• 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.

SE/30 stuck at boot with raster but no cursor

mrkirkby

Member
Hello everyone,

I'm looking for a bit of advice. I have an SE/30 board which I've recapped but I am having problems with further diagnosis on the above problem.

I'm aware of the dead mac scrolls and Larry's other book and have tried suggestions found around the web to replace the bourns resistor networks and SCSI chip.

Here's the story so far:

Recapped with tantalums. I've had them off a few times to check for shorts underneath and have gone through each adding solder mask where there's a risk.
Replaced all 3 resistor network chips with donor board parts.
Removed SCSI chip, Swim chip, Glue chip and UG12. Cleaned with IPA and re-soldered.
Completely replaced SCSI chip and UG12 with donor parts from another board.
Checked the reset line going to SCSI and SWIM. It's sat at 5V and drops to 0 when reset is pressed.
I've checked voltages at the power / video connector and all are present and within tolerance.
The whole board has been ultrasonically cleaned.

I think it's possibly a bad via or internal voltage missing. Any ideas where I could focus my attention? I guess checking voltages at each chip might be a good idea so I'll do that next. One thing I do have of note is when measuring resistance between +12 and -12, in one direction there is a low resistance value. Could possibly be related. I did try removing -12V via my ATX extender to see if it made any difference as that line only appears to head for the expansion slot but no luck.

If anyone has any experience with these beyond 'it's caps' my ears are open!

Many thanks in advance and happy to have finally registered here following decades of stalking ;)

@techknight, particularly want to thank you. I've taken your advice in the past from previous posts and come good with a few other boards.

All the best,
Andrew
 

Garrett B

Well-known member
Hi Andrew. I am in a similar position with my Mac Classic board. I'll let you know if I make any headway on mine that might be helpful for your SE. Good luck!
 

mrkirkby

Member
Hi Garrett,

Most posts I've seen so far point toward shorts in bad traces caused by tantalums or bad vias etc.

Just can't find it. I had a good hour long probe around earlier and still none the wise. Tried swapping the sony chips out too. I'm almost certain it's a via but where!?

It's driving me a little crazy!

Will let you know if I find the culprit too.
 

max1zzz

Well-known member
Do you get a chime? and when you say you get a raster, whet exactly is on the screen? is it just solid white or is there and visible pattern?
 

mrkirkby

Member
Yes to chime. Raster is the raster pattern drawn from the ROM as described in a bunch of reference sources. Usually accompanied by a cursor but it looks like SWIM isn't initialising as per the bourns problem described above so I have raster, but no cursor.

Looks like I'm perhaps missing some pins to the ROM perhaps. The area usually affected around the power connector is quite good on this board strangely and in general most of it has escaped corrosion save a couple of hot spots.

Think I'm possibly on my own with this as it's most likely a data line missing or internal trace gone bad.
 

techknight

Well-known member
Hello everyone,

I'm looking for a bit of advice. I have an SE/30 board which I've recapped but I am having problems with further diagnosis on the above problem.

I'm aware of the dead mac scrolls and Larry's other book and have tried suggestions found around the web to replace the bourns resistor networks and SCSI chip.

Here's the story so far:

Recapped with tantalums. I've had them off a few times to check for shorts underneath and have gone through each adding solder mask where there's a risk.
Replaced all 3 resistor network chips with donor board parts.
Removed SCSI chip, Swim chip, Glue chip and UG12. Cleaned with IPA and re-soldered.
Completely replaced SCSI chip and UG12 with donor parts from another board.
Checked the reset line going to SCSI and SWIM. It's sat at 5V and drops to 0 when reset is pressed.
I've checked voltages at the power / video connector and all are present and within tolerance.
The whole board has been ultrasonically cleaned.

I think it's possibly a bad via or internal voltage missing. Any ideas where I could focus my attention? I guess checking voltages at each chip might be a good idea so I'll do that next. One thing I do have of note is when measuring resistance between +12 and -12, in one direction there is a low resistance value. Could possibly be related. I did try removing -12V via my ATX extender to see if it made any difference as that line only appears to head for the expansion slot but no luck.

If anyone has any experience with these beyond 'it's caps' my ears are open!

Many thanks in advance and happy to have finally registered here following decades of stalking ;)

@techknight, particularly want to thank you. I've taken your advice in the past from previous posts and come good with a few other boards.

All the best,
Andrew
I was going to ask, i dont see where you specify the actual problem unless i overlooked it.

