🚀 go-pugleaf

I am having BIG problems trying to get a 3rd party disk working.
The disk in question is a Hitachi DK516C-16(about 1.3 Gb)
I have created a disktab entry like so:
	DK516C-16|Hitachi DK516C:\
		:ty=hitachi:ns#81:nt#15:nc#2179:\
		:pa#518805:ba#8192:fa#1024:\
		:pc#2606400:bc#8192:fc#1024:\
		:pe#419175:be#8192:fe#1024:\
		:pf#1006020:bf#8192:ff#1024:\
		:pg#503010:bg#8192:fg#1024:\
		:ph#159390:bh#8192:fh#1024:
I got the values for ns, nt and nc from rzdisk, in particular
	rzdisk -g current /dev/rrz0a

(Isn't it strange that 81*15*2179=2647485, and I have pc# = 2606400 ?
Well I am told by DEC that this is ok. Normal.)

I put this on the disk, using chpt(with correct options), newfs'd the a partition,
put it on again with chpt -a, and newfs'd every partition.
Sometimes when I fsck the a partition immediately after it says
BAD SUPERBLOCK, use another superblock. I have then seen it work on block 32(but not
always), and if I then check it again it with no b option, if fails, and then fails next
time I specify block 32 again.
Other odd things have happened (usually superblock trashing) with the other file systems
when I have been using different numbers in /etc/disktab, but I think the one above
is correct now. Can anyone send me one that works?

Also I have run "rzdisk -s 0 -1 /dev/rrz0c" and scanned the disk ok with no erors.
So I think there is something wrong with the a partition superblock, but I can't be
sure. Is there a way of getting it to use another all the time? How do I map bad blocks,
but should I do this if rzdisk says it's ok? Also, does fsck -b 32 tell the sytem to
always use this superblock?

Is there a known problem with DEC and big disks.

Any help greatly appreciated. I am getting desperate.
Thanks in advance,

Ross