• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Levco MonsterMac 2MB Teardown & Repair


Well-known member
About a year ago I bought a 1984 128k Mac that was listed as having been upgraded to 512k but had a green-and-black “Levco MonsterMac” sticker on the front, so I expected there might be some secrets in there. When I received it I found a port threaded through the external floppy opening - a telltale sign of course - and I booted it up and sure enough - it has a 2MB RAM upgrade! However the video was flickering and often died altogether - probably bad J4 solder joins. I put it aside
for awhile due to other priorities and finally opened it up today.



Here’s what I found: (the motherboard had to be pried out with a screwdriver due to the expansion board attached)


What a gorgeous upgrade board. It appears to clip directly onto the 68000, and to get power from the logic board. Separate soldered-on lines labeled “overlay” and “clock” also run down to the logic board. I love the hand-labeled ICs: “Sneezy”, “Dopey”, “Doc”, “Sleepy”, “GSB”, and “SCSI”. Also this upgrade appears to come with custom Levco ROMs? The SCSI ribbon cable, assuming that’s what it is, clips on at P5. As you can see from the original ‘Macintosh’ label, that is an original 128k motherboard lurking underneath.

Also hiding in there was something I’ve never seen before: this is a “Levco MacBreeze” piezoelectric fan (labeled as “motorless”), rather loosely attached to the floppy enclosure with velcro. It’s fairly quiet though does make a bit of a buzzing sound as it runs: when plugged into AC power, the white plastic bits wiggle back in forth, creating some perceptible airflow (though not a ton). The fan is powered with test clips attached directly to AC power on the analog board.

View attachment 1785FDBE-D346-4C1A-A123-B9983A55C126.MOV

Next step is to reflow some solder joins and see if I can get the video working consistently again - to be continued.


Well-known member
Yes @joshc this board actually got a rave review in a contemporary Macworld … I will try to dig up the reference soon.

Meanwhile tonight I did the requisite reflow of the solder joints at J4. “After” pic attached probably doesn’t exactly make me proud of my soldering skills but … trust me, it was much worse before (especially at pin 1), and now it works perfectly — video flicker totally fixed.


I also noticed the battery holder had snapped off the A/B


…so fixed it with some hot glue.

And there she is, showing off her 2MB of RAM while running the lovely Deluxe Music Construction Set (just try running that on an SE/30! … well, it crashes on mine, at any rate).


I noticed the foil RF shield was missing, so I found one one eBay and bought it. Also, in order to reflow J4 I had to peel off the foam sticky pads on the A/B and will have to replace it with some foam tape. Most importantly I still need to try out that SCSI connector! That will be the next step.


Well-known member
Deluxe Music Construction Set
Does it only run on 68000 CPUs?

I wonder if it's in the same vein as MusicWorks. Never got it to work on anything newer than my long-gone Plus. Haven't tried an SE though.

Nice work on the thing overall! That board is something else, and that fan is Neat™ but yeah, I agree to a replacement, had I one 🤣


Well-known member
Have one of those; but yours seems to be a newer build than mine is. I've got a Rev C, dated 1985, SN1093. Instead of a SCSI chip, I've just got an unpopulated space on the board for a 2-row header, labeled Levco Expansion port.

Mine had one of the mask roms die a long time ago, and just a few years ago, one of the members here loaned me his logic board & MonsterMac board to use as a reference to figure out why mine was hanging. So - as that's a different rev board, I would strongly suggest that you image those ROMs before bit-rot sets in.

I'm not 100% sure, but I think those PALs can be mapped out as well, and "de-compiled" back to the set of equations needed to program a new set.

I'm also pretty sure that's a scsi port that's snaking out the floppy port hole in the case.

It appears to clip directly onto the 68000, and to get power from the logic board.
So, doing a PDS slot the hard way. ;)


Well-known member
Great advice thakns @bdurbrow . I‘ve never imaged an EEPROM before, is there a recommended programmer for reading these old things?


Well-known member
What kind are they? I've had good luck with my TL866 II with the types I've had here: if you can find out what kind they are I'll check whether it supports them or not :).


