Jump to content
SiLic0ne t0aD

New Coleco RGB board?

Recommended Posts

In this same thread he's already been told. If he really paid 2k for someone to make one, they not only ripped him off but should have done their homework. The citrus board was adapted from published datasheets. I'm taking it further and will have something superior to any of the available RGB mods, except the F18. Since that mod adapts the digital signals and uses code to build RGB.

Share this post


Link to post
Share on other sites

Finally had the time to test Citrus' PAL RGB design and my Colecovision refuses to work at first try.

This Coleco was an "untested" unit I bought and since I had no power supply I did some work before installing the RGB mod:

 

- I managed to adapt a 12V3A DC transformer inside the case. That feeds a picoATX psu which provides the +5 and +12V.

- I did the 4164 RAM Mod as I missed the -5V line (and wanted to ensure that the ram would be OK)

 

And this was the result:

 

NkbxyDQRhQT2OLofOYiQ9RjdC0ip5ENcEGK_RhJa

 

Which finally ended in this:

 

6vPH9MQVU32D34PMBA5K9zguc0UL2B36jGj6hBgm

 

And unfortunately what I obtain when powering the console on is an intermitent black screen and sometimes some artifacts like this:

 

4fyj6fdaTxr1WEXfVTm4dm2Tc30zZpdy505WQFej

 

In any case I do not consider the RGB mod to be the cuplrit. I am unable to tune the Coleco on the TV via UHF, so there must be something else :(

By the way the voltages were measured and OK, the Z80 /RESET stays high when powering on and the RAM was tested.

 

Regards

Edited by Quickie
update

Share this post


Link to post
Share on other sites

ColecoVision is VHF (channel 3 or 4 depending on switch).

 

Anyway, your images are probably restricted by who can view, so all I see is a circle with a line through it.

 

-5V may be required to make the controllers work reliably.

 

 

Share this post


Link to post
Share on other sites

Hi,

 

Thanks for responding.

I thought that -5V wasn't needed for anything else than the old RAM ICs.

I read on another thread that the -5V could be required for the controllers, but that seemed to be another topic (or at least his author stated afterwards that should work fine without -5V).

 

Since I can't edit again my previous post, here are the pictures (apologies):

 

6NTFsotSxw2eDAzViLmPodvZxXREs6vnd9WRW8iN

 

w8oJ0r2RnBBWXeFi2JYsEwndA3YYvqqVJut3gbto

 

AnRPBasefhEF3cQMlw7imktRRyDDdryI75-Kc-6w

 

Update:

 

I replaced the Z80 processor and it seems that the console is not so dead...

 

WaQf8tHx1hX8jiRH5v3YUkXfs3JKBs2jpYXdjpWj

Edited by Quickie
New picture

Share this post


Link to post
Share on other sites

Whoa... there's a separate PAL version?  I wonder why they think they need that?  Anyway, with video chip apparently working now, looks like sync is still missing.  Check the path of the S wire for a signal.  Check that it connects to R4 on the right side.  Make sure R4 isn't shorted across.  Check the connection from S to U4 top left pin.  It also connects to U5 right, second pin from the top.  U4 left, second from top, goes to the VDC Y input.  Make sure that connection is good.  Between bottom left and top right of U4 should be VCC, 5V.

 

So, U4 is directly responsible for sync.  As far as U5 goes for this circuit, make sure it doesn't have any shorts.  It doesn't contribute to sync, but a short could zap your sync output.

 

Look for solder bridges all over the board, but especially in those areas.  Use a magnifying glass, since those are some pretty tiny chips.

Share this post


Link to post
Share on other sites
10 hours ago, ChildOfCv said:

And if all else fails, try just building the standard version.  I don't think NTSC or PAL are different enough at the stage where this board connects to warrant needing their own special versions of the circuit, and the 1.2 version of the board is much newer anyway.  https://oshpark.com/shared_projects/WhLjVooT

 

Hi,

 

I will try to check with detail the PCB I made and the soldering to check that there isn't any short, however when I soldered the components and installed the mod on the PCB I double checked that there wasn't anything unexpected.

 

In regards of the other NTSC board I have one half-done (I miss the resistors and other passive components). But I discontinued it when I realized that it was only valid for NTSC machines. I guess that this is because the TI-99 (and the Colecovision) do use a different VDP on each region: TMS9918A (NTSC) vs TMS9929A (PAL) with their own specs.

