Sperry Univac Core - Which computer and Capacity?
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
16K On Tue, Jun 22, 2021 at 12:16 PM Douglas Crawford via vcf-midatlantic < vcf-midatlantic@lists.vcfed.org> wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
Just count the wires? -andy
On Jun 22, 2021, at 5:47 PM, Douglas Crawford via vcf-midatlantic <vcf-midatlantic@lists.vcfed.org> wrote:
Hows ya know?
On 6/22/2021 12:20 PM, Bill Degnan via vcf-midatlantic wrote:
16K
On Tue, Jun 22, 2021 at 12:16 PM Douglas Crawford via vcf-midatlantic < vcf-midatlantic@lists.vcfed.org <mailto:vcf-midatlantic@lists.vcfed.org>> wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg <https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg> Thanks, DC
Its from 1970's and there are two planes. On Tue, Jun 22, 2021, 5:47 PM Douglas Crawford via vcf-midatlantic < vcf-midatlantic@lists.vcfed.org> wrote:
Hows ya know? On 6/22/2021 12:20 PM, Bill Degnan via vcf-midatlantic wrote:
16K
On Tue, Jun 22, 2021 at 12:16 PM Douglas Crawford via vcf-midatlantic <vcf-midatlantic@lists.vcfed.org> wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation?https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
Honestly I dont know for sure, but 32 is too much for this era board. It could be only 8K but it looks too new. What year did univac becime sperry univac? The board is only as old as that date, or less. Then if you can look at the silkscreen on the ttl chips for dates. Bill On Wed, Jun 23, 2021, 12:26 AM Douglas Crawford via vcf-midatlantic < vcf-midatlantic@lists.vcfed.org> wrote:
So each plane is known to be 8k? ... that was standard or typical? On 6/22/2021 5:53 PM, Bill Degnan via vcf-midatlantic wrote:
Its from 1970's and there are two planes.
On Tue, Jun 22, 2021, 5:47 PM Douglas Crawford via vcf-midatlantic <vcf-midatlantic@lists.vcfed.org> wrote:
Hows ya know? On 6/22/2021 12:20 PM, Bill Degnan via vcf-midatlantic wrote:
16K
On Tue, Jun 22, 2021 at 12:16 PM Douglas Crawford via vcf-midatlantic <vcf-midatlantic@lists.vcfed.org> <vcf-midatlantic@lists.vcfed.org> wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation?https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
On 6/22/2021 12:15 PM, Douglas Crawford via vcf-midatlantic wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
That would be the Drawing Number. The part number is one higher and is on one of the connectors. Listed here as a 8K memory board for AN/UYK-20( page 15) : UYK_20_20A_Technical_Summary_May91.pdf (trailing-edge.com) <http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/UYK_20_20A_Technical_Summary_May91.pdf> I never saw one without the covers. I remember them as being 16 bits wide, so the core planes must be 8 deep. -- -- DuaneCraps sdɐɹɔ ǝuɐnp
On 6/23/2021 8:33 AM, Duane Craps via vcf-midatlantic wrote:
On 6/22/2021 12:15 PM, Douglas Crawford via vcf-midatlantic wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
That would be the Drawing Number. The part number is one higher and is on one of the connectors.
Listed here as a 8K memory board for AN/UYK-20( page 15) :
UYK_20_20A_Technical_Summary_May91.pdf (trailing-edge.com) <http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/UYK_20_20A_Technical_Summary_May91.pdf>
I never saw one without the covers.
I remember them as being 16 bits wide, so the core planes must be 8 deep.
--
-- Several years ago I brought a UYK-20 Logic print Technical Manual to VCF, that someone gave me..I don't know where it is now. Bill Dromgoole might know. If someone wanted to hook it up it might have enough information to identify the pin connections. 8K X 16 would be so cool. -- DuaneCraps sdɐɹɔ ǝuɐnp
On Wed, Jun 23, 2021 at 02:23:06PM -0400, Douglas Crawford via vcf-midatlantic wrote:
That's a possibility. I have thought it would be cool to make a core work. If its not a whole lot of analog conditioning I suppose its something I'd consider. Just at a glance, I'd expect that its all digital on the connector, so there would be power and interface to deal with, only. Not too bad if that's the case. It does have two broke ICs- smashed to the point of decapped, but that's easy to deal with. Other than that its in great shape.
If all the driver logic is with the board should be digital interface. Looks like their may be a second layer of logic not seen in the picture. PDP-8/A is a sandwich with core mounted to base board with all the analog stuff. Does anybody know why it looks like only X & Y wires without sense/inhibit? https://i.ebayimg.com/images/g/RQUAAOSw1kxgujxP/s-l1600.jpg
On 6/23/2021 8:33 AM, Duane Craps via vcf-midatlantic wrote:
On 6/22/2021 12:15 PM, Douglas Crawford via vcf-midatlantic wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
That would be the Drawing Number. The part number is one higher and is on one of the connectors.
