• Please review our updated Terms and Rules here

Which floppy drive for RQDX3 and M9058?

D

DrCharles

Experienced Member
Joined
Sep 2, 2014
Messages
101
Location
West Plains, MO
The last piece of the puzzle... I now have everything to put a floppy on my 11/23+ except the physical floppy drive.

I've done a lot of reading and I could use an RX-50 dual 5.25 drive, or a TEAC 55-GFR 5.25 drive.
Right now I'm thinking of buying two 55-GFR's, one in the Windows box and one connected to the RQDX3.

The 3.5" drives are apparently usable with a custom breakout board like this one:
http://avitech.com.au/?page_id=645

And that page (and links) have lots of useful info. There is even a mini-breakout board:
http://avitech.com.au/?page_id=2402

But the one question I can't find an answer to is:

Can a 3.5" PC drive be used with the RQDXE (M9058.) breakout board, or is additional circuitry still required (and if so, what)?

I plugged one in for an experiment with the PC_standard "twisted" cable, and the RQDX3 doesn't know it's even attached. (yes, I hooked up power too) ;)
Anyone have experience with this combo?
thanks.

Edit: I found the schematic for hooking up a floppy to the RQDX3 directly. It can handle a 3.5" (as an RX33) but still doesn't say what the RQDXE (M9058.) wiring is. I can dig out the M9058 schematic too, and see what comes out of it on its 34 pin header. Probably just need to tweak a line or two. Hopefully.
 
Last edited:
Sorry about all the posts, but there is only 15 min. allowed for editing...

I'm out of town for the week, so won't be able to test my theory until the weekend, but I suspect I've been "jumpered" again ;)

From looking at the RQDXE schematic, comparing to the mini-BOB (http://www.avitech.com.au/mm-files/rqdx-floppy-mini-bob/rqdx-bob-floppy-schematic.pdf), it looks like the signals match up... and the RQDX3 can use a 3.5" drive as an RX33 unless I misinterpreted things.

Except for the twist vs. straight cable issue. I know the 3.5 HD drive was A: in the old PC I took apart, but now can't remember whether I used the twisted connector or the straight one. Didn't change any jumpers on the drive itself.

The RQDXE puts out Motor On via pin 16, and Drive Sel 1 on pin 10 (assuming the little "RX50" jumper is in place, of course. It's there in the ebay picture).
But that's a straight-cable hookup, NOT a twist, since the floppy header also wants pins 16 and 10 respectively.

So that would explain why the floppy motor never came on, if 10 and 16 were reversed due to hooking up the twisted connector, which I'm 100% sure I did!
Then there's the issue of whether it's DU0, 1 or 2. I have no MFM hard drives hooked up so it shouldn't be too hard to find.

Hopefully someone who's done this before can confirm... anyway I have six days to think about it :)
 
Last edited:
This has all been done before. You should study the photos in this album: http://www.vcfed.org/forum/album.php?albumid=376

All cables are to be straight, no twists. Also the FMPS are available for the RQDX3 and RQDXE. If you study them, they will answer your questions about signaling. You also need to have the floppy drives jumpered properly.

Yes, RX33 is a correct drive to use (as you know, RX50 works also). Yes a 3.5" HD floppy drive (RX23) will work also and will look like an RX33. There is discussion of this in this thread: See here: http://www.vcfed.org/forum/showthread.php?41116-Booting-Pdp-11-from-flash-USB . See pages 3 and 4.

Lou
 
You've referenced the RQDX3 Break Out Board at avitech.com.au. Be advised that that site belongs to one of the members of this site, Mal, who posts here frequently. There's another page at his site you should become familiar with on just the topic you're asking about: What floppy drives can you use with your RQDX3.

Take a look here. http://avitech.com.au/?page_id=2660

Pay particular attention towards the bottom of the page where he discusses the Teac FD-55 drives he's successfully jumpered to work with an RQDX3.
 
DDS said:
Pay particular attention towards the bottom of the page where he discusses the Teac FD-55 drives he's successfully jumpered to work with an RQDX3.
Click to expand...

I am, in fact, familiar with that page, which gives much useful data on 5.25" floppies. And if you look carefully at my first post in this thread, you'll find:

I've done a lot of reading and I could use an RX-50 dual 5.25 drive, or a TEAC 55-GFR 5.25 drive.
Click to expand...
:)

Good to know he's a member here. Nowhere in his material could I find the answer to the specific question I asked, which again was:

Can a 3.5" PC drive be used with the RQDXE (M9058.) breakout board, or is additional circuitry still required (and if so, what)?
Click to expand...


Thanks very much Lou N2MIY for the confirmation, and also the link. I must have skipped over that one, due to the title (... booting from flash USB) :tellme:
It's encouraging that it has been successfully done by others. Will report back this weekend when I can play with my PDP-11 again.
 
I'm still pulling my hair out... The drive I'm trying to use (as an RX33) is a TEAC FD-235HF 4240. I found jumper settings online at http://oldcomputer.info/media/disk35/3fd0020a.pdf

Made a straight cable from the RQDXE connector to the floppy, set the drive jumpers to DS0... and nothing. Same with DS1.
The motor never comes on when doing a DIR or INIT to any of DU0...DU3!
(It does spin for half a second when a disk is inserted).

