[Morpheus_79]

I forgot…

According to @marqs :

To program flash, a combined FPGA image must be first generated and converted into JTAG indirect Configuration file (.jic). Open conversion tool (“File->Convert Programming Files”) in Quartus, click “Open Conversion Setup Data”, select “ossc_pro.cof” and press Generate. Then open Programmer and ensure that “Initiate configuration after programming” and “Unprotect EPCS/EPCQ devices selected for Erase/Program operation” are checked in Tools->Options. Then clear file list, add generated file (output_files/ossc_pro.jic) and press Start after which flash is programmed. Installed/updated firmware is activated when programming finishes (or after power-cycling the board in case of a fresh flash chip).

… both of these checkboxes should be ticked:

[Morpheus_79]

@BoldFusion
According to these FAQs:

https://www.macnica.co.jp/en/business/semiconductor/support/faqs/intel/124561/

… go in Quartus Prime to “Tools” -> “Options” and select this option under “Programmer”:

[Morpheus_79]

@BoldFusion As I’ve written here:

OSSC Pro: Failed Firmware Update

… there are some problems with the very cheap ones. But for just a one-off usage it’s understandable, that no one wants to spend around $300 for an original Altera/Intel Blaster or even $70 for a Terasic one.

I would recommend buying an ultra-cheap Blaster for 5-9 bucks… like this one:

https://www.amazon.com/dp/B07F5H5LPZ

… and using it under Linux (since with its original firmware it has its problems under Windows).

OR using this custom firmware with it:

https://github.com/dougg3/CH55x-USB-Blaster

… like it’s described here:

Fixing a knockoff Altera USB Blaster that never worked

… and using it under Windows (or Linux)!

Alternatively you can buy the little more expensive WaveShare Blaster:

https://www.amazon.com/dp/B08CV3L6LF/

… and using it only under Windows (since it has its problems under Linux).

[Morpheus_79]

I’ve ordered a USB blaster and I’ll try reflashing the firmware via the JTAG port. If that fails I’ll send off an e-mail. Thanks!

Please be aware, that (aside from the very expensive original Intel/Altera USB blaster) a lot of cheap USB blaster clones are out there. Some work better than others – depending on the combination of driver version, Quartus Prime version and OS. I’ve found those two blog posts to be an intereresting read about that:

Fixing a knockoff Altera USB Blaster that never worked

Fixing more cheap Altera USB Blaster clones: CPLD adventures

[Morpheus_79]

5.3V is not perfect… but should be okay. Afaik no component of the OSSC is directly connected to the 5V line (maybe except the 5V output of the HDMI-Port). But the ICs receive their power from the voltage regulators U5 (AP7361C) and U6 (TLV70033). U5 is able to handle 6V max and U6 can handle 5.5V max. So 5.3V is within their specs.

[Morpheus_79]

Shall I check if the oscillator outputs the required frequency?

Yes.

The ‘display flashes + solid green led’ behaviour is somewhat similar to an OSSC without any firmware flashed onto it. I’ve seen similar behaviour on an OSSC with esd damaged flash memory… but since the op had nearly the same problem as you and fixed it by replacing the oscillator, it should be a better starting point. Especially since your OSSC sometimes works -> could be the oscillator delivering no stable clock.

If the oscillator is fine: check the voltages on every voltage regulator too…

[Morpheus_79]

Sadly those inconsistent faults are the hardest to find. Have you checked the oscillator?

[Morpheus_79]

Maybe it’s just a cold solder joint… maybe it’s a quality issue with the component.

But the assumption in this case is NOT: the oscillator is broken. It’s more or less: the FPGA doesn’t do anything – why is that? Could be a broken FPGA, or a broken oscillator,  or a broken flash, or a broken something else. Sometimes things simply break without an obvious reason.

 

[Morpheus_79]

You can check if the FPGA is working via JTAG connector and usb blaster. Since the problem sounds very similar to the one from the op you could check the oscillator via oscilloscope.

[Morpheus_79]

I don’t think there are plans. And even if there were: there are at least two BGAs to solder… with the Cyclone V being a BGA256, which is a pain in the ass to solder without (semi)professional equipment.

[Morpheus_79]

I’ve seen something similar on one OSSC i’ve built in the past ->  some green noise in the output:

… and some purple and green noise and lines in the test screen:

In that case there were some contact problems inside the HDMI port. So maybe there are some dirty pins inside the port… or there are some cold/broken solder joints between the HDMI port and the PCB.

[Morpheus_79]

it looks that the wire that connects to the chip leg actually is the leg that connects to ground?

No. Originally the leg isn’t connected to anything:

[Morpheus_79]

As far as i know:
The Supergun outputs RGB video… so there are separate lines for the red, green and blue color channel. It may be possible, that the Supergun is defective and sends an out of spec or garbage signal with to much voltage only over the red line, while the green and blue signals are fine. So in the best case the signal on the red color channel line is ignored be the OSSC or Framemeister (= blueish/greenish picture)… in the worst case some components are damaged on the OSSC or Framemeister (= blueish/greenish/no picture).

A damaged component on the OSSC would explain the blueish/greenish picture with your Xbox after using the defective Supergun. But as ‘marqs’ said: it does not explain, why the picture with your Hyper NeoGeo 64 is perfectly fine…

[Morpheus_79]

You don’t have to buy from Mouser. Shipping costs are expensive there for small purchases (especially from outside the US). For such purchases with just a few items i would recommend LCSC:

http://www.lcsc.com/product-detail/Video-Amplifiers_Texas-Instruments-THS7353PWR_C2652733.html

http://www.lcsc.com/product-detail/Multilayer-Ceramic-Capacitors-MLCC-SMD-SMT_Kyocera-AVX-06035C104J4Z2A_C597193.html

They ship directly from China… but the shipment is fast nonetheless (i’m from Germany and their shipments always arrive in around or less than a week)… and (unlike some other Chinese online shops) they only sell genuine components.

 

The SSF2X image at the top was output by Framemeister rather than OSSC, and I don’t know why it looks green here either.
SSF2X was not output with OSSC, and when I tried it with another arcade PCB (Fighters History), it was output in fragments (it was also greenish. SCART)
I connect it to the PCB with the same JAMMA, but only Hyper Neogeo64 outputs in perfect color.
What could be the cause…

There may be a problem with the output of the red channel, making the Framemeister ignoring it and causing the damage on your OSSC. But i can only assume. Nonetheless: if you are still planning on buying a new OSSC -> be careful with this exact setup since there could be something wrong…

[Morpheus_79]

Does that mean the OSSC is malfunctioning?

Yes.

What is “THS7353 video amplifier”??

I don’t know how similar the the Kaico version to the original PCB design compares… but on the original the THS7353 (U17) can be found here:

It’s a pretty cheap component:

https://mou.sr/3M74dPQ

… and (with a bit of experience) not that hard to replace using hot air.

There’s a lower probability, that C76 could be broken instead of the THS7353… which can be found here:

Since it’s even cheaper:

https://mou.sr/46NY3O9

… and easier to replace, it should be checked first.

[Author]

Posts