[N8VEM: 20071] New ECB backplane and PropIO running!

Discussion:

John Snowdon

2015-09-03 13:19:29 UTC

I'm almost finished putting my 8 slot backplane, PropIO v2 and Z180 Mark IV
system together.

All the boards are built, and today I've just uploaded the code to the
Parallax cpu on the PropIO board. So far I'm still waiting on finding a
method to write the AM29F040B 512k flash on the Mark IV (my old parallel
Willem board just won't work reliably in any of my modern systems, so I may
try and pick up a TL866 programmer so that I can do this sort of stuff
under Linux), so I don't have a finished system yet.

<Loading Image...

>

<Loading Image...

>

Anyway, after a little bit of work trying to sort out the null modem wiring
between the PropIO serial port and a cheap USB to serial adaptor I had some
success with the Parallax SimpleIDE tool to detect the microcontroller, but
I couldn't find a tool to send the code to the board (the parallax tool
under WINE doesn't work). I found 'bstl' instead, a native Linux Parallax
utility which does the same job as the Propeller Tool does in Windows.

I've ran the tool, uploaded the code and then reset the board. It beeps and
then after a second I can now see the startup messages and version number
across the bottom of the screen.

<Loading Image...

>

<Loading Image...

>

One thing I haven't seen, however, is anything on the activity led. The 5v
and 3.3v leds are okay, the speaker beeps and vga output works, but nothing
on the activity led. It's not a major issue, it could be a broken led as
they just came out of my bits drawer; so I'm not sure how old they are or
what the specs were.

John

--
You received this message because you are subscribed to the Google Groups "N8VEM" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem+***@googlegroups.com.
To post to this group, send email to ***@googlegroups.com.
Visit this group at http://groups.google.com/group/n8vem.
For more options, visit https://groups.google.com/d/optout.

John Coffman

2015-09-03 15:21:31 UTC

Permalink

John Snowdon

2015-09-03 18:08:12 UTC

Permalink

Hey John,

I've got a Tomsad Willem Pro 3. From what I can work out, it's a 'PCB 3'
design. It's this one:

Loading Image...

It appears to have an A18 jumper to the left of the dip switches, but no
combination I try ever works. I did at one point find that the dip switches
were labelled in reverse (dip 1 == 12, dip 12 == 1) to what the Willem
software showed them, that took a while to work out and then it worked well
for blowing 27c256 and 27c512 eproms for a good few years :-)

The biggest problem is that I no longer have a physical Windows machine to
run the programmer from. I kept an XP system around for years, but it
didn't make the cut when we moved house last year and everything else I've
got is Linux now. Virtualbox runs the software in a virtual Windows 7
image, and will actually read the few eproms I've thrown at it, but I can't
seem to do anything with this flash rom beyond reading it (it's full of
rubbish, so looks to have had something written to it previously) -
identify, erase and program all fail.

Something a bit more modern and that would work natively in Linux seems to
be the best option now.

John

Post by John Coffman
old parallel Willem board
The larger chips need a Willem board that has the "A18" jumper, labeled J3
on my board.
I have the 5.0E board, and use it for 512K flash (SST & AMD) and for
programming the EEPROM for the PropIO. In the latter case, I remember
discovering that either the s/w indicated the use of the wrong socket, or
the board was mislabeled; but I finally figured this out, and have
programmed a couple of those EEPROMs for different versions of the PropIO.
Once you get your first Mark IV flash memory programmed, updates can be
flashed in-place from CF card or SD card FAT16 file system.
--John

J. Alexander Jacocks

2015-09-03 19:46:17 UTC

Permalink

John,

There are not many programmers that have Linux software available (or Mac,
for that matter). You would probably be better served by installing a
second disk in one of your machines, and installing Windows XP on it. I
personally put a lot of effort in to trying to find a programmer that
doesn't require Windows, with very little success.

- Alex

Post by John Snowdon
Hey John,
I've got a Tomsad Willem Pro 3. From what I can work out, it's a 'PCB 3'
http://www.ezoflash.com/willem/62.jpg
It appears to have an A18 jumper to the left of the dip switches, but no
combination I try ever works. I did at one point find that the dip switches
were labelled in reverse (dip 1 == 12, dip 12 == 1) to what the Willem
software showed them, that took a while to work out and then it worked well
for blowing 27c256 and 27c512 eproms for a good few years :-)
The biggest problem is that I no longer have a physical Windows machine to
run the programmer from. I kept an XP system around for years, but it
didn't make the cut when we moved house last year and everything else I've
got is Linux now. Virtualbox runs the software in a virtual Windows 7
image, and will actually read the few eproms I've thrown at it, but I can't
seem to do anything with this flash rom beyond reading it (it's full of
rubbish, so looks to have had something written to it previously) -
identify, erase and program all fail.
Something a bit more modern and that would work natively in Linux seems to
be the best option now.
John

Post by John Coffman
old parallel Willem board
The larger chips need a Willem board that has the "A18" jumper, labeled
J3 on my board.
I have the 5.0E board, and use it for 512K flash (SST & AMD) and for
programming the EEPROM for the PropIO. In the latter case, I remember
discovering that either the s/w indicated the use of the wrong socket, or
the board was mislabeled; but I finally figured this out, and have
programmed a couple of those EEPROMs for different versions of the PropIO.
Once you get your first Mark IV flash memory programmed, updates can be
flashed in-place from CF card or SD card FAT16 file system.
--John
--

You received this message because you are subscribed to the Google Groups "N8VEM" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at http://groups.google.com/group/n8vem.
For more options, visit https://groups.google.com/d/optout.

John Snowdon

2015-09-04 07:44:40 UTC

Permalink

Hi Alex,

I've been looking in to this and the only one I've found that has a
supported, open-source utility is the Minipro TL866CS / TL866A:

http://minipro.txt.si/index.php?title=Main_Page

https://github.com/vdudouyt/minipro

There is another set of programmers from Batronix that have Linux/OS X
tools, but: 1) they're significantly more expensive and 2) the software
seems to be some giant .Net/Mono gui - the utility for the Minipro is a
nice simple C command line tool. The cheapest bundle seems to be around Â£35
(around Â£60 if you want all the optional TSOP and QFP adaptors), so it's
not a massive investment to try it out.

