Subject:
|
Re: memory not remembering...
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Sun, 26 Jan 1997 15:59:39 GMT
|
Original-From:
|
Fred G. Martin <fredm@media.mit.%ihatespam%edu>
|
Viewed:
|
1101 times
|
| |
| |
Ben -- the pin 27 R/W line of the RAM should be connected directly to
the R/W line of the 6811. I don't remember the pin # off hand -- it's
7 i think -- but just check the schematic.
check the continuity first on the board. if that's ok, then check all
the way to the chip leads. to get your probe into the 6811 sockt,
unfold a desk staple or grab a bit of wire to poke down into the 6811
pin.
if this connect is bad, it would explain your problems.
-Fred
In your message you said:
> I'm stumped.
>
> In the process of assembling my first handyboard,
> I have just reached the stage where I install the memory
> (all previous checkpoints passed).
> The computer successfully downloads the bootstrap,
> sets the config register correctly, and dowloads pcode_hb.s19
> file (at least, according to dl), but when I power-cycle the
> board, the IC banner does not appear on the LCD screen- just the
> row of 16 boxes.
>
> Browsing through the handyboard archives, I found the following
> suggestion by Fred for dealing with this specific problem:
>
>
> > Success on the first bootstrap download and failure to write into RAM
> > could easily be explained by a fault in the memory circuit. The
> > memory circuit, which includes the 'hc373, the 'hc132, and the RAM
> > chip itself, with about 27 signals going to the 'hc11, is by far the
> > most complicated subsystem on the board, and any single problem
> > anywhere in the circuit will cause the memory to not work, leading to
> > the download error described.
> >
> > It's tedious, but the way to debug this is to individually examine the
> > signals on the '373 and the ram chip using a scope or logic probe.
> > With the HB turned on (not in bootstrap mode -- just turn it on and
> > let it execute garbage), all of the appropriate signals (data bus,
> > address bus, R/W line, '373 clock line) should be oscillating. If any
> > are disconnected or shorted to power or ground it should be easy to
> > see with the logic probe.
> >
> > Then with an ohmmeter and the board turned off, check that there is
> > continuity where there is supposed to be (using the schematic as a
> > reference) and no shorts between address and data lines.
>
>
> Armed with an oscilloscope and an ohmmeter, I checked all the chips
> and connections in the memory circuit, and found everything was
> working as it should, except the R/W line for the ram chip (pin
> 27, according to my schematics). It does not oscillate but remains
> at +5V.
>
> What sort of oscillations should I expect on this pin? Do I need to
> be trying to execute code or transmitting data to see the
> oscillations?
>
> I tried reheating and resoldering this connection but nothing changed.
> (I reheated all the connections in the memory circuit just to be
> sure)
>
> I also noticed that this pin drops to ground when the board is turned
> off, while the chip power supply (the adjacent pin 28) remains at
> +5V, so the two of them aren't shorted together (which was my first
> guess at what was happening)
>
>
> Any thoughts or suggestions appreciated.
>
> -max
>
>
>
>
>
|
|
Message is in Reply To:
| | memory not remembering...
|
| I'm stumped. In the process of assembling my first handyboard, I have just reached the stage where I install the memory (all previous checkpoints passed). The computer successfully downloads the bootstrap, sets the config register correctly, and (...) (28 years ago, 26-Jan-97, to lugnet.robotics.handyboard)
|
2 Messages in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|