• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Recapped Macintosh Classic with checkered screen

kupouzar

Active member
I actually replaced the CNY twice:

1) CNY17-3 - ebay
2) CNY17G-3 - https://www.digikey.com.au/en/products/detail/vishay-semiconductor-opto-division/CNY17G-3/4072867

Same result for both - currently I have 2) installed.

Thank you for the tip, I will start on finding transistors/diodes to replace.
Okay, then let's suppose that the optocoupler is alright. I'd probably start with the DP3 and DP4 Zener diodes. Have you tried replacing those yet? They are 1n4148 and are connected to the CNY. If you don't figure this out, maybe try replacing the QP1 with a CNY75GB instead of the CNY17 as the final try.
 

kupouzar

Active member
Update: the Classic was in the garage all day turned off and it's gotten quite warm today (+22C). When I tried to switch it on - it just booted ok straight away.
It should be completely stable though… I've also found a schematic that could help you.
 

Attachments

  • 0002C0A1-65EE-4A1B-AE64-A0F172BB1BBC.png
    0002C0A1-65EE-4A1B-AE64-A0F172BB1BBC.png
    2.1 MB · Views: 51

xeiter

Active member
Okay, then let's suppose that the optocoupler is alright. I'd probably start with the DP3 and DP4 Zener diodes. Have you tried replacing those yet? They are 1n4148 and are connected to the CNY. If you don't figure this out, maybe try replacing the QP1 with a CNY75GB instead of the CNY17 as the final try.
yeah I have already replaced these.

Thank you for the schematics - I will try to make sense of it .... I am very new to radio electronics. Been going on common sense and logic so far without any fundamental knowledge of components etc.
 

kupouzar

Active member
It might be worth checking all of the components on the line leading to the CNY17, especially the pin 2. Maybe there’s something wrong with the amplifier consisting of QP3 and QP5.
 

xeiter

Active member
It might be worth checking all of the components on the line leading to the CNY17, especially the pin 2. Maybe there’s something wrong with the amplifier consisting of QP3 and QP5.

Update: changed QP3 and QP5 - same behaviour from Mac Classic.

After that, I played with temperature of DP14 which is on the same path as QP3 and QP5 and rapidly cooling down of DP14 seemed to cause instant drop of voltage from 5.1v to 4.5v.

I was trying to target DP14 with my freezing spray as well as could but cant be 100% sure that I only cooled DP14 and nothing around it (I had my analog board installed into chasis and could not reach well enough)

I am now waiting for DP14 component (LM431) in the mail and will swap it out once get (should arrive around 10 of June).
 

kupouzar

Active member
Update: changed QP3 and QP5 - same behaviour from Mac Classic.

After that, I played with temperature of DP14 which is on the same path as QP3 and QP5 and rapidly cooling down of DP14 seemed to cause instant drop of voltage from 5.1v to 4.5v.

I was trying to target DP14 with my freezing spray as well as could but cant be 100% sure that I only cooled DP14 and nothing around it (I had my analog board installed into chasis and could not reach well enough)

I am now waiting for DP14 component (LM431) in the mail and will swap it out once get (should arrive around 10 of June).
That actually makes sense. The LM431 is a shunt regulator which regulates its output voltage by diverting some of the incoming current to the ground. It is basically a Zener diode with some extra circuitry. It also seems that it is supposed to regulate the current flowing down from the LED in the CNY so it could definitely be the cause. If a diode gets warm, its forward voltage drops and thus the resistance decreases. In this case it means that more current could flow through the CNY after warming the regulator up. Then there's the PP1 trimmer to fine tune the current regulated by the LM431. Looks like we've just figured this out.
 

xeiter

Active member
That actually makes sense. The LM431 is a shunt regulator which regulates its output voltage by diverting some of the incoming current to the ground. It is basically a Zener diode with some extra circuitry. It also seems that it is supposed to regulate the current flowing down from the LED in the CNY so it could definitely be the cause. If a diode gets warm, its forward voltage drops and thus the resistance decreases. In this case it means that more current could flow through the CNY after warming the regulator up. Then there's the PP1 trimmer to fine tune the current regulated by the LM431. Looks like we've just figured this out.

Thank you for following my updates and giving very useful suggestions.

Fingers crossed for the DP14 fix. I will definitely post back once I swap it out.
 

alexGS

Well-known member
Thank you for following my updates and giving very useful suggestions.

Fingers crossed for the DP14 fix. I will definitely post back once I swap it out.

I’m excited about that too - thank you for updating us and best of luck!