Cheers,
John

Post by J. Alexander Jacocks
John,
There are not many programmers that have Linux software available (or Mac,
for that matter). You would probably be better served by installing a
second disk in one of your machines, and installing Windows XP on it. I
personally put a lot of effort in to trying to find a programmer that
doesn't require Windows, with very little success.
- Alex

John Snowdon

2015-09-28 13:25:52 UTC

Permalink

Well I finally got hold of another programmer - a cheap TL866cs and it
works brilliantly in Linux. I also took the opportunity to replace my
single, used AM29F040B with a tube of brand new SST39SF040 chips. The
combination worked perfectly.

I first ran through the TEST0 rom and have a clean bill of health *except*
the SD card WP & CD test. I'm still not 100% certain what I'm supposed to
do with P4, but I put jumpers on 3 & 4 and 11 & 12, based on the fact I'm
using an FCI socket and a post I found to the list which mentioned bridging
those pads. No matter what I do and whether I have a card in or not, the
test always comes back as CD = False, WP = True and Interrupt = No.

Secondly when using the UNA CPM image 1.00.17 I don't seem be to holding
settings in nvram; if I start the board up, set the date and time (ignoring
any other settings for now...) and see the message about the DS1302 being
updated, on turning the board off it loses all settings and reruns the full
setup "due to 0xEE".

Lastly, without turning the power off, if I run through the full setup I
need a few pointers on what I should be entering for some of the questions.
Is there a guide somewhere that details what each question means and what I
should be entering based on the mark IV and Prop IO boards in the system?
Enabling the Prop IO screen and keyboard in the setup doesn't seem to set
them as the default output device, should that be the case?

I can feel that I'm almost there, but just need a few pointers to resolve
these last few queries :-)

John Coffman

2015-09-28 14:36:34 UTC

Permalink

Post by John Snowdon
Well I finally got hold of another programmer - a cheap TL866cs and it
works brilliantly in Linux. I also took the opportunity to replace my
single, used AM29F040B with a tube of brand new SST39SF040 chips. The
combination worked perfectly.
I first ran through the TEST0 rom and have a clean bill of health *except*
the SD card WP & CD test. I'm still not 100% certain what I'm supposed to
do with P4, but I put jumpers on 3 & 4 and 11 & 12, based on the fact I'm
using an FCI socket and a post I found to the list which mentioned bridging
those pads. No matter what I do and whether I have a card in or not, the
test always comes back as CD = False, WP = True and Interrupt = No.

Assuming the FCI part number matches the configuration information, if
there is no change in the signals as the card is inserted and withdrawn,
this may be a soldering issue. The CD and WP outputs from the SDcard
socket are simple SPDT switches to ground. Check the trace continuity
to the CD/WP pins on the socket. Check that the pins short to ground on
card insertion/removal.

http://n8vem-sbc.pbworks.com/w/file/71455971/dual%20SDcard%20configuration.pdf

Post by John Snowdon
Secondly when using the UNA CPM image 1.00.17 I don't seem be to holding
settings in nvram; if I start the board up, set the date and time (ignoring
any other settings for now...) and see the message about the DS1302 being
updated, on turning the board off it loses all settings and reruns the full
setup "due to 0xEE".

What type of battery backup are you using? If you have a Supercap
installed, is the polarity correct? Are you configuring it to charge
up: (response) 2+2 for 2 diodes and 2K series resistor. Later you may
maintain it with 2+4 or 2+8.

Post by John Snowdon
Lastly, without turning the power off, if I run through the full setup I
need a few pointers on what I should be entering for some of the questions.
Is there a guide somewhere that details what each question means and what I
should be entering based on the mark IV and Prop IO boards in the system?
Enabling the Prop IO screen and keyboard in the setup doesn't seem to set
them as the default output device, should that be the case?

The PropIO peripherals are not supported by Setup. It is only supported
as the console device. It should be detected on the first boot after a
5-10 second delay (the SDcard LED flashes during the delay time).

A full initialization can be requested as one of the last Setup
questions. If NVRAM is invalid, a full initialization is always done.

--John

Post by John Snowdon
I can feel that I'm almost there, but just need a few pointers to resolve
these last few queries :-)

John Snowdon

2015-09-29 15:34:45 UTC

Permalink

Hi John,

Sounds like I need to recheck the continuity on the SD card socket. It
looked good to me, but there's not a lot of room to check those pins once
the other components are in, I'm quite prepared to hold my hands up for
that one if I've got some solder bridges ;-)

In relation to the DS1302, I'm using the supercap, which I'm sure is
installed correctly - though I'll confirm that this evening. However I'm
afraid you've lost me with regard to the 2+2, 2+4 information.

I think the NVRAM/battery issue is the root cause of the other issues as on
reboot I don't get the Prop IO initialised by the Mark IV and the full
setup routine runs again. I think sorting the supercap issue may rectify
those problems.

John Coffman

2015-09-29 16:19:15 UTC

Permalink

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
On 09/29/2015 08:34 AM, John Snowdon wrote:
<blockquote
cite="mid:6b70d793-9b63-4211-9d5f-***@googlegroups.com"
type="cite">
<div dir="ltr">Hi John,
<div> 
</div>
<div>Sounds like I need to recheck the continuity on the SD card
socket. It looked good to me, but there's not a lot of room to
check those pins once the other components are in, I'm quite
prepared to hold my hands up for that one if I've got some
solder bridges ;-)</div>
</div>
</blockquote>
I hate surface mount.  But with SDcard sockets there is no choice. 
Getting those SPST mechanical switches right is important.  GND is
on pin 3 and pin 6; but watch out for the weird numbering of the
socket pins.  Go over to the DualSD card docs for pix of mounted
sockets and .PDF's.  The MarkIV pages link over there. 
 
I just checked the schematic, and I noticed that if the VDD (+3.3v)
supply is at GND, you will see the CD and WP fail to toggle on card
insertion.  Check that VDD supply is around 3.3v. 
 
