5 Feb
2004
5 Feb
'04
12:25 a.m.
I had a spare moment yesterday so I decided to try and troubleshoot my
bad M7891-DC (128kw) in an 11/34a.
Looks like I have no +12v. The 12v regulator seems to be working as I
have 12v up the 2A fuse. But nothing after that.
I'm assuming the fuse has blown. Easy to replace, but I also suspect it
blew for a reason. (and replacing it might just yield another blown fuse
:-)
Anyone seen this before? Do they sometimes blow due to transients?
I'm tempted to apply 12v after the fuse with a bench supply that has a
current regulator and see what it draws. It's been a while since I
worked with "plain old DRAMs" but I assume the 12v is for them (Vdd).
-brad
5 Feb
5 Feb
12:57 a.m.
On Thu, 2004-02-05 at 14:25, Brad Parker wrote:
I had a spare moment yesterday so I decided to try and troubleshoot my
bad M7891-DC (128kw) in an 11/34a.
Looks like I have no +12v. The 12v regulator seems to be working as I
have 12v up the 2A fuse. But nothing after that.
I'm assuming the fuse has blown. Easy to replace, but I also suspect it
blew for a reason. (and replacing it might just yield another blown fuse
:-)
Hmm, usually the transistors blow to protect the fuse...
Anyone seen this before? Do they sometimes blow due
to transients?
Fuses do just blow sometimes. You could try replacing it and seeing
what happens, but make sure it's got the same rating. Another old audio
amp trick is to use a car indicator bulb (12v, 21W) or tail light bulb
(12v, 5W) across the fuse - if it's drawing a clearly silly current, the
bulb will light.
I'm tempted to apply 12v after the fuse with a
bench supply that has a
current regulator and see what it draws. It's been a while since I
worked with "plain old DRAMs" but I assume the 12v is for them (Vdd).
Be careful - my bench PSU, when set to 12v, actually produces 6v-0-6v
with chassis earth (and hence mains earth) being the 0v centre tap.
What type of DRAM chip is it? I seem to recall that the 4116 DRAMs used
in the ZX Spectrums tended to fail short-circuit between 12v and 0v.
This usually stopped the chopper PSU that provided +12v and -5v from the
nominal 9v input. This caused mad black-and-white flickering, because
the RAM was reading out random garbage and the 12v supply also fed the
colour encoder.
HTH
Gordon.
3:11 a.m.
New subject: M7891-DC 2a fuse on +12? fixed!
To follow up on my own post, replacing the fuse created a *hot* 12v
regulator. My DVM indicated that something was shorting Vdd.
Figuring it was a cap, I started pulling the + lines of the big caps.
On the second one the short went away.
Turns out one of the large Sangamo 47uf caps across Vdd was a dead
short. I removed it and now the +12 is happy and the board works.
(I'll replace it in my next digikey order)
Do old electrolytic's short out? The MS11 parts list just say "AL EL"
which I assume is aluminum electrolytic. For some reason I was thinking
it would be a tantalum (based on it's silvery look) but now I'm thinking
not. I think I'll replace it with a nice axial lead tantalum.
I know sometimes badly mfg'd electrolytics will leak, but I've never
seen a dead short - but my expeience is mostly with *new* products, no
ones that are so "experienced" as this one :-)
Normally I wouldn't post this sort of thing but I figured someone else
might find it interesting/amusing.
-brad
Brad Parker wrote:
I had a spare moment yesterday so I decided to try and troubleshoot my
bad M7891-DC (128kw) in an 11/34a.
Looks like I have no +12v. The 12v regulator seems to be working as I
have 12v up the 2A fuse. But nothing after that.
I'm assuming the fuse has blown. Easy to replace, but I also suspect it
blew for a reason. (and replacing it might just yield another blown fuse
:-)
Anyone seen this before? Do they sometimes blow due to transients?
I'm tempted to apply 12v after the fuse with a bench supply that has a
current regulator and see what it draws. It's been a while since I
worked with "plain old DRAMs" but I assume the 12v is for them (Vdd).
-brad
3:21 a.m.
New subject: M7891-DC 2a fuse on +12? fixed!
>>>> "Brad" == Brad Parker
<brad(a)heeltoe.com> writes:
Brad> Do old electrolytic's short out? The MS11 parts list just say
Brad> "AL EL" which I assume is aluminum electrolytic. For some
Brad> reason I was thinking it would be a tantalum (based on it's
Brad> silvery look) but now I'm thinking not. I think I'll replace
Brad> it with a nice axial lead tantalum.
Should be ok. Tantalum was uncommon back in the 1970s, except for
miniature capacitors.
And yes, electrolytics fail with age. It's common for old power
supplies to require replacement of all their electrolytics. (By "old"
I mean 1950s or so -- but 30 years is enough that it isn't a surprise
here either.)
paul
6:03 a.m.
New subject: M7891-DC 2a fuse on +12? fixed!
To follow up on my own post, replacing the fuse created a *hot* 12v
regulator. My DVM indicated that something was shorting Vdd.
Figuring it was a cap, I started pulling the + lines of the big caps.
On the second one the short went away.
You were lucky. Normally it's the last one on the board that turns out to
be shorted ;-)
Turns out one of the large Sangamo 47uf caps across
Vdd was a dead
short. I removed it and now the +12 is happy and the board works.
(I'll replace it in my next digikey order)
Do old electrolytic's short out? The MS11 parts list just say "AL EL"
I've had it happen. Remember that the plates of an electrolytic cap are
the metal anode and the electrolyte. The dielectric is the oxide film on
the surface of the anode. If that disolves back into the electrolyte,
then you can get a dead short.
I had this happen in the PSU module of a Canon CX laser printer. I was
testing various bits of it (it seemed fine), I left it for a few hours,
and when I came back I found a dead short on one of the unregulated power
rails. Turned out that the 10000uF smooth cap had decided to go
short-circuit at that moment. Thank %deity I didn't power it up like that!
-tony
7578
days inactive
7578
days old
4 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
brad＠heeltoe.com -
gordonjcp＠gjcp.net -
pkoning＠equallogic.com