I tried to buy freeze spray, but can’t get it here in New Zealand any more due to regulations.
 

xeiter

Active member
Update: received DP14 and DP11 in the mail and changed them. First DP14, then DP11. Unfortunately, exactly the same behaviour - low 5V (4.5v) and checked board + wobbly screen. Once the Mac warms up - voltage goes up and the mac boots up properly. Wobble also goes away after mac boots up and warms up even more.

Back to the drawing board.
 

xeiter

Active member
I changed my attention to CP21 (B 682K 08). I noticed there is a very small crack next to one of the legs and when I was using freezing spray on it - the voltage would drop instantly.

I do not know if this will lead anywhere but after I took it off the board and used freeze spray on it its capacitance was rapidly changing according to my multimeter. Just not sure what to replace it with as I cant find what this capacitor is exactly.

On the schematics it has .0018 written next to it. Maybe someone can give me a replacement part number for it? That would be very helpful.
 
Last edited:

kupouzar

Active member
Very strange... Yes, it could be this capacitor, ceramic capacitors also tend to be heat sensitive. According to the schematic it should be a 0.0018uF (1.8nF) ceramic capacitor. I can't really see anything else that could cause this issue tbh...

Btw, have you tried tuning the PP1 after the replacements again?

At this point you've already ruled out all of these green ticked components, am I right?
mac_classic_analog-2.jpg
 

kupouzar

Active member
I've also found this. There is some information about the CP21 in Italian and they say that the value is 6.8nF. I also found another Bomarc schematic and there's also 6.8nF. If you look at the bottom of the page, you'll notice that these two schematics are for two different revisions of the analog board. So replace it depending on which revision you have:
820-0395-A (early): 6.8nF
820-0395-B (late): 1.8nF

edit1: the Italian website also states that the value of this cap changes significantly if heat is applied
edit2: 682K sounds like 68pF*10^2 = 6800pF = 6.8nF with a tolerance K (10%)
 
Last edited:

alexGS

Well-known member
I've also found this. There is some information about the CP21 in Italian and they say that the value is 6.8nF.

Nice find! :)

I especially enjoyed (translated) “The acid infiltrates everywhere and every time you use the soldering iron, an olezzo of burnt cat piss will come out.”

However, do note that the page mentions the same problem after replacing those parts;
“I am therefore replacing the following components under test:

CP4 = 47uF 25V
CP5 = 10uF 50V
CP21 = 6.8 nF (capacity value that varies greatly with heat)
These three components were joined by hot glue and you have to unweld them together.
IP1 = TDA4605
QP1 = CY17G-3

Once these last replacements have been made, the machine starts but with low voltages at the beginning, and then gradually arrives at stable voltages and the machine is fully functional.”

So the successful step seems to be this afterwards:

QP2 = IRFBC40
CP40 = 22KpF

Specifically, the CP40 was out of tolerance but still very changeable in capacity (it was enough to hold it in your hand to see the capacity vary greatly).
Once these last two components were replaced, the machine started at the first shot in the two days of random ignition attempts.”
 

kupouzar

Active member
Nice find! :)

I especially enjoyed (translated) “The acid infiltrates everywhere and every time you use the soldering iron, an olezzo of burnt cat piss will come out.”

However, do note that the page mentions the same problem after replacing those parts;
“I am therefore replacing the following components under test:

CP4 = 47uF 25V
CP5 = 10uF 50V
CP21 = 6.8 nF (capacity value that varies greatly with heat)
These three components were joined by hot glue and you have to unweld them together.
IP1 = TDA4605
QP1 = CY17G-3

Once these last replacements have been made, the machine starts but with low voltages at the beginning, and then gradually arrives at stable voltages and the machine is fully functional.”

So the successful step seems to be this afterwards:

QP2 = IRFBC40
CP40 = 22KpF

Specifically, the CP40 was out of tolerance but still very changeable in capacity (it was enough to hold it in your hand to see the capacity vary greatly).
Once these last two components were replaced, the machine started at the first shot in the two days of random ignition attempts.”
Yes, the whole restoration is very accurately documented there 😆. I didn't have time to read the whole page though.

The QP2 is a FET so it should theoretically be more thermally stable but God knows what happens if it's bad. I thought maybe the QP2 was too far away for the freeze spray to reach but I bet even a slight breeze of an air duster should be enough to cool the faulty component down to the point the voltage drops.

So maybe try checking all of the suggested components with something that isn't as strong as a freeze spray.

I'm looking forward to hearing that it's fixed 😁.
 

xeiter

Active member
Very strange... Yes, it could be this capacitor, ceramic capacitors also tend to be heat sensitive. According to the schematic it should be a 0.0018uF (1.8nF) ceramic capacitor. I can't really see anything else that could cause this issue tbh...