<blockquote
cite="mid:6b70d793-9b63-4211-9d5f-***@googlegroups.com"
type="cite">
<div dir="ltr">
<div> 
</div>
<div>In relation to the DS1302, I'm using the supercap, which
I'm sure is installed correctly - though I'll confirm that
this evening. However I'm afraid you've lost me with regard to
the 2+2, 2+4 information.</div>
</div>
</blockquote>
"2+2" would be a complete reply to the prompt regarding the "Trickle
charge ..."  Note:  the response is prompted as "d+r" for diodes
& resistor. 
 
Also, make sure there is no solder bridge from the square pin of
SuperCap mounting pads to VBAT.  I doubt that this is the trouble. 
 
<blockquote
cite="mid:6b70d793-9b63-4211-9d5f-***@googlegroups.com"
type="cite">
<div dir="ltr">
<div> 
</div>
<div>I think the NVRAM/battery issue is the root cause of the
other issues as on reboot I don't get the Prop IO initialised
by the Mark IV and the full setup routine runs again. I think
sorting the supercap issue may rectify those problems.</div>
</div>
</blockquote>
Are you seeing the SDcard LED flash on cold startup?  Are you using
the correct board I/O address:  0x40  (off, on, on, on) for the P10
jumpers?  UART output depends only on the internal Z180 UART, which
is unaffected by the I/O address jumper setting. 
 
Which board revision do you have?  1.00 or 2.0.  Rev. 1 board needs
a wiring update, but it would have nothing to do with your problems. 
 
--John 
 
 
</body>
</html>



-- 
You received this message because you are subscribed to the Google Groups "N8VEM" group. 
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:n8vem+***@googlegroups.com">n8vem+***@googlegroups.com</a>. 
To post to this group, send email to <a href="mailto:***@googlegroups.com">***@googlegroups.com</a>. 
Visit this group at <a href="http://groups.google.com/group/n8vem">http://groups.google.com/group/n8vem</a>. 
For more options, visit <a href="https://groups.google.com/d/optout">https://groups.google.com/d/optout</a>.

John Snowdon

2015-09-29 20:07:18 UTC

Permalink

Thanks for the further info John.

I've got one of your v2.0 boards, the TEST0 rom image passes all tests
excluding the CD/WP and I get the sequences of LED flashes for each test as
described in the accompanying README file. I'm going over the soldering of
the SD card socket now to make sure I've not done anything stupid

<Loading Image...

>

<Loading Image...

>

Thanks for the set up response info re: the DS1302. I'm just looking at the
board now and I've definitely gone the supercap route (0.22F, 5.5v); fitted
as positive lead to square pad, negative lead to the round pad.

I'll recheck for any bad soldering before I go any further.

John

Post by Wayne Warthen
Hi John,
Sounds like I need to recheck the continuity on the SD card socket. It
looked good to me, but there's not a lot of room to check those pins once
the other components are in, I'm quite prepared to hold my hands up for
that one if I've got some solder bridges ;-)
I hate surface mount. But with SDcard sockets there is no choice.
Getting those SPST mechanical switches right is important. GND is on pin 3
and pin 6; but watch out for the weird numbering of the socket pins. Go
over to the DualSD card docs for pix of mounted sockets and .PDF's. The
MarkIV pages link over there.
I just checked the schematic, and I noticed that if the VDD (+3.3v) supply
is at GND, you will see the CD and WP fail to toggle on card insertion.
Check that VDD supply is around 3.3v.
In relation to the DS1302, I'm using the supercap, which I'm sure is
installed correctly - though I'll confirm that this evening. However I'm
afraid you've lost me with regard to the 2+2, 2+4 information.
"2+2" would be a complete reply to the prompt regarding the "Trickle
charge ..." Note: the response is prompted as "d+r" for *d*iodes & *r*
esistor.
Also, make sure there is no solder bridge from the square pin of SuperCap
mounting pads to VBAT. I doubt that this is the trouble.
I think the NVRAM/battery issue is the root cause of the other issues as
on reboot I don't get the Prop IO initialised by the Mark IV and the full
setup routine runs again. I think sorting the supercap issue may rectify
those problems.
Are you seeing the SDcard LED flash on cold startup? Are you using the
correct board I/O address: 0x40 (off, on, on, on) for the P10 jumpers?
UART output depends only on the internal Z180 UART, which is unaffected by
the I/O address jumper setting.
Which board revision do you have? 1.00 or 2.0. Rev. 1 board needs a
wiring update, but it would have nothing to do with your problems.
--John

John Snowdon

2015-09-29 21:01:02 UTC

Permalink

NVRAM now seems to be keeping settings - perhaps I just hadn't given it
enough time? Keeps time/date and device settings on power-off now.

Anyway, can't see anything immediately wrong with the SD card socket -
still not working. Will have a more detailed check tomorrow.

Apart from the SD card, one thing I can't seem to get working is
console/keyboard redirection to the Prop IO. It gets detected on start up
of UNA CPM, it does the normal 'cog' load screen on the monitor I've got
plugged in to it, displays version information (0.94), but as soon as I
type anything over the serial terminal and then hit return the vga output
goes screwy (blocks of colours, corrupted text etc), and after a few
attempts at this my monitor eventually loses signal. The serial terminal is
still responsive and working at this point. Press the mark iv reset and the
Prop IO immediately reboots and the screen appears again.

Post by John Snowdon
Thanks for the further info John.
I've got one of your v2.0 boards, the TEST0 rom image passes all tests
excluding the CD/WP and I get the sequences of LED flashes for each test as
described in the accompanying README file. I'm going over the soldering of
the SD card socket now to make sure I've not done anything stupid
<https://lh3.googleusercontent.com/-syCc9_tmR98/VgrsWUQPmVI/AAAAAAAAAEk/ha3ZTTXwm4c/s1600/P1050757-.JPG>
<https://lh3.googleusercontent.com/-ob3NuKTqTG8/VgrsZ5D0bbI/AAAAAAAAAEs/2bp_cxebOpg/s1600/P1050758-.JPG>
Thanks for the set up response info re: the DS1302. I'm just looking at
the board now and I've definitely gone the supercap route (0.22F, 5.5v);
fitted as positive lead to square pad, negative lead to the round pad.
I'll recheck for any bad soldering before I go any further.
John