I also moved the "RX50" jumper on the breakout board from 1-2 to 2-3. Still nothing.
RT-11 map shows the registers at 172150 and 172152, and I can read/write them.
SHOW DEV: DU confirms installed at the default 172150 with vector 154.

There may be a Pin 2 issue, but that'll only cause problems when I can actually access the drive...
I've read everything on the avitech.au pages and other places...

It appears that the RQDX3/RQDXE outputs a motor-on signal only on pin 16. The select has to be on pin 12, so that makes it DRV SEL 2 from the RQDX3, and drive B: (DS1) for the drive itself.

The twist cable not only swaps selects, but changes the motor-on between pins 10<->16.
Again, I'm using a straight cable.

What am I overlooking, before I get out the scope and start probing?
Perhaps this particular drive can only be used with a twist cable (to put the motor signal on the correct pin)?
 
Last edited:
Nothing coming out on the motor (16) or drive select (12) pins, at all, when I run an XXDP diagnostic or attempt to access the drive via RT-11.

When I jumper drive pin 12 to ground, the drive LED lights up.
Jumper pin 16 to ground, the motor starts. So the drive itself is not the problem.

Red light goes out on the RQDX3 self-test after about six or seven seconds, as it should.
Incidentally the controller has Microcode Version 3 according to XXDP - is that the right one?

Thought maybe a ribbon cable was backwards - but reversing one end of the 50 pin between the controller and breakout card causes self-test to fail.
The cable between the breakout and the drive is brand new and only fits one way (and pin 1 on each is labeled).

So now what... at this point I'm not even sure if it's software or hardware :confused:
 
I just found a really "Dumb & Dumber" mistake. :rolleyes:

I had installed the M9058 breakout board in the backplane UPSIDE DOWN :shock:
Never seen a board where the component side faced away from every other board DEC makes. Looking closely at the board it actually says "Side 1" on the solder side, and "Side 2" on the component side. Sigh.

That explains why I didn't have any signals coming out the 34-pin connector... because the buffer chips had no power.
Now they have +5 and ground, at least.

But it still won't work, and I think I have found the (other) reason why... the BDCOK L signal is supposed to be low when the DC supply is good. And it's not.
The POWER OK lamp on the front panel is lit, and everything else in the rack is working (+5 and +12 are correct). Without this signal, though, all the disk select drivers are disabled. I suspect it's generated in the power supply, so time to dig out THAT set of prints.

And the band played on...
 
BDCOK H (I typed it wrong before) is actually generated on the three-switch front panel. It is indeed going high (not low) shortly after power-up. So far so good.

That panel has a 10 pin ribbon cable that passes the signal through the power supply motherboard and right out again on another 10 pin cable. THAT one goes to the backplane, thence to BA1 on all rows. But somewhere between the front panel and the M9058, it's disconnected because it's not going high at E6 pin 5. The trace on the board is intact down to the BA1 finger.

Yes, I've already checked the ribbon cables for continuity. It may be a corroded or damaged trace on the backplane itself... easiest way is just to pull the box from the cabinet, then disassemble as needed.

Incidentally, I had made another major error - the breakout board can NOT go in "any slot" as some of the instructions say, it MUST be in the A/B slots since only those have the BDCOK H signal going to pins A1! At least on the H9276 backplane, that is.
 
I removed and reinserted the 10-pin ribbon cable connectors a few times... lo and behold, there is now a DC OK signal all the way to the breakout board, and E6-3 is going low which enables the drive select buffers.

BUT the #$@%^ drive still doesn't respond!!

All XXDP diagnostics fail ZRQA** through ZRQF** with errors basically saying there's no drive connected. Under RT-11 if I enter INIT DUx: (x=0..3) I can see the error red LED blinking on the RQDX3.

I'm out of ideas for now. :( Unless the RQDX3 or the M9058 are actually defective, either as purchased or because I damaged something plugging the latter in upside down.
 
It works! :)

Yet another self-inflicted injury. Despite checking the orientation of the 50-pin cable between the RQDX3 and the M9058 more than once, somehow I'd managed to get it upside down where it plugs into the controller. :roll: I thought that caused a failure previously, but apparently not. So two wrongs didn't make a right this time, either.

Anyway, with ALL the cables and cards plugged in the right way, the 3.5" drive starts up and RQDF?? formatted a floppy. Now running read/write tests on it.
But I think the worst is over.

I really wish DEC and others would have used a connector that's impossible to plug in backwards! :stern:
 
All the XXDP tests work and it recognizes the 3.5" HD floppy as an RX33. So far so good.

But RT-11 (5.01) does not see the drive as any of DU0..3.
I'm assuming there is something I have to CONFIG to make it work (DU is definitely installed)?
Or is my version of RT-11 too old for an RX33? :confused:
 
As I suspected (and confirmed on cctalk), V5.04 is the first one that supports an RX33.

I booted a slightly more recent RT-11XM pack which is 5.04 and it does notice the DU0: and I can read and write files to it. :)
Time to figure out how to make a newer version of RT-11SJ so I can run my TSX+ 6.50 on top of it. (TSX+ needs the single-job monitor, not the XM).

Then I have to start messing with PUTR and see if the Windows box can make a floppy readable by the PDP-11.
 
You must log in or register to reply here.
Back
Top