Btw, have you tried tuning the PP1 after the replacements again?

At this point you've already ruled out all of these green ticked components, am I right?
View attachment 57287

Yeah I did adjust PP1 but that didnt fix the cold start issue.

Correct about the ticked components - they have all been replaced as well as most caps and

My board is 820-0395-A (early) so CP21 is 6.8nF then.

I have also found this thread (same issue as mine) but fix was in a totally different part of the board ( 2 KOHM 1W “RF2” resistor that feeds the TEA2037 ) - https://68kmla.org/bb/index.php?threads/help-for-a-frustrating-recap-of-a-classic.42547/post-463620.

I will concetrate on my heat sensitive area for now though.


1685401819486.png

Re: freeze spray vs just air - I actually did notice that just me blowing onto the general area seems to bring the voltage down (not as fast as freeze spray).

Thank you guys for helping me here and for the info you found for me.... the quest continues. I hope to bring new update soon.
 
Last edited:

xeiter

Active member
Looks like my local electronics store has the ceramic caps I need... dont need to wait for it to me posted.

I will also try and get the CP40 too while I am there. As I have already changed the QP2 = IRFBC40 .... this will hopefilly be it (fingers crossed)


1685403402700.png
 

Attachments

  • 1685403355858.png
    1685403355858.png
    134.8 KB · Views: 19

xeiter

Active member
Update: replaced CP21 and CP40 .... same thing. Cold start - low voltage 😭

What I did afterwards is: take two paper straws and insterted one inside the other to make it long and started blowing on components one by one while the Classic was on.

What I found is that it was VERY sensitive to that straw air and voltage was instantly dropping from 5v to 4.6-4.7v but was climbing back up much faster than when I was using the freezing spray which made it easier to repeat the attempts of air blowing (I didnt have to use hair drier to bring the voltage back up - it climbed up by itself very fast) 😄

Anyway, I thought that if I was able to target better where I was blowing (with that long straw) I could find the faulty component.... and I was blowing VERY carefully to make sure not much air came out onto components.

I have not yet been able to find the exact component using this technique as the voltage drops alomost instantly from me blowing on different component (all in the same area) so I can't pin point the exact component because of that. Below is an approx imate area affected (ticked components have already been replaced).

Red area - affected by voltage drop when I blow on the components in the area, green ticks - replaced components.

I am going to do more troubleshooting with my straw tonight. Guys, I was blowing SOOOO little into the straw but was still seeing instant voltage drops. This is puzzling me as well as excites me to keep going.

1685433548684.png
 

xeiter

Active member
I am starting to suspect the optocoupler (CNY17) - it is in the middle of the air blowing square :). Perhaps I could have damaged it (twice) when soldering it in? I will get a socket and put a new CNY17 onto it.
 

xeiter

Active member
Could not make myself wait to get the socket until tomorow and resoldered a new CNY17 (did it quickly and carefully) - same issue.

Took off DP5 and DP7 diods (near all the caps) and cleaned the electrolyte that came out of the leaked caps under. When initially recapping, I missed this. My leaked caps damage was pretty bad - I had to scrape scrape things down to copper and apply proective layer to a large area where all the caps are.

Resoldered DP5 and DP7 - same low voltage issue. But things seem somewhat more stabe...ish. At least voltage climbs up faster maybe.

I might replace RP24, RP27, RP23, RP28 and RP10 next. Blowing (through a long straw and very very little bit) onto them (especially RP27 and RP24) causes an instant voltage drop and a very quick climb when stop blowing.

It makes me giggle when I imagine that you guys try to picture me with a straw in my mouth blowing on random legs and pins of components :D

I will do more clean up under these diods if required - they all sit flush on the board.

Maybe I should do a full AB wash? Never done that ... how do I do it?
 

kupouzar

Active member
I believe there is no point in replacing those resistors as they shouldn't really be that heat sensitive. If you apply heat, the resistance increases a little but that is definitely not noticeable if the changes are only in units of degrees. You could try cleaning the AB with an IPA spray. It might also be worth it to check the CP24.

Did you replace those DP5 and DP7 with new ones?

Have you tried blowing on the DP6 (MBRI045) near the transformer? That one is also on the 5V rail. I don't believe the slight breezes from your straw contraption would reach all the way to it though.

This is for sure a cap or a semiconductor. What soldering technique/tools are you using by the way? According to the datasheet, these optocouplers' soldering temperature should never exceed 260°C applied for more than 10s.
 
Top