SE/30 reset not working

[yoyomac]

Hi. I signed up to the forum as I could not find any cue on the web for the following problem.

I have an SE/30 that has worked perfectly well for many years until recently when I noticed that the Reset button (next to the Programmer's switch on the left side) had no effect anymore.

In details, when the button is pressed, the screen goes black (regular behavior) but when the button is released no reset/reboot/bong takes place, the screen comes back with the same content it had before (not even garbled) but the Mac though is now frozen (not suprising I guess). It seems the reset signal did not reach the CPU, or the startup circuit.

At this point my only option is to turn the power off and back on.

Other than that the Mac always boots fine, no garble, no checker, no stripes on the screen. Also it restarts perfectly fine when I click restart from the pull-down menu in the Finder.

Same behavior regardless of whether I boot from the hard drive (OS 7.1) or a floppy (OS 6.0. 8) . It is a hardware problem.

In doubt, I replaced the lithium battery (some old macs do use a kick from the battery at startup) but to no avail. Also I checked the physical tactile reset button on the board works fine.

Has anybody experienced something similar? Any idea where the problem could be?

Any suggestion welcome...

 

[Apostrophe]

Well if you don't mind my saying so, if your SE/30 boots fine when you use the power switch, then why use the programmer's switch at all?

-Apostrophe

 

[Dennis Nedry]

Well, using the power switch turns everything off and back on whereas the reset button leaves everything on. Maybe it's slightly better for the equipment to not cycle the power like that especially if you're doing a lot of resetting.

What happens if you press Apple + Control + Power on the keyboard?

Also check the logic board for leaky capacitors and clear goop from them on the board. This is a common problem with old logic boards which can be fixed.

 

[equill]

While an SE/30 can use a Desktop II or later ADB keyboard for input, it is not pseudo-soft power savvy as a Colour Classic is, ie it cannot use the power key on the keyboard for startup or shutdown.

The interrupt and reset switches both connect chosen points in the MLB circuitry to ground. If you get part of the expected result of pressing one of them, but not the whole result, it suggests an open circuit somewhere else. It is not beyond possibility that the open circuit is the result of a broken trace on the MLB, caused by the universal villain: leaked capacitor goo. If your SE/30's MLB has not already been reworked to replace its original capacitors, a thorough board wash followed by minute trace inspection and component replacement may be staring you in the face.

At the very least you should open the SE/30 to inspect the MLB for (1) the presence of modern tantalum electrolytic capacitors, mounted horizontally, or (2) the tell-tale smudges on the MLB surrounding one or more of the older aluminium electrolytic capacitors. In the words of trag, who keeps a supply of replacement capacitors for SE/30 owners, the faulty caps are those that look like little fuel tanks, mounted vertically.

These forums are stuffed full, pressed-down and running-over with discussions about capacitor leakage, so you will have no difficulty in getting info. about capacitor replacement.

de

 

[Dennis Nedry]

While an SE/30 can use a Desktop II or later ADB keyboard for input, it is not pseudo-soft power savvy as a Colour Classic is, ie it cannot use the power key on the keyboard for startup or shutdown.

I was wondering about this, I thought that key combo might not work on some Macs but I wasn't really sure.

 

[equill]

It's easy enough to discern in desktop Macs when you need to do so.

True 'soft power' (called active power-control by Apple) needs a keyboard power-key, and there is no AC mains-power toggle-switch at the rear of the Mac. There may also be a soft-power (indicating) switch at the front of the Mac. (Most Macs from the Mac II series onwards)

Pseudo-soft power Macs have an AC mains-power toggle-switch at the rear of the Mac, but can respond to a keyboard power-key. (Colour Classics/P250 and P275, and many of the all-in-one LCs, Performas, and 5nnn PowerPC Macs)

Passive power-control Macs have only an AC mains-power toggle-switch at the rear of the Mac. The toggle-switch must be used even if there is a keyboard power-key. (All Compact AIOs from SE to Classic II. Macintosh to Plus don't use ADB keyboards at all.)

de

 

[JDW]

How long was it between the time you last used the switch when it worked, and the next time you pressed it when it didn't work?

Switches do break (if not the switch itself, then the solder contacts on the logic board) and contacts do oxidize over the years. You need to carefully open your Mac and check the switch on the logic board itself. And I personally would love to see some crisp, zoomed-in photos of your switch when you do that (both top and bottom of motherboard).

 

[Scott Baret]

The keystroke will work on any Mac provided you have one of the following INITs installed:

Programmer's Key

Interrupt Button

These allow any Mac to use the Control-Command-Reset combination. I'm not sure about Interrupt Button, but I know Programmer's Key even lets you do that on a Plus, swapping the ` key for the reset key. You can also use your interrupt button this way by pressing Control-Reset (or ` on a Plus).

Both are public domain utilities and are pretty easy to find.

 

[yoyomac]

How long was it between the time you last used the switch when it worked, and the next time you pressed it when it didn't work?
Switches do break (if not the switch itself, then the solder contacts on the logic board) and contacts do oxidize over the years. You need to carefully open your Mac and check the switch on the logic board itself. And I personally would love to see some crisp, zoomed-in photos of your switch when you do that (both top and bottom of motherboard).

