2023 Advanced PCB Remake For the 800XL -- Revision 1.1

[Jfcatari]

@olix

From the docs, I would say no:

[reifsnyderb]

5 hours ago, olix said:

Can I use a 74HCT14 instead of a 74AHCT with IC U19?

What is the difference?

After skimming through this https://www.retrobrewcomputers.org/doku.php?id=buildtips:components_compatibility_and_substitution

 

I think that a 74HCT14 might work.  I never tried it.  If you have one on-hand, it won't hurt to try it.  But, if it doesn't work, you'll have to go with the 74AHCT14.

 

If you have a scope, it will be immediately obvious if it isn't working because you won't get a clock pulse coming out of the clock circuitry.

 

Edited March 17 by reifsnyderb

[_The Doctor__]

AHCT, is very fast, HCT is not as consistent or fast. You might get one that works and one that doesn't when using the normal HCT.

[+mytek]

Just my 2 cents...

 

When I did some evaluations of alternatives for the 74LS08 many years ago, I discovered that AHCT was indeed too fast, causing severe under and over shoot (aka: noise). This could be tempered with an inline resistor, but that was more like a band-aid. Whereas the standard HCT part was actually slower than the LS part previously in play. Hence the reason I settled upon using a 74F08 instead, which was faster than the LS or HCT equivalents, but not so fast to cause other issues.

 

EDIT: for the particular use case being talked about in this thread, a 74F14 would be my suggested replacement for U19.

[reifsnyderb]

10 minutes ago, mytek said:

Just my 2 cents...

 

When I did some evaluations of alternatives for the 74LS08 many years ago, I discovered that AHCT was indeed too fast, causing severe under and over shoot (aka: noise). This could be tempered with an inline resistor, but that was more like a band-aid. Whereas the standard HCT part was actually slower than the LS part previously in play. Hence the reason I settled upon using a 74F08 instead, which was faster than the LS or HCT equivalents, but not so fast to cause other issues.

 

EDIT: for the particular use case being talked about in this thread, a 74F14 would be my suggested replacement for U19.

A 74F14 wouldn't work as the 74AHCT14 is needed for the clock.  Two of the U19 inverters are also used for the reset circuit.

 

 

 

 

 

 

[+mytek]

1 hour ago, reifsnyderb said:

A 74F14 wouldn't work as the 74AHCT14 is needed for the clock.  Two of the U19 inverters are also used for the reset circuit.

Since a 74F14 is still a Schmitt Trigger inverter, albeit a TTL output type, it will have similar characteristics when used in an oscillator circuit but simply not go to the full positive output as an HC equivalent part. However I wouldn't think this would be a problem for the GTIA oscillator input pin since that works fine with the original 2 transistor oscillator circuit which also can't switch to the full +5V rail either. But this is all conjecture on my part, and this is about your board design, not mine, so you get to call the shots based on your experience with it.

 

On a side note, if the reset circuit is based on an RC delay, I would also think the 74F14 would be OK for that as well, once again because it's Schmitt Trigger in nature and the NMOS chips that need it work fine with TTL level outputs. Since my curiosity is now peaked, I will have to do some tests later today to verify for my own satisfaction and future reference.

 

EDIT: Hold your horses. This likely does not need to be high speed for your specific application. I've used 74HC14 chips for the same thing in the past (1088XLD), and most likely a 74HCT14 would also work, especially since you originally spec'ed a higher speed version of that without incident.

[CJ Atari]

6 hours ago, mytek said:

Just my 2 cents...

 

When I did some evaluations of alternatives for the 74LS08 many years ago, I discovered that AHCT was indeed too fast, causing severe under and over shoot (aka: noise). This could be tempered with an inline resistor, but that was more like a band-aid. Whereas the standard HCT part was actually slower than the LS part previously in play. Hence the reason I settled upon using a 74F08 instead, which was faster than the LS or HCT equivalents, but not so fast to cause other issues.

 

EDIT: for the particular use case being talked about in this thread, a 74F14 would be my suggested replacement for U19.

Could this be what is causing this clock?  Frequency is correct.  And thanks.

[+mytek]

5 hours ago, CJ Atari said:

Could this be what is causing this clock?  Frequency is correct.  And thanks.

That does kinda look like what I saw when running my experiments using an AHCT chip, but I can't say with certainty that is what's happening on your end. In the 576NUC+ I used a 74AHCT14 and didn't see any issues with using it as the oscillator stage. For all I know it might have looked like your waveform, but if it did the over/under shoot wasn't a problem for GTIA.

[olix]

Does anyone have a part number from Mouser or Digi-Key for a compatible SW1 switch: 
ROM Bank (was RF Switch)

 

I live in Germany and this switch was not installed on the PAL machines. So I don't have a sample to look for a replacement.