Then i followed further in the thread and i wanna make sure i actually got this right: you get a chime with a raster.

What raster do you get? got a pic? what chime? Death chimes or good chimes? or good then death?

If you get a good chime with a gray screen like a normal SE/30 that means 95% of the circuitry is functional and you are having problems either communicating with the SCC or SWIM.

Start there... check all lines to the SWIM on every single connection back to each point to each IC as indicated on the schematics. same thing with the SCC.

If that looks ok, replace the SWIM with a new one.
 

mrkirkby

Member
I was going to ask, i dont see where you specify the actual problem unless i overlooked it.

Then i followed further in the thread and i wanna make sure i actually got this right: you get a chime with a raster.

What raster do you get? got a pic? what chime? Death chimes or good chimes? or good then death?

If you get a good chime with a gray screen like a normal SE/30 that means 95% of the circuitry is functional and you are having problems either communicating with the SCC or SWIM.

Start there... check all lines to the SWIM on every single connection back to each point to each IC as indicated on the schematics. same thing with the SCC.

If that looks ok, replace the SWIM with a new one.
Ah sorry, I think I wrote the post while still in my head a little. You know how these things are when you're thinking problems through.

So for more clarity, the image shown is of the pattern you see immediately after a good mac's display comes on. Before the cursor appears shortly later, followed by the ? icon or Welcome to macintosh. Older technical reference books refer to it as the grey raster pattern.

The chime is a happy mac. Pressing the programmers switch does nothing. Reset resets. The machine seems to enter a halt state before SWIM has initialised as described in some of those books.

Totally agree on the 95% remark. I've many of these machines and have put the others back into action. This one's just escaping me. The only remedy is more probing I think. I have looked at the old schematics a few times as a reference with some of the probing I've already done.

Thanks for your input
 

techknight

Well-known member
Just make sure all the lines to the SWIM, and i mean ALL of them tone back to the GLU, CPU/ROM areas.

If that looks ok, then replace the SWIM.
 

mrkirkby

Member
Got good tones on all of them. Replaced the SWIM for another from a donor machine and same result.

Might be time for me to invest in a logic sniffer and have a look at the lines at the PDS. I wonder if I have a short somewhere on the board perhaps. What’s a little odd was that just after the re-cap, the system worked for one boot cycle then that was it. It showed Simasimac before that point so I’m not sure if it was ever good before this.

I have another one with somewhat similar symptoms although I have a cursor on that one. Will focus on this first though to save confusing myself.

Thanks for your help!
 

rikerjoe

Active member
I'm following this thread because my SE/30 just started showing the same symptoms - boot chime followed by the gray raster screen and no mouse pointer. It worked fine after a logic board recap and only just started showing this problem. Another possibly related problem is in recent weeks when the SE/30 was still booting I had to plug the keyboard and mouse into separate ADB ports. They wouldn't work if the mouse was plugged into the keyboard and the keyboard into one of the ADB ports. Keyboard and mouse work fine on other classic macs, and other keyboards showed the same problem. I was suspecting the +5V rail might be low when the gray screen and no mouse pointer showed up. Might all of these be related to a low +5V rail? I toned out the SWIM and SCSI and ADB chips and all tone out just fine. Voltage rail measurements resulted in the following: +12V measured +11.97, -12V measured -11.80, +5V measured +4.75, and -5V measured -5.20. The +5V rail seems low to me. I'm guessing this means I need to recap the power supply. I'll report back in case this works for me.
 
Last edited:

mrkirkby

Member
I'm following this thread because my SE/30 just started showing the same symptoms - boot chime followed by the gray raster screen and no mouse pointer. It worked fine after a logic board recap and only just started showing this problem. Another possibly related problem is in recent weeks when the SE/30 was still booting I had to plug the keyboard and mouse into separate ADB ports. They wouldn't work if the mouse was plugged into the keyboard and the keyboard into one of the ADB ports. Keyboard and mouse work fine on other classic macs, and other keyboards showed the same problem. I was suspecting the +5V rail might be low when the gray screen and no mouse pointer showed up. Might all of these be related to a low +5V rail? I toned out the SWIM and SCSI and ADB chips and all tone out just fine. Voltage rail measurements resulted in the following: +12V measured +11.97, -12V measured -11.80, +5V measured +4.75, and -5V measured -5.20. The +5V rail seems low to me. I'm guessing this means I need to recap the power supply. I'll report back in case this works for me.
Worth checking your bournes network packages toward the back. Take a look at attached. It was step one in my process so far but mine appears unrelated.
 