I've been using the switch regularly (like a couple times every week, mostly because of Netscape 2.0.2 hanging now and then). So I immediately noticed when the switch stopped working.

Initially I suspected the switch itself. I desoldered it, measured the resistance across the pins of the switch, and across the grommets on the motherboard where the switch sits. Then I soldered the switch back in place. No improvement.

I'm pretty confident the switch itself works fine.

Here's a summary of my ohmmeter tests:

* Reset switch (location S1 on motherboard):

- Switch mounted on board, pin-to-pin, button unpressed: 0.625 kOhms

- Switch mounted on board, pin-to-pin, button pressed: ~1 Ohm

- Switch desoldered off the board, pin-to-pin, button unpressed: infinite kOhms

- Switch desoldered off the board, pin-to-pin, button pressed: ~1 Ohm

- S1 grommet-to-grommet (switch removed) on board: 0.625 kOhms

* Programmer switch (location S2 on motherboard):

- Switch mounted on board, pin-to-pin, button unpressed: 1.425 kOhms

- Switch mounted on board, pin-to-pin, button pressed: ~1 Ohm

By the way the Programmer switch responds perfectly fine at all times.

Anybody care to check their values on their motherboard with a working reset switch? Let me know. It might help me locate the problem (open circuit hypothesis).

Next, I'm going to look real close for a broken trace on the motherboard and check the capacitors.

 

[yoyomac]

The interrupt and reset switches both connect chosen points in the MLB circuitry to ground. If you get part of the expected result of pressing one of them, but not the whole result, it suggests an open circuit somewhere else. It is not beyond possibility that the open circuit is the result of a broken trace on the MLB, caused by the universal villain: leaked capacitor goo. If your SE/30's MLB has not already been reworked to replace its original capacitors, a thorough board wash followed by minute trace inspection and component replacement may be staring you in the face.

That's what I feared... this one will get solved the hard way...

I was secretly hoping somebody would post a reply like "it-happens-to-me-all-the-time-all-you-need-to-do-is-replace-component-xy34"

I'm torn. Since this is the only (and minor) problem I have, I'm not sure I feel like raiding all the capacitors on the motherboard without actually knowing for sure it'll fix it.

 

[JDW]

I have SE/30 schematics and logic board photos. Using those, I created a detailed PDF this evening that shows the relevant connections of the RESET switch. I have uploaded this PDF to my docs.google.com account. However, Google has not yet implemented PDF sharing (poo-poo on your Google!), so I am forced to "invite" people to this PDF by email address. Therefore, if you would like to access this PDF, please PM me your email address and I will promptly issue the invites for you.

In the PDF I created, I show pertinent sections from two schematics: one drafted by Apple and another reverse-engineered by BOMARC. Both schematics are electrically the same. They are just drawn differently. I highlighted the relevant traces in yellow on both schematics.

At the bottom of my PDF you can see top/bottom photos of the Reset switch. Simply put, the Reset switch goes through resistor R5 to pin-6 of both Sony chips UB10 & UB11. The Reset switch merely feed Ground to both of those pins through R5. Hence, to test if the switch is at fault, one need only temporarily solder a little wire to the Reset Switch side of R5, reconnect cables to your logic board, then boot your Mac, and then touch the tip of the wire you soldered to ground somewhere on the logic board. You will be doing the same exact job as the switch would -- feeding ground through R5 to pin-6 of each Sony chip. If that causes a reset, you know your switch is bad, regardless of your DMM tests. If it doesn't force a reset, then perhaps R5 is bad, or the pin-6 connections to each of the SONY chips is bad, or perhaps the SONY chips themselves could be bad? Does your audio work fine (loud)?

Again, if you want to see the PDF (which really makes the above discussion crystal clear), please just PM me an email address.

 

[equill]

... same exact job as the switch would -- feeding ground through R5 to pin-6 of each Sony chip. If that causes a reset, you know your switch is bad ...

Entirely true (in the Bomarc schematics), but only part of the story given all the happenings at reset. The action above presumably invokes the chime 'I am resetting, o Master', but there is also the grounding of summat else by way of R11, and that is, presumably, the serious part of the reset: screen blanking followed by the rest of the chain of events. Even with the help of command-option-8, however, I could not, in a quick perusal, definitely trace the target of this action in my copies of Bomarc. What, in words, do the Apple schematics show?

de

 

[yoyomac]

In the PDF I created, I show pertinent sections from two schematics: one drafted by Apple and another reverse-engineered by BOMARC. Both schematics are electrically the same. They are just drawn differently. I highlighted the relevant traces in yellow on both schematics.

Victory! I got the problem fixed. The schematics got me started scrutinizing the motherboard and I found out that capacitors C6,C9 and C10 had been leaking.

I replaced the 3 of them and, miracle, the reset is now fully working again.

Thanks for all the pointers and the schematics, it helped.

I'll look into replacing all the other capacitors. However I must say my motherboard is pretty butchered, I'm glad the Mac is still working. My soldering iron I think is way too big, a generic Radioshack 64-2070 25W. I really made a mess desoldering the old caps.

Anybody has a suggestion for a budget soldering iron ($20-ish) more appropriate for motherboard work?