Please Help: 2600 Junior PAL + UAV Rev.D

[TheMontezuma]

Hi,

 

I wondered if anybody could help me with the UAV mod in my 2600 Junior PAL?

 

I configured the UAV for 2600. The picture is very good, but the colors seems to be wrong (as far as I can tell).

 

Edited December 9, 2018 by TheMontezuma

[+tf_hh]

I configured the UAV for 2600. The picture is very good, but the colors seems to be wrong (as far as I can tell).

 

Did you try to re-adjust the color using the color-pot? Same behavior like with all Atari computers :-)

[TheMontezuma]

Thanks for the tip, but I have already tried it.

The best what I could get is:

 

If I turn the pot further, the TV shows black screen...

 

And this is a reference (scrrenshot from a youtube video):

[+tf_hh]

Hmm. Maybe you grab the COLOR signal from the wrong point? It´s a long time ago since I built an UAV into a 2600 Junior, so I don´t remember really where I connect COLOR, but looking into Jerzy´s schematics:

 

I would suggest to remove that cap (C50). The 1k pull-up resistor must remain. Grab the COLOR signal anywhere between pin 9 of TIA and C50. The pad you´re used (R54) is not found in the trace of COLOR, but I think it´s a mistake in the schematics.

 

 

[+-^CrossBow^-]

I could be wrong as I have never installed a UAV into any of the 2600s. But on the 7800 the TIA color must be soldered to the TCOL pad in the lower right of the UAV board. Looks like you have it attached in the color where the Maria and 8-bits are attached. So trying moving that over to the TCOL point and see what happens?

[TheMontezuma]

Hmm. Maybe you grab the COLOR signal from the wrong point? It´s a long time ago since I built an UAV into a 2600 Junior, so I don´t remember really where I connect COLOR, but looking into Jerzy´s schematics:

 

2600_UAV.png

I would suggest to remove that cap (C50). The 1k pull-up resistor must remain. Grab the COLOR signal anywhere between pin 9 of TIA and C50. The pad you´re used (R54) is not found in the trace of COLOR, but I think it´s a mistake in the schematics.

 

 

 

R54 upper pad is connected to the PIN9 of TIA and C50.

 

I could be wrong as I have never installed a UAV into any of the 2600s. But on the 7800 the TIA color must be soldered to the TCOL pad in the lower right of the UAV board. Looks like you have it attached in the color where the Maria and 8-bits are attached. So trying moving that over to the TCOL point and see what happens?

 

The TIA color is connected to the "COLOR IN" UAV connector as it should be. There is no "TCOL" pad on my UAV.

 

 

Anyway it looks like my UAV is dead. Suddenly the picture was gone. I started to messure and realized that the VCC is only ~2V.

I found a short circuit between GND and VCC, so I desoldered UAV and realized that:

- 2600 is OK

- UAV shows 4,7 Ohm between GND and +5V pads

- Once plugged in a temporary socket on the 2600, the UAV board gets hot in the area of the GND, 5V pads

 

Not a happy day...

[+-^CrossBow^-]

 

R54 upper pad is connected to the PIN9 of TIA and C50.

 

 

The TIA color is connected to the "COLOR IN" UAV connector as it should be. There is no "TCOL" pad on my UAV.

 

Incorrect...

 

Look at your UAV again just above the Ground connection point? See that small hole that sale "Tcol" next to it? That is the TIA color input and likely where you should have wired in the connection for the color. On the 7800 installations, that is where the TIA colorburst gets connected and the Maria colorburst is connected to the Color In on on the UAV output section same as the 8-bit and 5200 setups. But I'm pretty sure that Tcol connection is specifically for the TIA color.

 

The UAV should not get hot or even warm to be honest. Double check where you are putting your +5 and ground connections from and make sure you don't have a small wire strand from the +5 or Gnd on the UAV touching the other and causing a short.

[TheMontezuma]

 

Incorrect...

 

Look at your UAV again just above the Ground connection point? See that small hole that sale "Tcol" next to it? That is the TIA color input and likely where you should have wired in the connection for the color. On the 7800 installations, that is where the TIA colorburst gets connected and the Maria colorburst is connected to the Color In on on the UAV output section same as the 8-bit and 5200 setups. But I'm pretty sure that Tcol connection is specifically for the TIA color.

 

The UAV should not get hot or even warm to be honest. Double check where you are putting your +5 and ground connections from and make sure you don't have a small wire strand from the +5 or Gnd on the UAV touching the other and causing a short.

 

Now I see what you mean.

You are right. I was only focused on the terminal connectors and the TCOL pad is obviously where I should wire the TIA color signal...

 

The UAV is wired where originally the 4050 chip was soldered. Ground and VCC is provided through the PINs: 1 (5V) and 8 (GND).

I will double check the connections against a short, but I'm afraid that is possibly too late for my UAV

[TheMontezuma]

I got info from Bryan

 

"T col" pad shall only be used for 7800

 

For 2600 color signal shall be connected to "Color In" connector.