Well-known member
BTW here’s a nice blurb on this guy from “MicroTimes” volume 3 number 5 … or is it an ad? https://archive.org/stream/microtimesvolume00bamp_27/microtimesvolume00bamp_27_djvu.txt
"Product Of The Month
November 1985


Now you can turn your friendly Macintosh
into a real Monster. Levco's MonsterMac
enables the memory in your computer to
run up to 30% faster than a standard
Macintosh. This proven upgrade also
provides 256K of PROM space, and the
first real Macintosh expansion port for
peripherals such as the new Levco
OverDrive SCSI internal 20 megabyte hard
disk. The whisper-quiet Levco MacBreeze™
piezoelectric fan is included to keep your
Macintosh running cool, even when the
demands are hot. MonsterMac also
provides a true motherboard emulation
mode, to functionally "turn off" the extra
memory for those rare programs that don't
understand a world beyond 512K.


Levco offers the MonsterMac with clip-on
convenience, for simple, quick installation
and no modifications to your Macintosh
motherboard. MonsterMac works with old
or new Apple ROMs and with either 128K
or 512K Macintoshes. Most importantly,
MpnsterMacs are available now\

"The MonsterMac receives the highest recommendation for
its speedy 2 megabytes, its compatibility with all the test
software, and the lure of a real bus expansion port."
MacWorld, February 1986


Well-known member
I would be quite surprised if those ROM's where actually custom, most likely they are just prgrammed with the Plus ROM's (this is what most of these kind of upgrades used) It's still worth dumping them and finding out though

Assuming they are replacing the system ROM's you should be able to use CopyROM to dump it form a running system


Well-known member
Those are advertisements; but still interesting, though!

On my machine, at least, there are custom ROMs that aren't just a copy of a Mac Plus ROM. There's a set that seems to do some initialization of the MonsterMac's hardware, and then there's the set on the motherboard. From what I can tell (I haven't tried disassembling the ROMs and figuring out what the code does from the disassembly, and it's been a few years since I worked on it, so take this with appropriate grains of NaCl) the early boot process (POST, finding a boot-able disk, and showing the "Happy Mac" icon) is handled by the set of ROMs on the MonsterMac board, which then may switch to the ROMs on the main logic board for running the system. During boot, at least for my machine, the "Happy Mac" icon has a couple of pixels added that make it look like it's got fangs (it's a Mac, that's a monster - get it?). Or, the set of ROMs on the MonsterMac board may stay active while the machine is in operation.

I don't think that there were Mac Plus ROMs - modified or not - originally on my version of the MonsterMac, just based on build dates: my version has copyright and chip manufacturing dates from early 1985, and the Plus wasn't released until early 1986. I think that they took the 128k or 512k ROMs, and modified them to support the 2mb of RAM, and added any other stuff needed. However, my Mac also has an 800k drive, so at some point it probably got a variant of either the 512ke ROMs or the Plus ROMs.


Well-known member
There was a design for some kind of SCSI expansion board published, maybe in BMUG newsletter? There was a thread on these forums about it a few years ago. The designer was a reasonably famous person in Mac hardware or Software circles back then.

IIRC, it predated the Mac Plus release and so included code to drive the SCSI port.

Could the Monster ROMs be something like original non-FAT Mac ROMs with some variant on that SCSI code embedded?


Well-known member
@bdurbrow so I checked and I see just a normal Happy Mac when booting up. Could it be they got less ”cute” with the later-rev boards/ROMs?

Tonight, inspired by some Larry Pina advice, I re-attached the piezoelectric fan with some 1/4” thick x 1” wide weatherstripping as a buffer (and Velcro holding the fan to the weatherstripping). This created a very well padded but secure connection which will reduce the vibration imparted by the fan on the rest of the unit ... these things are fairly quiet but vibrate a lot (you can feel the case vibrating if you rest your hand on it with the fan running inside).

Last edited:


Well-known member
OOPS. Well, I was excited to hook everything up and try out the SCSI connection. I hooked up my MonsterMac to a properly-terminated known-good 80 MB external SCSI HD. The Mac booted normally, bonged and showed a flashing question mark — hmm, it didn’t seem to be seeing the SCSI HD. So I restarted the Mac aaaaaand …. The second time, no bong. Just this:


Tragically, this now happens 100% of the time.

