HappyCactus

Hi Alex Well... now that you mention it, I never considered this possibility. I don't remember where I bought it or if they are my original cartridge. So it might be possible they are indeed NTSC games. I need to further investigate. Thank you! That would explain a lot of things. Federico

Hi, I'm fixing a 2600AP, probably from Hong Kong, revision is unknown ("Rev" is visible but no number), dated Nov 30, 1982. I installed the single transistor "composite mod" circuit (with BC547 instead of 2n2222/2n3904), by removing Q201, L201, R222, R209, C209 and RF modulator as described. I changed the values of the bias resistors to 4k7 / 6k8 because the voltage levels were very low (around 400mV). Now the levels are pretty ok (I'm in doubt if adding a 100uF series capacitor to the output of the buffer stage). Tested with the diag cartridge v.2.6P. Seems acceptable. But with Pitfall! the trees are blue instead of green. All games are same color, green seems pretty missing. Tweaking with the color adjustment potentiometer doesn't seem to fix. Any idea? Note we're talking about PAL version, Hong Kong -- Power capacitor is rotated (parallel to short side of the board) and the video buffers (A203) are missing. Thanks in advance. F.

Ok, I guess I had to check more carefully the documentation. Here http://kevtris.org/files/sizes.txt I found this: That answers to my quetsion. Thanks anyway.

Hello all, I'm a newbie in Atari 2600 programming though I have 30y of experience on programming at various levels, and experience with different ASMs and hardware. Currently I decided to get back to electronics design, and wanted to revive my old 2600. My plan is to have a very simple multicart with a GAL and a flash eprom (I know there are ready made solution like Uno cart and harmony, but I'm doing this for fun). I spotted a possible issue in the bootloader for the carts. My current hardware can theoretically handle 128 (7 bits) banks of 4kbytes, so I can load up to 127 4k minicarts (or 63x8k or even 16x32k, doesn't matter). The problem is that an assumption of the most basic bank switching scheme (F4/F8 styles) is that there's sort of continuity between the code in the old and new bank. We don't want that at PC+1 after the bank switching we execute garbage code or data. The bootloader code should, after selecting the cartridge to load, perform the bank switching and jump to the reset vector, either by jmp or rts. And loading any "non-bank-aware" code can't provide this assumption. There can be no rts or jmp in the context of the new bank. So my idea was to load the bank switching code in RAM, starting at some address at the bottom of the ram (opposite the stack), then jump in ram and perform the reset from that, avoiding the code pollution due to the bank switching. Would that work? Is there any previous experience with this? Or am I missing anything obvious ? Thank you in advance! kind regards F.

Oh, I didn't consider this. I was testing it in circuit, but I detached the RF Modulatore because I needed to connect the amplifier for the composite signal... and indeed I removed the pullup that is **inside** the Rf Modulator! I reconnected the modulatore and indeed the signal is perfect! Thank you, you saved me a lot of headaches! All works fine now.

Thanks, I like it. I'll probably do the investment. Thank you, very useful. F.

I found someone from China that ships for 3 dollars, I ordered 3 pieces just in case, if none worked, I'd be glad to buy from you (I already spent a ridiculous amount of money for the shipment of the transformer, but guys, I waited 40 years to play with it...) Thanks! F.

Excellent, any link? I've seen an old topic here but I wasn't able to find any further details, thank you in advance! F.

Ok so thanks to TrackHappy I was able to power my Intellivision NTSC version on. I've not yet applied the Composite Video modification because I first wanted to check that everything was ok, so I removed the RF modulator and checked the Video out signal. It's completely absent (0V) I checked the output of the U10 chip (AY-3-8915) and I found that though there are signals on the V1-V5 inputs, no output is present (R1-R4) they are all 0V. So I guess the chip is dead. What are my options? I've found a chinese seller that has some sample for a reasonable price, but of course I'd like to know if there are any practical alternatives, especially because the video signal (NTSC) can feed some modern display but not all. Thanks in advance. F.

Thanks Trackhappy, you had been very kind and it was a very pleasant transaction! I'll send you any update if I can resurrect it! thank you again!

Thank you Glenn, after all the offer is not unfeasible. I sent you a PM. Cheers Federico

I had a spare KB2040 Adafruit gave me for free with my last order. Why don't use it?

