128k/512k boot sequence

[Cinan]

Have done some of my own research.  Glass bead caps it appears.  I assume I can replace this with a ceramic cap of 4.7uf 35v ?

 

[Cinan]

I have taken C9 out of the board to test it.  it is just showing an open circuit on the multimeter when I test for capacitance.  I have a spare non-working Mac Plus board,  is there a glass bead cap on here that I can transplant,  too hard to read so I need some experience to tell me - please.

If I can't transplant can I replace C9 with an electrolytic or a ceramic cap ?  On the schematic it looks like it is polarised but on the board there is no +/- marking,  help !!!

 

[cheesestraws]

So I replaced LM311 without any change.  I am now thinking it might be C9 as per the image above.  In my understanding if C9 isn't working properly then the -ve won't be presented to pin 3 and therefore LM311 won't create the reset signal.  But if the reset switch is pressed (SW2) then it will complete the circuit and the negative presented to P3 and hence LM311 creates the reset signal when I press SW2 and the CPU reboots ??  is that logic right ?

Sorry for slow reply.  That's pretty much what I thought was going on in that circuit, too.  If I'd been thinking, the reset switch working suggests that the chip was fine.  Oh well. 

I assume I can replace this with a ceramic cap of 4.7uf 35v ?

This is probably what I would try in your position.  If I'm reading the circuit right, C9 is mostly there to set the time delay before the reset signal is being generated, so you probably have a certain amount of latitude in precisely what you replace it with.  With that in mind, I'd start with whatever I had in the cupboard that had about the right specs, which probably would be a ceramic of some description.

 

[Cinan]

No change unfortunately,  I used a tantalum and then an electrolytic. the sequence is as follows :

1. power on - vertical bars

2. reset switch - boot ok with bong.

3. power off

4. power on - boots ok with bong

5. power off

6. power on - boots ok with bong

point 4& 6 only work if they are within a few seconds of 3&5 leading me to believe that there is a capacitance issue somewhere as once the charge has been lost it goes back to point 1.

I can't see other capacitors on the schematic that would be causing that.  So unless there are any suggestions looks like I am using the reset switch as the 2nd power on switch.

 

[cheesestraws]

I wonder if the power supply is taking a while to become stable when the machine is first "switched on", either for thermal or capacitance reasons.  The reset signal to the CPU will only work if the power supply lines are nice and solid at that time.  On later machines, I believe, the reset signal is actually not generated until the power rails are stable, but on this one it just looks like a timer.  I don't suppose you happen to have another one of these machines or a Plus do you?  It would be useful to try this board out with another analogue board/PSU to rule this out

 

[Cinan]

Yes I have already tried that.  I have used this LB with a 128k analog with same response and the 128kbLB with this AB and it works fine.  So it's something on the LB.

 

[cheesestraws]

Ah.  Well, that's irritating .  I'm not sure how you would troubleshoot this further without an oscilloscope or way to look at fast signals, I'm afraid.  I'll keep thinking...