Listed here as a 8K memory board for AN/UYK-20( page 15) :
UYK_20_20A_Technical_Summary_May91.pdf (trailing-edge.com) <http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/UYK_20_20A_Technical_Summary_May91.pdf>
I never saw one without the covers.
I remember them as being 16 bits wide, so the core planes must be 8 deep.
OK very cool now we got to the bottom of it! :) You are right- 1 higher part as connector P2 says "90536 ASSY 7128082" Which is a direct match to page 15: "Core Array 8k" That would be 8K of the word-size of the computer - which I read is 16 bits. Which comes full circle to Bill Deg ... claiming it was 16K... which I figure he was thinking bytes, and if so, correct! 1991 publication on that document, and the computer had a long run with the Navy of course. so no wonder the board artwork is hand routed but the board is populated with 79 - 82 date code ICs. Wikis put the development in the early 70s. Pretty neat pedigree for this board. So great to know what it was from. Thanks Duane! Excellent. If you would like to see the real thing without the covers I can bring it to the museum. I'd love to have you look at it, and we can check the geometry of the core planes to see if they are 8 deep as you suggest. DC
<snip>
Listed here as a 8K memory board for AN/UYK-20( page 15) :
UYK_20_20A_Technical_Summary_May91.pdf (trailing-edge.com) <
http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/UYK_20_20A_...
I remember them as being 16 bits wide, so the core planes must be 8 deep.
OK very cool now we got to the bottom of it! :) You are right- 1 higher part as connector P2 says "90536 ASSY 7128082" Which is a direct match to page 15: "Core Array 8k" That would be 8K of the word-size of the computer - which I read is 16 bits. Which comes full circle to Bill Deg ... claiming it was 16K... which I figure he was thinking bytes, and if so, correct! 1991 publication on that document, and the computer had a long run with the Navy of course. so no wonder the board artwork is hand routed but the board is populated with 79 - 82 date code ICs. Wikis put the development in the early 70s. Pretty neat pedigree for this board. So great to know what it was from. Thanks Duane! Excellent.
If you would like to see the real thing without the covers I can bring it to the museum. I'd love to have you look at it, and we can check the geometry of the core planes to see if they are 8 deep as you suggest.
Boards like this from various manufacturers are on the walls at Kennett Classic too. BIll
Here s a link to a manual on bitsavers: bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/N00039-73-C-0431_AN_UYK-20_Technical_Manual_Volume_1_Jan1974.pdf <http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/N00039-73-C-0431_AN_UYK-20_Technical_Manual_Volume_1_Jan1974.pdf> looks to be preliminary information; it has a lot of hand drawn figures Memory description starts on page 3/49. paragraph 3-78 There was a control board and a data board for each group of 4 memory cards. On 6/22/2021 12:15 PM, Douglas Crawford via vcf-midatlantic wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
-- DuaneCraps sdɐɹɔ ǝuɐnp
The board has bottom row (nearest connectors) of TI ICs label SN97674J. Two of these have an extra designation "SN J55325J". There are two smashed ICs in this row that would need replacing which I presume are the same ICs as the rest of the row. I'm not finding anything thing relevant on web searches, internet archive, or any recent sales on ebay. Any recommendations on where else to look? On 6/24/2021 9:17 AM, Duane Craps via vcf-midatlantic wrote:
Here s a link to a manual on bitsavers:
bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/N00039-73-C-0431_AN_UYK-20_Technical_Manual_Volume_1_Jan1974.pdf <http://bitsavers.trailing-edge.com/pdf/univac/military/an_uyk-20/N00039-73-C-0431_AN_UYK-20_Technical_Manual_Volume_1_Jan1974.pdf>
looks to be preliminary information; it has a lot of hand drawn figures
Memory description starts on page 3/49. paragraph 3-78
There was a control board and a data board for each group of 4 memory cards.
On 6/22/2021 12:15 PM, Douglas Crawford via vcf-midatlantic wrote:
Board is identified as 90536-7128081-00 G Web hits on this are thin. Anyone know what Univac machine this is from and its capacity? I suppose count of the wires H x L = # of bits of storage, and one side is the word size of the computer? Confirmation? https://i.ebayimg.com/images/g/kw4AAOSwwlBgujxR/s-l1600.jpg Thanks, DC
On Mon, Jul 05, 2021 at 04:54:11PM -0400, Douglas Crawford via vcf-midatlantic wrote:
The board has bottom row (nearest connectors) of TI ICs label SN97674J. Two of these have an extra designation "SN J55325J".
Could be custom labeled. I think the thread said schematics were available. Can you tell from the schematic?
On 7/5/2021 8:56 PM, David Gesswein via vcf-midatlantic wrote:
On Mon, Jul 05, 2021 at 04:54:11PM -0400, Douglas Crawford via vcf-midatlantic wrote:
The board has bottom row (nearest connectors) of TI ICs label SN97674J. Two of these have an extra designation "SN J55325J".
Could be custom labeled. I think the thread said schematics were available. Can you tell from the schematic?
Ah, good call. Let me check.
participants (5)
-
Andrew Diller -
Bill Degnan -
David Gesswein -
Douglas Crawford -
Duane Craps