 

I thought that, being the PAL RGB mod an untested one perhaps the sync line could require of some extra adjustment by increasing the value of 75Ohm on Sync.

In any case I've PM'ed @citrus3000psi to check if he can confirm this.

 

Thanks for your help!

 

 

Share this post


Link to post
Share on other sites

I'd still try the NTSC one though.  The 9928 and the 9929 have the same pinout.  The signaling is different on the R-Y and B-Y outputs due to the phase reversal.  Well, perhaps that's the one sticking point.  It would make a displayable image with color either way, but perhaps the colors will be off.

 

Anyway, yeah check out the sync output.  You could even remove U5 entirely until you get a stable image, then add it back.

Share this post


Link to post
Share on other sites

Well I reverse engineered the circuit.  It's basically this project, but Citrus forgot to put the .1uF capacitor between composite sync input of the LM1881 and the Y input.  So it's possible that the 1881 never detects sync because of voltage bias.  To test that theory, you should cut the trace to U4 pin 2, then scrape away enough mask to solder another capacitor over the gap you made, being careful to not short it with anything else.  I drew a red line over the approximate location.

 

citruspalfront.thumb.png.e25cae7a104f5e42f7ceb1261627b12e.png

  • Thanks 1

Share this post


Link to post
Share on other sites
45 minutes ago, ChildOfCv said:

Well I reverse engineered the circuit.  It's basically this project, but Citrus forgot to put the .1uF capacitor between composite sync input of the LM1881 and the Y input.  So it's possible that the 1881 never detects sync because of voltage bias.  To test that theory, you should cut the trace to U4 pin 2, then scrape away enough mask to solder another capacitor over the gap you made, being careful to not short it with anything else.  I drew a red line over the approximate location.

 

citruspalfront.thumb.png.e25cae7a104f5e42f7ceb1261627b12e.png

Yeah! Big Thanks for your time and help!

I will try it asap and post here the result.

 

I remember seeing that project back in the day in Hackaday, but I completly forgot about it and I did not associate it with this.

Share this post


Link to post
Share on other sites

Well, I've tried your suggestion, however that trace is so tinny and so near to a resistor that I traced both sides of it (U4:2 to U1:3) and cutted it on the base of U4 pin 2, where there is somo more space, soldered the 0.1uF cap on the U4 pin 2 leg and soldered a wire on the other end of the 0.1uF cap:

 

Here the pad cutted:

 

_I5QALs4QwQeCUUxzMZgcA9Aovgv93_HuSLP2-81

 

And here with the capacitor soldered and wired to U1. Not the best soldering job on the world, but does its job (and there are no shorts):

 

osOZb_7F_R_kgcGIwp9GBGcN7-_pFP-9uEQFbahx

 

Although it does not work fine yet it seems to present the image more continuously:

 

sCAUOP_cFpQsIvf7dNZYmWCvgT4cNscZtyExz-dX

 

It's almost there...

Share this post


Link to post
Share on other sites
4 hours ago, Quickie said:

Yeah! Big Thanks for your time and help!

I will try it asap and post here the result.

 

I remember seeing that project back in the day in Hackaday, but I completly forgot about it and I did not associate it with this.

Yeah this one reorganized the layout to solder to the VDC chip, but other than the lack of that capacitor and jumpers, it looks identical.

  • Thanks 1

Share this post


Link to post
Share on other sites

So, one thing to ask though:  Did you insulate the 4066 from the resistors and capacitors underneath it?  I can't tell from the pictures, so that concerns me.

 

Also, what do you have for measurement equipment?  A logic probe or oscilloscope would help a lot.  It would be nice to see if the sync and burst outputs from the 1881 are happening.

  • Thanks 1

Share this post


Link to post
Share on other sites
10 hours ago, ChildOfCv said:

So, one thing to ask though:  Did you insulate the 4066 from the resistors and capacitors underneath it?  I can't tell from the pictures, so that concerns me.

 

Also, what do you have for measurement equipment?  A logic probe or oscilloscope would help a lot.  It would be nice to see if the sync and burst outputs from the 1881 are happening.

Hi,

 

I insulated the 4066 in this unappropiate package with hot glue from the passive components underneath. Good point :)

I can recheck there if there might be some undesired contact underneath.

