• Please review our updated Terms and Rules here

Extended Arithmetic Element (EAE) re-implementation using current TTL ICs?

I would be interested also. Like Roland, my interest would be for a relatively 'authentic' clone as opposed to something with an FPGA or CPLD.

But beggars can't be choosers!

Dave
 
This would be a drop-in replacement board set for the latest rev. M8340 and M8341? Yes, I'd purchase at least one board set.

I'm also interested in purchasing a couple Omnibus-USB serial I/O boards (either Hachtman or McLeod style).
 
vrs42 said:
Well, I do have drawings for the BM8L on my site (but not for either flavor of the BA08).

Vince
Click to expand...
Correction: *Building* a BM08/L is a problem. I could just gut one of the pdp8/L's, remove all the wiring, and rewire it as a BA08 or BM08. I have a full set of pdp8/I cards, so probably enough to handle the card side, but that would be a shame.

I do also have a broken 3U size block of flip chip wire wrap blocks about the size of a DW08 but it's not a DW08. I think it was some weird custom thing from the pdp12, but I could strip the wires, and wire it up to be enough of a BM08 to allow a plug into my pdp8/E chassis which could then be the Omnibus. Another "sin" but possibly less bad than gutting an L.

Of course at that point we start thinking about shared memory and wiring both pdp8/L's to the same Omnibus as the worlds first SMP pdp8.....
 
Have been thinking about making a modern version of a BM08 or BA08 to use with the 8/L, but it’s low on the priority list. A real one would be much more fun.
 
czunit said:
Correction: *Building* a BM08/L is a problem. I could just gut one of the pdp8/L's, remove all the wiring, and rewire it as a BA08 or BM08. I have a full set of pdp8/I cards, so probably enough to handle the card side, but that would be a shame.

I do also have a broken 3U size block of flip chip wire wrap blocks about the size of a DW08 but it's not a DW08. I think it was some weird custom thing from the pdp12, but I could strip the wires, and wire it up to be enough of a BM08 to allow a plug into my pdp8/E chassis which could then be the Omnibus. Another "sin" but possibly less bad than gutting an L.

Of course at that point we start thinking about shared memory and wiring both pdp8/L's to the same Omnibus as the worlds first SMP pdp8.....
Click to expand...
Well (predictably) I'd built the replacement BM8L thingy with a CPLD (though TTL is also quite doable) on a PCB with a lot of box headers. I'd use my cable replacement cards to cable the box headers to the vintage gear so that I didn't have to modify it.

If it fit (and a BM8L might not, though the NVRAM MM8I will), I'd consider breaking up the board into single-height chunks that will fit into the original cable slots. Then a single ribbon to provide the missing interconnect to stitch the cards back together into a single unit.

I suspect the price would be astonishingly low (probably less than $100, unless gold fingers were involved) and no irreversable mods to the vintage gear.

(I have proto-board outlines which provide 3, 4, or 5 single-height modules that are tab-routed to snap apart, so that I can order the modules together as a single board from JLPCB or the like.)

Vince
 
I looked at this a little, and it seems to me like the hard bit for closely mimicing the original layout is replacing the two DEC ROMs in the M8340. Though I suppose there are still a few bipolar ROM chips around.
 
vrs42 said:
I looked at this a little, and it seems to me like the hard bit for closely mimicing the original layout is replacing the two DEC ROMs in the M8340. Though I suppose there are still a few bipolar ROM chips around.
Click to expand...
I don't think the IM5600 ROMs would be hard. The identical 82S23 bipolar PROMs are still available, the contents are documented on page 13 of the engineering drawings and I have a programmer which can program the 82S23.

The harder bits are the weird 8235 and 8266 MUXs, these are rare and expensive, so it would make sense to figure out an alternative implementation of that part of the circuit.

The 74H53 is another odd IC. It is different from the standard 7453. One of the AND gates has 3 inputs whereas the standard 7453 has only 2 input AND gates.

I wonder what the 97401 OC NAND is but suspect it is just a selected 7401 which could be replaced by a 7439 (or 7438 which has a different pin-out).

The more exotic ICs on the EAE boards are still available online including Ebay, but some prices are eye watering. :cry:

Tom
 
I see that you've looked at it more than I have. Good catch on the H53.

Checking my inventory, I have at least 10 tubes of H53, and at least 1000 7439. I didn't go through the whole "ICs in tubes" tub, so there may be more. I didn't come across any 8235 or 8266 tubes in the first few inches, though.

Vince
 
thunter0512 said:
I don't think the IM5600 ROMs would be hard. The identical 82S23 bipolar PROMs are still available, the contents are documented on page 13 of the engineering drawings and I have a programmer which can program the 82S23.
Click to expand...
Each of the ROMs can be replaced by a GAL... 16V8. Actually ATF16V8 is available now. I don't think this is cheating because one programmable device is replace by another.

