Portable / Powerbook SWIMs

[desertrout]

In an effort to nail down a potential replacement SWIM for my 5120, I began looking at SWIMs on other boards of a similar vintage. I have narrowed down the potential options to:

344S1029: the direct replacement, what I pulled off my 5120

344S1029-A: seen on 5126, PB100, 140, 170

343S1029-A: seen on PB100, 140, 170, 145B, 165, 180c (and I assume all others prior to the 150).

The SE/30, IIvi, LC, LCII, Classic, IIsi etc. all use the 344S0061-A, which is also PLCC-44 like the above but I'm not sure if they would be suitable as a replacement if power is a factor. My understanding is that SWIM is 100% backward compatible, even with IWC... and seeing that the PB100 / 140 / 170 are seen with two versions of the SWIM, one of those versions is found on the 5126, I'm venturing that early PB SWIMs would work just fine as a replacement on the 5120.

Does anyone have any specific knowledge about this? I haven't been able to find anything in search.

 

[techknight]

They are all the same. ive used desktop logic board SWIMs on Portables without issue. 

 

[desertrout]

They are all the same. ive used desktop logic board SWIMs on Portables without issue. 

Interesting, and awesome - thanks!  I'll source a donor board.

I'd be very curious to know the differences between the PN's.

 

[techknight]

Who knows. All I know is ive tried to use them interchangeably, and I have not had any ill-effects. 

 

[desertrout]

LOL - I suppose that's all that matters ultimately. If it works, it works.

Still, even if it's ultimately trivial, there was an intentional choice to put 344/343S0061-1/A in desktops, and 344/343S1029-A in laptops. I'm not going to lose sleep over it... probably.

 

[techknight]

I am going to assume its probably power consumption related. 

 

[desertrout]

Yeah I can't imagine it'd be anything else (floppy drives are floppy drives), and probably less of an issue for the Portables than for the PB's. Anyway, a seller on eBay is offering 343S0061-A's for $10 (or make an offer, which I did), so one's on the way. I'll update once it's in.

 

[Stephen_Usher]

Could the desktop ones be NMOS and the laptop ones CMOS?

 

[techknight]

If that relates to power consumption, then its entirely possible. Unless we have internal apple documentation, then we have no way of knowing, unless you know how to do certain experiments? 

Regardless, ive used them interchangeably and luckily have had no ill-effects as of yet. 

 

[Stephen_Usher]

Yes, CMOS is less power hungry, but it also has a lower voltage for logic high and has a lower maximum current throughput. CMOS driving TTL/NMOS chips can cause them damage due to excessive power draw. Also the TTL chip can sometimes not see the logic '1' values. There are special CMOS chip types which can interface with TTL/NMOS chips.

 

[desertrout]

FWIW, I can confirm that the 343S0061-A works just fine in the Portable (5120), as expected.

 

[desertrout]

FYI, here's the word from the Portable Developer Notes that confirms it is CMOS, with some additional power management-related instructions (page 6-10) (http://mirror.informatimago.com/next/developer.apple.com/documentation/Hardware/Developer_Notes/Macintosh_CPUs-68K_Portable/Portable.pdf):

The SWIM chip is a static cell CMOS design and as such has a negligible DC component to the current drawn during operation (the major contributing factor is the AC component). For the AC component, three factors determine the amount of current drawn:

• Frequency of operation

• Number of gates switching

• Capacitance to be charged when switching

The strategy for power management on the SWIM chip is to control the frequency of the clock (power is left on to this device even during the sleep state). Specifically it is either on and switching at 16 MHz or it is off (DC). Savings by using this technique have been measured to be about 35 mA. The state of the clock is controlled via the power manager processor through the Normandy ASIC. The normal state of the clock is to be off (thus providing the lowest average power). However, because the floppy drive is not able to interrupt the system when a disk is inserted (and subsequently notify the system to turn the SWIM chip clock on) the system must periodically turn the clock on and check for a disk insert event. Of course, while a disk is inserted and the drive is in operation, the clock should be on.

There ya go!

 

[gijsmans]

FYI, here's the word from the Portable Developer Notes that confirms it is CMOS, with some additional power management-related instructions (page 6-10) (http://mirror.informatimago.com/next/developer.apple.com/documentation/Hardware/Developer_Notes/Macintosh_CPUs-68K_Portable/Portable.pdf):

Thanks, great documentation. 

 

[desertrout]

NP. Combined with @Stephen_Usher's comment, I think I should probably start looking for donor boards to get the proper 343S1029-A...