N* update - replaced 7805 regulator

[billdeg]

(this is a copy of a post I sent to the N* Yahoo Group)

There's an old thread about this project, but I am starting a new one.

1) Not getting 5V to the A drive (less than 2V). Traced the problem to the 7805 inside heatsink, 6030B, the front-most 7805. I replaced the 7805 and now I get the correct voltage to everything.

2) I checked every card, they all seem to be getting the correct voltages (12 or 5)

ZPB processor (also have a ZPB-2)
MDS-AD3 (also have a MDS)
IMS 4mhz 16K RAM set for top memory (I believe - checking to confirm)
N* 16K RAM bd (8000-Bfff)
n* 32K RAM bd (0-7fff)

3) I installed an in-circuit z80 emulator to check the memory (read / write test) - no errors through 7FFF. I can't write to anything above that, but the RAM is there, unsure why but I assume that this is part of an initial lockdown of RAM by design, as part of the bootstrap process (I have to confirm this). It's possible also that the emulator is not allowing the ROM to finish what it needs to do. Not worried at this point. I checked location E800 and what appears to be bootstrap code is located in this location (which is where the manual says it should be). I don't at the moment have a copy of the bootstrap code, but I assume I can find this on the web to confirm it's loading into memory. Looking for that now.

4) When I power on the machine, both drives spin but neither engage the boot disk. After maybe 20 seconds the drives stop spinning. I verified that the drive dip switches were correct from when the system worked. I also tried other similar drives (using Tandon 100-2A's). It's possible that I have the wrong settings,. but this is what worked before:
A DRIVE: Tandon 100-2A jumpering TP11's 2nd (facing from the front) and inserting a T-RES terminator in TP7
B DRIVE: MPI model 52 with G jumpers on: 2-5-7, terminator installed (not needed).


I must be getting close. I will let you know if I make more progress. .

 

[modem7]

It's possible that I have the wrong settings,. but this is what worked before:
A DRIVE: Tandon 100-2A jumpering TP11's 1st and 3rd (facing from the front) and inserting a T-RES terminator in TP7
B DRIVE: MPI model 52 with G jumpers on: 2-5-7, terminator installed.

If these drives are daisy chained, shouldn't there be only one terminator (on the drive at the end of the cable)?

What happens with only the TM100 connected? I'm unfamiliar with your system but I would have expected the use of a straight-through cable (no twist) together with the TM100 jumpered for DS0 (wire in position 2) rather than DS1 (wire in position 3). DS1 is what I expect to see in a PC system for the first drive (not that other systems don't use that too).

terminator in TP7

Very minor: Note that TP7 is Test Point 7, the pin sticking up out of the board, right next to the terminator socket. On my TM100s, the terminator socket is designated as 2F.
Same with the jumper socket. It's not TP11.

 

[billdeg]

Tandon 100-2A setting for Northstar Horizon Drive Assignment

Tandon 100-2A setting for Northstar Horizon Drive Assignment

First of all, the system is working again. Using my z80 in-circuit emulator, I confirmed that the disk controller bootstrap program was loading into memory. I researched the dip switches for the drive and determined that my dip switch socket was one place off. Moving the socket and resetting the dip switches, plus setting my terminal to 9600 b (for the left/corner serial port - voila I am back in business.

As long as there is one terminator, it does not matter where it is in the chain. It does not seem to matter if there is more than one terminator either, at least from my experimenting.

Anyway, I did some research and for the record here are the drive assignment jumpers for a Tandon 100-2A. This will work for a Northstar Horizon using the MDS controller

pin 2 jumpered to 15 = drive 1 (or whatever you call drive A/boot drive)
pin 3 jumpered to 14 = drive 2 (drive B)
pin 4 jumpered to 13 = drive 3 (C)
pin 5 jumpered to 12 = drive 4. (D)

 

[billdeg]

Very minor: Note that TP7 is Test Point 7, the pin sticking up out of the board, right next to the terminator socket. On my TM100s, the terminator socket is designated as 2F.
Same with the jumper socket. It's not TP11.

Thank you for the correction.

 

[modem7]

As long as there is one terminator, it does not matter where it is in the chain.

Just like with SCSI, OEM documentation will indicate that the terminator should be placed at the ends of the cable, otherwise signal reflections may occur and may interfere with card to device communications. If you're happy with the terminator being mid-cable, that's fine, as long as others reading this know that there may be a risk of intermittent operation.

It does not seem to matter if there is more than one terminator either, at least from my experimenting.

Over a year ago, I collected a 5150 for a friend (eBay purchase). It had dual TM100s and would not boot. The cause turned out to be that both TM100s had terminators. And so maybe certain floppy controllers are better at dealing with two terminators than others.

 

[billdeg]

I agree, the first thing to do if you're having trouble with a set of drives and you're sure the dip switches are correct is to move the terminator to the end of the line, and remove all extra terminators. As a general rule.

For the N* MDS it does not seem as much to matter, and this is actually mentioned in the documentation.

Bill

 

[MikeS]

As mentioned, it is always recommended that the terminator goes at the far end to avoid reflection problems, and that only one be installed to minimize load on the FDC drivers.

Your original post that "it doesn't matter" is about the same as saying speed limits don't matter because you went for a drive at 100mph and didn't get caught.

m

 

[billdeg]

It does not matter in the case of my system. There is no performance or other issue.

 

[Druid6900]

Regardless of what the drive manufacturer's specs might say, I've never had a problem with having NO terminators on the drives. More than one, yes, but 1 or zero, no.

Perhaps if you are working in an environment that would turn your chromosones into jello, it might be advisable to terminate the signal cable.