Skip to main content
Home Forums Designing a portable RAM card Designing a portable RAM card
Thread

Designing a portable RAM card

Designing a portable RAM card Hardware 204 posts Oct 4, 2014 — Nov 21, 2018
You used it primarily as a RAM disk, but 100% of it can be used as regular RAM, correct?

I'm 99.99% sure that it can, but I want to be certain as that .001% is bothering me :) .

I will definitely get one of these.

c

yea. I chose to go RAM disk to make sure the RAM was good. 

This Portable Ram Max out card is awesome... i want one for my 2 portables.

My 5120 and 5126.

the 5120 already has a stock 1meg ram card in it.  The 5126 has no ram expansion only one meg ram onboard.

i need at least 2 megs of ram to run 7.1, ( pretty much my favorite OS )

So i will move on one Ram Card, $80 bucks as soon as i get the cash.  And i don't mind getting a little cash back in circulation with mike

Getting him some R&D moneys paid back.

i want one, possibly 2

then i can finally play with my portables.  [:)] ]'>

Backlit needs the /AS jumper. Without it it only sees 5MB. with it I can get all 8MB.

However there is an odd bug. When an application goes into expanded RAM the backlight flickers on and off rapidly with RAM accesses.

If I dont use the /AS wire and cap it at 5mb its perfectly fine.

But my /AS wire was a horrendously long clip lead plus I didnt have /DTACK hooked up either, so that could have been an issue.
Did you ever solve this? Backlight flickering when you access memory above 5MB? I would doubt that the lack of a /DTACK signal has anything to do with it - if the CPU needed /DTACK but it wasn't present, it should just freeze up, or else there might be some external logic that forces a bus error. Either way I think you'd see major software problems, not just a flickering backlight.

My guess is you may have bus contention. Something else is mapped into the memory region above 5MB, and now your RAM card and that something else are trying to drive the bus at the same time. This causes a sudden increase in the power supply current, making the supply voltage droop, and causing the backlight to visibly flicker. The only problem with my theory is that if you had bus contention, you should also also see noise/bad data on the bus, leading to software errors. Maybe the new RAM chips just have more powerful drivers, and overwhelm whatever other thing they're fighting with.

Very nice work!

it was a bad RAM IC. I swapped the RAM, and all is well.... This same card was giving me weird issues as well on the non-backlit, memory manager errors when going to about this macintosh. 

But the Backlit is capped at 8MB, while the non-backlit is at 9MB. My guess is the last MB of RAM is occupied by the backlight controller. 

Replacing the RAM IC fixed it all. My new PCBs will have a 7MB cap jumper. to disable the last meg of RAM for the backlit portable, if it becomes an issue. 

I bought one of these off ebay! Thanks! It works great! :) My 5120 has been entirely recapped by a genius at the soldering gun, waiting on a cable to mount the scsi2sd, with the 8 megs of ram, and working on a new battery solution. I hate the Macintosh (in general) but this baby is just .. way, way too cute. And it looks nice next to the IIc+.

Only thing is, system 608 crashes when trying to access "About the Finder". I got worried for a minute after I installed the card, but after booting 701 and 71 and seeing the entire 9 megabytes the problem definitely lies within system 6. Is it just a case of "There's no way a portable can have 9 megs of ram, I shall now crash the Memory Manager" (rhetorical question I guess, as I won't be running 6 anymore)

Run it without the memory expansion and set the memory control panel and then boot with extensions off. I'm sure there is something in the extensions causing problems.

Then install the memory expansion back in and see if the boots with extensions off and then extensions on.

You need some sort of system enabler. Been there with the same issue.

If you had a version of Virtual Compact that run on Portables (maybe it's hackable?), you could stuff as much RAM on that thing as physically possible.  All the extra RAM-as-RAMdisk would then become virtual memory storage.  So there'd be almost no hit on performance, and you'd have a butt load of RAM.

None of these work on the portable. which is 68000
Sorry.  I did not realize that they had that limitation.   But thinking on it, I realize I've never tried them on anything earlier than a 68030 based machine.   Well, poo.

Yea, well it happens. The problem with all the testing software is they either run full 32-bit addressing, or at the minimum require CFM-68K which again requires an 020+ one of the biggest reasons I want the 020/030 running in my portable. 

Seems the best way for me to test is just create a 100% RAM disk to utilize maximum RAM, copy all my HDD contents over and boot from RAM. if everything is ok without bombing out there there, then its a good bet that the RAM is ok. 

So far out of the 12 I just assembled, I have 1 that has another bad RAM IC. 

Seems the best way for me to test is just create a 100% RAM disk to utilize maximum RAM, copy all my HDD contents over and boot from RAM. if everything is ok without bombing out there there, then its a good bet that the RAM is ok.
Probably so. I can't think of a better option.

So far out of the 12 I just assembled, I have 1 that has another bad RAM IC.
That doesn't sound too bad. Probably higher than the failure rate ought to be, but 1 in ~50 sounds managable. Much better than the possible 1 (or 2) in 12 at the beginning.

mp.ls