Free PowerBook 1400c 166MHz upgrade...

[Phipli]

Oh, gosh, this is amazing! My 'spare' PowerBook 1400, the one that barely works (in a sense); has a faulty PCMCIA connector and I think the drive connector doesn't work, in fact has the Motherboard that will boot the 166MHz CPU! I get a chime and an image on the (now) cs display. There's no drive attached, so it just sits there with the floppy disk ?ing at me!

This means, in theory I can do the upgrade! It also means I have a significant amount of work ahead of me to swap the motherboards!

That's nuts: I've been basically treating that one as the rubbishy spare one and downgrading it over time: removing 16MB of RAM; swapping its 1400c display for the other PB's cs display; despising it for the meagre space it's taking up. And now it turns out it's the one I need! Hoarders of the world, rejoice!

-cheers from Julz

Congratulations

Should be quite a speed bump. Might even match the speed of a 100MHz 8100 after the upgrade

 

[Snial]

Congratulations

Thanks!

Should be quite a speed bump. Might even match the speed of a 100MHz 8100 after the upgrade

LOL! I'm not sure anything could match the speed of a 100MHz 8100, those things go like the blazes!

 

[Phipli]

LOL! I'm not sure anything could match the speed of a 100MHz 8100, those things go like the blazes!

An 8100/110?

 

[GRudolf94]

Interested in good images of the boards when you get them out, @Snial - even though they have separate assembly numbers (820-xxxx), I'm not aware of a single case where Apple used different PCBs for versions of machines. See, for instance, G5s not upgrading regardless of which CPU module you install, but boards being the same etch (630-xxxx), or MDDs being strapped for 133/167MHz bus.

Maybe we can spot the presumable difference in strapping/populated parts that makes 166s the odd duck.

 

[Phipli]

I'm not aware of a single case where Apple used different PCBs for versions of machines.

630 vs 630 DOS

Possibly 8100/80 Vs 8100/100, unless they had preplanned for the 601v, I forget.

Off topic, I was just trying to see if I could think of any because it does feel like something Apple would do. Especially with machines that don't have a CPU daughter board.

 

[GRudolf94]

630 vs 630 DOS

Possibly 8100/80 Vs 8100/100, unless they had preplanned for the 601v, I forget.

Off topic, I was just trying to see if I could think of any because it does feel like something Apple would do. Especially with machines that don't have a CPU daughter board.

IIRC 8100 mentions some different resistors on the CPU power supply rail between 110 and 80/100. 630 DOS is just a 580

 

[Phipli]

IIRC 8100 mentions some different resistors on the CPU power supply rail between 110 and 80/100. 630 DOS is just a 580

The data we have in the 8100 is from the 100MHz design though, I was trying to remember what the situation is on the 80 board

 

[Snial]

An 8100/110?

In my pedantic defence, it won't match the speed of an 8100/100 if it goes faster ;-) !

 

[Phipli]

In my pedantic defence, it won't match the speed of an 8100/100 if it goes faster ;-) !

If I generate lots of interrupts through ADB at a tuned frequency, I can make it as close as I like

 

[Snial]

Interested in good images of the boards when you get them out, @Snial - even though they have separate assembly numbers (820-xxxx),

Sure, I'll take photos! I have an iPhone 7, so by today's standards I'm not sure the images qualify as 'good'.

I'm not aware of a single case where Apple used different PCBs for versions of machines. See, for instance, G5s not upgrading regardless of which CPU module you install, but boards being the same etch (630-xxxx), or MDDs being strapped for 133/167MHz bus.

Maybe we can spot the presumable difference in strapping/populated parts that makes 166s the odd duck.

Indeed, then maybe the earlier motherboard can be hacked. I understand that both boards are 33.3MHz, so my first thought was that it's less likely to be a timing issue. I wondered if the issue is that the earlier board can't deliver enough power to the 166MHz CPU, or if the 166MHz CPU doesn't allow enough power for the motherboard. The latter doesn't entirely make sense to me, but I think I read in a reddit post or in the PB1400 service manual that you get a click instead of a boot and a chime if there isn't enough power going to the CPU. Which is what I kept hearing when I plugged in the 166MHz CPU to the 117MHz board.

-cheers from Julz

 

[Snial]

If I generate lots of interrupts through ADB at a tuned frequency, I can make it as close as I like

Squirm, squirm, I'll have to think a bit before I come up with a suitably obtuse retort for that one ;-) !

 

[Phipli]

Sure, I'll take photos! I have an iPhone 7, so by today's standards I'm not sure the images qualify as 'good'.

Generally the most important thing is good lighting. Taking photos tends to capture the interesting details best in good natural light (e.g. a window). Less blur and flash reflections.

Another thing is wipe the lens with a clean rag and IPA before taking important photos. You'd be amazed

Squirm, squirm, I'll have to think a bit before I come up with a suitably obtuse retort for that one ;-) !

You could comment on the granularity of the interrupt?

 

[CircuitBored]

@Snial Would you still like the video-out board from my 1400c? I offered it at the UK meetup and am just now realising that I never followed up on it. Drop me a line if you'd like it.

 

[Snial]

@Snial Would you still like the video-out board from my 1400c? I offered it at the UK meetup and am just now realising that I never followed up on it. Drop me a line if you'd like it.

That's very kind of you! Are you sure you don't want it for your own 1400c? It might be worth waiting until I've done the mobo swap, because I could end up with two dead PB1400s :-O !

 

[Snial]

Well, I've managed to get down to the 166MHz logic board. It was all OK, except that the PCMCIA flaps got in the way of finally removing the logic board and when I did, I broke the eject button cap (or whatever it's called). I'll aim to avoid that when I take the other PCB out.


The ROMs are correct for the 166MHz board. From the PowerBook 1400 service manual:

"
Part Number 661-1188 has ROM U3 (Low, 341S0203) and ROM U4 (High, 341S0204). It works in PowerBook 1400, 117 MHz and 133 MHz units.
Part number 661-1381 has ROM U3 (Low, 341S0364) and ROM U4 (High, 341S0365). It works in PowerBook 1400, 117 MHz, 133 MHz, and 166 MHz units."



Sorry about the glare, I decided not to take the plastic cover off. The board looks OK to me, the PB1400 it came from is somewhat tarnished.

Next up, my other PB1400.

-cheers from Julz

 

[3lectr1cPPC]

I believe the only difference between the two boards are the ROMs, nothing else. Mine looked identical when I had a 117 and 166 board side by side.

 

[Snial]

I believe the only difference between the two boards are the ROMs, nothing else. Mine looked identical when I had a 117 and 166 board side by side.

So, a 117MHz & 133MHz board could be hacked to support 166MHz CPUs if we could replace the ROMs (e.g. with Flash)?

 

[3lectr1cPPC]

Perhaps? Not sure. My guess that the ROMs are the only difference goes on visual appearance alone pretty much.

 

[LaPorta]

While you have it apart, recap it. Same goes for the screen.

 

[MacUp72]

That is the 166mhz board? I cant remember seeing on mine that slim aluminium rail on the top of the board back..have to look that up