats indefinitely or until I press a different key, which does the same thing. I've tried pressing every key on the keyboard including the "keyboard" button at the top (not sure what that even does) to no avail.

Does anyone have any pointers on this? I'm going to open it up and re-seat a keyboard cable if it has one, and look for bad capacitors, etc, but I wanted to see what you guys thought.

Thanks!

[Keatah]

The simple stuff. Check for debris in the keyboard. OR keys that might not repeat. Check for loose chips and reseat them. Once the basics are done, report back. We can then troubleshoot to a component level.

Edited by Keatah, Fri Dec 30, 2011 4:36 AM.

[JamesD]

If the keys work at all I don't think messing with the keyboard cable will help. On a regular Apple II,II+ they had a separate chip for handling the keyboard you could swap out, but I'm not sure about the IIc. I think the IIc had all the logic integrated into custom chips. There are just some things that might not be fixable on the IIc.

[Yoder]

Thanks guys. I reseated all socketed ic's as well as the kbd cable to no avail. I see no damaged traces, etc. I'm gonna replace the electrolytic cap near the kbd connector. And just reiterate, ALL keys repeat forever...

I'll update after the cap replacement.

Thanks...

[Keatah]

Bullshit! A //c is eminently repairable. It might be a bit more difficult because of size and more integration, but nonetheless still repairable.

Now, What happens if you have no keyboard connected? You'll prolly' need a dos disk that drops you to basic or something..

I'm also going to assume the repeating key eventually beeps and "Errors out" with a "\" and then a new "]" prompt comes and the process repeats.

Also what happens if you try playing a game that uses something like a key for firing missiles. (dropping bombs won't work) it's gotta be missiles or lasers. Does it also repeat?

Check C46 and R6 and possibly the de-bounce C45..
It is the AY-3600 that does the keypress detection and it is the any-key-down line output of that IC that (may) cause a problem. The keyboard is directly connected to this chip.

It could also be the 8-1 mux or MMU or IOU. I wouldn't go swapping the MMU or IOU just yet, not until we've exhausted other options. It could also be the 2716 keyboard map rom as well.

Be sure all these chips have good solid logic on them.

Look online for the //c reference manual volume 1.

I'm gonna take a good guess and say we have a problem with the decoding of the $C01x address because that clears the keyboard strobe (bit7 at $C00x).

If the IOU is seeing AKD(any-key-down) for too long a time then it strobes itself for auto-repeat. This is AKD on the ay-3600 connecting directly to the IOU. I bet your problem lies there, or something affecting how that line behaves. We're talking about pins 5 and 16 on the ay-3600, and pins 31 and 32 on the IOU.

There's also a chip that controls the serial port at the ass end of the machine. If somebody blew it up then it could also cause a problem.

You will need schematics and some time to learn the keyboard circuit.
Beginning on p229 of reference manual volume 1 you can learn something about the keyboard circuit.
And chapter 6 of inside the apple //c will explain more how it works, an incredible read if you want to know exactly what happens when you press a key. P168 is golden for this issue!

You can get all this shit right over here -- http://apple2online....=1_15_Apple-IIc

Edited by Keatah, Fri Dec 30, 2011 12:49 PM.

[Keatah]

Ohh regarding the capacitor.. Do you know what it does? Have you looked at it on the schematic? And why are we changing it?

And does it behave this way in 80 column mode?

The keyboard switch, it's supposed to change between qwerty and dvorak style.. IIRC

Isn't this a truckload of fun! Or what!!

Edited by Keatah, Fri Dec 30, 2011 12:57 PM.

[Yoder]

Ok, I attempted booting into ProDOS from a floppy w/the keyboard disconnected, but it actually doesn't seem to finish its boot sequence. I get the screen full of colored tiles (the one that flashes on a normal boot). I also tested each key and all of them repeat, but there are a few (G,YUIOP[) that are not working so I'm going to see if these are connected to a common trace that may be damaged or something.

About the capacitor (C60). It's one of the few electrolytic capacitors, it's right next to the kbd connector, and it appears that it's connected to it by a trace, so I assume it has some role. I actually just found a schematic in the Apple IIc Reference manual I dug up online (haven't found a reference to C60 yet). It looks like C46 and the resistor network next to it handle kbd scanning rate, and C46 modulates the key bounce mask which was addressed in this post about a IIe: http://www.appleii.i...y-how-can-i-fix I don't think ceramic capacitors go bad, but C46 looks a little "fat". I do know that resistors can also go bad, so maybe that's worth looking into also. The C45 cap is a funky poly-style one that has 223 on one side (223 pF?) and A5K or R5K on the other side. I've actually similar style capacitors go bad on Atari's I've fixed. I'll have to dig a bit deeper.

