LCII Sony Superdrive Issue

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
Hey all, a while back, while doing some cleaning, I managed to damage one of my 3.5" Superdrives. It's taken me some time to narrow down the issue. I had just cleaned it and was testing disks, so I had it connected, just setting loosely in the chassis, I lifted the front to insert a disk, and I suspect it shorted something out when it touched the metal magnetic shielding beneath it.

The symptom:
Upon boot, the heads make no movement whatsoever, inserting a disk results in no movement, but eventually the OS wants to "initialize" the disk, which does nothing, so you can do nothing but eject, but that does work (at least i'm pretty sure eject worked... it's been a couple of months now!)

What I've tried:
  • Initially I suspected I'd shorted the controller TA7774P, but I've swapped that with a known good one, and no difference in the result.
  • While waiting for 3d prints to finish, I decided to just start probing the board with the multimeter looking for shorts, and I found one:

Where I've landed is a 3 legged component labelled "B•6", and "Q4" silkscreened on the PCB:
IMG_2825D copy.JPG


Legs 1 and 3 have a short between them. All the other B•6 components on the board do not, additionally my known good drive shows no short across the same legs. When that component is in place, I also have a dead short between either end of the C17 capacitor just to the left of it in the photo. With that B6 component removed, I get no such short.

By now, you've probably guessed, I don't have a ton of technical experience with this, but I can troubleshoot... sometimes! What I'm looking for is the name of this component so I can source a new one... From my searching, it appears to be some kind of barrier diode, but, again, I'm not well versed in this.

Any help would be appreciated 😀
 

retr01

Senior Tinkerer
Jun 6, 2022
2,473
1
796
113
Utah, USA
retr01.com
@JDW did an excellent video on the Sony Superdrive. In the description of this video, there is a link to a Mouser cart listing the capacitors. However, almost all of them are unavailable. I am hoping @JDW will check and see.

As for the specific capacitors, I searched for a list for the Sony 2MB (1.44 MB) drives and didn't have much luck searching.

Let's wait and see if someone will jump in.

 
  • Love
Reactions: JDW

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
The capacitor is fine, unless the 3 legged black thing labeled b6 is a capacitor, but i would assume it'd be labeled with a "c" on the silk screen then. The issue, I suspect is with this component, not the capacitor.
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
@JDW did an excellent video on the Sony Superdrive. In the description of this video, there is a link to a Mouser cart listing the capacitors. However, almost all of them are unavailable. I am hoping @JDW will check and see.

As for the specific capacitors, I searched for a list for the Sony 2MB (1.44 MB) drives and didn't have much luck searching.

Let's wait and see if someone will jump in.

Mouser troubles me greatly. I spent a lot of time updating carts. And sometimes I’ll update a cart only to find one of the capacitors I updated then goes out of stock the next day, even though they were hundreds in stock at the time I replaced it! It’s crazy.
 

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
@JDW do you what the component is I have circled in the OP or how i would go about finding out what it is?

Your videos have always been so helpful, and i was able to bring a mac plus back for my stepdads retirement thanks to you!
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
@JDW do you what the component is I have circled in the OP or how i would go about finding out what it is?

Your videos have always been so helpful, and i was able to bring a mac plus back for my stepdads retirement thanks to you!
Thank you for your kind words, and I apologize I overlooked the content of your opening post.

I don’t have a schematic, and I don’t have that particular drive in front of me right now. But since you were able to swap out a chip on the PCB, you would be experienced enough to do the following.

I think it is either an NPN or PNP transistor. It’s probably an NPN. But we don’t know the part number and therefore we don’t know which pins are the emitter, the collector, or the base. But using your multi meter, you should be able to probe the other parts that are marked the same to determine if they have one leg that is consistently tied to the ground. That would be the emitter. Then probe to see which pin has a base resistor on it (typical values ranging between 1k and 10k ohm), and that would be the base. The other pin would be the collector.

Once you know that, you could carefully and delicately desolder one of the known good – not shorted — transistors and treated like it is an NPN and test it as a switch (by soldering tiny wires to it), adding a base resistor of 1k or 4.7k) to see if it works as an NPN transistor normally would to switch ground from the emitter to the collector when 5V is applied to the base resistor (and open circuit with no voltage applied to the base resistor).

For example, you could wire an LED and a 330 ohm resistor in series with the transistor’s collector, where the transistor would switch the LED on and off on the LED’s ground leg, depending on whether or not you apply 5 V to the transistor’s base resistor.

That test would confirm it is indeed an NPN transistor, and then pretty much any SMD NPN transistor that will fit the pads would probably work as a good substitute.
 
  • Like
Reactions: mitchkramez

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
@JDW thanks so much for your reply! I'm pretty new to electronics troubleshooting, so this would be a great test to get to understand how the transistor functions in the circuit! Thanks for the detailed description too 😀 That’s helpful for me just getting my head around all of this.