Attachments

  • 5F773C00-D48A-47EC-A2BB-7516765A21E2.jpeg
    5F773C00-D48A-47EC-A2BB-7516765A21E2.jpeg
    423.1 KB · Views: 43

mrkirkby

Member
Worth checking your bournes network packages toward the back. Take a look at attached. It was step one in my process so far but mine appears unrelated.
Should also add there’s more on this in larry’s repair secrets book detailing the se and SE30 along with expected resistance values measured in circuit.
 

rikerjoe

Active member
Should also add there’s more on this in larry’s repair secrets book detailing the se and SE30 along with expected resistance values measured in circuit.
Awesome, thanks for the info. One more data point to add. I had a "doh" moment and decided to test my SE/30 board in a working SE machine. Ended up with the same result - chime, gray raster screen and no progress to mouse pointer. Voltages measured similarly on the SE/30 board: +12V rail measured +12.06, -12V rail at -11.91, -5V rail at -5.04, and +5V rail at +4.81. I will do some measurements on my working SE board as a control case and check on likely suspects such as the ones you suggested in about a week once I get a chance to work on the SE/30 board again. Good luck in your hunt.
 

mrkirkby

Member
Made some progress tonight.

After a ton of probing in all of the wrong areas, a number of pins on the ADB chip were found not to be making contact. It's the first I've seen this problem related to that specific part, but sure as hell once re-soldered the mac now gets to the blinking disk.

I still have some SCSI problems in that the machine won't boot from SCSI either internal or external but I'm sure some further probing will trace the problem easily enough.

I'll update for a complete view but I'll leave the note here should anyone have the same adventure. I saw it boot an OS from my floppy emu earlier which is great to see.

One less on the land fill.

Thanks to all who have helped so far.

A
 

mrkirkby

Member
Worth checking your bournes network packages toward the back. Take a look at attached. It was step one in my process so far but mine appears unrelated.
So take a much closer look at your ADB chip following what I found above. I'd bet it's probably just needing a bit of solder. Probe the pin to the board at each trace and see if you're missing any. Given you were loosing ADB, I'd say you have a solid bet it's the same issue. Let me know if you want any further help. Some of the repair books I've read mention the chip is more sensitive to heat than others, so maybe avoid blasting it with hot air if you have such tools. It's actually just a PIC under the hood but I'd be very surprised if you could find a binary dump.

Cheers,
A
 

rikerjoe

Active member
I checked my ADB chip again from a continuity standpoint - checking both the solder joints at the pads for continuity as well as via the schematic to neighboring locations - and continuity checked out fine. I also checked all three of my resistor packs as a control case for RP10, measuring crosswise pins in circuit, and all measured 45 ohms. I checked my voltage rails again, and my +5V rail is now measuring +4.6V. Unloaded the power supply is putting out +5.02V on the +5V pin. Both of those numbers seem a little low - unloaded I would expect +5.2V, and loaded +5.0V, but I don't know if running slightly low on the +5V rail would lead to boot chime but not getting through the initialization sequence, getting stuck on the gray roster prior to the mouse pointer appearing. Sigh. I'll do some more testing with my SE which boots fine to see what I can learn, and perhaps that will give me an insight into the problem with my SE/30.
 

rikerjoe

Active member
It's not a power supply problem. I used my SE board in my SE/30 and it booted just fine. I believe I narrowed the culprit to a faulty filter, 155-0007-E located at L13 on the SE/30 board, and marked as L1 LNET on the schematics. I think the pin numbering is incorrect on both the Apple SE/30 original schematics as well as the re-engineered schematics, which threw me for a loop. Regardless, I did some internal continuity measurements on the filter and compared versus the filter on my SE board marked 155-0007-C. Opposite pins (1-8, 2-7, 3-6, and 4-5) show continuity on both filters. However, the working filter on the SE also gives continuity crosswise on the central pins, 2-6 and 3-7. This does not happen on my SE/30. I wish I could find a data sheet on this filter to confirm the above, but not luck so far. Looks like I will need to track down a replacement. Wish me luck!
 

rikerjoe

Active member
I thought I'd provide an update...with a good outcome. I decided to follow the advice provided by @mrkirkby and double-check the ADB chip. I had one or two connections that were suspicious enough that I decided to remove the ADB chip, clean it and the pads, and reinstall. Sure enough, that did the trick. I get a mouse cursor almost instantaneously after power, and successful booting after that. Many thanks for the tip, @mrkirkby.
 
Top