• Please review our updated Terms and Rules here

RL8A (Disk controller for the RL01/02) UV-EPROM Dump

pbirkel@gmail.com

Veteran Member
Joined
Apr 7, 2013
Messages
526
Location
Silver Spring, MD, USA
All;

With the recent upgrade of the RL02 Emulator (wuhoo!) I've been contemplating getting an M8433 RL8A (hex wide) Disk Controller for the RL01/02 into operation on an Omnibus backplane.

Unfortunately my M8433 came with an uncovered 2708 and the admonition that it had been ambient-UV erased. Since its last-known environ was a SWUS location with limited protection from the Sun I can readily understand why that might be so. It also came with lots of fine red dust attesting to the limited "environmental controls" under which it had been stored. Alas ...

Does anyone have a dump of that EPROM? It's the only one on the board, sitting in the upper-left in socket E133. No possibility for confusion!

A net-search under various guises consistently ... fails.

My thanks to anyone who can supply such an image "dead or alive"!

paul
 

Mal

Experienced Member
Joined
Apr 3, 2008
Messages
91
I have an untested RL8A. The window on the 2708 on my board is also uncovered, so the status of the EPROM contents is unknown.

I read the EPROM this morning. Here's a link to the file -> www.avitech.com.au/mm-files/roms/rl8a/RL8A.BIN

The 16-bit checksum (as reported by the Promlink software when the EPROM was read in my Data I/O 29B with UniPak 2) was 0xACB3.
 

Terry Kennedy

Experienced Member
Joined
Jul 7, 2017
Messages
241
Location
New York
The 16-bit checksum (as reported by the Promlink software when the EPROM was read in my Data I/O 29B with UniPak 2) was 0xACB3.
Chances are that this EPROM's contents will look "wrong" even if they are correct. It is either sequence control for the discrete stuff on the board, or used to translate block addresses into C/H/S mapping (a trick I used on an S-100 board I designed years ago). So if it looks like a repeating sequence of bits, it is probably correct.

One thing that would be helpful is to report the DEC firmware part number on the EPROM. In later products, this is usually a paper label covering the window. This will have some or all of the following: 2-5-2 part number, date code, checksum, copyright. The 2-5-2 part number would be something like 23-339E5-00 (this is one of 2 EPROMs on the RQDX3). The prefix and suffix are often dropped to save space on the label, so you'd have 339E5. 23 indicates programmables, and the last 2 characters of the middle group indicate the part type and capacity. E indicates a UV-erasable part and the next character specifies the capacity, so E5 is a 27128, E4 is 2764, etc. The Ax parts are bipolar PROMs, for example. On a board of this vintage, the part number will probably be rubber-stamped on the package near the window, on the side opposite the manufacturer's labeling. Red or red-orange ink is common, white was used occasionally, and black was sometimes used on white ceramic packages like the 1702.

If the EPROM on this board controls the discrete logic on the board, as opposed to disk address decoding, it may well differ depending on the etch and ECO level of the board. That's why it is important to record the DEC firmware part number as part of the image file.

The parts list for the M8433 level ML005 (CS REV F) calls out E129 as a 23-040B7-00. If the 2708/8708 on your board is not in position E129, then this isn't relevant. A different drawing identifies the EPROM as being in position E133 and part number 23-012E2-00.
 

pbirkel@gmail.com

Veteran Member
Joined
Apr 7, 2013
Messages
526
Location
Silver Spring, MD, USA
Mal: Thx!

Terry: Mine is marked in red ink on the white ceramic carrier as: "WB8031 040B7". That doesn't correspond to any DEC identification system previously encountered, although the dddAd-format second-component is a common format, so perhaps "23-040B7-00" is meant -- in which case have you accidentally interchanged E129 and E133? Or perhaps the documentation is confused. Can you share the URL for the specific documentation that you're examining? If there was a more traditional marking on the window-label, well that's long gone.

E129 is a traditional 16-pin DIP socketed blown-fuse ROM (PN indistinct) that is marked "SG8026 219A1". Search for that identifier (e.g., as "23-219A1-00") doesn't turn up anything -- for example in this list: http://www.pcjs.org/devices/roms/dec/

I wonder if that corresponds to what Mal has in E129? The PCB is identified as "5013116 B", so perhaps it's a Rev B? Only other PCB markings state: "104160" and "M8433".

Thanks for the followup guys!
 

Terry Kennedy

Experienced Member
Joined
Jul 7, 2017
Messages
241
Location
New York
Terry: Mine is marked in red ink on the white ceramic carrier as: "WB8031 040B7". That doesn't correspond to any DEC identification system previously encountered, although the dddAd-format second-component is a common format, so perhaps "23-040B7-00" is meant -- in which case have you accidentally interchanged E129 and E133?
040B7 is definitely the same 23-040B7-00 from the parts list. WB8031 would probably be facility WB, 1980, week 31.

Or perhaps the documentation is confused. Can you share the URL for the specific documentation that you're examining? If there was a more traditional marking on the window-label, well that's long gone.
Here: http://dustyoldcomputers.com/pdp-common/reference/drawings/modules/m/m8433.pdf
Page 1 (mask/drill diagram) shows E129 near the upper left corner as a 24-pin package. Page 9 (schematic) also shows E129 near the top left, and also says "8708" and "23040B7".
http://www.textfiles.com/bitsavers/pdf/dec/pdp8/RL8A_Schematics.pdf is a more expanded version of the prior PDF. Mask/drill diagram on PDF page 15, schematic on PDF page 23, BOM on PDF page 28.

E129 is a traditional 16-pin DIP socketed blown-fuse ROM (PN indistinct) that is marked "SG8026 219A1". Search for that identifier (e.g., as "23-219A1-00") doesn't turn up anything -- for example in this list: http://www.pcjs.org/devices/roms/dec/
That's a standard small bipolar PROM. The schematics I linked call it E135. It is the write precompensation PROM on PDF page 21.

I wonder if that corresponds to what Mal has in E129? The PCB is identified as "5013116 B", so perhaps it's a Rev B? Only other PCB markings state: "104160" and "M8433".
Normally PCBs are 54-series. I forget what 50 is - it may be one particular etch layer.

Thanks for the followup guys!
No problem. I just wished my posts were approved faster - this is sort of like trying to communicate with the Voyager 1 space probe. :(
 

Mal

Experienced Member
Joined
Apr 3, 2008
Messages
91
I've attached a pic of the markings on the 2708 EPROM installed at location E133 on my M8433. Mine also includes the marking "040B7". The text adjacent to that marking (7919) is presumably the date code.

The I.C. 4 locations below E133 on my board (which I assume is E129) is an N82S123N. Below the manufacturer's markings there are additional markings (presumably by DEC) that read "7943 219A1". The 7943 is perhaps the firmware date (or the date on which the chip was programmed?). I've attached a picture of the markings on this IC as well.

I can't see much in the way of etch/ECO information on my board, except that "004 E" is stamped on the rear side of the handle. Text on the solder-side PCB trace layer says:

RL01
DISK
CNTLR.

104160
M8433

5013116 B

Hope that helps.


20170803_114724 (Medium).jpg20170803_115117 (Medium).jpg
 
Top