Building Rolands new Atom design...

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 8:25 am

Early this morning I was reading on your new Atom site section modifications and then I noticed something.

I did put the texts on that page on top of this photo... (see picture)

Of course I had made my Atom identical to yours on the photo's... :D
Attachments
SD Card holder.PNG

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 10:53 am

Hi Wim,

You have made the mod in the right way. The pin numbers of the SD card socket are just not logical. The right-most pin is number 9, then comes 1, 2, 3 etc.
Schermafbeelding 2015-02-07 om 11.48.35.png
I'll add that to my documentation too.

Greetz,
Roland
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

PhilYoung
Posts: 203
Joined: Sun Jun 12, 2011 4:55 pm
Contact:

Re: Building Rolands new Atom design...

Post by PhilYoung » Sat Feb 07, 2015 11:01 am

Hi Roland
roland wrote:
Before you fit the CPLD's, connect a 5V power supply and check if the output of the LM3940 is about 3.3V. If it's to high for whatever reason, the CPLD's will get destroyed. The biggest problem is to get them out of their socket, that might get damaged and replacing the socket will be a hell of a job.
I can confirm that removing a PLCC socket is a beast of a job (without a proper de-soldering station that is). Probably best to check it isn't 180 degrees out of alignment in the first place I would advise.

No harm done, except to my pride.

Cheers,

Phil Young

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 11:12 am

I think/hope I found the problem with the MMC. :D

This morning I bought a new SD card the one I did read about in a other topic, ATOM FPGA.

Roland told he uses this type in his new Atom design, so I bought the exact same so I could eliminate this cause.

download/file.php?id=14983&mode=view


It also didn't work... :(

Then I suddenly remembered me that when I was building the PCB a few weeks ago.

They had given me some wrong capacitors and resistors so I had to exchange them.

See picture below... :shock:

So today I am going to get me two 1.8K resistors and hoping this will solve the problem... :)

Ps. I have already exchanged R29 for 1.8K, this one was 18K too... :(
Attachments
1.8K vs 18K.PNG

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 11:18 am

roland wrote:Hi Wim,

You have made the mod in the right way. The pin numbers of the SD card socket are just not logical. The right-most pin is number 9, then comes 1, 2, 3 etc.
Schermafbeelding 2015-02-07 om 11.48.35.png
I'll add that to my documentation too.

Greetz,
Roland
I thought about that, but I better ask for making sure... :)

User avatar
hoglet
Posts: 8901
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: Building Rolands new Atom design...

Post by hoglet » Sat Feb 07, 2015 11:29 am

Wim,
multiwizard wrote:ISo today I am going to get me two 1.8K resistors and hoping this will solve the problem... :)
The wrong value resistors here will certainly cause problems.

Roland's values are carefully chosen to drop 5V PIC output signals to 3.3v levels, to be compatible with the 3.3v card.

With a 18K instead of 1.8K, the signals will be reduced to about 0.8v which is way too low to work.

So, hurry down to the electronics shop to buy these!!!

Dave

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 11:46 am

Damn, I asked for a details picture, I get one and I don't see you have the wrong resistors :oops:

It could be a problem:

The resistors are mend to reduce the voltage at the inputs of the SD card. That should be about 3.3V.

With the resistors 1k8 and 3k3 you get at the input (3300 / 5100) * 5V = 3.3V
If you use 18k and 3k3, the maximum input voltage is (3300/21300 ) = 0.77V

In real life the output voltage will be a bit less than 5V so the maximum input voltage will be lower than 0.77V
Schermafbeelding 2015-02-07 om 12.42.29.png
Source: SD Card Projects Using the PIC Microcontroller
So probably your 18F4525 will never see a logic '1' from the SD Card.....
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 11:48 am

hoglet wrote: So, hurry down to the electronics shop to buy these!!!
Ah Dave, you gave the answer while I was writing an explanation for Wim so he understands why it goes wrong.

I guess he will be at the electronics shop by now :lol:
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 12:28 pm

roland wrote:
hoglet wrote: So, hurry down to the electronics shop to buy these!!!
Ah Dave, you gave the answer while I was writing an explanation for Wim so he understands why it goes wrong.