Post by Wayne Warthen
Hi John,
Sounds like I need to recheck the continuity on the SD card socket. It
looked good to me, but there's not a lot of room to check those pins once
the other components are in, I'm quite prepared to hold my hands up for
that one if I've got some solder bridges ;-)
I hate surface mount. But with SDcard sockets there is no choice.
Getting those SPST mechanical switches right is important. GND is on pin 3
and pin 6; but watch out for the weird numbering of the socket pins. Go
over to the DualSD card docs for pix of mounted sockets and .PDF's. The
MarkIV pages link over there.
I just checked the schematic, and I noticed that if the VDD (+3.3v)
supply is at GND, you will see the CD and WP fail to toggle on card
insertion. Check that VDD supply is around 3.3v.
In relation to the DS1302, I'm using the supercap, which I'm sure is
installed correctly - though I'll confirm that this evening. However I'm
afraid you've lost me with regard to the 2+2, 2+4 information.
"2+2" would be a complete reply to the prompt regarding the "Trickle
charge ..." Note: the response is prompted as "d+r" for *d*iodes & *r*
esistor.
Also, make sure there is no solder bridge from the square pin of SuperCap
mounting pads to VBAT. I doubt that this is the trouble.
I think the NVRAM/battery issue is the root cause of the other issues as
on reboot I don't get the Prop IO initialised by the Mark IV and the full
setup routine runs again. I think sorting the supercap issue may rectify
those problems.
Are you seeing the SDcard LED flash on cold startup? Are you using the
correct board I/O address: 0x40 (off, on, on, on) for the P10 jumpers?
UART output depends only on the internal Z180 UART, which is unaffected by
the I/O address jumper setting.
Which board revision do you have? 1.00 or 2.0. Rev. 1 board needs a
wiring update, but it would have nothing to do with your problems.
--John

John Snowdon

2015-09-30 12:02:32 UTC

Permalink

Well, silly me.

It turns out I had the wire bridges for the SD card on the wrong pins. I
had the card detect link on 5+6 when it of course should have been 3+4.
Still having bother tracking down the right pads for write-protect (11+12
seemingly don't work for me).

Still, the SD card on the Mark IV now detects a card - I've partitioned a
2G card with a 500MB 0x32 partition and the rest as FAT32. Will's UNA CPM
seems to find it okay:

<Loading Image...

>

<Loading Image...

>

If anyone is wondering why I'm using Windows 95 Hyperterminal on an ancient
Thinkpad, well, it's because my cheap FDTI clone USB to serial adaptor
burned itself out the other day! Literally! It had a black patch on the
casing and very, very hot! I think I'll go for a brand name one to replace
it!

Next challenge is to try and find an old IDE cable to test the Mark IV IDE
functionality via a small (64MB) Compact Flash card.

Once I've tested that, I'll move on to why the Prop IO doesn't appear to be
working once it is initialised.

G. Beat

2015-09-30 14:51:39 UTC

Permalink

INTEK22 (eBay seller) based in Plano, TX is selling excess/surplus inventory.
I made an offer (accepted) for a couple of those FTDI factory cables.
New from Mouser, they are $30.00 and he is selling for 1/2 that price, with free shipping.

gb

John Snowdon

2015-10-01 21:16:24 UTC

Permalink

Hi again,

Now that the Mark IV seems to be running okay, I thought I'd take another
look at why the Prop IO (v2) doesn't appear to be working as intended.

On power-on, I've ran the setup routine and enabled the Prop IO display and
keyboard (I don't get a prompt for asking for Prop IO SD support, is that
okay?) in addition to ROM, RAM and SD and IDE master. When I reboot I get
the normal UNA BIOS messages about CPU type and speed, detecting storage
devices (RAM, ROM, SD on the Mark IV) and then a line that states:

"Configuring the Prop IO board..."

Then a blank line, then the usual prompt to choose a boot device. Looking
at the ROM code distributed with Will's UNA cpm 2014-08-01 release, I can
see that it might say some further information on detecting the board at
that point:

00003c90 77 18 00 2e 00 dd e1 c9 43 6f 6e 66 69 67 75 72 |w.......Configur|
00003ca0 69 6e 67 20 74 68 65 20 25 73 20 62 6f 61 72 64 |ing the %s board|
00003cb0 3a 0a 00 25 73 53 63 72 65 65 6e 20 69 73 20 38 |:..%sScreen is 8|
00003cc0 30 78 33 35 0a 25 73 4b 65 79 62 6f 61 72 64 20 |0x35.%sKeyboard |
00003cd0 69 73 20 54 54 59 0a 00 55 73 65 20 74 68 65 20 |is TTY..Use the |
00003ce0 50 72 6f 70 49 2f 4f 20 73 63 72 65 65 6e 00 55 |PropI/O screen.U|
00003cf0 73 65 20 74 68 65 20 50 72 6f 70 49 2f 4f 20 6b |se the PropI/O k|
00003d00 65 79 62 6f 61 72 64 00 dd e5 dd 21 00 00 dd 39 |eyboard....!...9|

I'm having trouble tracking down where this code is in the source for UNA
to try and work out why I'm not seeing anything more, but my grep-foo fails
me; I can only find it in the 512K-ROM.bin file.

Anyway, regardless of that, it appears that my board isn't getting detected
and prints nothing more about the Prop IO. I've verified that I have the
three jumpers on the board set to:

JP1: 1-2
JP2: 1-2
P6: All jumpers on except 0x40

The Propeller eeprom is loaded and VGA output works, as my initial post
shows. I've tried changing to a different PropIO address, I've also tried
changing the memory and IO wait states of the Mark IV to see if that makes
a difference (I'm running 1 memory and 3 IO wait states, but backed them
both right off to the maximum to see if things were too fast). I am running
an 18Mhz crystal at x2 in setup, so 36MHz, could that be too fast to talk
over the bus?

I suppose it's possible I've built the backplane incorrectly, but there
doesn't appear to be anything I could have done wrong there as its all
passive components.

Any ideas folks?

John

John Coffman

2015-10-02 04:41:43 UTC

Permalink

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
John, 
 
After a lot of looking at files on the Wiki, you may ignore my
earlier conclusion that the ROM image on the Wiki was corrupted. 
However, there may be some confusion over what image you are using. 
A lot of history is retained in archival sub-folders. 
 
