DC319 DLARTs and T11s (was T-11 ....)

Allison ajp166 at bellatlantic.net
Fri Jan 16 17:14:40 CST 2009

>Subject: DC319 DLARTs and T11s  (was T-11 ....)
>   From: "Bob Armstrong" <bob at jfcl.com>
>   Date: Fri, 16 Jan 2009 07:25:17 -0800
>     To: <cctalk at classiccmp.org>
>>A while ago, Choctaw Bob ]bob099 at centurytel.net] wrote:
>> [Schematic of Peter McCollum's hand built T11 SBC]
>>	Schematic here  http://www.geocities.com/saipan59/dec/t11.jpg
>  I just got around to looking at this (sorry - been busy :-) and while it's
>nice and simple, it uses a 6850 UART for the console.  It's easy enough to
>interface the 6850 (or any other modern UART for that matter) to the T11
>bus, but it's not going to be even remotely PDP-11 compatible.  If you want
>to have any hope of ever running any real PDP-11 software, you're going to
>need a DL11/KL11 compatible console interface.

DLARTs are hard o find.  I have a dozen t11s and one DLART.

>  For the T11 that means you also want a DC319AA DLART (a "DL-11 compatible
>UART chip") chip, and after looking around those seem to be far more
>difficult to come by than the DCT11 CPU chips.  Does anybody have a pile of
>DC319 chips handy, or know where they can be found?  Unlike the T11 chips,
>which can be harvested from a lot of fairly common and not very valuable DEC
>boards (e.g. RQDX3s), I don't know of any good donor for DLARTs.  There are
>a couple on my FALCON SBC-11/21 board, but I'm hardly willing to take it
>apart for this project :-)

Actually a 6402 (two or three are better) and the logic to do a ful DL isn't bad.
With at least two then you can do a Tu58/TU58sim as a disk and have a real 
bootable 11.

You also want at least 28KW of ram (2 32kx8 rams will do it) and TUboot in a
small eprom mapped at boot.

>  If you can't find a DC319, then the alternative is to build a DL11
>compatible interface using a standard UART like the 6402 and a handful of
>discrete logic.  I think the minimum you could get away with would be to
>implement the DONE and INTERRUPT ENABLE bits (bits 6 and 7) for both the
>receiver and transmitter CSR, and then to implement the standard vectors at
>60/64.  Oh, and of course the data registers for the transmitter and
>  The KL11 also implements a MAINTENANCE bit, a BUSY bit, and a READER
>ENABLE bit, but a) we have no reader, b) I doubt (I'm hoping) much software
>ever looks at the BUSY bit, and c) probably nothing uses the MAINTENANCE bit
>except the diagnostics.

Try to find the logic for DLV11J as that has basic serial. You do not need 
all the bits but most are handy and you also need to generate a valid vector 
for console (60Q) and also for any other port.

>  ISTR that DEC had an "official" standard written down for exactly what was
>required of a DL-11 compatible serial interface, but I can't find it

Microcomputer handbook.  I'll try to find something more.



More information about the cctech mailing list