 

I do have an oscilloscope. I was advised that I could check the sync wave on the LM1881 and on the Sync pad.

For me that seems like a good action plan ;) Let me grab some time and I post both results.

 

Regards

Share this post


Link to post
Share on other sites

Hello,

 

Apologies for not being able to update this earlier, but it has been impossible :(

Well, I had some time to do the testing, and although I guess that I would have done something bad, there is something interesting:

 

- This is the screenshot of the Sync pad of the RGB PCB on the Oscilloscope with AutoSet scale:

 

cOfdTLrIrvF02qJWa5rimi6DGsLHnSHnRQpwVJ75

 

The same wave from the same pad of the RGB PCB , but this time with the scale defined after adjusting the probe and with measured data (seems that there is no signal there...):

 

NKaexY3rCWrzuO9ri3yvZc__ENo30L4L1etLL8qq

 

And finally I captured the signal on the LM1181 pin 1 (Sync Output), which looks completely different:

 

6OF8ZtubFMliNhPT86R7Zf-R8ujqmQjkkaKpv-iG

 

Alexis, the creator of the TMS9929A RGB and Component adapter of Hackaday (Kudos for him!) has gently advised me that perhaps removing the 82 ohm resistor on the Csync output could enlarge the sync pulses and make them recognizable for my TV.

To be sincere I don't know if after the data I provided that explains it all, but I'll give a try...

 

 

Edited by Quickie

Share this post


Link to post
Share on other sites

Weeell!! I did some try/catch and soldered directly the sync pin of the RGB DIN connector to the LM1881N pin 1.

 

This is what I get:

 

jemttKoIj0HUySwVNL1G64LaA58s1xEh4SI_45tb

 

It's stable, altough there is some noise... this is almost there!

Share this post


Link to post
Share on other sites
1 hour ago, Quickie said:

Hello,

 

Apologies for not being able to update this earlier, but it has been impossible :(

Well, I had some time to do the testing, and although I guess that I would have done something bad, there is something interesting:

 

- This is the screenshot of the Sync pad of the RGB PCB on the Oscilloscope with AutoSet scale:

 

cOfdTLrIrvF02qJWa5rimi6DGsLHnSHnRQpwVJ75

 

The same wave from the same pad of the RGB PCB , but this time with the scale defined after adjusting the probe and with measured data (seems that there is no signal there...):

 

NKaexY3rCWrzuO9ri3yvZc__ENo30L4L1etLL8qq

 

And finally I captured the signal on the LM1181 pin 1 (Sync Output), which looks completely different:

 

6OF8ZtubFMliNhPT86R7Zf-R8ujqmQjkkaKpv-iG

 

Alexis, the creator of the TMS9929A RGB and Component adapter of Hackaday (Kudos for him!) has gently advised me that perhaps removing the 82 ohm resistor on the Csync output could enlarge the sync pulses and make them recognizable for my TV.

To be sincere I don't know if after the data I provided that explains it all, but I'll give a try...

 

 

With a resistor you should expect a scaled down signal though, not a dead line.  I bet that either a trace didn't make it, or one of the components wasn't quite soldered down.  Anyway, cool that you got it working with a direct connection!

  • Thanks 1

Share this post


Link to post
Share on other sites
On 11/29/2019 at 5:37 PM, Quickie said:

Weeell!! I did some try/catch and soldered directly the sync pin of the RGB DIN connector to the LM1881N pin 1.

 

This is what I get:

 

jemttKoIj0HUySwVNL1G64LaA58s1xEh4SI_45tb

 

It's stable, altough there is some noise... this is almost there!

Great!! So, does it mean that is correctly working finally? Hope that you also success with TI99/4A ;)

Share this post


Link to post
Share on other sites

Hi all,

 

I hope that I will be able to check components and traces located between LM1881N pin 1 and the sync pad tomorrow, and as @ChildOfCv states, most probably one of the resistor edges could not be properly solded to the pad.

If that is the case I expect to get rid of the current interference that is being presented.

 

@Papalapa The results on the Coleco are very promising, so, having in mind that the source of inspiration of this particular board was originally designed for a Ti-99/4a I have no doubt that it will work as well :)

Don't worry, I'll keep you posted in VdR ;)

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...