The zip file on the Wiki you should be using is
"unacpm-2014-08-01.zip".  The file inside is "512K-ROM.bin". 
 
The actual file is under "N8VEM Firmware and Registry", but it is
linked to from one of the pages under "ECB Processor Boards / ECB
SBC Mark IV". 
 
This release file is the one to use.  It DOES correctly find the
Propeller I/O v2 board, as long as the device code for the board is
0xA8.  With this ROM, you CANNOT configure the PropI/O--if the board
is found, the configuration is already done by the time you get to
that question in Setup.  YOU CANNOT CONFIGURE THE BOARD FROM SETUP. 
 
At first startup, if the PropI/O is plugged in, the SD card LED will
flash until the PropI/O is recognized.  All output will go to the
PropI/O; there will be no serial output.  If you get serial output,
then the PropI/O is not being recognized as being present. 
 
--John 
 
 
 
 
 
On 10/01/2015 02:16 PM, John Snowdon wrote:
<blockquote
cite="mid:1bf41ae1-12cc-4b2f-9081-***@googlegroups.com"
type="cite">
<div dir="ltr">Hi again,
<div> 
</div>
<div>Now that the Mark IV seems to be running okay, I thought
I'd take another look at why the Prop IO (v2) doesn't appear
to be working as intended.</div>
<div> 
</div>
<div>On power-on, I've ran the setup routine and enabled the
Prop IO display and keyboard (I don't get a prompt for asking
for Prop IO SD support, is that okay?) in addition to ROM, RAM
and SD and IDE master. When I reboot I get the normal UNA BIOS
messages about CPU type and speed, detecting storage devices
(RAM, ROM, SD on the Mark IV) and then a line that states:</div>
<div> 
</div>
<div>"Configuring the Prop IO board..."</div>
<div> 
</div>
<div>Then a blank line, then the usual prompt to choose a boot
device. Looking at the ROM code distributed with Will's Â UNA
cpm 2014-08-01 release, I can see that it might say some
further information on detecting the board at that point:</div>
<div> 
<div class="prettyprint" style="border: 1px solid rgb(187,
187, 187); word-wrap: break-word; background-color: rgb(250,
250, 250);"><code class="prettyprint"></code>
<div class="subprettyprint">00003c90
77 18 00 2e 00 dd e1 c9 43 6f 6e 66 69 67 75 72 |w.......Configur| 
00003ca0
69 6e 67 20 74 68 65 20 25 73 20 62 6f 61 72 64 |ing the %s board| 
00003cb0
3a 0a 00 25 73 53 63 72 65 65 6e 20 69 73 20 38 |:..%sScreen
is 8| 
00003cc0
30 78 33 35 0a 25 73 4b 65 79 62 6f 61 72 64 20 |0x35.%sKeyboard | 
00003cd0
69 73 20 54 54 59 0a 00 55 73 65 20 74 68 65 20 |is
TTY..Use
the | 
00003ce0
50 72 6f 70 49 2f 4f 20 73 63 72 65 65 6e 00 55 |PropI/O screen.U| 
00003cf0
73 65 20 74 68 65 20 50 72 6f 70 49 2f 4f 20 6b |se the PropI/O k| 
00003d00
65 79 62 6f 61 72 64 00 dd e5 dd 21 00 00 dd 39 |eyboard....!...9| 
 
</div>
</div>
 
I'm having trouble tracking down where this code is in the
source for UNA to try and work out why I'm not seeing anything
more, but my grep-foo fails me; I can only find it in the
512K-ROM.bin file.</div>
<div> 
</div>
<div>Anyway, regardless of that, it appears that my board isn't
getting detected and prints nothing more about the Prop IO.
I've verified that I have the three jumpers on the board set
to:</div>
<div> 
</div>
<div>JP1: 1-2</div>
<div>JP2: 1-2</div>
<div>P6: All jumpers on except 0x40</div>
<div> 
</div>
<div>The Propeller eeprom is loaded and VGA output works, as my
initial post shows. I've tried changing to a different PropIO
address, I've also tried changing the memory and IO wait
states of the Mark IV to see if that makes a difference (I'm
running 1 memory and 3 IO wait states, but backed them both
right off to the maximum to see if things were too fast). I am
running an 18Mhz crystal at x2 in setup, so 36MHz, could that
be too fast to talk over the bus?</div>
<div> 
</div>
<div>I suppose it's possible I've built the backplane
incorrectly, but there doesn't appear to be anything I could
have done wrong there as its all passive components.</div>
<div> 
</div>
<div>Any ideas folks?</div>
<div> 
</div>
<div>John</div>
</div>
</blockquote>
 
</body>
</html>



-- 
You received this message because you are subscribed to the Google Groups "N8VEM" group. 
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:n8vem+***@googlegroups.com">n8vem+***@googlegroups.com</a>. 
To post to this group, send email to <a href="mailto:***@googlegroups.com">***@googlegroups.com</a>. 
Visit this group at <a href="http://groups.google.com/group/n8vem">http://groups.google.com/group/n8vem</a>. 
For more options, visit <a href="https://groups.google.com/d/optout">https://groups.google.com/d/optout</a>.

John Snowdon

2015-10-02 07:59:41 UTC

Permalink

Hi John,

The release I've been using is indeed the unacpm-2014-08-01.zip (and the
included 512K-ROM.bin file), linked to from here:

http://n8vem-sbc.pbworks.com/w/page/90328904/BIOS%20releases%20--%20Mark%20IV

That's the file I programmed to the flash, but your comment about not being
able to configure the board from setup has now got me doubting what I've
done - the setup routine when I run it definitely asks if I want to enable
PropIO screen and keyboard (two seperate questions). It sounds like I've
got the wrong one in that case.

Is it the Mark IV SD card led that is supposed to flash while detecting the
PropIO, or the one on the PropIO board itself?