 

So I actually did it right in the first step, but must have done something else wrongly...

[+-^CrossBow^-]

Ahh.. okay. Again I've not installed a UAV into a 2600 only 5200 and 7800 (many actually). But again given how it is needed for the TIA colorburst I thought it would be the same for the 2600.

 

I've not see rev C boards but I thought those were also 7800 compatible? But I don't think the TCol was added until Rev D so how was TIA handled on the rev C boards then? Hmm...

[adamchevy]

Has anyone successfully done a UAV mod to a 2600 heavy sixer?

[radventure]

@TheMontezuma Did this ever get a resolution? I’ve just finished fitting a rev d UAV + IAB into a PAL 2600 jr (well, I still have to work out top case refitting, but it’s useable)

 

It is exhibiting the same colour issues.... while the picture is sharp, colour is pretty much on par with the screenshots in this thread. Colour pot adjustment has been performed.

 

My UAV board isn’t getting hot or anything though. I’m using svideo output to a retrotink mini, connected to a Samsung Syncmaster LCD. 

[alex_79]

Is there a single jumpers configuration for use with any 2600 console?  

No surprise that it doesn't work. I don't know about NTSC versions, but the cd4050 is surely NOT wired the same in PAL 6-switch and PAL jr consoles. The inputs for the lum and sync signals are on different pins, and the remaining 2 inputs are used for different functions: on the 6 switch they buffer the joystick fire buttos, while on the Jr (PAL-B) one is part of the cpu reset circuit, while the other one is connected to the 4.43 Mhz oscillator circuit (but the relative output is unused). There's also a PAL-I jr that uses a different board and the pinout might be different there too.

 

 

2600-2600A-2600jr_PAL.pdf

 

I would not power on that console until after tracing the board and fixing the connections, as there's the risk of damaging the mod board and/or the console itself.

Edited February 11, 2021 by alex_79

[radventure]

@alex_79 instructions were followed as per the 2600 jr specific parts of the included UAV guide, but yes - there is nothing indicating variations in jumpers for 2600 models or for a PAL B jr. I don't have the expertise to know more than the guides for these sorts of products. I have a rudimentary understanding of electronics, and can solder, desolder, replace through-hole style components, perform basic testing etc, but I'm not sure where to start on tracing functions.

 

I'll try reaching out to the The Brewing Academy for extra guidance also.

[TheMontezuma]

21 hours ago, alex_79 said:

Is there a single jumpers configuration for use with any 2600 console?  

No surprise that it doesn't work. I don't know about NTSC versions, but the cd4050 is surely NOT wired the same in PAL 6-switch and PAL jr consoles.

 

@tf_hh @Bryan

Do you know if cd4050 wiring in NTSC and PAL 2600jr consoles is the same?

If the UAV is only designed to work with the NTSC 2600jr, it would explain the wrong colors...

 

Edited February 12, 2021 by TheMontezuma

[+tf_hh]

4 hours ago, TheMontezuma said:

Do you know if cd4050 wiring in NTSC and PAL 2600jr consoles is the same?

If the UAV is only designed to work with the NTSC 2600jr, it would explain the wrong colors...

 

Well, it´s hard to give a 100% valid answer, because I never have had a NTSC 2600 in my hand. But I have successfully installed several UAV into 2600 Heavy Sixer, 2600 Light Sixer, 2600 Darth Vader and the Junior.

 

I didn´t trust the PAL schematics, as I never have seen an genuine Atari one, only the Jerzy files. Although Jerzy does a good job, sometimes it´s only copy & paste from earlier versions without checking every detail and there are differences. Years ago I sent some infos, but I never got a feedback. Unfortunately I change the email client five years ago and old mails are gone. So I give up and since them I use them only as an orientation, not more.

 

Speaking about the 2600 consoles I´ve been radical and remove all non-necessary parts in the video and audio circuit, especially the transformatio-to-HF circuit for audio carrier (RF usage). This results in massively less noise at the whole system. To make the correct connections, I test from TIA to the resistors etc. they might be correct the continuity and use then the pads for grabbing the signals - to prevent soldering directly at the chip´s pins. It´s only five signals, this will take a minute or two.

 

For sure it works. But wrong colors can´t be the effect of a wrong wiring, because there´s just one color signal. Wrong color intensity (luminance) I would accept as a result of mixing up the LUM lines. CSYNC and COLOR must be correct, otherwise you would see nothing (CSYNC wrong) or just black´n´white (COLOR wrong). Your examples shows wrong hue, not wrong intensity. There must be another problem...

 

[alex_79]

17 hours ago, tf_hh said:

I didn´t trust the PAL schematics, as I never have seen an genuine Atari one, only the Jerzy files. Although Jerzy does a good job, sometimes it´s only copy & paste from earlier versions without checking every detail and there are differences.