Thanks, it helps a lot. I have not yet decided how to proceed. I have a few other side project, including the USB disc controller adapter, that draw my attention. The two transformer solution would be the easiest but surely not very satisfying from the engineering point of view. But rebuilding the power supply board is more expensive, more time consuming and definitely not the most attractive, especially because I have no idea if my console is in working condition or not. We'll see next week what is my positioning. Maybe a simple voltage partitioning would be enough for the unregulated power rails? is it known how much current they sink? Thanks!

Hi Everybody, I had a spare KB2040 and a spare original Intellivision Disc Controller so... Here is a project to use the controller with any emulator running on platform supporting USB. The hardware is simply trivial: no soldering, no external parts other than the KB2040 that can by bought from adafruit for 8.95USD - well, you need tto solder at least a line of male strips, but that's not rocket science. I published the Arduino firmware on github. I provide instructions for building, flashing and configuring jzintv through a Keyboad Hack File. Hope that helps 🙂

Indeed, I assumed the schema above (taken from the service manual) correspond to the real assembly. My transformed is not working, I tested primary and secondaries with my tester in Ohmmeter mode, secondaries present some low resistance while primary is open circuit. I suspect the coil is burnt, though no damage is visible from outside (but it's sealed). If a fuse is present, it's probably gone. And from safety stand point, that would make sense - any appliance must have a safety fuse, even in the 70ies , as far as I know. Well, no. It's an ordinal NTSC sample, I just added a traveller transformer 230VAC to 115VAC (75VA). That's what I did, but I guess I explained very bad, sorry my fault. At the moment I found a PCB transformer, 20VA, 9V-0-9V. My idea is to put two of them, bring 5 wires with something like molex power connectors and bring them to the external. The idea of designing a new power supply is still there, but I don't even know if the logic board is ok so it would be a waste of time and resources if it was broken. General schema: single 18V PCB transformer or maybe even better single 18VDC (or 15V) plug, Traco power or other similar DC/DC converter with dual output 5V/12V, Charge pump (TC962) with Zener (Input to TC962 at 5V from the output, will result in -5V, then original 2.1V zener to reduce it to -2.1V). Only 16V unregulated and the P5 (5.6V unregulated?) are still needed, and I don't know how precise and what current rating they need.

Hi, thanks for your comments. Yes, I didn't notice that needing to generate -2.1V through a zener, I'd need a negative voltage from the "lower" legs of the diodes bridge. I found in the technical manual the characteristics of the transformer. Unfortunately it's very difficult to find a proper transformer with two secondary, One 18V the other 9V-0-9V. I guess my best bet is to have two separated transformers, but I fear it would too big to fit in the case. It would be external and I'll need a proper connector. The original transformer has apparently a fuse, unfortunately it's not very easy to open. Other options are to redo the power supply unit, this may not a big issue unless, again, for the -2.1V.

Hello, I just received my first Intellivision 2609 from US (I'm in EU) and I found out the transformer is probably dead. Just went through a 220V to 110V transformer, attached the plug to it but no voltages at the output of the power supplies. It seems the transformer is gone - no voltage to any output (tested while disconnected from the Power Supply board). Unfortunately it seems difficult to find any compatible transformer with reasonable prices. If I am not wrong, the original transformer is a 110V to 18V/9V, independent outputs. On the other hands, it seems dual transformers with 18V are very easy to find for less than 20Euros. So I thought that with minimal modifications to the power supply PCB it should be feasible: - use a 220V to 18V dual single output transformer (bonus: get rid of the external 220V to 110V). - Change all 16V rated capacitor with 25V rated. - Edit: remove the whole rectifier bridge actually feeding the 7812. Connect the 7812 to the same input as the 7805. - Change 7805 (and 7812= with Tracopower or any switching alternative - Change R1 220Ω 1/4W with 440Ω 1/4W - Remove R2 and connect 5V output (regulated) to the P5 pin (old 5.6V unregulated). I have some doubt about the potential noise injected by the switching power supply, though previous experiences from a similar modification to a Zx Spectrum wasn't so bad. (I'll probably keep the NTSC version, as it seems my TV is able to decode composite NTSC and PAL signals) Any thought, caveat, previous experiences? Thanks in advance. EDIT: Also a single winding transformer may fit F.

I forgot to mention that the hardware and firmware will probably be under FOSS and CC licenses.

Hi Astal4, Funny I just started similar project with the same objectives and same devices, only I'm using VHDL instead of Verilog. I just started with the Intellivision version, but switched to atari 2600 because it's easier and the bank switching pretty straightforward. My plan was to have an option to connect a microcontroller to program the whole cart through USB or Bluetooth, but still thinking. If you want to exchange some idea, I'm available for discussions 🙂 Thanks Federico