WTB Intel 7110 Bubble Memory Subsystem or Chipset
allisonportable at gmail.com
Thu Feb 22 17:14:45 CST 2018
On 02/22/2018 03:08 PM, Mark J. Blair via cctalk wrote:
> I'm about to acquire a couple of 1980s-vintage military surplus AN/UGC-137A terminals (i.e., glass TTYs with some local message preparation and storage capabilities) which have a bubble memory subsystem. They use plug-in cartridges containing 256 kbytes of storage in the form of two Intel 7110 1 Mbit bubble memory chips and their 7242 formatter/sense amplifiers.
> One of the cartridges contains the one and only copy of the terminals' firmware, which I believe they need to load up at each reboot. Naturally, extracting the contents of that irreplaceable cartridge for archival, and potential future emulation, is going to be a very high priority for me. I have a few different approaches in mind for accomplishing that. One approach would be to remove the two memory devices from the critical cartridge in order to dump their contents in an independent bubble memory subsystem.
> With that in mind, I'd like to get my hands on a working Intel 7110 bubble memory subsystem, or the parts to build one myself (i.e., a complete 7110/7220/7230/7242/7250/7254 chipset that I could make a board around).
What you plan is risky. You first need to know how they organize the
data in each of the loops.
The problem is did that interleave the two bubble or are they addressed
seperately. Both possibilities
were the case. Each BM required its own CPG, FSA and drivers but could
share the 7220 BMC.
> Might anybody here have what I need available for sale or trade? I might be able to use some arbitrary old computer or other device that has a subsystem based around the Intel 7110, or a development kit such as the Intel BPK-72, or a chipset to make my own board.
I got two of them back in the 80s, they are now part of a CP/M Z80
system I built back then.
Not much storage and sorta slow and power hungry.
> If I can't acquire or make the hardware to dump the memory chips outside of their native system, then I think my next option would be to passively snoop the host bus interface of the Intel 7220 controller I expect to find inside the terminals as they perform their initial firmware load, so that I can reconstruct the cartridge contents from the trace data.
The best and lowest risk point is to snoop is at the data bus
interface. Logic analyzer or something fast enough to
grab the data. The 7220 chip set gave a nice bus interface with a
fairly simple command set. Its also the side of
the device thats well documented.
I may have a few of the basic bubble memory units 7110 as they were
socketed. No extra CPD, FSA, Driver devices,
or BMC 7220.
> The terminals were made by the Librascope division of Singer, and brochures can be found here:
> I already have the critical cartridge in hand, and I posted some pictures of it on Twitter:
More information about the cctalk