Anyways, it looks like the IIc is thinking that I'm pressing the key longer than one second, initiating the repeat, (from below) or maybe it's not getting the key release signal? I suppose I could have a bad AY-3600 keyboard decoder or maybe it's the IOU that's interpreting the AY-3600 signal?

Will try to post the manual pics in the next post...it's giving me issues when I click "Post"...

Oh, I was wondering if that was Dvorak mode with that key! Still trying to get it switched to 80-column mode in DOS, but I booted Hitchiker's guide and said "Y" for 80-column and it's still doing it...but this seem software driven as opposed to switch-driven. Fortunately I could type a "Y" by switching to Dvorak style keyboard :-)

Ok, off to my multimeter and a beer!

Actually this is a lot of fun! Thank you for the help, guys!

[Yoder]

Ok, I'm trying to post the figures from the ref manual. I had to attach it. It shows the schematic and has a description...pretty cool.

Oh, and from reading here: http://staff.bcc.edu...tor_Coding.html that 223 C45 debounce capacitor is really 22 X10^3 pF or .022 uF :-) It looks like debouncing just addresses an issue with a few extra characters being displayed (Wikipedia)...I don't think a bad cap here would cause an indefinite repeat. It seems that it may be a key release issue, to me.

Attached Thumbnails

• 

Edited by Yoder, Fri Dec 30, 2011 2:37 PM.

[The Usotsuki]

Keyboard is a QWERTY/DVORAK switch.

[Keatah]

C60 is related to a -12V > -5V voltage regulator 7905 UB13. It serves to reduce noise. In a roundabout way it could remotely possibly affect the issue. But I don't believe so.

Be sure ay3600 has +5v and -12V and a good ground. Don't forget and check RP1 pullup resistor network.
Right now I believe UB16 is bad or C45 connected to it. That would be on pin 31. Check them. And make UB16 the first chip you swap.

Read the theory of ops to understand what happens when a key is released and/or held down. That's where U need to look.

Sams computerfacts has some troubleshooting directions supporting what I am saying. And the Ay-5-3600 datasheet also says something similar.

Work it through. Thrilling!

Edited by Keatah, Fri Dec 30, 2011 6:41 PM.

[Yoder]

Thanks Keatah. Yeah, I realized that the C60 cap is a filtering cap...I'm not going to mess with it for now. I checked the 7905 and I'm getting -11.34 in and -5.0V out. Then I checked pins 15, 27, and 30 of the keyboard encoder for ground, -12V and +5V, respectively and it was good. The only thing I'm seeing is that -11.3V, which I guess comes out of that big voltage converter under the shield. Hopefully that's close enough to -12V for the encoder and the other components.

I also changed the C45 capacitor with a new .022 uF one I had in my toolbox...no help.

I checked the RP1 resistor network, although I really don't know how to do it the right way. I measured resistance between each post and got 1000 ohm for all of them except two or so that had 500 and 650 ohms resistance. Is that network suspect? I wonder if I can pull one out of a bad sega console I have? I'll have to check.

UB16 is not labeled on the board. Is that the AY-5-3600? Unfortunately, it's not socketed, which is a bummer. Looks like they're not common anymore and they're $20 and up from what I've found. They only support 90 key switches, I believe, so I'm sure these aren't used much anymore. I think I'll just get another IIc. There's on on ebay with a bad keyboard....maybe I'll consider it.

BTW, I have about 12 keys that aren't working at all...but I verified that the switches are good. I traced the series of switches on the keyboard and found that there are 4 strings of keys that do not work, and these each lead to a distinct pin on the encoder and the traces to it are good. I bet there is an internal short in the decoder. Good news is the power supply I made seems to be working OK :-)

Thanks for your help! If anyone has a IIc around that they want to let go, let me know :-)

[Keatah]

The resistor network is probably good. Everything more and more is pointing to the Ay-5-3600 KBD decoder IC. Especially since you say 12 keys are not working.. Good job on the tracing. I say we stop right here and make a decision about the IC. Either replace it or not. Keep this //c for parts or fix it. There's little point in going further.

One thing, as a final attempt what happens if you disconnect the kybd while the power is on and and short the appropriate pins to make a connection and simulate a keypress?

ohh, and 11.3 volts is just fine in this circuit.