Code: 
  assign R1_Y1 = ~((~E & ~D & C & ~A) | (E & ~D & C & ~B & A));
  assign R1_Y2 = ~((~E & D & C & ~B & A) | (~E & D & C & B & ~A) | (E & ~D & ~C & B));
  assign R1_Y3 = ~((~E & D & C & ~B & A) | (~E & D & C & B) | (E & D & ~C & ~B & A));
  assign R1_Y4 = ~((~E & ~D & C & ~B & ~A) | (E & ~D & ~C & B & A) | (E & ~D & C & ~B & A));
  assign R1_Y5 = ~((~E & ~D & C & ~B & ~A) | (E & ~D & ~C & B) | (E & ~D & C));
//=================================================================================================
// inverse sum-of-products requires too many product terms for a 16V8, use non-inverse instead
//  assign R1_Y6 = ~((~E & ~D & ~C & B & ~A) | (~E & ~D & C & ~B & ~A) | (~E & ~D & C & B) |
//                   (D & ~C & ~B & A) | (~E & D & C & ~B & A) | (~E & D & C & B) |
//                   (E & ~D & ~C & B & A) | (E & ~D & C & ~B & A));
//=================================================================================================
  assign R1_Y6 = ((~D & ~C & ~B) | (~E & ~C & B & A) | (~E & ~D & C & ~B & A) |
                  (~E & D & ~C & ~A) | (~E & D & C & ~B & ~A) | (E & ~C & B & ~A) |
                  (E & ~D & C & B));
  assign R1_Y7 = ~((~E & ~D & ~C & B & ~A) | (~E & ~D & C & B) | (~E & D & ~C & ~B & A) |
            (~E & D & C & ~B & A) | (~E & D & C & B) | (E & ~D & ~C & B & A)|
            (E & D & ~C & ~B & A));
  assign R1_Y8 = ~((E & ~D & ~C & ~B & A) | (E & ~D & C & ~B & A) | (E & ~D & C & B));

  assign R2_Y1 = ~((~E & ~D & ~C & ~B & A));
  assign R2_Y2 = ~((~E & D & C & ~B & A) | (~E & D & C & B));
  assign R2_Y3 = ~((~E & ~D & ~C & B) | (~D & C & ~B & ~A) | (E & ~D & ~C & B) |
            (E & D & ~C & B) | (E & D & C & ~B & ~A));
  assign R2_Y4 = ~((~E & D & C & ~B & ~A));
  assign R2_Y5 = ~((~E & D & ~C & ~B & ~A) | (E & D));
  assign R2_Y6 = ~((~E & ~D & B) | (~E & ~D & C & ~B & A) | (~E & D & ~C & ~B & A) |
            (~E & D & ~C & B & ~A) | (E & B) | (E & ~B & A));
  assign R2_Y7 = ~((~E & ~D & ~C & B) | (~E & D & ~C & ~B & A) | (~E & D & ~C & B & ~A));
  assign R2_Y8 = ~((~E & ~D & C & ~B & A) | (~E & ~D & C & B) | (E & B) | (E & ~B & A));
 
gwiley said:
Each of the ROMs can be replaced by a GAL... 16V8. Actually ATF16V8 is available now. I don't think this is cheating because one programmable device is replace by another.
Click to expand...
I think GALs are a later technology. The bipolar PROMs (82S23) are the same vintage as the EAE/KE8E so might be a better match.
 
vrs42 said:
I see that you've looked at it more than I have. Good catch on the H53.

Checking my inventory, I have at least 10 tubes of H53, and at least 1000 7439. I didn't go through the whole "ICs in tubes" tub, so there may be more. I didn't come across any 8235 or 8266 tubes in the first few inches, though.

Vince
Click to expand...
Oh wow. That is quite some stock of 74H53 and 7439.
10 tubes of H53 and 1000 7439 will last you a few life times.
You are sitting on a small fortune of ICs given current Ebay prices. :)

I always buy a few extra ICs whenever I order to have some as spares, but I can't afford to buy in bulk.
 
thunter0512 said:
Oh wow. That is quite some stock of 74H53 and 7439.
10 tubes of H53 and 1000 7439 will last you a few life times.
You are sitting on a small fortune of ICs given current Ebay prices. :)

I always buy a few extra ICs whenever I order to have some as spares, but I can't afford to buy in bulk.
Click to expand...
I actually went through the rest of the tub later in the afternoon, and found another 8 tubes of H53. Nothing else of interest for this project, though I do have many tubes of N8233 and N8242.
 
Vince,

Please count me in for at least 2 maybe 3 of your EAE boards/sets.

With all this talk of TD8E controllers how hard would it be to make a TC08 TTL clone? I know the TC08 required both a KA8E (Positive I/O) and a KD8E (Data Break) boards. The ideal TC08 redux would be a single board with KA8E & KD8E on board. Or maybe a two board set. Kind of like the RK8E which included the Data Break in the three board set.
 
You must log in or register to reply here.
Back
Top