John,
After a lot of looking at files on the Wiki, you may ignore my earlier
conclusion that the ROM image on the Wiki was corrupted. However, there
may be some confusion over what image you are using. A lot of history is
retained in archival sub-folders.
The zip file on the Wiki you should be using is "unacpm-2014-08-01.zip".
The file inside is "512K-ROM.bin".
The actual file is under "N8VEM Firmware and Registry", but it is linked
to from one of the pages under "ECB Processor Boards / ECB SBC Mark IV".
This release file is the one to use. It DOES correctly find the Propeller
I/O v2 board, as long as the device code for the board is 0xA8. With this
ROM, you CANNOT configure the PropI/O--if the board is found, the
configuration is already done by the time you get to that question in
Setup. YOU CANNOT CONFIGURE THE BOARD FROM SETUP.
At first startup, if the PropI/O is plugged in, the SD card LED will flash
until the PropI/O is recognized. All output will go to the PropI/O; there
will be no serial output. If you get serial output, then the PropI/O is
not being recognized as being present.
--John
Hi again,
Now that the Mark IV seems to be running okay, I thought I'd take another
look at why the Prop IO (v2) doesn't appear to be working as intended.
On power-on, I've ran the setup routine and enabled the Prop IO display
and keyboard (I don't get a prompt for asking for Prop IO SD support, is
that okay?) in addition to ROM, RAM and SD and IDE master. When I reboot I
get the normal UNA BIOS messages about CPU type and speed, detecting
"Configuring the Prop IO board..."
Then a blank line, then the usual prompt to choose a boot device. Looking
at the ROM code distributed with Will's Ã UNA cpm 2014-08-01 release, I can
see that it might say some further information on detecting the board at
00003c90 77 18 00 2e 00 dd e1 c9 43 6f 6e 66 69 67 75 72 |w.......Configur
|
00003ca0 69 6e 67 20 74 68 65 20 25 73 20 62 6f 61 72 64 |ing the %s board
|
00003cb0 3a 0a 00 25 73 53 63 72 65 65 6e 20 69 73 20 38 |:..%sScreen is 8
|
00003cc0 30 78 33 35 0a 25 73 4b 65 79 62 6f 61 72 64 20 |0x35.%sKeyboard |
00003cd0 69 73 20 54 54 59 0a 00 55 73 65 20 74 68 65 20 |is TTY..Use the |
00003ce0 50 72 6f 70 49 2f 4f 20 73 63 72 65 65 6e 00 55 |PropI/O screen.U
|
00003cf0 73 65 20 74 68 65 20 50 72 6f 70 49 2f 4f 20 6b |se the PropI/O k
|
00003d00 65 79 62 6f 61 72 64 00 dd e5 dd 21 00 00 dd 39 |eyboard....!...9
|
I'm having trouble tracking down where this code is in the source for UNA
to try and work out why I'm not seeing anything more, but my grep-foo fails
me; I can only find it in the 512K-ROM.bin file.
Anyway, regardless of that, it appears that my board isn't getting
detected and prints nothing more about the Prop IO. I've verified that I
JP1: 1-2
JP2: 1-2
P6: All jumpers on except 0x40
The Propeller eeprom is loaded and VGA output works, as my initial post
shows. I've tried changing to a different PropIO address, I've also tried
changing the memory and IO wait states of the Mark IV to see if that makes
a difference (I'm running 1 memory and 3 IO wait states, but backed them
both right off to the maximum to see if things were too fast). I am running
an 18Mhz crystal at x2 in setup, so 36MHz, could that be too fast to talk
over the bus?
I suppose it's possible I've built the backplane incorrectly, but there
doesn't appear to be anything I could have done wrong there as its all
passive components.
Any ideas folks?
John

John Snowdon

2015-10-02 21:08:07 UTC

Permalink

With a great deal of help from John we managed to track down the cause of
the PropIO not being detected. It was, of course, the board IO being set
incorrectly. Set to 0xa8 it now works:

<Loading Image...

>

One thing I have noticed however, is that there appears to be some signal
degradation or transmission errors, only when I use slot 1 and slot 8 on
the ECB backplane (either way round); I get a garbled display of the
character display (though not of the PropIO rom text itself, obviously):

<Loading Image...

>

It's only these outer-most sockets that seemingly cause a problem, if I
move either card towards the other by just one slot, then the issue goes
away.

One more thing, I still don't seem to get the SD card detected when
inserted into the PropIO. I had some minor soldering issues on the Mark IV,
so I'll take a closer look at the socket again; it could just be a bad
contact.

Many thanks to everyone who has helped so far!

Regards,
John

Post by Wayne Warthen
Hi John,
The release I've been using is indeed the unacpm-2014-08-01.zip (and the
http://n8vem-sbc.pbworks.com/w/page/90328904/BIOS%20releases%20--%20Mark%20IV
That's the file I programmed to the flash, but your comment about not
being able to configure the board from setup has now got me doubting what
I've done - the setup routine when I run it definitely asks if I want to
enable PropIO screen and keyboard (two seperate questions). It sounds like
I've got the wrong one in that case.
Is it the Mark IV SD card led that is supposed to flash while detecting
the PropIO, or the one on the PropIO board itself?