I guess he will be at the electronics shop by now :lol:
viewtopic.php?f=41&t=5594&start=2160#p104957 :lol:

Look much much better now... :D
Attachments
P1090075.JPG

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 1:01 pm

Yesss,,, \:D/ it all/most works ok now,,, \:D/

It is having 8-bit in HD quality... :D

For me it is a fully qualified Atom... :D

I can really recommend this Super Atom Computer to everybody who would like to have a Atom. :idea:

For now I like to thank everyone so far for helping me solving any problems, lots of thanks guys... =D>


Roland do you have by any chance the pin out of the Joystick (P1003)


Kind Regards, Wim... :-)
Attachments
P1090085.JPG
P1090092.JPG
Last edited by Multiwizard on Sat Feb 07, 2015 4:23 pm, edited 1 time in total.

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 1:12 pm

Hi Wim,

I'm glad you solved it =D> And do you still think that you don't know much about (digital) electronics? By working with it, you'll learn a lot :lol:

The pinout is in the schematic diagram. That's all I have. I never connected a joystick to it.

Have fun with your Atom!

Greetz,
Roland
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
1024MAK
Posts: 9698
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Building Rolands new Atom design...

Post by 1024MAK » Sat Feb 07, 2015 1:36 pm

@ Wim - =D> =D> =D>

:D :D :D

Mark

User avatar
oss003
Posts: 3178
Joined: Tue Jul 14, 2009 11:57 am
Location: Netherlands
Contact:

Re: Building Rolands new Atom design...

Post by oss003 » Sat Feb 07, 2015 1:54 pm

Ok Wim =D> =D> =D> =D>

Nice job, now I can start with my Atom ..... :lol:

Greetings
Kees

User avatar
hoglet
Posts: 8901
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: Building Rolands new Atom design...

Post by hoglet » Sat Feb 07, 2015 3:37 pm

Wim,

Great that you now have the AtoMMC interface working.
multiwizard wrote: Indeed, the blue lines are gone... :D
The jumper to disable the SID is temporary work around. But at least is proves the cause. I'm wondering why Roland doesn't also have a problem.

Did I read that you have used a screened cable, and that made no difference?

How exactly did you connect the cable shield on the GODIL end?

Could you post a close up picture of this? Looking at the pictures you posted, I'm starting to suspect a wiring error, and that you have the blue signal running down the screen of the SID Cable. :shock: I think it should go down the purple wire.

Could you also try removing the jumper, so the lines come back, and then try completely disconnecting the SID cable from the GODIL. What happens to the blue lines then? Have you lost the blue signal completely if you do this?

Dave

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 5:43 pm

hoglet wrote: The jumper to disable the SID is temporary work around. But at least is proves the cause. I'm wondering why Roland doesn't also have a problem.
Well, the Atom has been designed for chaos. And I kept that design in my Atom. So the wires from Godil to the connector are just like the keyboard wires all over the Atom, being kept at their place by the capacitors of the RS232 level converters :lol:

Wim has all wires arranged in a way that makes me drool. But his audio cable is not shielded like mine. So I send him a piece of thin, shielded coax cable for the audio signal.
I guess that will make his blue lines disappear.

To be honest, I have some thin blue lines also but with a bit of tweaking my monitor I can make them disappear.
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
hoglet
Posts: 8901
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: Building Rolands new Atom design...

Post by hoglet » Sat Feb 07, 2015 5:54 pm

If you look at the picture in this post, and read the words carefully (which I obviously did not do), the it looks the the screen of the SID cable is connected to E7, which is the blue signal:
http://www.stardot.org.uk/forums/viewto ... 30#p104736

I think you need to try the following:
- Connect E7 back up to the purple wire in the ribbon cable at both ends
- Somehow connect the screen of the screened cable to ground, which is E1, at the GODIL end. Ideally also connect it to ground at the other end. Roland, how did you do this exactly?

Dave

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 6:04 pm

hoglet wrote:Wim,

Great that you now have the AtoMMC interface working.
multiwizard wrote: Indeed, the blue lines are gone... :D
Could you post a close up picture of this? Looking at the pictures you posted, I'm starting to suspect a wiring error, and that you have the blue signal running down the screen of the SID Cable. :shock: I think it should go down the purple wire.

