32k Flash Cart Rev. 2

[Cupcakus]

Thank you all for your input in the previous thread. I have addressed some of your concerns with this new prototype. I have designed and checked it, however I have not constructed it to see if it works. It should work fine however. "Should" being the key word :-)

 

#1. There was a concern about the Atari having to be near the computer for this to work. A power supply option was presented. I was concerned about the price increase in this product if everyone needed a power supply too. So I tried the battery route... it wasn't practicle as I expected. I ended up with this solution.

 

The flash cart will use the standard Atari 9v 500mA adapter! :-) We all have at least one of these, if not more. They are inexpensive if you want a spare, and you are not required to buy one with the cart, as you should already have one.

 

#2. EPP support, this wasn't stated in the previous post but the current proto uses SPP parallel, which is slow and is now producing unreliable results.

 

#3. I was also able to dramatically simplify the design of the circuit by not including any I/O chips for the Parallel port, and using just simple counting IC's for programming the chip.

 

Below is an image of the digram I made up for the new revision.

 

The bottom 24 holes will be connected to where the 2732 pins would be on an ordinary homebrew 2600 PCB. While this is a finished EEPROM programmer, it is not a flash cart yet. I still have to complete the bankswitching stuff, and connect the 2600 PCB to the whole thing.

 

The Atari shouldn't even know it's not talking to a 2732, if all goes well. By seperating this circuit from the cart PCB, I've bought myself a little more room inside the cart.

[church]

Looks good. I'm really looking forward to seeing this project completed.

[Cupcakus]

I guess what I'm fishing for is an "I Like", or "I dislike" the idea of using the Atari console power adapter. :-)

[CPUWIZ]

Like! Just like other accessories (ROM scanner etc.), it makes perfect sense to use the Atari PSU and I have like 9 of them.

[church]

Oh, I definatly like, I was going to suggest it till I read in your other thread that you were looking to use a 12v adapter rather than a 9v.

[jeff_a]

Very good idea, I definately like it. I'm really excited about this! You rule.

[JCroniger]

I like it. Everyone should have one of these!!

[LinkoVitch]

I didn't say earlier, but are you using Flash Chips or an EEPROM?

 

I remeber you saying about needing the 12v for the EPROM programming, but how about using a Flash based unit which accepts 5v programming? (or is this what you have gone for?)

 

I am building (trying to build) something similar for Jag, and am scratching my head about sending a 20us pule to the write pin of the chip, the specs say must be at least 20 us but no more than 30us.. how have you done this?

 

And does your software run under Windows? or is it DOS?

 

Good to see you are making progress, and the Atari Power I like also.. perhaps a passthru cable into the mix would be handy?

[Cupcakus]

I didn't say earlier, but are you using Flash Chips or an EEPROM?

 

I remeber you saying about needing the 12v for the EPROM programming, but how about using a Flash based unit which accepts 5v programming? (or is this what you have gone for?)

 

I am building (trying to build) something similar for Jag, and am scratching my head about sending a 20us pule to the write pin of the chip, the specs say must be at least 20 us but no more than 30us.. how have you done this?

 

And does your software run under Windows? or is it DOS?

 

Good to see you are making progress, and the Atari Power I like also.. perhaps a passthru cable into the mix would be handy?

 

I am using a 28C256 EEPROM. The Atmel version of that EEPROM has 32 bits of additional space for placing EEPROM identification information into it. I thought I might be able to use that for something, but it turns out I won't be able to. 12v was required to perform I/O on that section of the EEPROM.

 

Only 5v 50mA is required to write the chip normally.

 

Unless I'm reading the Datasheet incorrectly, the Write pin on this EEPROM has to be dropped LOW and held for at least 100ns and then brought back up. The address is latched as it goes down, and the data is latched as it comes up. The Write pin (Pin 1) on an EPP Parallel port handles this operation perfectly. 120us times are required if I was doing a page write to the chip. However, I am only writing the chip one byte at a time so this is not necessary.

