PDA

View Full Version : DG Nova 4



Conmega
October 7th, 2015, 07:01 PM
So I finally got around to completing my 2400 mile round trip to Minneapolis, MN to pick up the DG Nova 4 I purchased about 6 months ago.
This system is in an original DG rack with the main chassis and expansion chassis for the disk controller and extra serial ports.
This system has the RL01 (IBM 5444) style disk sub system which includes a fixed 5mb platter and a 5mb removable pack.
It came with a fair bit of documentation and 3 disk packs, one which has been in the drive for what looks like since it was removed from service.

Here are the photos I have gotten of it so far:
http://imgur.com/a/WF3r0

The rack and Cipher tape drive is something I got for free on the way up which I am working on hooking to my PDP 11/23. That's another project :)

Anyway the reason I don't have better photos at the moment is because its actually still in the back of my truck, I am going to take it out and tear it down tomorrow and take piece by piece down into the basement and reassemble the rack down here. Since getting the whole thing down together will be nearly impossible with the weight.
The two other packs look beautiful inside so as long as the drive and heads are ok I should be-able to bring it online after a good cleaning.
One pack is also labeled 'Master Backup' in-case if the fixed platter is bad and needs to be replaced, which the docs show how to do.
If I can't get the drive online or everything has been wiped and has no data on it I will be a bit out of luck as there is no other media drive on the device like tape drive or floppy drive.
Hopefully I can get it all online in the near future, its the first disk pack drive I've personally owned so if you guys have any tips feel free to let me know!

-Connor K

NeXT
October 7th, 2015, 08:20 PM
I'm so glad to see that you finally managed to get it back to HQ. That thing had been on ebay for months and I was getting worried. Have you found a way to unlock and release that loaded pack yet?

Qbus
October 8th, 2015, 06:09 AM
Although I am far away from getting my military Nova up and running I am tending to think the plan is the paper tape version of DG Nova Basic loaded thru the con port may be the answer. The problem there is finding someone who has the paper tape or at least an image of the program. I have a copy (PDF) of the “Fundamentals of Mini Computer Hardware” by Data General that’s dealing with the Nova 2 but it may prove useful if you need a copy of it. The manual gives a good insight to structure and opp codes for the early systems but don’t know how much would apply to something as late as the four. If we keep up all this DG activity maybe we can have a section just for technological dead ends like the Nova family?

firebirdta84
October 8th, 2015, 11:01 PM
Commega, this is fantastic. Thanks for sharing these pictures. I wish you well with your restoration journey on this.

I would love to jump in here, and say that I am just realizing the relationship between a machine that I am trying to reconstruct and the Data General Nova.

I've posted about the Microtech machines here before, and I'm finding some contacts who used to work at the company back in 1981-1983.

From them I'm learning that the Microtech that I'm restoring likely used the Point 4 processor, which was a Data General Nova-related machine.

http://MicrotechM1.blogspot.com

Compared to many of you here, I'm still a newbie to a lot of this, but I enjoy the conversations here, as everyone has always been very helpful to my projects over the years.

I'd really love to see some Novas, and eventually Point 4s, in operation again.

I'm in contact with Bruce Ray at http://novasareforever.org about all of this also.

I'm very happy to share any of the software and hard drive structure that I'm attempting to extract from these old backup tapes (partially successful so far!).


If we keep up all this DG activity maybe we can have a section just for technological dead ends like the Nova family?
Yes, Qbus, I'd love to contribute and participate, particularly if the Microtech really ends up fitting here.


I'm always interested in Data General talk.
Doug G, I grabbed this quote from you in another thread, but I it seemed appropriate here.

Thanks, everyone.
-AJ

Conmega
October 9th, 2015, 11:38 AM
NeXT: Thanks! And no I have not yet since its still in the back of my truck, I tried to remove the pack while it was back there but I felt resistance so I just left it alone, I have a feeling the heads either wiggled into place since they weren't obviously locked or I'm just not trying hard enough, but I would rather be safe than sorry!

Qbus: I saw you posting about your Rolm 1602, looks like a very cool machine and I wish you much luck with it! I appreciate the offer for the PDF but I think I have all the info I need for op-codes and such in the docs I got with the machine :)
DG was very good with their docs it seems, like with the disk subsystem they show how to program to access it and use it with custom code which is very very handy, not something I was expecting to see a whole section for in the manual for the disk subsystem.