Could you also try removing the jumper, so the lines come back, and then try completely disconnecting the SID cable from the GODIL. What happens to the blue lines then? Have you lost the blue signal completely if you do this?

Dave
Hi Dave & Roland, this is the only picture I have of it...
Attachments
GODIL Closeup.PNG

User avatar
hoglet
Posts: 8901
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: Building Rolands new Atom design...

Post by hoglet » Sat Feb 07, 2015 6:07 pm

If the screen of the SID cable is connected to E7, rather than ground, that's definitely incorrect, and will be making the blue lines worse.

Dave

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 6:23 pm

roland wrote:But his audio cable is not shielded like mine. So I send him a piece of thin, shielded coax cable for the audio signal.
I guess that will make his blue lines disappear.

To be honest, I have some thin blue lines also but with a bit of tweaking my monitor I can make them disappear.
Hi Roland, can you upload me a close-up picture of your GODIL wiring? :D

User avatar
oss003
Posts: 3178
Joined: Tue Jul 14, 2009 11:57 am
Location: Netherlands
Contact:

Re: Building Rolands new Atom design...

Post by oss003 » Sat Feb 07, 2015 6:54 pm

Hi Wim,

here is a close up of the wiring: download/file.php?id=14369&mode=view

E8 = SID output
E1 = SID GND/shield

Hoglets description of making up the audio/video cables can be found at: https://github.com/hoglet67/AtomGodilVi ... dio-cables

Greetings
Kees

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 7:08 pm

Correct, it's not very clear but if you take a closer look you can see that the shield of the white cable goes to pin E1 below the video cables.
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 7:45 pm

Dave, Kees, Roland,

it did do the trick, Roland, not even a thin blue line... \:D/

Thanks again you all... :D
Attachments
P1090119.JPG

User avatar
oss003
Posts: 3178
Joined: Tue Jul 14, 2009 11:57 am
Location: Netherlands
Contact:

Re: Building Rolands new Atom design...

Post by oss003 » Sat Feb 07, 2015 7:50 pm

Wim, is the SID also working now?

User avatar
1024MAK
Posts: 9698
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Building Rolands new Atom design...

Post by 1024MAK » Sat Feb 07, 2015 7:55 pm

Err, unless a screen / shield is also being used as the only return / ground / 0V line, it is normally best practice to only connect the screen / shield at one end. This is normally the source end.

For VGA signals, normally each video signal (Red, Green, and Blue) should each have their own screen. But in short runs, often just an overall screen is okay.

Audio cables are best run separately, but should also be in a screened cable.

And no, I have not yet worked out how I am going to do mine yet...

I've just spent a large chunk of the day hunting for the correct parts on suppliers web sites.

The biggest headache is trying to find the correct connectors (correct pin-out and spacing) for some of the connectors :twisted:

Hopefully I have ordered the correct ones for audio out and VGA out. I should have a mini DIN suitable for the mouse socket already.
I have tried one 2.1/5.5mm power connector that I have in stock - it does not fit, so tomorrow I will have a look at a small selection I got last year and see if any of those fit.
I have some old PCB mounting 7 pin DIN sockets, but the lead out is wrong. As is the two that Farnell UK stock :twisted: So the cassette tape I/O connector will have to wait until I can think of a "work around".

I did have a quick look at the 2x32 pin DIN connectors used for the expansion connector. But gave up, as there are various different types and configurations. So if anyone can please provide some more information, that would be helpful. Also, apart from adding stuff that I don't have (disk packs), what is it normally used for?

Oh and for information for other people, 74LS45, 74HC45 and 74HCT45, 4 to 10 line decoders are not available. But 7445 are still available.

Mark

User avatar
1024MAK
Posts: 9698
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Building Rolands new Atom design...

Post by 1024MAK » Sat Feb 07, 2015 8:05 pm

Just to add to what I said, I have ordered my 4MHz oscillator modules from eBay, as it was easier finding them on eBay than Farnell (which appeared to have a poor selection). And Rapid do not have any in stock (currently out of stock for the "14" pin version).

Mark

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 8:07 pm

oss003 wrote:Wim, is the SID also working now?
Will this do?... :D

https://www.youtube.com/watch?v=A2tvTx6 ... e=youtu.be
Last edited by Multiwizard on Sat Feb 07, 2015 8:23 pm, edited 1 time in total.