 

It runs under Windows, a special device driver is required for NT based windows OS's (NT,2000,XP)

[Cupcakus]

I just got a horrible feeling I'm doing addressing wrong....

 

Does the Atari address the PROM in reverse?

 

Example

 

The processor fetches the instruction at $001, is A0=1 or is A11=1

 

Does addressing start from the highest address pin back to zero, or from the first address pin to the highest? I thought I knew, but I'm having second thoughts.

[CPUWIZ]

A0=1 == $0001

[cschell]

Cupcakus,

 

Just a quick question, and I'll preface it with the terrible disclaimer of "No offense intended."

 

Are you aware of the use / need for bypass capacitors?

 

I ask because they don't appear in your circuit, and when the Mangia cart was being discussed you were surprised to see one there, stating it was not needed.

 

Chad

[Cupcakus]

Cupcakus,

 

Just a quick question, and I'll preface it with the terrible disclaimer of "No offense intended."

 

Are you aware of the use / need for bypass capacitors?

 

I ask because they don't appear in your circuit, and when the Mangia cart was being discussed you were surprised to see one there, stating it was not needed.

 

Chad

 

I am in no way a master at this craft. I have no professional training in these matters, so if things are not apparent, I will miss them. :-) Why would I need a bypass cap? The other proto is working without it, but maybe that is why my results are not always reliable.

 

Where does it go? And what is it's function?

[cschell]

Simply stated: Bypass capacitors are capacitors between power supplies and ground. They provide a low resistance AC path to ground. In ideal circuit analysis it is assumed that DC supplies provide a perfect AC path to ground, but real supplies do no possess this property. Thus you need the bypass capacitors.

 

In digital circuits they are mostly there to remove noise from your power lines. The noise comes from various sources including drivers switching on and off, power line noise, inductive and capactive coupling from other traces, etc.

 

Lack of bypass capacitors can cause seemingly correct circuits to fail in odd ways. It could very well be a source of some of your intermittent problems.

 

If you look at the datasheet for your voltage regulator (say a 7805), you'll see that there is a capacitor between the input and ground, and one between the output and ground. These are bypass capacitors. Often the exact choice of value or type is not critical, although the faster the circuit, or the more noise resistance required the more becomes involved in proper selection. Values of 10-100 nF are fairly common. Datasheets often provide suggested values in the typical circuit diagrams as well.

 

It is also not uncommon to place a bypass capacitor next to every IC, as you want the path to ground to be as close to the supply as possible. Again the data sheets will tell you if this is critical for particular components. Most memory circuits do require them fairly close to the chip.

 

And yes, this is why there is a capacitor on the Mangia cart.

 

Chad

 

Edit: Fixed cap value.

[Cupcakus]

Simply stated:  Bypass capacitors are capacitors between power supplies and ground.  They provide a low resistance AC path to ground.  In ideal circuit analysis it is assumed that DC supplies provide a perfect AC path to ground, but real supplies do no possess this property.  Thus you need the bypass capacitors.

 

In digital circuits they are mostly there to remove noise from your power lines.  The noise comes from various sources including drivers switching on and off, power line noise, inductive and capactive coupling from other traces, etc.

 

Lack of bypass capacitors can cause seemingly correct circuits to fail in odd ways.  It could very well be a source of some of your intermittent problems.

 

If you look at the datasheet for your voltage regulator (say a 7805), you'll see that there is a capacitor between the input and ground, and one between the output and ground.  These are bypass capacitors.  Often the exact choice of value or type is not critical, although the faster the circuit, or the more noise resistance required the more becomes involved in proper selection.  Values of 10-100 uF are fairly common.  Datasheets often provide suggested values in the typical circuit diagrams as well.

 

It is also not uncommon to place a bypass capacitor next to every IC, as you want the path to ground to be as close to the supply as possible.  Again the data sheets will tell you if this is critical for particular components.  Most memory circuits do require them fairly close to the chip.