AJ: Thanks for the luck! I think I've stumbled across your website before and I wish you the utmost luck with the Microtech, it seems like an interesting and odd machine :)
I'm actually not surprised it uses something from DG! Amazingly I've been learning a lot of company's seemed to use DG architecture, I guess DG sold it for a price to a company to use in their systems?
For example I've learned that some Nixdorf computers apparently used DG architecture and even similar disk subsystems. This is because I have a disk pack labeled Nixdorf I got off ebay awhile ago and it has the same sector holes cut in the base plate as the packs for the Nova! So in theory it should work in the Nova's drive assuming its servo info is the same. This is good news too because the seller I got it from has 3 more of those packs up, so if by chance they do work I can get myself a small hoard of packs. Although I may hunt for someone with a Nixdorf system running so they read the packs to make sure there isn't something valuable on their software or OS wise for Nixdorf systems. But as far as I can tell no-one has a running Nixdorf with a disk subsystem :/

I should also probably get in contact with Bruce Ray to see if he has any suggestions to help me get the system online and to check and see if any of the docs I have are new to him and see if he wants them scanned.
The amazing thing is the technical docs for the disk subsystem appear to be in their original packaging from when they were shipped so they are VERY nice, a good candidate for scanning if they haven't been scanned yet.
Also good because they are meant to go in binders so they are all loose pages :)

-Connor K

firebirdta84
October 9th, 2015, 05:49 PM
I just started a new thread about another DG Nova connection, Dan Paymar, if anyone here is interested:

http://www.vintage-computer.com/vcforum/showthread.php?49463

firebirdta84
October 9th, 2015, 05:59 PM
For example I've learned that some Nixdorf computers apparently used DG architecture and even similar disk subsystems. This is because I have a disk pack labeled Nixdorf I got off ebay awhile ago and it has the same sector holes cut in the base plate as the packs for the Nova! So in theory it should work in the Nova's drive assuming its servo info is the same. This is good news too because the seller I got it from has 3 more of those packs up, so if by chance they do work I can get myself a small hoard of packs. Although I may hunt for someone with a Nixdorf system running so they read the packs to make sure there isn't something valuable on their software or OS wise for Nixdorf systems. But as far as I can tell no-one has a running Nixdorf with a disk subsystem :/

Yes, I found an older thread here about the Nixdorf and it's DG Nova relation. I asked a question, but unfortunately, the original poster has gone silent for quite some time. IT's here:

http://www.vintage-computer.com/vcforum/showthread.php?23706

The current owner of Point 4, Don Burden, looks like he even posted there.


The amazing thing is the technical docs for the disk subsystem appear to be in their original packaging from when they were shipped so they are VERY nice, a good candidate for scanning if they haven't been scanned yet.
Also good because they are meant to go in binders so they are all loose pages :)

Yes, that is fantastic! I also may find the contents of those technical docs helpful for my Microtech project, so please do see if you can get those archived.

Thanks again!
-AJ

Conmega
October 22nd, 2015, 01:56 AM
Alrighty so a bit of an update with good news and bad news.

Good news is I have the disk drive cleaned up from its previous inhabitants...
Bad news is the bottom of the fixed platter (Which I had to remove to clean under) looks like this:
http://i.imgur.com/cNJspiM.jpg
Doesn't look like a head crash, and the head looks fine, but none the less I can't use this platter.

Other good news is looking through the docs more I learned that these types of drives of the 5MB per platter variant are basically Diablo 44 drives, or at-least in the docs they say 'Diablo 44 or 6045 series' which is the series of drive this is. Other good news is while poking around in the drive and talking on #classiccmp I have learned that even though these are RL01/02 style drives they do NOT have servo data on the platter! This means the 4 packs I got off eBay should most definitely work after a format. Which is especially good since I need to tear open a pack now for a replacement fixed platter.
This all also means the head assembly has a glass reticule on it for tracking its position.
This is where the REALLY BAD news comes in...
My glass reticule is cracked!!! http://i.imgur.com/LNgDKTi.jpg
It is a thin crack so it might be small enough it doesn't interrupt the light, but most likely I now have a 160lbs paper weight :(
Basically if this is the case there are two options to try and save it, take out and try and get the glass reticule reproduced by a company that does glass printing or something, but this is very unlikely to be-able to be done with the fine details of the lines, and even if it can be done I have to glue it back on in the exact right position meaning I basically have to align the heads...
Other option is to find a new head assembly... Which I only know of possibly 2 of these drives in existence myself sooo...
With that elephant now standing in the room I also will note I'm not close to powering up the drive anyway because I am awaiting replacement caps for these guys in the mail:
http://i.imgur.com/Iq9PoM0.jpg
Since they obviously have failed in bad ways leaking electrolyte onto the board in the past...
So enough about the very sad disk drive...

