24 Jul
2011
24 Jul
'11
10:54 a.m.
So I started checking caps but I've realised I'm out of my depth when it
comes to analogue electronics. I'll come back to it though.
For now I've substituted a modern peecee PSU (with the intention of
putting the proper DEC PSU back in place when I'm sure the rest of the
system is running).
I also built a small circuit to bring BPOK and BDCOK up at the right
times (http://home.windstream.net/engdahl/powerup_reset_circuit.htm).
What I'm not seeing is any terminal output. I discovered there's a small
2A fuse on the quad serial card (DLV11-J) which had blown. Replaced this
and now have the correct -12V on that card. Still no activity on Tx
though. I've double checked all the jumpers on that card, port 3 is set
for the console address, 9600 baud, RS232. I've checked the Tx pin on
the UART and no activity there also, so it's not a line driver issue.
Bear in mind I'm new to the PDP-11 so all this is taking some time since
I'm learning as I go ;) Digging a little further I found the ODT is
actually implemented as microcode on the CPU (M7270/KD11). There's a
pair of jumpers on the CPU card that determine the boot method and I've
changed these to boot straight to ODT. I'm trying to build the simplest
possible system at the moment to get anything working. If I've
understood corectly then just having the CPU+serial card should get me
the '@' ODT prompt. It doesn't. I've tried other combinations of
CPU+memory+serial+BDV etc.
Other things I've checked : The CPU has all 4 clocks and correct
voltages. The UARTs have clock and correct voltages. The bus lines are
being pulled high (by the CPU card I think).
Probing further the bus signals BSYNCL, BDOUTL etc. all seem to be
pulsing so _something_ is happening. What I have noticed and doesn't
seem right is during a BDOUTL cycle some of the data/address (BDALxL)
lines are oscillating at high frequency, i.e. for the whole period that
BDOUTL is low the BDALxL lines oscillate (around 20-30 times per BDOUTL
period). Does anyone have any ideas what could be wrong here?
Cheers,
James
On 13/07/2011 19:27, Tony Duell wrote:
On 09/07/2011
21:13, Tony Duell wrote:
Thanks Tony, you caused me to double check things...and it looks like
there's a couple of spikes on the ground of about 4Vp-p and 3Vp-p @
Hang on a
second. What is the 'scope probe ground lead connected to, and
what is the 'scope probe tip connected to?
If the 'scope ground lead is connected to the 0V rail and you are seeing
4V spikes on something else that's supposed to 'ground' (i.e. also
connected to the 0V rail) then either you have a very serious ground
bounce problem or there's an open-circuit somewhere.
Looks
like the +12V startup and -12V (actually -15) are fine. I'm
Have you checked
these with a 'scope? Excessive ripple on a supply line
could cause exactly the fault you are seeing.
Ho 32Khz which was throwing all my measurements out.
I'm guessing a cap,
does that sound about right?
32kHz sounds to me like the switching frequency. If
you're getting ripple
at that freqeucny, I would suspect the capactiros on the DC output side
of hte rectifiers.
-tony