 

And yes, this is why there is a capacitor on the Mangia cart.

 

Chad

 

Ahhh :-) Now I remember... Back in the day when I learned this, the book told me it was required to reduce RF interference, which is required by law in Part 15 of the FCC yadda yadda. And scince I never gave a hoot about the FCC's concerns with my home projects I just avoided them.

 

I have studied the Datasheets quite a bit for all the IC's I use, and I don't remember seeing any cap mentioned, but nonetheless, I will put a bypass cap on all IC's in this circuit and on the main power line. I remember reading that a .1uF cap was all that was necessary, however you suggest 10uF? Quite a difference... You're the expert, what do you think?

[cschell]

Oops,

 

You're right .1 uF is good.

 

Chad

[Cupcakus]

This is exactly why I posted the "schematic" here, if you or anyone else notice something I missed, speak up so I can fix it :-)

 

I will not be offended! :-)

[borq]

Hello,

 

how is the project going, i'm very interested in your flash cart.

 

just a message to get the post in the picture again.[/b]

[Cupcakus]

Everything is moving very smoothly! I am in phase 1 of protoyping right now....

 

And Thanks to generous members of the community such as Chad Schell for sharing a great deal of his techniques and knowlege used for his Cuttle Cart, to help me with my own design.

 

and Eckhard Stolberg for his seemingly endless knowlege of 2600 bankswitching techniques :-)

 

and Albert Yarusso for getting me to take a little more time with the project and make it much better.

 

The new prototype supports ALL bankswitching schemes, except for the 2k of RAM access on E7 carts. Which according to sizes.txt is only Burgertime. I apologise to all Burgertime lovers out there :-)

 

I've managed to get everything else to work... on paper anyway... including superchip RAM, and CV RAM+....

 

The design no longer needs external power... so it should just be plug and go.

 

The design no longer needs any dipswitches, all settings are stored electronically.

 

And as an added bonus I have started work on a "Multi" cart function of the cart that will allow you to place up to 7 2k/4k (Non Superchip) games on the cartridge, and play them from a menu driven interface!!!

 

 

Before everyone starts going nuts! :-) This is just on paper right now... I am soldering the first prototype as we type. This prototype must be tested to death, then I have to finish the PCB design and order PCB prototypes, and those have to be tested to death. Then I have to order and assemble all of the production flash carts. So we're still looking at some time before this is available... but hopefully not to much longer.

 

Also keep in mind that all of the above stuff works in "Theory" and in PC simulations... but I won't know until I try it on the real thing. So some of the features I have stated above might not make the final cut... but I am willing to put forth the effort to get it all to work.

[jeff_a]

you are my hero.

[MegaManFan]

And as an added bonus I have started work on a "Multi" cart function of the cart that will allow you to place up to 7 2k/4k (Non Superchip) games on the cartridge, and play them from a menu driven interface!!!

 

:o

 

Cupcakus, you're truly beautiful. I know you're a long way from the finished product yet, but I can't WAIT to get one and try it out!!!

[+Atarius Maximus]

The flash cart will use the standard Atari 9v 500mA adapter!

 

That's a great idea!

 

This is WAY cool, I would definitely be interested in one of these!

 

AM

[Rob Mitchell]

"I am not worthy! I am not worthy! I am not worthy!"

-- Wayne and Garth ..

 

Rock On! Game On!

 

Amazing to think that your proto may be the best product in a long time .. and satisfy the people who missed out on the Cuttle Cart project!

 

I want two!

 

Rob Mitchell, Atlanta, GA

[LinkoVitch]

Before everyone starts going nuts! :-) This is just on paper right now... I am soldering the first prototype as we type.  

 

Now THATS an impressive feat! soldering AND typing, or are you just melting the keys with the end of the soldering iron?

 

Great news, hope your prototypes don't have any escaping blue smoke, and run beautifully.

[Cupcakus]

Why else would we have toes?