Macintosh IIsi PRAM problems ...

[franklyn]

I successfully repaired my Macintosh IIsi ...

IIsi fault ...

I have a IIsi that works before fine and now it take some time until the sound of dead sounds up. The motherboard looks good except for a little oxidation on the four 74ALS245 chips. I desoldered these all and cleaned everything up and checked connections and soldered on four new chips but the...

68kmla.org

... but I noticed that the computer has a problem with the PRAM, it always loses the settings it has made after switching it off and rarely but not.

What could be the error and where should I look?

 

[JC8080]

You're sure the coin cell battery is good, and the installation is sound, getting power to the right places?

 

[franklyn]

Yes I think so ...

I change the cell fro a new one (maybe a 3V cell is to low?)

 

[franklyn]

I took a closer look and it turns out that the pram stays active for a while after it is switched off but is lost overnight. I'm now going back to the original 3.6V battery...

 

[JC8080]

I took a closer look and it turns out that the pram stays active for a while after it is switched off but is lost overnight. I'm now going back to the original 3.6V battery...

I wonder how often modern 1/2 AA batteries need to be changed to be safe.

 

[robin-fo]

Is it possible that you have a short on your logic board which drains your battery?

 

[franklyn]

The same behavior also occurs with the original 3.6V battery. If the computer is connected to the mains but is switched off, everything seems to be fine. However, as soon as I switch off the mains and the computer is left for a while, the PRAM is erased. The battery is not discharged. The transistor Q7 cannot be defective, but what about the double diode?

 

[nathall]

Interestingly enough, I have the exact same problem with mine. It’s the only “defect” after a full recap and thorough cleaning, so I’ve learned to live with it. It’s always connected to power so it only loses settings if I have an outage. I don’t bother to keep a battery in it any more. Be interested to hear if you figure it out, though.

 

[franklyn]

I don't know what the error could be. The battery voltage is present at 68HC05 on pins 13 and 27.

 

[Boctor]

I have this as well on my good IIsi. I have a broken motherboard where this circuit works as intended, but the test pads near the oscillator and the MCU do not have a pulse on the good board. I've been checking continuity, resistances, hertz, etc for hours and would be curious to know if anyone finds the solution. Surely the MCU itself can't have gone bad in such a specific way, right?

Edit: Just heatgunned the MCU from a board where the test pads show the correct hertz and stuck it on, no change whatsoever. Another component in the circuit or a continuity issue is more likely at play here.

 

[Boctor]

You are not going to enjoy hearing this, but I found the break in the circuit, at least on mine. It was between R33 and C107+C105 (back side.) I did this IIsi in 2017 or so, before I was anywhere near as good at neutralizing and washing PCBs, and some residual electrolyte was trapped in my seemingly-shiny solder joints. The trace next to R33-- which is also connected to the pad of the nearest SMD cap, C34-- has begun to rot because of my 21-year old self's negligence. It is no longer connected and needs a little solder bridging action. I would never have caught this, if not for compulsively testing the RTC PRAM battery circuit.

If in doubt, check continuity between C34's north-facing pad and R33's south-facing one, please. It may just save your machine's life if you made a goof like I did. Otherwise, the first page of the Bomarc IIsi schematic is your best friend.