I am suddenly very paranoid that I fried my accelerator by, perhaps, attaching the internal SCSI ribbon cable backwards at P5. Actually I now realize it’s really unclear which way is the RIGHT way to connect the internal SCSI ribbon, which will happily attach either way... I thought I was reconnecting it the same way it had been installed, but now realize I’m uncertain - should have been more careful.


Any sage advice? What are the odds a wrongly-connected SCSI cable would fry my machine? I’m feeling pretty bummed and not optimistic this is reparable.

(It’s not obvious how to remove the accelerator from the mobo in order to test out just the mobo and other internals, but I didn’t try very hard yet.)


Well-known member
I don't think that's unfixable.

The pattern on the screen is an indicator of the processor either not running, or being unable to perform writes to RAM. That checkerboard pattern is what RAM chips tend to look like when power is first applied, but nothing has been written yet (one block tends to go high, while the next goes low).

The SCSI chip on the MonsterMac board has the ability to hold the bus either by continuously asserting it's IRQ line; or by leaving it's data line drivers in output mode when they aren't supposed to be. It's possible, but less likely, that some of the inputs on the CPU side of the SCSI chip are now shorted to ground or +5v. However; I would tend to expect that flipping the connector around backwards just would have fried some of the output drivers on the SCSI chip that were actually connected to the ribbon cable; and not propagated across to the whole chip.

The MonsterMac on my machine is connected to the main logic board by a set of header pins that go down into a regular IC socket on the main logic board. If removing and re-installing the board stack from the chassis disturbed that connection, it could very well cause this.

Otherwise something else could have failed and it's just co-incidence. At this point I would go probing about the boards with an oscilloscope looking for abnormal or missing activity. I'd also want to have on hand some spare chips (68000 and 5380 in particular) to do some elimination testing. While I was at it, I'd put that 5380 in a socket also.

Both the 68000 and 5380 are still available as new-old-stock and used parts on eBay. Same with ram chips, and PAL/GAL chips. If it's a PAL/GAL, then some reverse engineering would be needed to figure out a new program for them to do the job (it doesn't have to be an identical bitstream, just a work-alike. Fortunatally these boards aren't that complex).


Well-known member
Thanks @bdurbrow I was in bad need of some expert encouragement there and that was just the trick! Coming soon: a whole bunch of poking around with my scope on this board …

One more question for you … perhaps I am missing something super obvious but, since you have one of these boards, how do you know which way the SCSI ribbon is supposed to be connected??


Well-known member
My board doesn’t have SCSI. When the machine was in daily use way back when, we had a serial-port attached 20MB hard drive called a PCPC MacBottom (I swear, was this thing made by twelve year old boys?). Eventually, that drive died, along about the same time as the PCPC company did, so I don’t even have the dead carcass anymore. Nowadays I use a BMOW FloppyEmu.

However, what I would do is download the data sheet for the AM5380 SCSI chip, and look up the pin out for the 25 pin SCSI port on Macs, and trace which pin on the D-sub connector goes through to the correct pin on the chip.


Well-known member
Also; there’s pin numbering on the PCB. However, they may have flipped the header around in the PCB CAD software that they used when making the board; I’d double check with a multimeter against the pin outs to be sure.


Well-known member
Oh yes I remember the MacBottom and their phone number from old Macworld ads, 1-800-MAC-BUTT (very memorable to 10-year-old me). Always wanted to stumbling on a still-working one.

Well, not sure if this is good or bad news but … I completely removed the accelerator completely and still have exactly the same problem. So whatever is wrong is not related to the SCSI chip. Did I toast some part of the motherboard somehow? (Maybe “good” news if it means I can get this accelerator working on a different 512k motherboard….)

By this way this is an interesting motherboard. The RAM is socketed and a Mux on a very tiny PCB was soldered in next to the CPU. Doesn’t look like any 512k motherboard I’ve seen. Was this an aftermarket 512k upgrade?



The accelerator is indeed connected with some side-facing header pins, it doesn’t seem like it would be super secure, I wonder if this dates to before the Killy clip became de rigeur?