PAL 3.546895MHz Crystal Alternative Solutions

[Beeblebrox]

5 hours ago, Overange said:

With the information acquired above it could be possible to redesign @mytek PalBurst Board to integrate the 14.187576MHz Crystal and a /4 Circuit to achieve the 3.546895MHz OSC Signal, therefore a one trick kit conversion from NTSC to PAL

I'd definitely be interested in buying such a board.

[+mytek]

5 hours ago, Overange said:

With the information acquired above it could be possible to redesign @mytek PalBurst Board to integrate the 14.187576MHz Crystal and a /4 Circuit to achieve the 3.546895MHz OSC Signal, therefore a one trick kit conversion from NTSC to PAL?

Anyone is free to do so

 

Schematics, ect. are all HERE.

[Mathy]

Hello guys

 

Doesn't @Dropcheck already offer such a board?

 

Sincerely

 

Mathy

 

[+mytek]

24 minutes ago, Mathy said:

Hello guys

 

Doesn't @Dropcheck already offer such a board?

 

Sincerely

 

Mathy

 

Dropcheck's board had the same circuit as my PALburst board, only with a larger footprint. It was aimed at the 1200XL for PAL conversion, something that was never a possibility before her board existed. Long before I made the PALburst board, her board became unobtainium. And since I wanted to be able to offer a PAL version of the 576NUC+, a new board was required and it had to be small in order to fit the tight quarters.

 

So the idea behind both of these upgrades, was to change the main crystal to match PAL and then add the GTIA piggyback color burst board to provide a PAL GTIA chip with the PAL color burst frequency phase locked to that main oscillator.

 

Only the 1200XL and the 576NUC+ need something like this, since most of the other A8 systems (including the 1088 series) have the circuit provided on their motherboards, with the NTSC variants simply requiring plugging in the missing components and soldering it up.

 

I thought I was experiencing Déjà vu...

 

Click on this link and then scroll down to see Dropcheck's board vs mine.

 

 

[Overange]

6 hours ago, mytek said:

Anyone is free to do so

 

Schematics, ect. are all HERE.

 

@mytek On it

Currently, this project exists solely in the realm of theory and prototyping. 

 

The core of the circuit employs a series resonant oscillator, which is a fairly straightforward design. To some extent, the third logic gate serves to isolate its output.

 

The subsequent phase entails dividing the clock frequency of 14.187576MHz by a factor of 4 to attain the desired PAL 3.546895MHz output.

 

Following that, the focus shifts to the Colour Burst Circuit, which draws inspiration from the NUC576 PAL Colour Burst Version.

Initially, I assembled a preliminary design with a direct plug-in configuration.

However, I anticipate that external modifications may be necessary to effectively isolate the existing main board oscillator circuitry.

 

 

Consequently, I intend to draft an alternative design that incorporates the isolation of external oscillator circuitry.

One potential drawback of this approach is that it might result in a higher placement off the Main board.

 

Another option under consideration is designing it as a compact PCB with attached flying leads.

 

I am open to any thoughts on this idea?

 

[Overange]

Bit of rework and this could be the not under GTIA version.

 

 It measures Apx 27mm x 25mm

 

This version can be wired into the W2 ( W10 on 1200XL )NTSC Link using OSC and PAL as the points of injection.

 

Existing OSC or if fitted PAL circuitry will be to be disconnected.

 

+5v and GND to be hooked up from another source.

 

 

 

[thorfdbg]

A difference between PAL and NTSC is that the color carrier is at a different frequency. This was done in PAL to avoid some leakage of the chroma signal into the luma signal which causes dot-crawl on B&W TVs. Compared to NTSC, PAL uses 5/4th of the luma dot frequency for the chroma carrier, and thus the main clock is first divided by four by two flip-flops (74LS74) and then used to lock-in a crystal of 5/4th the frequency for the chroma carrier. This also explains why the chroma carrier cannot be generated by an oscillator, but rather requires a crytal. The cheap lock-in (phase-locking) of the chroma carrier to luma will not work with an osicllator as the signal to lock on never reaches the internal crystal.

[macsonny]

On 9/15/2023 at 7:21 AM, Piotr D. Kaczorowski said:

 

 

0.70 PLN = 0.16 USD .....   Are you looking for a better price?

 

 

Are you able to sell these and send to Australia?

[Beeblebrox]

34 minutes ago, macsonny said:

Are you able to sell these and send to Australia?

That is the ntsc system clock crystal modern equivalent for XL's , so no good for your Pal setup, (following on from conversations in your other thread about sourcing a pal sys clock crystal for your 130XE).

 

To quote from overange's post earlier:

 

XL Based Systems
PAL = 3.546894MHz - Obsolete
NTSC = 3.579545MHz - Still Produced - Mouser Part No 710-830003056B 

 

Freddie/XE Systems
PAL = 14.187576MHz - Obsolete - may find old Chinese stock
NTSC = 14.31818MHz - Still Produced - Mouser Part No 815-AB-14.31818B2

 

PAL Colour Boost = 4.433618MHz - Still Produced - Mouser Part No 520-HCA443-20X

 

[Piotr D. Kaczorowski]

On 1/28/2024 at 12:10 AM, macsonny said:

Are you able to sell these and send to Australia?

Sure. If still available.  Send me priv to petertheatarian at gmail.com 

[macsonny]

On 1/31/2024 at 7:51 AM, Piotr D. Kaczorowski said:

Sure. If still available.  Send me priv to petertheatarian at gmail.com 

Message sent.

[Piotr D. Kaczorowski]

16 hours ago, macsonny said:

Message sent.

 

On Monday, I will verify the availability and send you the details.