The Ay-5-3600 has a got a basic MPU inside it (more or less) and has a masked rom. The information in it is programmed to Apple specs. It's a custom chip, so speak, but no big secret. It's amusing to imagine Apple sending off the program on a stack of 184 punchcards to General Instruments!

Attached Files

•  AY-5-3600.pdf   499.05K   13 downloads

[Yoder]

Keatah, thanks for all of your input. I agree completely. I thought of jumping some pins on the connector to simulate a keypress, but I checked the continuity of the keyboard harness out to the female connector and it's good, so I'm confident that I'll see the same thing if I short pins on the mobo connector.

OK, I was wondering if the AY-5-3600 was custom. I figured they have to have the keyboard matrix in there. So that most likely means that only the encoder from a IIc will work (as opposed to a IIe, etc).

At this point, this project is at a hold stage. I'll either pull an encoder from another dead IIc (I'm not good at desoldering large chips, though) or I'll use this as a donor. It has several good roms, a good drive, keyboard, etc. I'll probably start retrobrighting the case and keys this weekend so that I'll be ready to roll when I get another board!

And I saw the thread on your haul! Congrats! That's amazing! I'll post back in this thread when I get the IIc up and running...maybe it will help someone in the future.

Oh, and thanks for mentioning the computerfacts book. Were you reading the one for the IIe or is there a IIc pdf floating around? It's only 60 pages, but there's some great info in there! I have a C64 project, so I may look through the Commodore version of that book. I think the one C64 I have (firtst one I bought last year) may become a donor too. I pulled the PLA to fix another one with two bad ICs, and the SID is dead...plus the bottom half of the case was damaged in shipping.

Thanks again...

[Keatah]

I looked through the //c version -- http://apple2online....=1_15_Apple-IIc -- There are other docs that say the same thing. I also had personal experience where 6 keys would continually auto-repeat on a //e and it was the encoder too.

The sams computerfacts are kick ass for monitors and printers, especially assembly and dis-assembly! The number of parts in a printer is ridiculous!

By the way is retrobright available as a pre-made mix or a kit?

Edited by Keatah, Sat Dec 31, 2011 10:54 AM.

[madmax2069]

Keatah, on Sat Dec 31, 2011 10:51 AM, said:

By the way is retrobright available as a pre-made mix or a kit?

I beleve its only offically available as you buy the ingredients you mix. There are people selling kits, but they are not endorsed by Retrobright.

as taken from Retrobrights web site

Quote

It has come to our attention that some individuals are offering for sale "kits" similar to some of the Retr0bright recipes on auction and other sites. Please be aware that these are not endorsed by this Wiki in any way and the Wiki organisers shall not be held liable for any loss or injury caused by the use of these "kits".

Hydrogen Peroxide is classified as a hazardous substance and, as such, is not accepted by many postal services and couriers without declarations and special handling procedures. It also has other uses besides hair bleaching and Retr0bright, so we strongly advise caution on where and how much of this you buy.

This site's founders and authors do not sell Retr0bright for these very reasons and do not endorse or recommend any resale of premixed Retr0bright. If you see it for sale, it is not with our blessing or consent and we strongly advise caution: if you order some for delivery and it leaks in transit, you and the supplier could have some interesting questions asked of you.

We are also aware of something similar to Retr0bright being sold on eBay, with the seller using a picture taken from this Wiki. This is nothing to do with any of us, and we encourage you to ask the seller if they are complying with the Creative Commons License for the picture as we believe that consent has not been granted for this use, but, more importantly, also whether they are fully declaring the contents of their packages to their courier and are following best practice on shipping hazardous goods.

Edited by madmax2069, Sun Jan 1, 2012 12:01 AM.

[Yoder]

Cool, thanks for the link to the IIc book! I hadn't come across that page... Definitely a great book.

Yeah, I replaced the POKEY chip in my Atari 800 XL recently too. Not a single key was working on that one....

As far as retrobright...i just do it DIY. I've found that just using 40 volume creme H2O2 from Sally's works perfectly. I'll post pics today!

Edited by Yoder, Sun Jan 1, 2012 12:28 PM.

[Yoder]

Ok, the retrobirght is done. This ugly IIc now looks like it did in the 80s.