Yes, Jerzy Sobola's schematics are not official, so you should always verify the connections on the actual board you're working on, but they're a great reference noneless. There are different revisions of the schematics, too: I think those posted above are more recent than the ones in the AtariAge site.

Also note that official PAL schematics for the 6 switch console can be found in the PAL service manual.

Atari_2600_PAL_Service_Manual.pdf

 

Anyway, (many) years ago I did a s-video mod on one of my junior consoles, and later on moved that same mod into a 6-switch (link). In both cases I tapped the signals from the CD4050 and the pinout matched Jerzy Sobola's schematics. Of course it's always best to double check, as there might be different board revisions.

 

 

 

Quote

For sure it works. But wrong colors can´t be the effect of a wrong wiring, because there´s just one color signal. Wrong color intensity (luminance) I would accept as a result of mixing up the LUM lines. CSYNC and COLOR must be correct, otherwise you would see nothing (CSYNC wrong) or just black´n´white (COLOR wrong). Your examples shows wrong hue, not wrong intensity. There must be another problem...

 

If you compare with the youtube video above, then yes, there are hues differences, but the video shows the color of the NTSC game running on a NTSC console.

The PAL game on PAL console looks like this:

 

It seems to me that colors are correct (considering a little variance due to the color pot adjustement) and the differences are indeed due to wrong luma connections.

I don't know how this specific mod works, but if the LUM and SYNC are connected to a resistor ladder like in the Atari board, then I think swapping sync with one of the luma inputs could still produce a stable picture (the resistor value would be wrong, but the resulting signal might still be within tolerances).

 

 

Edited February 13, 2021 by alex_79

[radventure]

I'll have to get a picture of the image on my display, but in Pitfall! - I only get one shade of green for both the treetops and the background (the background is a lighter green in all imagery I've seen, but not on my Jr with mod). The background and treetops are both the darker green used for the treetops, no separation. The brown tree trunks are as expected. As previously stated - the quality is excellent. I've adjusted the colour ('color' for you US folk) pot and there's is no position that corrects the tree colours either.

 

I'm loading the PAL rom via a HC encore.

Edited February 17, 2021 by radventure
correction

[alex_79]

Load "Air Sea Battle",  press reset to start the game and flip the TV type switch to B&W.

The sky will show all 8 luma values in order from black to white.

If there are less than 8 shades and/or if they out of order then the luma pins are not conencted correctly.

 

BTW, is the CD4050 still on the board, with the mod pcb mounted on top of it, or has it been removed?

Edited February 17, 2021 by alex_79

[radventure]

The CD4050 is removed. I’ll check the BnW output as soon as I can in the morning.

[radventure]

[+DrVenkman]

You're missing a couple LUM bits, looks like, which would explain the missing/odd colors. 

 

Keep in mind, the UAV was designed initially for the A8. As it happens, TIA works similarly enough to GTIA that it can be used in both 2600's and 7800's. The current revision D has been thoroughly tested in NTSC 7800's and 2600's but I doubt seriously is Bryan had a PAL 2600 handy when he designed it. 

 

If this was my system, I'd go back to first principles. I'd use my multimeter to trace out the LUM bits from the PAL TIA outputs to their corresponding pins on the 4050 socket, then work out the necessary jumper settings. Actually, if it were me, I'd get hold of a new, bare UAV board and wire up the signals directly to the UAV through holes and skip the jumpers entirely. 

[radventure]

I *might* have already kicked another project off, thinking the UAV was straightforward and was done....... I’ll try and do some tracing when I can get some time. See where that takes things. 

[alex_79]

I had some pictures of one of my Jr board on my PC (PAL-B, board CO21503, rev. F1). I've highlighted the traces going  from the luma and Sync TIA pins to the CD4050, and they match J.Sobola's schematic:

 

board (top)

 

bord (bottom)

 

 

You might be able to fix the mod with minor modifications, by removing all the jumper plugs and soldering short pieces of wire to connect the pins as in this picture.

 

 

There's still the problem that with the CD4050 removed, the gate in the CPU reset circuit (circled in blue in the schematic) is missing. Since your console is still working without it, I guess it's not strictly necessary, but if it was me I'd follow @DrVenkman suggestion and I would remove the mod board, reinstall the CD4050, and then simply connect the mod using wires to the needed signals (those pullup resistors connected to the TIA pins are perfect for this).

 

Verify that your board matches the one shown here by checking continuity between pins marked with the same colors. If it differs trace it to find the right spots where to pick the signals.

Note that I'm not familiar with the UAV board as I never installed one myself, so use these info at your own risk!

 

Edited February 18, 2021 by alex_79

[+DrVenkman]

I’ve put an earlier Rev C UAV into one of my NTSCS 2600’s and a Rev D into one of my NTSC 7800’s. No need to even mount the thing into the 4050 socket if you start with a bare board or unplug it from the socket. Just run the LUM and COL signals straight to the corresponding through hole in the two rows of vias in the center. If you leave it out of the socket, you need to wire up the +5V and GND pads on the UAV.