I'll end it with some good news :)
I got the CPU up and running and am able to get into DG's version of ODT called VC (Virtual Console).
I was originally worried about getting the CPU online since the fuse for the PSU was blown but after looking over the inside of it I couldn't see anything causing any problems so I threw it back together, put a new fuse in, re-seated the boards in their correct positions since when I got it the boards where everywhere and fired it up and got a console!
Sadly it seems I have some ram issues as when I clear any memory location I get something like 002000 in that location instead of the expected 000000.
And since I don't have something like an extender board I will have to get creative with troubleshooting it...

Well in any case progress has been made in the sake of progress.

Oh and side note about scanning docs, apparently the company that owns Data General now, EMC, attacks anyone who posts scanned docs up soooo sadly I can't really scan anything without putting myself in a line of fire.
Although if you ever decide to pop onto either the #vc or #classiccmp IRC channels I am around on there (as Conmega of-course) and we may be-able to discuss and trade knowledge there.
There are also some other knowledgeable people on the DG topic on #classiccmp that I have been talking to about my system so yea.

Anyway its 6am here and I think I should get to bed before I destroy my schedule any more then I already have...

-Connor K

Conmega
October 22nd, 2015, 07:22 PM
Woot!
Small update, I did not have bad ram!
There was just a jumper missing that should have been in place!
The system now passes its self test and ram zeros out as it should!
I also have learned that it seems I have the 4/X CPU, although I am not 100% on the differences.
I also seem to have 128K of ram!
The CPU board I learned is jumpered to boot at device code 33 which is the disk drive, so it was booting from it as expected.
The CPU board's serial port also supports a WIDE range of bauds along with 20ma current loop! So I now have a use for my ASR33 :D
These bauds include:
50, 75, 110, 134.5 (I have no idea what that's about or for?), 150, 200, 300, 600, 1200, 1600, 2400, 4800, 9600, 19200

Now to just wait for the caps I ordered for the disk drive, after that I will hopefully find out whether or I have a trashed drive...

Well now its time to go play around with some assembly in VC (Virtual Console)!

-Connor K

Conmega
October 27th, 2015, 07:19 PM
So I swapped the fixed platter with a platter out of one of the packs I got off eBay and torqued it to the spec of 10 in-lbs.
That went well, I went to power the drive up for the first time after putting one of the other eBay packs in the top as-well to test, since I want to avoid destroying any of the packs that came with the system in-case they have any valuable info on them.
And a cooling fan spun for half a second then nothing, blew the fuse...
So I put another fuse in and tried it one more time just to check and it powered up, but I get no 'ON' light which indicates good power.
So there is a power problem, I was checking voltages, +5v is fine, -15v was low at -17, +15v was high at +17v, and -5v was a little low if I remember correctly.

As for the +27v and -27v lines I think they are the problem. Before I powered the drive up the second time I was checking the bridge rectifiers and apparently I forgot to plug the one for the 27v bridge rectifier back in.
I plugged it back in and blew another fuse, so obviously this is likely to be were the problem is. I unplugged it again and it turns on again as it was doing.

I am about to look into the 27v failure, it seems to be after filtering and rectifying because the rectifier seems fine and so do the caps.
I am getting schematics for everything as I type this so I hope I can figure this out soon, the suspense of powering this thing up is killing me ;)

-Connor K

LeftyX
December 11th, 2015, 12:45 PM
[QUOTE=Conmega;387178]Woot!
Small update, I did not have bad ram!
:
The CPU board's serial port also supports a WIDE range of bauds along with 20ma current loop! So I now have a use for my ASR33 :D
These bauds include:
50, 75, 110, 134.5 (I have no idea what that's about or for?), 150, 200, 300, 600, 1200, 1600, 2400, 4800, 9600, 19200
:
The 134.5 baud is for the IBM 2741 (Selectric mechanism) terminals -- I know, I had to support a bunch of them :-)

--LeftyX

Qbus
December 14th, 2015, 12:03 PM
If you have an ASR-33 with the paper tape option I may be able to set you up with copies of my Rolm/Nova diagnostics tapes. The diagnostics and memory exerciser I have is all for the Nova 1000 or 1200 but would speculate that the Nova 3 and 4 may be backwards compatible. Let me know if you get your paper tape system up and running and I can see if I can run off a couple copies of the Nova diagnostics but the big question would be if the system is backward compatible. Have you tried running any of the software from the Nova 1000/1200 manual?