User avatar
hoglet
Posts: 8901
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol
Contact:

Re: Building Rolands new Atom design...

Post by hoglet » Sat Feb 07, 2015 8:12 pm

=D> =D> =D> =D> =D> =D> =D> =D> =D>

100% perfect

You also have a real Atom SID, don't you? I'd be interested if you can tell much difference.

Dave

User avatar
roland
Posts: 3624
Joined: Thu Aug 29, 2013 8:29 pm
Location: Born (NL)
Contact:

Re: Building Rolands new Atom design...

Post by roland » Sat Feb 07, 2015 8:17 pm

1024MAK wrote: The biggest headache is trying to find the correct connectors (correct pin-out and spacing) for some of the connectors :twisted:
That's exactly why I changed some of the connectors on the new MK2 boards. I have Farnell part numbers in the bill of materials so future Atom-builders will not have that problem.
1024MAK wrote: Hopefully I have ordered the correct ones for audio out and VGA out. I should have a mini DIN suitable for the mouse socket already.
VGA is normally not a big problem. For the audio out you can either take your board to the local electronics shop and try some sockets. A usable type might be Pro Signal MJ-352W-C but I don't know where to get it. I can try my local electronic store for a few of them as that is the place where I got mine.
1024MAK wrote: I have tried one 2.1/5.5mm power connector that I have in stock - it does not fit, so tomorrow I will have a look at a small selection I got last year and see if any of those fit.
In the first place I used the power connector of my old Atom, but that didn't survive the removal from the board. So I bought a new one that didn't fit exactly but with some bending, pushing and pulling it fitted at last.
1024MAK wrote: I have some old PCB mounting 7 pin DIN sockets, but the lead out is wrong. As is the two that Farnell UK stock :twisted: So the cassette tape I/O connector will have to wait until I can think of a "work around".
Maybe I can do you a favour with the tape I/O connector of my very rare Acorn Atom "Yellow Edition" that I received yesterday....
1024MAK wrote: I did have a quick look at the 2x32 pin DIN connectors used for the expansion connector. But gave up, as there are various different types and configurations. So if anyone can please provide some more information, that would be helpful. Also, apart from adding stuff that I don't have (disk packs), what is it normally used for?
The external connector is normally used for a disk pack, MDCR, connection to an 19" rack with loads of expansions (Wim has several of them). The internal connector is mostly used for additional memory or an additional graphics card. These connectors are not really necessary as the main board has already most features.

Examples:

External connector: Farnell 1096900 HARTING 0902 164 6921 PLUG, DIN41612, R/A, B, 64WAY
Internal connector: Farnell 1096907 HARTING 0902 264 6825 SOCKET, DIN41612, B, 64WAY
1024MAK wrote: Oh and for information for other people, 74LS45, 74HC45 and 74HCT45, 4 to 10 line decoders are not available. But 7445 are still available.
IIRC the 74LS145 is a good replacement that is also still available.


Edited: added 64p connector part numbers
Last edited by roland on Sat Feb 07, 2015 8:30 pm, edited 1 time in total.
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
Multiwizard
Posts: 1751
Joined: Wed Jan 11, 2012 9:03 pm
Contact:

Re: Building Rolands new Atom design...

Post by Multiwizard » Sat Feb 07, 2015 8:30 pm

hoglet wrote:You also have a real Atom SID, don't you? I'd be interested if you can tell much difference.

Dave
It's weird but the real AtomSID sounds warmer, it is the same difference between a Lp and Cd... :?

User avatar
1024MAK
Posts: 9698
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Building Rolands new Atom design...

Post by 1024MAK » Sat Feb 07, 2015 8:31 pm

Thanks for the help Roland.

Yes, I was looking at the pin-outs for the 7445, the 74xx42 and the 74xx145 and at a quick glance, the pin-outs for all three look the same. I intend to study the data sheets later. I have ordered a 74HCT42 and a 74LS145 as well.

This is not the first time that I have encountered such a situation with "standard" 74 series logic chips.

Mark

Edited to move a misplaced "a" around. Took me two goes :oops:
Last edited by 1024MAK on Sat Feb 07, 2015 8:36 pm, edited 2 times in total.

Post Reply