[olix]

Today I received my package with the 5 circuit boards. It was a collective order for some ABBUC club members in Germany. I would like to say a big thank you to Brian. 

Not only did he do a great job with the development of this board, but he was also very helpful in arranging a cheap shipping to Germany.

@reifsnyderb: Thank you very much! 

 

 

[+scorpio_ny]

9 hours ago, olix said:

Does anyone have a part number from Mouser or Digi-Key for a compatible SW1 switch: 
ROM Bank (was RF Switch)

 

I live in Germany and this switch was not installed on the PAL machines. So I don't have a sample to look for a replacement.

Hi Olix 

 

I had ordered one from Digikey to see if it would viable. I still have not received it yet. Here is the part number:

https://www.digikey.com/en/products/detail/apem-inc/MHSS1105/1949465
 

When I get it, I will let you know if it works.

 

 

[+scorpio_ny]

Hi @reifsnyderb,

 

I finally built the 800XL remake and I have an interesting issue. When I try to boot to basic with the RF switch set to the left, it always goes to rom self test. When I try to boot a cartridge, it still goes to self test. When I try the switch on the right, it looks like it tries to get into basic and it just hangs at the blue screen. Cartridges work fine. I just started to test with some multi-carts and it seems to be working fine. When I ran Simcheck 1.5, it recognized it had as an Atari 400/800 CPU with 48K of RAM which was very strange!

 

Here is the kicker: I am building this system to see how much I can build it without original Atari components. I had already built the Byte Attic's Sally adapter a while ago for this project (before you made this board). I decided to give it a try and guess what? It worked! I was able to boot into basic when I set the switch to the left (Atari) or right (Altria Basic). So I am at a loss what gives. Could it be some sort of timing issue? The other Sally CPU's were Synertek models and one that was manufactured in Mexico. 

 

Since it worked with the CPU board I was going to use anyway, I pressed on and installed a Pokeymax and the Sophia 2. It was already using the modern 6520 chip too. And they all worked! I am sure there will be some compatibility issues with some software or hardware down the road. But it is nice to see that we can  are close to building an Atari with none of the original custom chips. Hopefully, there will be more cost efficient replacements down the road.

Here it is playing the Last Squadron loaded from a Side 3

 

@olix: The switch I ordered was too small.

Edited March 23 by scorpio_ny

[reifsnyderb]

5 minutes ago, scorpio_ny said:

Hi @reifsnyderb,

 

I finally built the 800XL remake and I have an interesting issue. When I try to boot to basic with the RF switch set to the left, it always goes to rom self test. When I try to boot a cartridge, it still goes to self test. When I try the switch on the right, it looks like it tries to get into basic and it just hangs at the blue screen. Cartridges work fine. I just started to test with some multi-carts and it seems to be working fine. When I ran Simcheck 1.5, it recognized it had as an Atari 400/800 CPU with 48K of RAM which was very strange!

 

Here is the kicker: I am building this system to see how much I can build it without original Atari components. I had already built the Byte Attic's Sally adapter a while ago for this project (before you made this board). I decided to give it a try and guess what? It worked! I was able to boot into basic when I set the switch to the left (Atari) or right (Altria Basic). So I am at a loss what gives. Could it be some sort of timing issue? The other Sally CPU's were Synertek models and one that was manufactured in Mexico. 

 

Since it worked with the CPU board I was going to use anyway, I pressed on and installed a Pokeymax and the Sophia 2. It was already using the modern 6520 chip too. And they all worked! I am sure there will be some compatibility issues with some software or hardware down the road. But it is nice to see that we can  are close to building an Atari with none of the original custom chips. Hopefully, there will be more cost efficient replacements down the road.

Here it is playing the Last Squadron loaded from a Side 3

 

@olix: The switch I ordered was too small.

Hello!

 

It's good to see you got it working!  I also see that you are using a non-Atari crystal.  I had nothing but problems with non-Atari crystals.  What's the part number on the crystal you used?

 

I don't know what to tell you about the ROM issue.  It does sound like some sort of timing issue, though.  Either that or a bad connection somewhere.  I never ran into that sort of problem.

 

Best Regards,

 

Brian

 

 

 

[+scorpio_ny]

57 minutes ago, reifsnyderb said:

Hello!

 

It's good to see you got it working!  I also see that you are using a non-Atari crystal.  I had nothing but problems with non-Atari crystals.  What's the part number on the crystal you used?

 

I don't know what to tell you about the ROM issue.  It does sound like some sort of timing issue, though.  Either that or a bad connection somewhere.  I never ran into that sort of problem.

 

Best Regards,

 

Brian

Hi Brian!

I was not aware that you had issues with non-Atari crystals and that would make a lot of sense on why I was having issues. I guess I lucked out that the Sally adapter somehow compensated for it. The crystal was recommended by @Jfcatari and here is the part: https://www.digikey.com/en/products/detail/cts-frequency-controls/MP036S/67661 

 

So, does this mean it is recommended that the 800XL remake should use original Atari donor parts to lessen the chances of issues?