John,
After a lot of looking at files on the Wiki, you may ignore my earlier
conclusion that the ROM image on the Wiki was corrupted. However, there
may be some confusion over what image you are using. A lot of history is
retained in archival sub-folders.
The zip file on the Wiki you should be using is "unacpm-2014-08-01.zip".
The file inside is "512K-ROM.bin".
The actual file is under "N8VEM Firmware and Registry", but it is linked
to from one of the pages under "ECB Processor Boards / ECB SBC Mark IV".
This release file is the one to use. It DOES correctly find the
Propeller I/O v2 board, as long as the device code for the board is 0xA8.
With this ROM, you CANNOT configure the PropI/O--if the board is found, the
configuration is already done by the time you get to that question in
Setup. YOU CANNOT CONFIGURE THE BOARD FROM SETUP.
At first startup, if the PropI/O is plugged in, the SD card LED will
flash until the PropI/O is recognized. All output will go to the PropI/O;
there will be no serial output. If you get serial output, then the PropI/O
is not being recognized as being present.
--John
Hi again,
Now that the Mark IV seems to be running okay, I thought I'd take another
look at why the Prop IO (v2) doesn't appear to be working as intended.
On power-on, I've ran the setup routine and enabled the Prop IO display
and keyboard (I don't get a prompt for asking for Prop IO SD support, is
that okay?) in addition to ROM, RAM and SD and IDE master. When I reboot I
get the normal UNA BIOS messages about CPU type and speed, detecting
"Configuring the Prop IO board..."
Then a blank line, then the usual prompt to choose a boot device. Looking
at the ROM code distributed with Will's Ã UNA cpm 2014-08-01 release, I can
see that it might say some further information on detecting the board at
00003c90 77 18 00 2e 00 dd e1 c9 43 6f 6e 66 69 67 75 72 |w.......
Configur|
00003ca0 69 6e 67 20 74 68 65 20 25 73 20 62 6f 61 72 64 |ing the %s board|
00003cb0 3a 0a 00 25 73 53 63 72 65 65 6e 20 69 73 20 38 |:..%sScreen is 8|
00003cc0 30 78 33 35 0a 25 73 4b 65 79 62 6f 61 72 64 20 |0x35.%sKeyboard |
00003cd0 69 73 20 54 54 59 0a 00 55 73 65 20 74 68 65 20 |is TTY..Use the |
00003ce0 50 72 6f 70 49 2f 4f 20 73 63 72 65 65 6e 00 55 |PropI/O screen.
U|
00003cf0 73 65 20 74 68 65 20 50 72 6f 70 49 2f 4f 20 6b |se the PropI/O k|
00003d00 65 79 62 6f 61 72 64 00 dd e5 dd 21 00 00 dd 39 |eyboard....!...
9|
I'm having trouble tracking down where this code is in the source for UNA
to try and work out why I'm not seeing anything more, but my grep-foo fails
me; I can only find it in the 512K-ROM.bin file.
Anyway, regardless of that, it appears that my board isn't getting
detected and prints nothing more about the Prop IO. I've verified that I
JP1: 1-2
JP2: 1-2
P6: All jumpers on except 0x40
The Propeller eeprom is loaded and VGA output works, as my initial post
shows. I've tried changing to a different PropIO address, I've also tried
changing the memory and IO wait states of the Mark IV to see if that makes
a difference (I'm running 1 memory and 3 IO wait states, but backed them
both right off to the maximum to see if things were too fast). I am running
an 18Mhz crystal at x2 in setup, so 36MHz, could that be too fast to talk
over the bus?
I suppose it's possible I've built the backplane incorrectly, but there
doesn't appear to be anything I could have done wrong there as its all
passive components.
Any ideas folks?
John

John Coffman

2015-10-02 22:05:24 UTC

Permalink

John Snowdon

2015-10-02 23:02:27 UTC

Permalink

Haha, I'll stop puzzling over that then! :-D

Post by John Snowdon
One more thing, I still don't seem to get the SD card detected when
inserted into the PropIO. I had some minor soldering issues on the Mark IV,
so I'll take a closer look at the socket again; it could just be a bad
contact.
UNA does not support the SDcard socket on the PropI/O. The socket is not
wired to return either Card Detect or Write Protect.
--John
Many thanks to everyone who has helped so far!
Regards,
John

John Coffman

2015-10-02 22:16:09 UTC

Permalink

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
On 10/02/2015 02:08 PM, John Snowdon wrote:
<blockquote
cite="mid:525afe74-3053-401d-b211-***@googlegroups.com"
type="cite">
<div>One thing I have noticed however, is that there appears to be
some signal degradation or transmission errors, only when I use
slot 1 and slot 8 on the ECB backplane (either way round); I get
a garbled display of the character display (though not of the
PropIO rom text itself, obviously):</div>
<div> 
</div>
</blockquote>
John, 
 
This would be the first example of signal reflection on any N8
backplane, including the 12-slot backplanes.  I have only noticed
slot sensitivity on my S-100 backplane, never on any of the N8
backplanes.  I have slot-sensitivity on the S-100 at 6Mhz.  AFAIK,
the N8 backplanes are good up to about 16Mhz.  The new,
experimental, KISS-68030 CPU board can access Memory on a 4MEM board
at 16Mhz and 1 w.s.  This is making memory references across the
backplane.  The SBC-188 uses 4MEM memory in a similar fashion up to
16Mhz. 
 
You probably noticed that @ 36Mhz, you need to run the memory at 1
wait state.  Does the number of I/O wait states affect the garbled
characters you are seeing?  At 36Mhz, I always run at 4 I/O wait
states.  I/O speed is not the issue; error-free operation is. 
 
--John 
 
 
</body>
</html>



-- 
You received this message because you are subscribed to the Google Groups "N8VEM" group. 
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:n8vem+***@googlegroups.com">n8vem+***@googlegroups.com</a>. 
To post to this group, send email to <a href="mailto:***@googlegroups.com">***@googlegroups.com</a>. 
Visit this group at <a href="http://groups.google.com/group/n8vem">http://groups.google.com/group/n8vem</a>. 
For more options, visit <a href="https://groups.google.com/d/optout">https://groups.google.com/d/optout</a>.

John Snowdon

2015-10-02 22:59:31 UTC

Permalink

I've tried 3 and 4 IO wait states and it doesn't appear to make any
difference - I still get garbled text if the two boards are at the extreme
ends of the backplane (slot 1 and 8), but not if they're in slots 1 and 7,
or 2 and 8. Could it be the quality of components I'm using?

It's not a major issue for me to be honest - I can't see myself needing to
use all 8 slots; perhaps another a couple, but not all the other 6.

John Coffman

2015-10-03 01:51:34 UTC

Permalink

John Snowdon

2015-10-03 08:51:35 UTC

Permalink

Yes, all LS series logic devices - I followed the bom on the site for the
Mark IV, ECB and PropIO v2, the only deviations may be some minor variance
in caps (perhaps some ceramics in place of metal film where I wasn't able
to source them as easily), and leds (which I had a big stock of already, so
just used them), I also used MAX-series line drivers as recommended on the
list as pin for pin upgrades over the RS422 drivers as listed.