The first pic is after removing all the residue from the stickers that had been on this thing. The "IIc" logo was a casualty unfortunately. It doesn't look bad, though. This was about 1.5 to 2 hr in the Tampa sun with probably 25-50 cents' worth of Creme 40 peroxide from Sally's. The majority of the whitening happened in the first hour or less. It turned out nearly perfectly. Now I just need to get the guts working :-/

Attached Thumbnails

• 
• 

[Blackbag]

That looks great! So, what's your formula for your DIY retrobright solution using the Creme 40 from Sally's?

[Yoder]

That's it! Just Creme 40! It seems to do a great job without adding oxy clean. You just have to really keep after it because it starts to dry and then can leave streaks where it overbrightens. I've used it on NESs, SNESs, Apples (this IIc and a IIgs), Commodores (64C, C128, and an A500), and Ataris (800XL and an 800). I'm hesitant to use it on a C64 since they're kinda dark and more apt to streak. I have a broken case I can test it on, though.

I really should consider adding some xantham gum to it to help the dehydration issue. That stuff is just so expensive that I haven't done it yet...it's like $14 a bag at the grocery store :-/ Plus it makes it more work. I really like that I can just brush this stuff on and go with it.

[Keatah]

When retrobrighting //c's and iigs'es is it common to damage the printed logo?

[madmax2069]

Keatah, on Mon Jan 2, 2012 1:17 AM, said:

When retrobrighting //c's and iigs'es is it common to damage the printed logo?

That would depend on how strong the mix is, If the Hydrogen Peroxide percentage is too strong or you have too much of it in the mix then it will over brighten and streak (bloom) the case and will have a bleaching effect on the printed logo (also damage the plastic). This is the reason I use a lower strength Hydrogen Peroxide in my mix, Yeah it takes longer to obtain the desired effect and you might have to re-coat it but its far better to do it this way. Being that the printed logo is light already it makes it that much more harder to keep from having a bleaching effect on it.

its a trade off, you mix it to where the Hydrogen Peroxide is strong the less time it takes to get the desired effect, but the more chance you will have on bleaching the printed logo and over brighten and streak the case, You use a weaker Hydrogen Peroxide or less of a stronger Hydrogen Peroxide in the mix and the longer it takes to get the desired effect but the less of a chance you will have of over brighten and streak and bleaching the printed logo.

If you leave it on the case for too long it will also lead into the same thing. If you do not put on even coats it will also lead to streaking. If you allow it to dry in spots you will run into spots that is more yellow then others.

Edited by madmax2069, Mon Jan 2, 2012 7:38 AM.

[madmax2069]

Sorry about the dbl post, I never use Hydrogen Peroxide that's over 10%, stronger is not better in this case.

If you look at the Problems and Pitfalls section at #3 label fade http://retr0bright.w...ms+and+Pitfalls, long periods of treatment can cause label fade, constantly check and reapply if needed. the following link should also be looked over http://www.classic-c...rial-on-c64.htm

Edited by madmax2069, Mon Jan 2, 2012 8:36 AM.

[Yoder]

Notice that the IIc logo was gone before retrobrighting. It came off when I used orange goo-gone to remove the sticker residue that was on it.

40 Volume Creme peroxide is 12%, BTW. I've never had it damage any labels, silkscreen, etc...but I'm sure it's possible. Yes, you never want whatever you're using to dry because the concentration of H202 increases as it does. I saw a little streaking in couple of spots in this IIc, but they all bleached out upon re-application of peroxide.

[madmax2069]

Yoder, on Mon Jan 2, 2012 8:53 AM, said:

Notice that the IIc logo was gone before retrobrighting. It came off when I used orange goo-gone to remove the sticker residue that was on it.

Ouch, Yeah I also learned my lesson of using goo gone to remove sticker residue on certain materials, If its a paper sticker I would use a wet cloth to dab and lay on the sticker and eventually rub it off with my fingers and finish removing the residue with a dry towel(takes some elbow grease), The metal IDschool tags are a PITA at first, usually they will peel off but leaving the glue/adhesive behind. I have seen some use a form of hot glue, sometimes it can be peeled off with a bit of work. some use the same kind of adhesive as tape and can be just as much a PITA to remove if not more.

I will never use goo gone on any sort of project like this again. lesson learned.

I have heard good things about Micara Graffiti remover in removing the sticker residue and even marker pen marks (it will remove paint), but i fear from using it from my past experience with such stuff as goo gone.

Edited by madmax2069, Mon Jan 2, 2012 10:38 AM.

[Yoder]

Yeah, I used the orange oil version which is less aggressive, but it it still softened the IIc logo to a point where it just smeared off. I was bummed, but I have realize that the case looks 1000X better even though it's missing that logo.