Someone else noticed that the trace it’s on is silkscreend with 12V(C) just above that component, do you think this is a 12v component?
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
Here are the LED test circuits I was pondering earlier...

1658037194613.png


Closing the switch causes the LED to light in both cases. But that assumes it is a BJT transistor. It very well could be a MOSFET transistor instead.

Yes, that 12V marking means 12 volts...
1658037242820.png


In light of that, it's probably a PNP. Further evidence for that is from this Aliexpress PNP with the same markings on it...


But it's important to keep in mind that a pair of BAT54 diodes often come in that same SOT23 case too. That's why a schematic would be handy. Sure, you could use a meter in beeper (continuity) check mode to trace out a simple schematic showing all connections in and out of that master SOT23 part, and that may allow us to determine if it is an PNP or diode pair. It just would take a little time to do that. But so would desoldering a known good part and doing the diode test.
 

robin-fo

Tinkerer
Feb 17, 2022
99
50
18
Switzerland
while doing some cleaning
Did you also clean the R/W heads? If yes, are you sure that you didn‘t pull the upper head too much upwards and thus have permanently bent the bronze strap/spring? I destroyed two drives this way and the symptoms are the similar…
Insert a disk into the disassembled drive and check if the upper head really touches the surface of the disk.
 

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
@JDW Just wanted to follow up here with you and everyone, that I got the drive working... well... mostly?

I didn't have a resistor to test as you suggested, and I was curious why a resistor is needed for the test?

I took a risk and ordered a 2SB815B6-TB from a UK company on ebay. After installing it, my drive came to life... for the most part.

I was able to get it to read and write 1.44mb disks, but not 800k disks. I couldn't get it to read any known good disks however, 1.44 or 800k. I must have spent 2 hours tweaking the ZTS alignment with no luck. This was with known good heads from another drive, that work perfectly in the other drive!

The 1.44 disk I got to work was repeatable in that it would format successfully and I could fill it with data, and read that data back off of it after ejecting. I wondered if there wasn't something quite right about the transistor I picked up in that maybe the 800k disks just take a little more oomph to read/write or something...

Anyway, I'm happy that it even kind of worked, and the drive actually spun up and read disks, even if it wasn't a complete fix! If anyone has ideas on what would cause the behavior described - I'd love to hear it!
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
Glad to hear the transistor swap worked. Testing an NPN BJT transistor with a resistor on the "base" is important to protect the transistor from damage. Without a base resistor, the transistor can be damaged. That is not true of a MOSTFET though, since those are voltage controlled. BJT transistors like an NPN are current controlled, which is why a base resistor is necessary to limit current into the transistor via the base. However, using too large of a base resistor can limit the current to such an extent that the transistor won't switch well, if at all. That is why my earlier post recommended values between 1kΩ and 4.7kΩ, to be safe, yet also to ensure adequate current flow.

So it seems the transistor really was the problem, and the read/write issues afterward are stemming from misalignment, which is hellacious to fix. Even a fraction of a millimeter misalignment can cause read/write errors. I'm not the most experienced person when it comes to fixing misalignment, so perhaps someone more experienced than I can chime in about that. But it is encouraging to hear that the transistor swap did get the drive somewhat working again.
 

mitchkramez

New Tinkerer
May 18, 2022
40
20
8
South Dakota of all places
@Paolo B - Unfortunately I've referenced that document and had many successful realignments in the past (well, ok, 3 successful), but I could not get this one to re-align at all. Maybe I got lucky with my other ones and this one is just crazy difficult though!

@JDW - as usual, thank you for your patience in helping me understand that process better. I was pretty thrilled to a) have found a component that didn't seem to be behaving right with nothing more than a multimeter, and b) having replaced it have it spring back to life! Your help and willingness to detail these processes has proven helpful to me so many times as a "new tinkerer", so again, thank you.
 
  • Like
Reactions: JDW

Paolo B

Tinkerer
Nov 27, 2021
258
144
43
Nagoya, Japan
Unfortunately I've referenced that document and had many successful realignments in the past (well, ok, 3 successful), but I could not get this one to re-align at all. Maybe I got lucky with my other ones and this one is just crazy difficult though!
800/1400 is a matter of single or double density, sectors per track. Requirements for track positioning (and therefore alignment) should be the same.
The drive knows about the disk media (normal or high density) from the sensor which detects the corresponding slit in the floppy case the moment it gets loaded. Default position means high density. If there‘s no slit in the disk, the sensor is triggered and that gives the signal that it’s 800 k disk only.
Maybe there’s something there which is not OK. I would check the level downstream the sensor before and after inserting a 800k disk.
 
  • Like
Reactions: JDW