It's not a problem, I can easily work around it as long as I'm aware of the
limitations. I've impressed myself with what I've been able to do so far -
it's beyond anything I've ever done before (a couple of DIY power amp kits
and a little bit of 8/16bit computer/video game repair work here and there)
so the fact that there seem to be a few little niggles is not something
that really bothers me. The fact that a computer I soldered together myself
(mostly) works without issue is more than impressive enough for me :)

Post by John Snowdon
I've tried 3 and 4 IO wait states and it doesn't appear to make any
difference - I still get garbled text if the two boards are at the extreme
ends of the backplane (slot 1 and 8), but not if they're in slots 1 and 7,
or 2 and 8. Could it be the quality of components I'm using?
It's not a major issue for me to be honest - I can't see myself needing to
use all 8 slots; perhaps another a couple, but not all the other 6.
John,
(Backplane8 167mm [new] version). slots 1 & 8
I/O 2 w.s. -- prefectly
good output
Are you using LS components for your I/O receivers. LS240, 241, 244, 245
have hysteresis on the inputs. Other logic families (notably HCT) do not.
This might make a difference.
In the last 6 years I have heard of no one with slot sensitivity on the N8
ECB bus.
--John

Wayne Warthen

2015-09-03 23:23:12 UTC

Permalink

Hi John,

The activity LED on the PropIO V2 will not do anything until it receives
commands from another card (in your case, the Mark IV). So, everything is
normal so far.

Thanks,

Wayne

Post by John Snowdon
I'm almost finished putting my 8 slot backplane, PropIO v2 and Z180 Mark
IV system together.
All the boards are built, and today I've just uploaded the code to the
Parallax cpu on the PropIO board. So far I'm still waiting on finding a
method to write the AM29F040B 512k flash on the Mark IV (my old parallel
Willem board just won't work reliably in any of my modern systems, so I may
try and pick up a TL866 programmer so that I can do this sort of stuff
under Linux), so I don't have a finished system yet.
<https://lh3.googleusercontent.com/-0m7SJ_iRsxg/VehIHwhC02I/AAAAAAAAAD8/x-8zGaULT1A/s1600/P1050446-.JPG>
<https://lh3.googleusercontent.com/-yks4ADa7BfU/VehIEIxufYI/AAAAAAAAAD0/6NHg_djh8Ao/s1600/P1050445-.JPG>
<https://lh3.googleusercontent.com/-CHQLjUT6Jlk/VehIPg3RwPI/AAAAAAAAAEE/4VXiros7YfA/s1600/P1050449-.JPG>
Anyway, after a little bit of work trying to sort out the null modem
wiring between the PropIO serial port and a cheap USB to serial adaptor I
had some success with the Parallax SimpleIDE tool to detect the
microcontroller, but I couldn't find a tool to send the code to the board
(the parallax tool under WINE doesn't work). I found 'bstl' instead, a
native Linux Parallax utility which does the same job as the Propeller Tool
does in Windows.
I've ran the tool, uploaded the code and then reset the board. It beeps
and then after a second I can now see the startup messages and version
number across the bottom of the screen.
<https://lh3.googleusercontent.com/-AUfHW0ltqro/VehIThdOUoI/AAAAAAAAAEM/SvTk3Ln7Zcw/s1600/P1050450-.JPG>
<https://lh3.googleusercontent.com/-9H88yWWmitI/VehIWUFGWeI/AAAAAAAAAEU/NlnhS6G094A/s1600/P1050451-.JPG>
One thing I haven't seen, however, is anything on the activity led. The 5v
and 3.3v leds are okay, the speaker beeps and vga output works, but nothing
on the activity led. It's not a major issue, it could be a broken led as
they just came out of my bits drawer; so I'm not sure how old they are or
what the specs were.
John

John Snowdon

2015-09-04 07:37:26 UTC

Permalink

Ah, my misunderstanding. I thought this was an activity indicator for
commands or code being sent to the Propeller; so I was waiting for it to
flash while uploading the .spin file :-)

Post by Wayne Warthen
Hi John,
The activity LED on the PropIO V2 will not do anything until it receives
commands from another card (in your case, the Mark IV). So, everything is
normal so far.
Thanks,
Wayne

Post by John Snowdon
I'm almost finished putting my 8 slot backplane, PropIO v2 and Z180 Mark
IV system together.
All the boards are built, and today I've just uploaded the code to the
Parallax cpu on the PropIO board. So far I'm still waiting on finding a
method to write the AM29F040B 512k flash on the Mark IV (my old parallel
Willem board just won't work reliably in any of my modern systems, so I may
try and pick up a TL866 programmer so that I can do this sort of stuff
under Linux), so I don't have a finished system yet.
<https://lh3.googleusercontent.com/-0m7SJ_iRsxg/VehIHwhC02I/AAAAAAAAAD8/x-8zGaULT1A/s1600/P1050446-.JPG>
<https://lh3.googleusercontent.com/-yks4ADa7BfU/VehIEIxufYI/AAAAAAAAAD0/6NHg_djh8Ao/s1600/P1050445-.JPG>
<https://lh3.googleusercontent.com/-CHQLjUT6Jlk/VehIPg3RwPI/AAAAAAAAAEE/4VXiros7YfA/s1600/P1050449-.JPG>
Anyway, after a little bit of work trying to sort out the null modem
wiring between the PropIO serial port and a cheap USB to serial adaptor I
had some success with the Parallax SimpleIDE tool to detect the
microcontroller, but I couldn't find a tool to send the code to the board
(the parallax tool under WINE doesn't work). I found 'bstl' instead, a
native Linux Parallax utility which does the same job as the Propeller Tool
does in Windows.
I've ran the tool, uploaded the code and then reset the board. It beeps
and then after a second I can now see the startup messages and version
number across the bottom of the screen.
<https://lh3.googleusercontent.com/-AUfHW0ltqro/VehIThdOUoI/AAAAAAAAAEM/SvTk3Ln7Zcw/s1600/P1050450-.JPG>
<https://lh3.googleusercontent.com/-9H88yWWmitI/VehIWUFGWeI/AAAAAAAAAEU/NlnhS6G094A/s1600/P1050451-.JPG>
One thing I haven't seen, however, is anything on the activity led. The
5v and 3.3v leds are okay, the speaker beeps and vga output works, but
nothing on the activity led. It's not a major issue, it could be a broken
led as they just came out of my bits drawer; so I'm not sure how old they
are or what the specs were.
John