 

Best,

Piotr

 

 

[Simius]

You can always use cheap and easily available 17,73447MHz crystal and divide it by 5 and by 4 obtaining at once both, synchronous 3.546894MHz and 4.4433618MHz signals.

https://www.mouser.pl/ProductDetail/IQD/LFXTAL011002Bulk?qs=e4%2FAndAAwgI7Ljo2GaL2HQ%3D%3D

Edited February 3 by Simius

[macsonny]

4 hours ago, Simius said:

You can always use cheap and easily available 17,73447MHz crystal and divide it by 5 and by 4 obtaining at once both, synchronous 3.546894MHz and 4.4433618MHz signals.

https://www.mouser.pl/ProductDetail/IQD/LFXTAL011002Bulk?qs=e4%2FAndAAwgI7Ljo2GaL2HQ%3D%3D

I worked out I need the PAL = 14.187576MHz. Any easy way of achieving that?

[Overange]

7 hours ago, macsonny said:

I worked out I need the PAL = 14.187576MHz. Any easy way of achieving that?

@Simius's focus lies on enabling the 3.546894MHz signal to serve as the timing signal for the system. 

All Pal-based Atari's utilised 3.546894MHz, including the later XE line. Nevertheless, the Freddie achieved this by dividing its crystal clock signal of 14.187576MHz by 4.

 

Simius suggests the possibility of utilising a single crystal to generate both required clock signals in a PAL-based system: the main clock signal and the colour burst signal.

 

Considering the XE's 14.187576MHz crystal and the potential future challenge of sourcing it, an alternative is to acquire an Amiga Crystal rated at 28.37515MHz and employ a circuit to divide it by 8.

 

It's intriguing to note that Jay Miner, the designer of the Atari and Amiga Chipset, to show that both chipsets have links to the magic number of 3.546894

One might wonder in an alternate reality if Jay Miner never departed Atari and instead created the 16-Bit 'Atari A500'?!?!?!?

[Simius]

10 hours ago, macsonny said:

I worked out I need the PAL = 14.187576MHz. Any easy way of achieving that?

I have several of them so I can send you one. PM sent.

[TZJB]

1 hour ago, Simius said:

I have several of them so I can send you one. PM sent.

 

I expect that you do due to a minimum order quantity of 1000? 😀

[Simius]

I didn't order production. I bought in the store when were available.

[TZJB]

4 hours ago, Simius said:

I didn't order production. I bought in the store when were available.

 

I guessed as much, just joking with you. That doesn't seem so bad. Purchasing 1000 crystals is a lot of investment.

[_The Doctor__]

9 hours ago, Overange said:

@Simius's focus lies on enabling the 3.546894MHz signal to serve as the timing signal for the system. 

All Pal-based Atari's utilised 3.546894MHz, including the later XE line. Nevertheless, the Freddie achieved this by dividing its crystal clock signal of 14.187576MHz by 4.

 

Simius suggests the possibility of utilising a single crystal to generate both required clock signals in a PAL-based system: the main clock signal and the colour burst signal.

 

Considering the XE's 14.187576MHz crystal and the potential future challenge of sourcing it, an alternative is to acquire an Amiga Crystal rated at 28.37515MHz and employ a circuit to divide it by 8.

 

It's intriguing to note that Jay Miner, the designer of the Atari and Amiga Chipset, to show that both chipsets have links to the magic number of 3.546894

One might wonder in an alternate reality if Jay Miner never departed Atari and instead created the 16-Bit 'Atari A500'?!?!?!?

There would be an Atari in just about every hand about now. That's what would have happened. They could have done the Amiga without keyboard legally but didn't, Trammy said no. It would have eventually unified a sector of market under that architecture. That is sort of what the clones did for other architectures. The prospect of both companies making the machine would have lead to one or the other eventually folding in to the other, or the proliferation to other manufacturers as well. Atari would still be here and it's bench innovative products would have likely been revisited and came to fruition. Puffer is Pelaton, Phones et el. etc etc. You know the drill.

 

[Overange]

Indeed, it's fascinating to consider the alternate paths that technology could have taken.

 

While Atari's potential dominance is an intriguing possibility, it's worth noting that in reality, Apple emerged as a significant player in the consumer electronics market, particularly with its innovative approach to personal computing and later with products like the iPhone.

Apple's success demonstrates the importance of vision, timing, and execution in shaping the trajectory of an industry. The interplay between different companies and their decisions ultimately shapes the course of technological evolution, highlighting the complexity of innovation and competition in the tech sector.

[Chri O.]

7 minutes ago, Overange said:

While Atari's potential dominance is an intriguing possibility, it's worth noting that in reality, Apple emerged as a significant player in the consumer electronics market, particularly with its innovative approach to personal computing and later with products like the iPhone.

Steve Jobs was a brilliant thinker, an inspiration, and a leader, but back in 1974 he was an 18-year-old dropout soldering resistors onto Atari's arcade games. 😁

[Overange]

3 minutes ago, Chri O. said:

Steve Jobs was a brilliant thinker, an inspiration, and a leader, but back in 1974 he was an 18-year-old dropout soldering resistors onto Atari's arcade games. 😁

It's kind of funny how things circle back to Atari 🤔

[Chri O.]

Question how accurate this PAL 3.546894MHz clock must be ?

[Overange]

2 minutes ago, Chri O. said:

Question how accurate this PAL 3.546894MHz clock must be ?

 

The accuracy of the clock is crucial for maintaining proper synchronization and timing within the system.

Even small deviations from this frequency can lead to issues such as graphical glitches, audio distortions, or even game malfunctioning.

Generally, for reliable operation, the clock accuracy should be within a tight tolerance range, ideally around ±0.1% or better. This ensures that the system operates as intended and delivers a smooth gaming experience without any noticeable disruptions.