[reifsnyderb]

1 minute ago, scorpio_ny said:

Hi Brian!

I was not aware that you had issues with non-Atari crystals and that would make a lot of sense on why I was having issues. I guess I lucked out that the Sally adapter somehow compensated for it. The crystal was recommended by @Jfcatari and here is the part: https://www.digikey.com/en/products/detail/cts-frequency-controls/MP036S/67661 

 

So, does this mean it is recommended that the 800XL remake should use original Atari donor parts to lessen the chances of issues?

My main issue was with color.  I was unable to get color unless I used the Atari crystal.  Otherwise, using non-Atari parts, where possible, is fine.

[+scorpio_ny]

Just now, reifsnyderb said:

My main issue was with color.  I was unable to get color unless I used the Atari crystal.  Otherwise, using non-Atari parts, where possible, is fine.

Ah ok. It is funny you mentioned that. When I first powered it on, it was only black and white too. I was able to correct it by adjusting the color pot. I had to do an initial very large adjustment until I saw some color. Then, I had to do very, very small turns to get the color right.

[Jfcatari]

Glad to see you got it working...Your desk looks a lot like the rat's nest I have on mine.

 

@scorpio_ny Could you please do me a favor?

 

Could you test the composite video out on your remake build using a tube based TV or a tube based composite monitor?

Do you have color just using composite video?

 

I see you have a RetroTink in the mix which is taking svideo and converting it to HDMI.

[+scorpio_ny]

10 hours ago, Jfcatari said:

Glad to see you got it working...Your desk looks a lot like the rat's nest I have on mine.

 

@scorpio_ny Could you please do me a favor?

 

Could you test the composite video out on your remake build using a tube based TV or a tube based composite monitor?

Do you have color just using composite video?

 

I see you have a RetroTink in the mix which is taking svideo and converting it to HDMI.

Hi @Jfcatari,

 

I personally do not own a CRT based display. I know someone that does and maybe I can take my computer over to test. 

 

The only thing is that it now has a Sophia 2 installed. What are looking for to see?

[Jfcatari]

I am tracking down an issues with composite video being only B/W when I connect to a tube monitor or CRT.

 

When I use

• SDVIDEO to HDMI convertor, I get color.
• Composite to VGA convertor, I get color
• Straight composite to tube TV or CRT, I get only B/W

 

[+scorpio_ny]

18 minutes ago, Jfcatari said:

I am tracking down an issues with composite video being only B/W when I connect to a tube monitor or CRT.

 

When I use

• SDVIDEO to HDMI convertor, I get color.
• Composite to VGA convertor, I get color
• Straight composite to tube TV or CRT, I get only B/W

 

Ok! Which clock crystal are you using? @reifsnyderb mentioned that he had a lot problems with the non-Atari sourced clock crystal getting only B/W. I had the same issue until I adjusted the POT.

Edited March 23 by scorpio_ny
Change words

[Jfcatari]

I am using the same crystal that you are using that I suggested a little back.

 

I am getting color and did have to adjust the pot for it to work, BUT I only get B/W when using composite video to a tube TV/CRT,  I always get color when using the converters from:   composite to VGA, svideo to HDMI,  and composite to HDMI

 

Working with Brian, we are thinking the problem might be back at the GITA or clock side.  

I have GITA's and Atari clocks that I can swap in.

 

I wanted someone else to confirm that they are seeing what I am seeing.  Brian is using an Atari clock.

 

[+scorpio_ny]

5 minutes ago, Jfcatari said:

I am using the same crystal that you are using that I suggested a little back.

 

I am getting color and did have to adjust the pot for it to work, BUT I only get B/W when using composite video to a tube TV/CRT,  I always get color when using the converters from:   composite to VGA, svideo to HDMI,  and composite to HDMI

 

Working with Brian, we are thinking the problem might be back at the GITA or clock side.  

I have GITA's and Atari clocks that I can swap in.

 

I wanted someone else to confirm that they are seeing what I am seeing.  Brian is using an Atari clock.

 

It will be an interesting test since I am using the Sophia 2 instead of GTIA.

[+scorpio_ny]

Hi @Jfcatari,

 

I am going to go to try with a regular TV. (I am hoping that the TV has composite in 😄). But one thing, I set my Retrotink for composite video input and made sure that the wiring was for composite (no svideo connection) and tested it with my LCD display. I got a color screen (and tons of jail bars 😱)

[+scorpio_ny]

Hi @Jfcatari,

 

I have the same issue with a CRT. No color.p

[Jfcatari]

Nice,,,maybe i am not crazy..So we might be on to a crystal issue!

 

Do you have atari crystal to test with?

 

I have but not had a chance to swap out